PerformanceHealth Improvement

Less is Not the Same as More

2010-11-11T12:19:00.001-08:00

The title of this posting is clearly a self-evident statement, but despite this, there is a clear tendency for some, perhaps many, to forget its truth and significance. A few years ago, I noted a research organization had adopted an approach in writing its headlines on particular research reports with claims that the reports described how to achieve desirable outcomes such as: "Reduce sales costs by six times"; "Cut overhead by 4x", etc.

While such outcomes are desirable, they are also mathematically impossible! While it is a relatively simple thing, mathematically speaking, to increasecosts by as much as two, four, six or any number of times, it is wholly different with respect to reducing costs or any other typical performance measure. Once you reduce costs as much as onetime, you have eliminated them entirely, and the only way to get past zero in such a case would be to translate what were once costs into revenue. One times anything always yields the same number you started with, so nothing real can be reduced to below zero.

More recently, in fact, just last night, I listened while a highly respected newscaster on ABC News noted that the dosage of one airport full-body scan involved as little as 2,000 times less radiation than a chest x-ray, and 200,000% less than a CT scan. ["Pilot Rebellion: Pilots Refusing to Use Full Body Scanners or Submit to Patdown" ABC News with Diane Sawyer, Nov 10, 2010 (abcnews.go.com)] Again, while it is a simple matter for the dosage of one application to involve 200,000 times as i.e. 200,000 times moreas another, it is impossible for it to involves 200,000 times less.

The math is pretty simple. If a given treatment involves 1 unit of radiation or some other toxic or dangerous content, and another involves 200,000 units of the same risk, then that which involves the higher amount does create 200,000% or 200,000 times more risk. But if the desire is to describe how much lower the low-risk number is than the high-risk counterpart, the best that can be said is that the risk of the low-risk option is only 1/200,000 as great, not 200,000% or 200,000 times less.

When you start with the higher figure and wish to describe how the lower relates to it, you can never get below a 100% or one times reduction, because everything below that is in negative space. 200,000 divided by one equals 200,000, but one divided by 200,000 equals .000005, meaning that one is 99.9995% less than 200,000, not 200,000% or 200,000 times less.

To get from 200,000 to one, it is necessary to get rid of 199,999, or 99.9995% of 200,000. To get rid of 200,000% of 200,000 would require eliminating 200,000 times 200,000 or 40,000,000, and where would all that extra "stuff" come from? It may be understandable for news media and marketers to look for the most eye- or ear-catching way of describing differences, but in the case of more vs. less, the rules of mathematics severely limit what is true regarding differences involving "less". 100% less or one times less is usually the absolute limit.

Predictive Analytics in PHI Targeting

2010-10-11T10:27:00.000-07:00

There is a long-standing difference of opinion regarding the selection of “targets”, i.e. the individual members of a population that are the best prospects for successful PHI intervention. Originally, there was little selection by PHI sponsors, only the self-selection of individuals who were interested enough to enroll in particular interventions. Recruitment efforts were usually either focused on the sickest members, for disease management, or on the population as a whole for wellness efforts.

Selecting the sickest members involved a kind of predictive analytics, but only the most basic assumption, namely that members who had the costliest diagnoses and past history were likely to have the best prospects for saving money. The reality, however, has always been different from this “common sense” assumption. For one thing, people who have the highest costs in one year are unlikely to have similarly high costs in the next, merely due to “regression to the mean”, the tendency for outliers to gravitate toward a more normal level of costs after on high-cost year.

For another, those people with the highest costs or costliest conditions may not be the most likely to respond to interventions. Many may be so firmly habituated to unhealthy behaviors, irreverence toward medical or prescription treatments, or the advice of professionals as to be highly resistive to change. The choice that PHI sponsors have is whether to focus more attention on those who are clearly “problem children” when it comes to managing their own chronic conditions – or to select those most likely to manage them well.

Pharmacy benefit managers have adopted predictive analytics in order to identify those most likely to be non-adherent to their medications, and thereby more likely to develop new or recurrent problems. Once identified, they are targeted for phone or mail reminders, or offers of easier methods for refilling prescriptions, e.g. by mail rather than having to find a convenient pharmacy. [“Express Scripts, Other PBMs Go Data-Diving to Predict Health Problems” Wall Street Journal Health Blog Oct 11, 2010 (blogs.wsj/com/health)]

By contrast, a growing number of employers are leaning toward promoting wellness in order to prevent chronic diseases, rather than waiting until they arise and need management. Having found that healthier employees are also more productive better performers, keeping them healthy, plus improving the health of those at risk of chronic disease can make a lot more sense in the long run. [J. Greene “Wellness Programs Healthy for Biz and Workers, UM Researcher Says” Crain’s Detroit Business Oct 10, 2010 (www.crainsdetroit.com)]

Focusing on those who are already healthy, or even those who are at risk but not yet afflicted with chronic diseases, can make sense, but it still omits what may be the most important element of predictive analytics. Both sponsors and clients of PHI should be devoting far more attention to predicting precisely who is most likely to respond favorably to the particular interventions available for application to them. Some vendors, including HealthMedia, Inc., already seek to predict who has the levels of motivation and self-confidence that will help promote their success. But we need far more science that matches interventions to particular attitudes, preferences, etc. that predict exactly who is most likely to respond positively to particular interventions, and thereby improve the ROI of sponsors’ investments.

Predicting the riskiest targets has a long history, but without also predicting the most promising targets for interventions, it can virtually ensure that PHI results will be sub-optimal.

How Many "Diseases" Are There?

2010-07-29T10:22:00.000-07:00

It has long been the practice of clever advertisers to invent and name dreadful conditions, for which they offer a cure, profitably, of course. A similar tendency seems to apply to “medical problems”, where new and often complex syndromes, conditions, difficulties, and diseases seem to emerge regularly and often. It is one thing to generate the fear of “halitosis” and social ostracism that logically follows, however, as compared to inventing diseases and eliminating responsibility for them. At least part of the continuing dramatic inflation in sickness care costs is likely to be traceable to the increasing number of sicknesses that are being “discovered” in populations. And the less that government and commercial insurers agree to pay sickness care providers per service, the more there is an incentive to “discover” more conditions needing service.

If overweight/obesity -- along with sexual perfidy, gambling, and a host of other unhealthy and “anti-social” behaviors – are defined and treated as diseases, rather than correctable behavior problems, the potential for correcting them cost-effectively is diminished significantly. I recall when “behavioral health” was invented as a euphemism substitute for “mental illness”, which became politically incorrect in the socially sensitive 60s. But now, it has taken on the additional meaning reflecting the extent to which our behavior contributes to, threatens or harms our health. And the extent to which we are deemed responsible for such behaviors is a key element in social and economic policy with respect to them.

As a practical matter, if unhealthy weight is deemed a disease, or at least a condition significantly affected by “genetic pre-disposition”, which is easier to prove, then it becomes problematic for employers, insurers, or anyone else to discriminate between people of healthy vs. unhealthy weight. While clear behavior differences, such as smoking vs. non-smoking has survived as a reason for charging people more for their insurance, even for refusing employment, it is clearly at high legal risk to reward or punish people for behaviors that are affected by “diseases” or genetic differences.

While there is all but universal agreement that consumers need to become more responsible for their health behaviors and health/sickness care use, achieving an increase in responsible health behaviors will be much more difficult if there are no methods acceptable or legal for discriminating among healthy vs. unhealthy behaving consumers. Any diminution in or limitations on the consequences of health behaviors will obviously diminish consumer accountability. While employers are reportedly increasingly interested in using disincentives to influence employee health behaviors, the effects thereof would disappear if they became illegal or grounds for civil litigation. [M. Brill “Shape Up or Pay Up” MarketWatch.com July 29, 2010]

It may make sense to at least recognize that people differ in their innate ability to control their behaviors. Employers already are offering greater-value incentives for behavior changes generally recognized as more complicated or otherwise difficult to change, for example. But if health behaviors are widely deemed as caused by diseases, even this mode of discrimination might become prohibited, or at least risky for employers.

Today I read about what may be a particularly egregious example of this potential. Apparently a woman had been charged with killing her newborn babies in Europe, and at her trial, a psychiatrist testified that she suffered from “pregnancy denial”. If people cannot be held accountable for murdering babies, because a disease is suspected, it would seem unlikely that they can be so held for not keeping fit. [“Couple in Custody After 8 Babies Found in Plastic Bags” MSNBC.msn.com July 29, 2010]

Evaluations of Little Value in PHI

2010-07-29T09:45:00.000-07:00

As the popularity of comparative cost-effectiveness research increases in reactive medical care, it is understandable that proactive health management is getting similar attention. Two recent articles point out some of the problems with such research when applied to wellness and chronic disease management, however, though the tendencies involved are horribly common.

Echoing the consistent conclusions of government-sponsored studies of disease management, a macro-analysis of studies conducted by the Center for Studying Health System Change concluded that: “…return on investment for wellness initiatives is uncertain…”. [“Wellness Programs Don’t Necessarily Improve Health or Lower Costs” Wall Street Journal Health Blog July 29, 2010 (blogs.wsj.com/health)] While most such studies are limited in their scope to examining reductions in sickness care costs as the only source of ROI, the idea of attempting to reach general conclusions about proactive health efforts is absurd in the first place.

Sickness care studies require rigid control studies, where the medical intervention is the same for all patients in the intervention group, and a control group ensures that the only difference between the intervention group and the control group is that one got the intervention and the other didn’t. By contrast, when comparing wellness interventions, the intervention is almost never the same for all, or even a majority of the people in the intervention group. Wellness interventions vary all over the map, with respect to what kind of health goals are involved, what methods for coaching are applied, etc. There is no real possibility of reaching a general conclusion about “wellness” from studies involving widely varying interventions, anymore than there would be if a similar attempt were made to evaluate the ROI from “medical care” in general.

Instead of wasting money, time and media attention on general conclusions based on the small proportion of research meeting clinical standards on wellness interventions, it would be far better to use more widely-focused wellness studies to identify precisely which among the host of interventions being used actually do work, which are most cost-effective, and therefore, which should be emulated. A similar approach should probably also be used with “complementary and alternative medicine” (CAM), which includes such an incredible array of non-medical interventions as to make the idea of reaching general conclusions about it ridiculous from the outset.

Another all-too-common silliness applies to media and industry reports about particular types of health plans. In another recent article, the efficacy of Medicare Advantage health plans was reported with respect to reducing patient re-admissions, a goal of both payers and hospitals, as threats of non-payment for presumably avoidable re-admissions make them worth avoiding. [J. Lubell “Advantage Plans Helping to Reduce Readmissions, AHIP Says” ModernHealthcare.com July 28, 2010]

The conclusion that Advantage plans reduce re-admissions was apparently based on cross-sectional comparisons of rates between populations that were enrolled in such plans compared to populations enrolled in traditional fee-for-service plans. While such comparisons are always of some interest, they absolutely do not show that Advantage plans cause readmissions to go down. They merely describe different rates in the different populations, e.g. 16.7% in Advantage vs. 20.5% in Texas. Unless analysis shows that patients in Advantage plans had lower re-admission rates after they enrolled in such a plan, compared to what their rates had been before they enrolled, plus that there were no other possible causes of the lower rates likely to have caused such a difference, the comparison of rates between two different population is no more than an interesting phenomenon.

It is understandable that people and organizations with “an agenda” pre-disposing them to find positive, negative, or equivocal results will do so, whatever the data show. I recall an example when I was a doctoral student where the director of a research project reported a conclusion exactly opposite what the data showed, reflecting her strong, though erroneous convictions, rather than the data. When economic self-interests are involved, similar bias is possible. And the media seem to prefer interesting, even startling results when reporting findings to careful analysis of the data underlying such results.

Evaluation should focus on delivering sound and useful conclusions that will advance the causes addressed, rather than, as Mark Twain noted, yielding “…wholesale returns in conjecture for trifling investments in fact.” Far too many evaluations of little or no value are routinely carried out, and unfortunately published.

Macro and Micro Applications of Engagement in PHI

2010-06-24T10:46:00.000-07:00

The current fixation on employee engagement is likely to produce some significant changes in how workforces are managed. In turn, improvements in overall management should lead to their own positive impact on employee productivity, performance, retention, and value to their employers. Moreover, the concepts and methods used in promoting and sustaining overall employee engagement should be adaptable to achieving similar improvements in engagement in PHI initiatives as well.

One of the latest discussions of employee engagement comes from Paul Herr, author of Primal Management (AMACOM Books 2009). He describes an emotion-focused rather than traditional rational management approach to managing employees, with the potential to double productivity by doing so, given the average of only 30% of employees really engaged with their jobs and employers. He than describes how five basic human “appetites” can be used as the basis for achieving such results.

The five factors he describes are sources of “emotional reward units” (ERUs) that greatly influence, if not fully determine employee motivation levels. These are: 1) self-protection/security; 2) cooperation/work relationships; 3) skills/competency; 4) skill deployment/achievement; and 5) innovation/discovery. Such intrinsic rewards have the advantage of being something management can enable (or prevent/destroy), and therefore use in its efforts to improve engagement and thereby employee value.

These appetites have the added advantage of being alternatives to simplistic, rational motivators such as pay and bonuses, which have the disadvantage of adding one dollar in cost to employers for every dollar of benefit they add to employees. To the extent that employers can add ERUs that cost less than a dollar for every dollar of value they add to employee reinforcement, affecting their presence, productivity, performance, retention, and overall worth to the employer, these five appetites can be more efficient in yielding desired improvements in engagement, effort and results than cash payouts.

This same potential exists with respect to how employers motivate employees toward participation, enthusiastic engagement, completion, and success in PHI initiatives. Each of the five can be translated into ways that PHI initiatives either deliver ERUs as motivators, or enable employees to gain ERUs, thanks to participation and success in such initiatives. In turn, PHI participation can add rewards through its impact on employee appetite satisfaction.

The self-protection/security factor, for example, is directly influenced by employee health and risk status. Poor health not only reduces health-related quality of life (HRQoL); it also affects employees’ ability to secure and maintain gainful employment, to protect their ability to earn a living. It also affects their potential for “being all they can be” in terms of satisfying or improving upon the other four appetites. The fact that health risks and productivity impairment factors handicap employees in their execution of their job functions carries over into their ability to devote the energy and effort needed to achieve PHI goals and objectives.

Cooperative relationships are at least potential elements in PHI interventions. Engaging employees with similar PHI challenges and goals in social networks, such as “buddy systems”, support groups, etc. can be rewarding by themselves, in addition to helping PHI participants remember and carry out necessary behavior changes. In turn, adding to employees’ social networks, by itself, is known to have positive impact on their health and engagement with their jobs and employer.

Skills/competency is normally the domain of the Human Resources employee training and development function, but individuals’ levels of skill and self-confidence/self-efficacy are also key to their likelihood of participating and succeeding in specific PHI initiatives. This dimension is one of the better predictors of both employee participation and success in such initiatives, and can be useful in deciding how much to invest in recruiting individuals to the right initiatives. Having confidence in one’s ability to overcome barriers to behavior change is a key element in readiness to change, and in PHI enrollment.

Skills deployment/achievement is likely to be as important in maintaining both PHI participation and health behavior changes. When self-efficacy is accompanied by explicit and clear progress toward goals, and eventually by achievement of significant improvement in health, it yields benefits in terms of self-esteem, and often the respect of one’s peers. Empowering employees to take charge of their own health, and proving to them that they can succeed at it, can yield its own set of rewards and ERUs, in addition to the added value it has for employers, and whatever rewards it offers in return.

Innovation/discovery can also be an inherent element of PHI participation experiences, when employees learn something new about themselves, or show they can master a new capability. And since the skill/competency of self-management is a key element of PHI initiatives, the “innovations” in their own lives that employees are able to introduce and demonstrate can have significant added value when applied to other health or personal challenges.

Because the ultimate aim of PHI is to improve the overall performance and value contributions of employees, it makes sense to incorporate ways of increasing engagement in both general employee relations efforts and PHI initiatives. And by the same token, it is likely that employees’ success in PHI will supplement other engagement “engineering” efforts in improving their overall value to their employers, and thereby the ROI of PHI, itself.

The Case for Stress in PHI

2010-05-20T12:14:00.000-07:00

The emphasis in managing the health of populations is usually to eliminate health risks or diseases and injuries. But in many situations, the best thing to do about stress may not to be to reduce it, but to actually increase it. While no smoking is easily the best status for employee health, no stress is not. The challenge is to identify and develop an ideal level of stress, rather than to eliminate it entirely.

This is actually the aim of many efforts aimed at eliminating health risks – to achieve among individuals and populations an ideal level of specific health indicators – blood pressure, cholesterol, sugar for example, as well as body weight, heart rate, etc. The only time these measures are zero is usually when someone is dead, after all. The same is true for stress, where a truly “no stress” situation is likely to be achieved only when people are comatose or dead.

The usual state of stress, particularly in high-stress societies such as the US, and in the many high-stress jobs that most people have, as well as in the high-stress situation of the present recession, is unhealthily high. But without some stress, some “pressure” being present, workers are unlikely to work hard, or perhaps at all. Of course, there will probably always be managers whose motto is “I don’t suffer from stress, I give it to others”. But rather than an atmosphere of fear and trembling, it is usually better to have one of “eustress”, or positive-impact stress, rather than distress, where effects are negative.

A recent discussion of the Renaissance makes the case that the kind of stress imposed by societies and patrons in Northern Italy in the 15th century was about perfect. The competition for work, as well as the willingness of “employers” to open such competition to a wide range of different “solution providers”, meant that large numbers of individuals from varying professions were invited to compete. All were usually stressed to design a solution to a problem to which nobody knew the answer, or to compete with their peers through side-by-side comparisons of past work. This often led to masterpieces, though occasionally to mistakes such as the leaning tower of Pisa. [B. Ferrari & J. Goethals “Using Rivalry to Spur Innovation” McKinseyQuarterly.com May 2010]

Similar approaches have been used in Research and Development efforts, where competing teams in “skunk works” units face deadlines to come up with better solutions. The galvanizing of human energy and effort that was achieved by all parties during World War II showed what people are capable of when threats to their survival motivate them. While competition as a zero-sum game has been criticized for its negative effects on individuals, it has also been responsible for exceptional efforts and results. There is no way to realize the best, by definition, unless there is competition.

As long as the rivalry involved is “friendly” rather than cutthroat, and as long as there is the potential to combine ideas and even solutions in the ultimate “winner”, the stress of competition often yields better results, merely because it stimulates effort and attention, compared to a “no worries” alternative. After all, the goal in PHI, as in management in general, ought to be to produce the greatest value outcomes, and healthy stress is likely to be necessary to stimulate the effort to do so. Moreover, changing behavior, per se, is almost sure to create stress, as well as improve health and performance.

The Heisenberg Uncertainty Principle in PHI Evaluation

2010-05-06T09:54:00.000-07:00

When I began my career in proactive health management 20 years ago, while VP for strategy and marketing at a multi-hospital system in Denver, I approached the Coors Brewing Company as an example of employers interested in the idea, at least in worksite wellness. I learned that while Coors had a strong interest and major investment in wellness, it evaluated its investments entirely in terms of health insurance expenditure reductions that resulted.

I felt then, and have become increasingly convinced since, that evaluating investments in workforce health should include attention to a far-ranging set of impacts that they can have on overall business performance. The only real limit to such impacts is that they must logically come from changes in human capital asset value and overall performance, as opposed to new technology, for example. But expanding the scope of evaluations brings with it an unavoidable complication, an uncertainty about which and how much impacts should be attributed to PHI investments, as opposed to any number of other causes.

In effect, there is a “Heisenberg Uncertainty Principle” at work. The original principle applied to quantum mechanics, noting that it is impossible to simultaneously determine both the position and momentum of matter. The more precisely that one can locate the position of something, the less precisely one can be about its velocity, and vice versa. (www.wikipedia.org/wiki/Uncertainty_principle)
A similar principle applies to evaluation: the more broadly we look at dimensions of value that PHI interventions could have affected, the less certain we will be about whether changes detected were caused by such interventions.

Given this reality, there are two main “solutions” to the problem: 1) compete with other causes, particularly when there are other functions, departments, investments, interventions, etc. that are likely to or have claimed credit for changes noted; or 2) cooperate with them to reach a consensus on sharing the credit. There is a third option, of course, namely to take a “scientific” approach to discovering the truth about causation and credit – but that is unlikely to be acceptable in most situations.

While government insurers have taken a scientific approach to evaluating Medicare and Medicaid interventions into beneficiary health, they are the exception. Employers normally avoid such approaches for good reasons. Random or matched pair assignment to “intervention” vs. “control” populations is both too much trouble/cost and sure to be unpopular among employees. In one of the few examples where this was tried, a state education system randomly assigned school districts to intervention vs. control conditions, but as soon as it realized how much money it was saving through the intervention, it assigned all the control populations to the intervention condition. Science could not trump economics.

An alternative to rigorous scientific study might well be one of measuring the full "value chain" of effects that PHI requires to achieve its results. If it can be shown that specific interventions are linked to specific changes in participants' knowledge and attitudes about healthy behaviors, particular changes in related health behaviors, in health status indicators, then in productivity, performance, and value -- this would make a stronger case for a causal relationship than citing value indicators alone. Arguably, this is not "proof", but at least it is "circumstantial evidence".

Competing with other causes is sure to be contentious, since there are likely to be a number of other “silos” that could claim credit for improvements in workers’ productivity, performance, and value to the employer. And the broader the range of value dimensions and specific performance indicators involved, the more causes will have grounds to join the fray.

Consider the following list of value impacts proposed for evaluating the workforce training and development function: increased sales/revenue growth; decreased labor costs; improved customer satisfaction, quality, on-time delivery, productivity, cycle time, employee satisfaction/morale; reduced waste, worker injuries, turnover. [D. Brown “The True Value of Learning” Skillsoft.com 2008] Every one of these dimensions should at least be considered as possible effects of PHI interventions as well.

Human Resources, compensation practices (e.g. pay-for-performance), benefits other than health, and a wide range of morale boosting policies or processes may well deserve and demand credit for improvements in the value that employees deliver. A host of other investments (e.g. capital equipment, information technology, supplies, work environment, etc.) may also insist on their share of credit. Competing with them on a “political” rather than scientific basis could create far more harm than good. PHI may or may not have high enough credibility to get a fair share of the credit, and lose out in competing for future investment. On average, well-designed evaluations of PHI have shown ROI levels in the 2:1 to 5:1 range, which should allow plenty of room for sharing credit.

Cooperation, e.g. by either sharing equally in the credit or negotiating something other than equal shares, may work better, as long as agreement/consensus can be achieved amicably, quickly, and credibly. Of course, if the total value improvement noted does not more than equal the total costs of interventions competing for credit, cooperation may break down. Naturally, CEO and CFO support will be essential for such cooperation to work. Supporting this possibility is the likelihood that there will be plenty of credit to share, particularly in the long haul, as examples of specific interventions have already demonstrated.

Initiating a pay-for-performance system, replacing a uniform hourly wage system for windshield installers at Safelite, for example, resulted in an immediate 44% improvement in productivity, at only a 10% increased cost in wages. Empowering corporate office employees to set their own hours and choose their own place to work improved productivity by 35% and cut annual turnover from 16.7% to zero. Neither example provided any information on any other interventions that might have claimed credit for some of the value added, but had any done so, there would clearly have been plenty of room to share the credit.

There is actually another source of "credit competition" that may have to be considered -- public, community, and governmental interventions that may be affecting the health of entire populations, in addition to those members of the population that are employees or dependents thereof. IBM, for example, has recently initiated a major research effort to identify factors that affect health, as the basis for investments by anyone interested in population health. ["IBM Launches Research Effort to Build 360 Degree View of Factors Affecting Human Health" IBMMay 6, 2010 (www-03.ibm.com)]

Fortunately, while investments by sponsors other than the employer may well deserve a share of the credit, the IBM project should also identify which are the most cost-effective interventions in population health. Armed with such information, PHI should be able to compete effectively with any other sources of impact. Broad-based interventions are already increasing with recent health reform, and these investments may also deserve some credit. They will probably focus mainly on reducing sickcare costs, however, and thereby leave plenty of room for PHI investments more focused on workforce productivity, performance, and value.

The majority of employers still seem to rely on their faith in the clear connection between employee health and productivity, performance, and value when making PHI investment decisions. On the other hand, far better decisions, particularly about which among competing human capital interventions, as well as which health management problems and solutions to invest in, can be made if there is more and better data on the full range of value impacts such interventions might have. It may be true that “When ignorance is bliss, ‘tis folly to be wise”, but data, analysis, and understanding of causes and effects are likely to be a far better basis for making investment decisions. And cooperation seems likely to be both more welcome and advantageous than competing for credit when effects are known.

Shifting Sickcare Providers to Health Improvement

2010-05-04T09:28:00.000-07:00

It seems clear, to me at least, that large numbers of providers, perhaps eventually a majority of them, will be shifting from traditional “sickcare” to managing the health of populations, for their own as well as the world’s good. Some will shift entirely to a proactive health focus, but most will probably retain a mixed health and sickness focus. This shift will be slow in coming, especially where medical care is decentralized as much as in the US, though may occur swiftly where there are national systems that can help manage the process. The reasons for this shift are the same almost everywhere – the cost of preventable illness and injury have become or are becoming unsustainable.

In the US, there may well be the most varied approach to shifting the health care “paradigm”. We have so many different approaches to medical and health care already, and all will be competing to be among the survivors. These include:
• Traditional modest-sized practices
• Large group practices
• “Concierge” practices
• Retail clinics
• Worksite clinics
• Group practice insurers
• Integrated health systems
• Complementary and alternative medicine providers

The broadest shift that will have to be part of structural change will be focusing the attention of physicians on the costs of care, rather than solely on the effectiveness thereof. While physicians have traditionally been insulated from cost issues, even trained to consciously avoid considering costs when choosing treatments, there will be a major shift toward comparative cost effectiveness as a key if not chief consideration. This is especially true for health management, in contrast to sickcare, since there is little justification for a “damn the torpedoes” attitude in selecting solutions.

Already, medical schools are introducing courses or at least discussions of the costs of care into medical education, and residencies are required by accreditation standards to do the same. [S. Okie “Teaching Doctors the Price of Care” KaiserHealthNews.org May 4, 2010] While emphasis in such education and training is on the costs of sickcare at the moment, health management will clearly emphasize selection of the most cost-effective solutions out of necessity, at least in the long run as payers learn how best to select and pay providers.

The growth of concierge practices has already been promoted by the potential for health management to be a separate basis for charging patients annual fees. And fees have tended to focus on affordable to middle class levels, in contrast to the only the wealthy need apply approach of the earliest examples fifteen years ago. {L. Stein “Concierge Doctors Say an Annual Fee Can Mean Better Care” St. Petersburg (FL) Times May 4, 2010 (www.tampabay.com)] Where early examples focused on luxury levels of access/availability and amenities, most of the current practices focus on better health management, and have the data to prove it. (www.mdvip.com)

The idea of patient-paid health management is only half the story, of course. Physicians and hospital-physician integrated systems are developing and testing medical homes. While most seem to be starting out focusing mainly on reducing patient re-admissions and managing chronic illness, they will surely end up as “comprehensive health management homes”, though with another name, since already the “medical home” sounds too much like a nursing home to consumers.

Physicians are sure to recognize, eventually at least, that the best approach to both managing chronic illness patients and healthy or at-risk patients is to lead broad-based team approaches and larger organizations. Physicians, themselves, are not only untrained in, but too expensive for all but a leadership role and motivator in proactive health management. [D. Brown “New Health Care Law Might Make Your Doctor More Informed, Efficient, Responsive” Washington Post.com May 4, 2010] As payers insist on the most cost-effective, proven ROI approaches thereto, practices will have to be in a position to deliver them.

Retail clinics are likely to enjoy at least an early lead in PHI, since there are so many of them already conveniently located, reasonably priced, and staffed by nurse practitioners or physician assistants who can do the job so much more inexpensively than physicians. Already, many of the existing examples are adding disease management and wellness services to routine sickness care, in order to both increase and smooth out demand for their services. Where they are located near small to medium sized employers, they may be enlisted by organizations in workforce health improvement efforts, as well.

Worksite medical clinics are already growing, among large employers at least, as part of their continuing, indeed growing efforts to improve employee health and productivity. As employers stretch their goals and evaluations to include performance and value measures, and gain a more complete picture of their ROI from such investments, these clinics figure to grow even more. Despite the economic recession and reform legislation, most employers plan greater emphasis on workforce health and performance. [“The Road Ahead: Under Construction with Increasing Tolls 2010 HewittAssociates.com May 2010]

Pre-paid group practice insurers, such as Kaiser Permanente and Group Health organizations, have been investing in proactive health for some time, though not as long as the original “Health Maintenance Organization” model anticipated. As long-term bearers of risk for healthcare costs, as well as providers, they recognized the value of prevention, wellness, disease management, etc. earlier than other insurers, and have made effective use of their expertise in attracting payer clients. Even insurers have taken on the challenge, recognizing the interests and intentions of their employer clients.

A significant element of the development of health management will be discovering and developing an appropriate role for “complementary and alternative medicine”. Consumers already have a far stronger attachment to CAM approaches to health as well as sickness than science supports, but providers will certainly include proven approaches where they fit demand for cost-effective results. Already, large numbers of physicians and hospitals have created “integrated medicine” programs that include CAM solutions in which they have confidence, and evidence relative to what truly works most cost-effectively is growing, as it is in traditional medicine.

It is far too early to predict, at this stage, which provider organizations and approaches to health management will survive longest and dominate the future. Payers are generally still unsophisticated with respect to what the full value of outcomes they can achieve, and to evaluating what they are already getting. As they increase their own skills and understanding, they will stimulate far more informed competition which will weed out less cost-effective providers in the market.

With respect to health management, we are much like the early 20th century when automobiles, bicycles, airplanes, railroads, and other means of getting us from place to place all competed for the same market, and hundreds of different options were often available within each category. We will eventually reach a far more concentrated group of providers and set of options, though it will take some time. But having in mind “The Gretsky Principle” of going where the puck is going to be instead of where it is now should help providers and customers alike make the process of creating a truly cost-effective solution easier and earlier.

Rewards + Recognition = Reinforcement in PHI

2010-04-30T10:34:00.000-07:00

While most of the talk on the topic of increasing participation and engagement in population health efforts focuses on incentives as the best solution, it is clear that what is really needed is reinforcement. While “incentives” cover only monetary or other rewards that cost money to deliver, “reinforcement” includes, by definition, anything and everything that succeeds in promoting the adoption and continuation of a given behavior.

There are two practical problems with the use of incentives:
1) they cost money, thereby automatically, and often significantly reducing the benefit/cost ratio likely to emerge from PHI investments; and
2) they have limited, if significant impacts on behavior, especially over time and for continued vs. one-time behavior changes

Offers of rewards, plus their delivery when conditions are met, have proven helpful in particular elements of PHI, such as completing health risk assessment (HRA) surveys and biometric screening tests. They seem to work with behaviors such as quitting smoking, though they have also prompted “cheating”, among participants = falsely claiming to have quit, or not reporting relapses. This means that incentives are wasted. They are also subject to “inflation” effects, requiring continuous increases in the amounts of rewards involved in order to maintain the same results.

There is also the problem that rewards tend to apply only to the cognitive or “thinking” appeal of behavior change, leaving out the potential that affective or “feeling” appeals might have significant, even greater impact. Reinforcement can include, even focus on emotional impacts, in addition to rational appeals. When coupled with their low or no cost, non-monetary reinforcement has a lot to recommend it.

In a recent example of emotional, brain-science-based behavior change, students at Stanford University were subjected to an indirect, “stealth” appeal to justify their adopting healthier eating habits. By stressing the environmental and social benefits to others of eating local produce and fresh foods in general, a significant increase in the eating of locally-grown fruits and vegetables was achieved. By contrast, students only exposed to the usual personal nutritional arguments for eating healthier food actually slipped a bit in their habits during the same period. [S. Ipaktchian, et al. “Stealthy Leads to Healthy in Effort to Improve Diet, Study Shows” Stanford Medicine Apr 26, 2010 (med.stanford.edu/ism)]

Reinforcement can include, but not be limited to monetary rewards, with greater impact than either recognition or incentives alone. The key is to learn what precise levels, types, and frequencies of reinforcement work best for which individuals or segments. Differences across the traditional “generations” of people, such as baby boomers plus generations X and Y, may not be consistent or significant enough to add greater efficiency or effectiveness, however, so recognizing individual differences, along with changes in response over time, will probably work better.

Recognition can include a wide range of communications, symbolic gestures, and sources. Rather than being limited to supervisors and managers, recognition can come from customers, fellow employees, “buddy” or “team” participants in specific behavior change efforts, social support groups, family and friends, for example. Just as “360o” evaluations of performance are growing in use, so the full range of people who are interested in and willing to support behavior changes in people can be invited to become “recognizers”.

Reinforcement can change in form as well as amount over time, and be modified as to schedules for its application as well. “Dynamic Intermittent Reinforcement” has been shown to be effective in promoting adherence to prescription medication regimens, for example. [M. Kalayoglu, et al. “An Intermittent Reinforcement Platform to Increase Adherence to Medications” American Journal of Pharmacy Benefits July 2009 91-94]

Moreover, various artificial intelligence approaches can continuously analyze the “dose/response” history of intermittent reinforcement to determine and continuously update what is the most effective/efficient type and schedule of reinforcers for individuals, segments and populations. By learning through experimentation and experience, dynamic reinforcement can keep the costs thereof within pre-determined limits based on the predicted and determined financial benefit vs. cost of reinforcers in use on an ongoing basis, rather than wait until a given behavior change effort is complete or the budget year has ended.

One of the biggest limitations of even the most comprehensive and complex customization approaches to behavior modification is that most are not truly dynamic. They may reflect vast amounts of information gained from initial HRA surveys or screenings, but base all customized communications on that information alone, rather than continuously revise communications based on individual participant histories of response thereto. Dynamic approaches can not only revise communications when initial efforts are not as effective as hoped, but revise reinforcers throughout a year, rather than wait until its end.

By taking advantage of the full range of reinforcements, and the dynamic potential of computerized digital information systems, along with the full and growing range of wired and wireless communications, PHI sponsors and providers have a far greater chance of discovering and applying optimal behavior change interventions, and thereby achieving optimal results and ROI. The key is to start with a broad vs. narrow view of what is reinforcement and how it can be used, rather than myopic assumptions about incentives and reasons that modify behavior.

Critical Masses in PHI: Three Levels

2010-04-14T13:57:00.000-07:00

While the idea of “critical mass” owes much to the development of the atomic bomb, it is equally useful in considering how to plan and manage the health of populations. Until there are enough resources, participants, and financing available, there is likely to be little hope of success. But in PHI, there are three different levels of “mass” that are likely to be essential, relating to the country, the employer, and the individual involved therein.

National and International Mass

At the national level, there has been a slow but fairly steady growth in the numbers and sizes of key requirements. Employers, insurers, advocacy organizations, governments, and consumers have joined in both demanding and supplying resources and financial support. In many cases, insurers are working with employers and governments, for example, in pursuing their common interests relative to reducing the incidence and prevalence of disease and injury, and of factors that limit worker productivity, performance, and value to employers.

As the globalization of industry continues, there is growth in global solutions to health and performance improvement. Increasingly, PHI suppliers are exploring opportunities in other countries, such as vielife.com, which started in the UK, but is now involved in the US as well as Brazil, in addition to other countries in Europe. The US Center for Preventive Medicine has recently begun offering services in the UK. Multinational companies are multiplying their opportunities by “exporting” successful initiatives into international locations.

While any attempt to aggregate multiple organizations brings with it extra efforts and expense in coordination or integration of efforts, it should also spread the costs of PHI strategies and initiatives over multiple sponsors, and reduce the costs of individual sponsors. In addition, it can take advantage of the special capabilities and resources of different categories of organizations, such as health care providers, employers, insurers, governments, and advocacy organizations.

Employer Mass

Each employer wishing to invest in PHI strategies and initiatives may or may not enjoy sufficient “mass” on its own. Those that do may be able to choose among virtually all the different approaches to improving the health of their workforces, from onsite medical clinics to self-insurance, for example, where smaller employers would not be able to afford such options. But where locations are conveniently close, multiple employers may cooperate in sponsoring a single onsite clinic or intervention, for example, rather than relying on remote alternatives. And nearby employers may find it possible to enlist the services of retail clinics as a close equivalent of an onsite option.

By joining with their own insurers and nearby health care providers (e.g. hospitals that offer PHI services, or concierge physicians with significant proactive health capabilities) employers may also be able to create more focused efforts than are possible with traditional diversity of workforce residence locations. Employers may directly contract with healthcare providers in medical home or similar proactive efforts, rather than directly sponsoring their own medical care.

Cooperating with local agencies, such as YMCA and YWCA organizations, fitness centers, etc. can add significantly to the mass available to employees, while reducing costs to both employers and workers. Local churches, even barber and beauty shops have proven to be useful allies in reaching populations for screening and communications efforts. Local branches of national advocacy organizations focused on particular diseases or health risks may provide support for employer efforts.

Here again, involving such outside organizations will require extra effort, but should also reduce the costs that employers much bear alone. Enlisting employees and dependents, themselves, as part of the solution, e.g. via social media promotion of participation, support groups for participants, and volunteers in screening efforts, can also help to expand the scope while reducing the costs of efforts by employers.

Individual Mass

Closely related to what employers can look for as sources for added mass in their efforts are the resources that individual employees and their families can enlist. The family is the first and most logical place to start, since all members share in the costs of “unhealth”, and the benefits of improved or maintained health. Co-workers, or even just local residents who are pursuing similar health goals can be sources in “buddy system” or other forms of support, in team competitions, for example.
By discovering, then making employees more aware of the community resources available, employers can help individual employees put together their own critical masses of support. As such sources prove themselves, or even when they merely figure to add significantly to success, employers may make it easier for employees to take advantage of such resources, by negotiating discounts, paying some portion of the costs, etc. in addition to promoting awareness thereof.

The nature of PHI is such that a systematic approach, involving as many parts and members of the system that affects individual and population health in efforts to improve and maintain both, is likely to be a better option than any one “silo solution” or series thereof. By identifying and enlisting the resources of others who are interested in or benefit from improved workforce and insured population health, critical mass is far more likely than through the efforts of individual organizations thinking and acting alone.

What Should Be Targeted for Change in PHI?

2010-04-12T10:19:00.000-07:00

The two most significant decisions in planning health improvement efforts relate to selecting whom to target for such efforts, and what change(s) to aim for among such people. The two choices are strongly connected, of course, since the selection of people is usually based on the health and risk conditions they have and the behaviors they have adopted that cause or improve them.

One of the biggest problems inherent in the selection of what is to be changed is the built-in effect of how the impact of particular conditions or behaviors among people is measured. In the vast majority of situations, the impact of a given factor is identified and analyzed based on the measured health care costs among insured members plus the estimated value contribution losses among workers who are affected by each such factor.

If we seek to estimate the “losses” associated with obesity, for example, we identify all insured members or employees who are identified as obese, then count or estimate the total losses for all such people. Combining “excess” health care costs along with excess value losses (compared to non-obese counterparts) may well suggest that obesity “accounts” for hundreds of dollars in such costs, and perhaps thousands in total losses.

Unfortunately, the total burden of obesity thus estimated will normally be many times the true impact of obesity, itself, rather than an accurate measure of its effects. This is because most people, whether or not they are obese, tend to have far more than one single factor that affects the total costs of each individual. In one example of total costs and losses, for example, the grand total of all factor-specific estimates turned out to be over three times the total actual costs and losses for the population.

This over-counting is a natural consequence of the way impacts of individual factors are counted. If a large number of obese members of the insured population or workforce also have poor fitness levels, unhealthy diets, smoke, don’t get enough sleep, etc., the actual losses of each such member will be counted as caused by obesity, but also as caused by each of the other factors identified as health risks or value impairment factors. The more risks are identified, the greater the over-counting will tend to be.

The over-counting can be easily recognized at the population level, since it will result in the total risk/impairment factor losses being far greater than the known total of actual health care costs and value impairment losses for the population. But the over-counting of individual factor costs and losses will make it appear that there is far greater potential for recovery of such costs and losses available in the typically factor-specific interventions that PHI relies upon to generate its financial benefits.

For example, research has found that people who don’t sleep enough tend to overeat. In a controlled study, on after subjects got only four hours of sleep, they consumed 22% more calories, on average, than on days when they got eight hours, roughly 560 more calories. Over time, this would yield roughly a pound a week in weight gain. [A, Harding “People Get Hungrier When They’re Starved for Sleep” Reuters.com Apr 9, 2010] If sleep deprivation and obesity are commonly found in the same people, both factors will tend to yield significant overstatement of their costs, losses, and potential for gains.

Moreover, because health and impairment factors often occur together, it may well be that a typical “silo” intervention focusing on only one factor may not have anywhere near the desired effect unless an associated factor is also addressed. People who have unhealthy sleeping habits may simply not be successful in reducing their calorie intake until they first increase their overall duration of sleep. Or, for that matter, it may be that sleep interventions will not work until those affected achieve weight reductions, since obesity is associated with sleep apnea, for example.

In addition, when the costs and losses linked to single factors are thus overstated, the effect may well be that the wrong factors are selected for change. It might well be, for example, that because overweight/obesity is typically the most prevalent single cost/loss factor, it will be most often selected for intervention. Yet it may also be other factors commonly found to co-exist with weight problems, perhaps depression/anxiety, poor fitness, etc. that are far more responsible for the effects attributed to weight. If these factors are not also addressed, there may prove to be relatively little effect from weight management interventions.

And if that isn’t enough of a problem, weight management is known to be one of the most difficult factors to achieve lasting success in. It might well be that improving the fitness levels, sleep deprivation, or emotional problems among obese members of the population will have far greater success and positive impact than would weight management efforts.

While it is almost always easier to think about and deal with one problem at a time, the fact that health risks and value impairment factors rarely occur one at a time should be recognized in PHI planning. Unless the real impacts of individual factors can be determined via statistical analysis, and the real potential of such factors for successful intervention identified, there may be far more unrealized expectations and disappointed hopes than there should be. While overstating the size of the problem and the potential for gains has been common in PHI, there are plenty of true costs, losses and potential to justify intervention without exaggerating any of these.

Individual Insurance Mandate: As Bad(?) as Social Security

2010-04-05T09:17:00.000-07:00

seems to be an unusually high degree of objection to the idea of an individual mandate for Americans to buy health insurance under the recent health reform legislation. This is almost a perfect issue for self-proclaimed champions of “liberty” and professional media “commentators” to rail against. How dare the government force (with penalty of a fine beginning in 2014) individuals to buy health insurance!

From an actuarial perspective, i.e. from the point of view of insurance, itself, it is far better for populations to be entirely insured, than for individuals to have the ability to buy or not buy whenever they please. The purpose of insurance is to “spread the risk” over as large a number of people and as long a period of time as possible, in order to keep the costs to individuals who are drawing benefits therefrom as low as possible.

Health insurance really only makes insurance sense when it covers unexpected, catastrophic events such as death, auto accidents, floods, fires, etc. that would create severe burdens on people if they had to pay the full costs involved at one time. It spreads the risk, the costs, the timing, etc. over far greater numbers, thereby keeping the right-now costs to individuals bearable.

Without insurance, in many cases, we’d have to bear serious damage to our economic well-being whenever a crisis occurs. Without insurance, many of us would be dependent on public or private charity to pay, or burden others who have to pay more in order to make up for our failures. To some degree, everyone benefits from insurance when they have it, through the sense and reality of security and protection it affords.

But there is always a degree of unfairness about insurance, whenever the catastrophic event involved arises with greater predictable frequency or severity among some more than others. Why should those of us who are above-average safe drivers have to pay for careless young and senile old drivers? Why should those of us who live healthy lives have to pay for those who life unhealthy ones?

Differential prices for differential risk have been adopted in most insurance situations, and are often included in employer-sponsored health insurance. But one thing that most negatively affects the equity and costs of insurance in a population is the ability of individuals to choose to buy coverage only when they feel like it.

In insurance jargon, this is an open invitation to “adverse selection”, i.e. it offers opportunities for people to buy only when they are sure they will have a “covered event” in the immediate future. While this is harder to do with auto insurance, it is pretty simple for most people with respect to health insurance. If women could wait until they are pregnant to buy health insurance, the costs of maternity care would end up being spread only among pregnant women, forcing all to pay the average costs per case, rather than sharing across the entire population.

The best example of “forced insurance” aside from “compulsory” auto insurance (proven ineffective, adding costs for “uninsured driver” coverage to those who obey the law, while not stopping uninsured drivers from driving) is the Social Security System. While not truly insurance, it is compulsory for all employees, including self-employed earners, and covers both those unexpectedly disabled and those who reach eligibility age. By sharing costs across a huge population and an entire working lifetime, it makes the costs bearable, and adds the “sharing” effect of requiring those who earn more to contribute more, while also enabling them to collect more in benefits.

Back during the Depression, only a modest proportion of the population could expect to qualify for retirement benefits under Social Security, though almost everyone wanted to be sure they were not impoverished if they were lucky enough to survive that long. The system can expect serious trouble in the future, of course, because there are too many people receiving benefits and too few paying in “premiums”, as the age mix of the population has shifted markedly since the 1930s.

Compulsory health insurance is little more than an added “social security” insurance program, with private health insurance taking the place of government except when federal and state governments become the insurer of last resort for low-income segments of the population. The idea of all contributing to ensure the well-being of almost all is nothing new in this country, nor is it in most other countries of the world. If people are allowed to choose whether and when to buy health insurance, the idea gets lost and the benefits diminished.

It is understandable that many people object to being “forced” to buy insurance, even though the penalty for not doing so is set to be considerably less than the predicted cost of insurance. On the other hand, there are few among us who would not want to have others share in the costs of expensive sickness care if a catastrophic illness or injury affected us. But if people can choose whether and when to buy insurance, they will be seriously reducing the benefits for all, including themselves, if such a need ever arises for them.

Opting out of health insurance forever would make almost all of us dependent on the kindness of others, while enabling us to choose when to buy and drop insurance amounts to “gaming the system” to the detriment of our neighbors. In Massachusetts, for example, people can make such choices, and they are ruining the system for their peers, with “short-term” opters-in generating average costs of $2200 per month while insured, compared to contributing only $400 in premiums. [K. Lazar “Short-Term Customers Boosting Health Costs” Boston Globe Apr 4, 2010 (www.boston.com/news)]

It can be argued, of course, that by not insuring consumers, we give them far more “skin in the game” with respect to leading healthier lives. But when spreading the risk over all members of the population, as well as the complete lifetime of each is compared to enabling free choice of whether and when to be insured, the greater good for the greater number clearly lines up with compulsory coverage. Even political conservatives like Mitt supported the idea, while governor of Massachusetts, at least. In the long run, requiring those who can afford to do so to buy health insurance is the best way for a health insurance system to work for all. [J. Cohn “Why Americans Should Support an Individual Mandate” KaiserHealthNews.org Apr 5, 2010]

Solving the Health Crisis: Timing and Responsiblity

2010-04-05T08:05:00.001-07:00

There are two separate, though related, issues of timing when it comes to solving, or at least ameliorating the health crisis affecting the world. First is the basic question of when in the development of disease and injury should we focus our efforts. The US, and to a great extent, the world’s “health system” is chiefly focused on reactive health care, on diagnosing and treating states of “unhealth” after these arise. But most experts agree that the best solution, in at least most cases, is to proactively intervene before unhealthy occurs, to reduce the incidence and prevalence of unhealthy in the first place, wherever it can be accomplished cost-effectively. [R. Pear “New Health Initiatives Put Spotlight on Prevention” New York Times Apr 4m 2010 (www.nytimes.com)]

It is not that this is a new idea. Public health experts have long argued for “moving upstream” vs. spending virtually all efforts on rescuing those already afflicted. But the “system” is almost wholly devoted to rescuing, and a huge portion of the economy is dependent on revenue generated thereby. While the pharmaceutical industry has embraced prevention to some degree, what will happen to all those hospitals and physicians if we won a “war” on sickness?

Moreover, prevention is by no means the most cost-effective option in all cases, nor even in most cases, at least so argue many health gurus. And even if prevention is the best option, there remains the question of who should pay for it? Everyone seems convinced that one essential is that consumers should take greater responsibility for managing their own health. But this is a silly philosophical argument unless we figure out how to achieve such a “consummation devoutly to be wished”.

Except for the sick care industry dependent on the failure of preventive efforts, virtually every employer, insurer, government, and consumer would be better off if there were less sickness around. But employers, at least, are striving mightily to shift the costs of health insurance to government, taxpayers, and employees where possible. Insurers have only lately begun to invest in prevention, since they do not have the responsibility for paying the costs of individuals’ sickness long enough, in most cases, to make investments that don’t pay off for years.

Government seems to be persuaded that prevention, from primary to secondary to tertiary, doesn’t pay off well enough – and it is often as short-term focused as are insurers. Employers have learned that healthier employees are more productive and deliver greater economic value to their employers thereby, so most are investing separately in prevention via wellness and health promotion, even while reducing their investments in sickness care insurance.

The second timing issue relates to when in the lifetimes of people, is the best time to invest in prevention. On the one hand, the federal government seems to believe that waiting until people are eligible for Medicare is too late, while covering prevention for Medicaid mothers and children makes sense. Increasingly, it is being recognized that “prevention” is best adopted early in the life of individuals, whereas epidemics of obesity, diabetes, heart disease, and other expensive chronic conditions are initiated early. [L. Carroll “Mid-Life Health Ills Take Root in Tots” MSNBC.com Apr 5, 2010]

If babyhood and young childhood represent the most cost-effective opportunities for prevention, at least for instilling healthy diet and exercise habits, who will take responsibility for that? Parents are increasingly absorbed in generating enough income to survive and live the life they wish, while relying on day care and schools to “develop” their children properly. Schools are hardly able to meet their traditional educational roles, much less add on responsibility for producing healthy children.

While insurers share an interest in promoting healthier populations early on, the only time it makes sense for them to invest in prevention of unhealthy in children is when such children are listed as dependent “members” of the health plan for which the insurers are responsible. If the payoff from prevention among children is many years, often decades off, what economic sense does it make for insurers, or employers for that matter, to invest now?

It can be and has been argued forcefully that consumers ought to be the ones responsible for investing at least time and effort in the health of their children and themselves. But how can all or most of us be persuaded, encouraged, motivated, empowered, and reminded to do so at the most appropriate time and place, while using the most cost-effective methods? If it “takes a village” to raise a child, it certainly takes a large part of a country and its economy to achieve real health in its population.

Already, there seems to be great willingness to shift responsibility for preventing mistakes in sickness care to patients and families, to shift as much responsibility for the care of elderly to family members. Shifting more of the burden to consumers for prevention and health promotion will always seem attractive to institutions that would have to bear the responsibility otherwise. But can it be done?

We should be able to agree that the timing of care should be shifted to before rather than after people are sick, wherever cost-effectiveness is greater with earlier interventions. We should also be able to agree that timing interventions as early in the life of individuals should be practiced wherever the most cost-effective results come from such interventions. But until we focus on devising and implementing ways to enlist the most cost-effective coalitions of who’s responsible, we will be simply arguing, rather than acting.

Gaining Indirect Value from PHI Investments

2010-03-20T08:47:00.000-07:00

The emphasis of PHI is on productivity, performance, and employee value contributions improvement as much as on health per se. For this reason, it makes sense to consider the direct and indirect effects of pay-for-performance systems on both health and performance. The Health as Human Capital Foundation blog (hhcf.blogspot.com) offers a number of discussion of the impact of employers’ compensation and benefits policies and practices on worker absences, productivity, performance, and turnover, for example. It also includes examples of the indirect effects of P4P on health, as workers who are paid more for higher productivity or better performance tend to take better care of their health, as a result.

Analyses by HHCF staff have found that property values are strongly linked to obesity rates, for example, suggesting that people with more money take better care of their health. [“Money Matters. What Do Skinny People in Big Houses Have to Do with Flu Shots and Bonus Pay?” Entry 11 - 2008. May 26 2008] Incentive systems that augment the value to workers of better self-management of their own health are linked to better health status. “There are Wellness Incentives, and then There Are Incentives that Increase the Importance of Being Well” Entry 13 – 2007 June 17, 2007] By contrast, explicit health improvement efforts often fail to yield desired results. [“Getting Real: The Reasons Companies Rarely Find Actual Dollar Savings with Health-Improvement Programs” Entry 26 2009 Dec 21, 2009}

Paying employees explicitly and significantly for higher productivity and better performance can serve as a strong motivator to get them enrolled and significantly engaged in health improvement efforts sponsored by employers or insurance plans. And because P4P incentives are only paid to those whose health improvement efforts actually yield value to the employer, it is a simple matter to ensure that incentives only go to those who succeed in their health and performance improvement efforts, rather than wasting money on incentives that yield only participation in particular elements of a PHI initiative. Employers can scale performance incentives to ensure that they yield a positive ROI simply by making them gainsharing amounts.

Enabling employees to gain greater compensation, whether raises or bonuses linked to performance can also have an indirect impact through altering the mix of low, average and high performing workers in the employee population. When Safelite switched from hourly pay to “piece rate” compensation for its windshield replacement staff, for example, it found that turnover rates among low performers went up dramatically, while rates for high performers went down. Over time, such an effect will significantly alter the proportions of high vs. low performers, and thereby yield added value for employers. [E. Lazear “Performance Pay and Productivity” American Economic Review 190:5 Dec 2000 1346-1361]

The use of P4P systems will also tend to alter the mix of recruits that employers attract, whenever there are similar competing jobs available that do not pay more for high performance. As the mix of job applicants shifts toward higher percentages of top vs. average or low performers, this will add a separate, though related impact to that affecting current members of the workforce. Examples such as Safelite demonstrate how profitable such effects can be – it achieved a 44% improvement in productivity for only a 10% increase in overall compensation paid to its workers.

Yet another indirect effect is the tendency for P4P systems to motivate workers eligible for performance-based pay or bonuses to stimulate efforts among “back-office” support staff. In the Safelite example, a common problem that had accepted as part of the support system was that of inventory staff delivering the wrong windshield to installers. This meant having to make two trips for installations at customers’ locations. Once the P4P system was introduced, installers began pressuring the inventory staff to get it right the first time, since every mistake cost the installers valuable time. As a result, errors declined dramatically, adding to customer satisfaction, as well as to installers’ and the company’s revenue and profits. [B. Hall, E. Lazear, et al. “Performance Pay at Safelite Auto Glass” Harvard Business School Dec 6, 2001 (Case Studies 9-800-291 & 292]

Such indirect impact can be enhanced by including support staff in the P4P system, though this was not part of the Safelite example. It would be a relatively simple matter to include performance pay or bonuses to support staff based on their delivering the correct items to installers, though this would tend to reduce the amount of money available to pay installers. In a gainsharing system, the more people are eligible for P4P compensation, the further gains must be shared, though in the Safelite example, with 44% greater revenue potential vs. only 10% more compensation, there would have been plenty of room for more workers to share, while still enabling the employer to keep the lion’s share, or at least an equitable share, of the gains.

While the usual approach to PHI is to initiate health and wellness programs separately, it may work better if P4P systems are introduced or augmented at the same time. To the extent that performance incentives will serve as sufficient motivation for many workers to engage in healthier behaviors, they will reduce dependence on participation incentives and reduce overall costs of PHI strategies. Moreover, because performance incentives can be scaled to the level of performance and value improvement achieved, they can increase the likelihood that PHI investments will yield a positive and admirable ROI.

How Much Would Prevention Really Save?

2010-03-18T10:52:00.000-07:00

There have been many champions of “prevention” who insist it should be the major focus for health reform. Estimates have varied as to how much of current sickness care expenditures could be eliminated if prevention were universally applied and adopted – somewhere between60% and 80% used to be common, though one recent author claims: “Today, preventable illness accounts for close to 90% of all health care costs. [E. Kasen “Preventive Care Strategies: The Carrot and the Stick” Employee Benefit News Mar 1, 2010 (http://benefitnews.com/news)]

Unfortunately, all such champions are focusing on the wrong question. While it seems to be the case that only 5% of our national health expenditures are devoted to prevention, the real question is how much difference “prevention” would make, in practice, to sickness care costs. The classic argument against any universal adoption of prevention (whatever that means) is that it typically has to be applied to a lot of people in order to make a difference to as many as one person.

Answering the question of how much would prevention save requires estimating:
1. how many people would have to be “treated” with a given preventive intervention
2. how much that would cost, and
3. what precise difference to sickness care costs, plus perhaps losses of productivity and performance attributable thereto, would be made

Cost-effectiveness analyses of a wide range of preventive interventions have yielded a wide range of results, in terms of both expenditure savings and quality-adjusted life years (QALYs) gained through preventive interventions. Flu shots tend to show up on the positive side, with low cost per shot and high levels of sickness care and lost productivity savings.

But a host of other preventive interventions cost a good deal more than they save, sometimes orders of magnitude more, and end up costing as much as millions of dollars per QALY added for the people affected. In effect, preventions is subjecting itself to the same myopic approach that medical treatment usually applies in judging whether and which treatment should be used in sick patients.

The typical approach is to look solely at effectiveness, the degree of positive impact a given treatment has, compared aeffective, in this sense, it is part of quality care; if, by chance, there have been comparative effective studies made, the most effective treatment should be used.

Back in the “old days”, i.e. before individual differences (usually genetic) were recognized as affecting which treatments work best, and even which work at all, practitioners could justify their efforts entirely based on whether or not they were proven to be effective. Arguments have been common about which are most effective, and practitioners have insisted on their rights as professionals to make decisions on an individual patient basis, even when science indicates otherwise.

With the move toward the personalization of sickness care, the need to recognize personal differences has become greater. It even applies to preventive efforts. A recent study indicates that when weight management diets are selected based on genetic data, they are significantly more effective than when such personal differences are not recognized. [“DNA Test Could Predict the Most Effective Diet” BBC News Mar 5, 2010 (www.healthandwellnessassociation.com/news)]

This complicates the determination of effectiveness and cost-effectiveness of preventive treatments of all kinds, since it means that separate calculations must be made for as many different segments of the population are affected by individual differences. But the item most frequently omitted from calculations of the relative cost-effectiveness of prevention is the cost of getting providers and patients to adopt preventive methods, particularly those that require repetition or must be sustained, perhaps repeated over long periods.

It is one thing to lose weight for example, but what is the cost – to both sponsors of weight loss efforts and those who participate therein – of doing so? With weight loss, how much more expensive will the right diet be, and how much will the exercise cost? How much time, effort, and expense will be incurred by practitioners of various kinds in coaching, nagging, or whatever roles will be required?

And particularly with weight loss, how long will the effects of interventions be sustained? A recent study of 757 hospital workers who participated in a voluntary 12-week wellness program focused on diet and exercise found that obese patients averaged a 3.0% weight loss after the program was completed, but only a 0.9% loss after one year. Overweight patients lost an average of 2.7% of their body weight at the end of the program, but only 0.4% after one year. [R. Preidt “Workplace Wellness Programs Work” Health & Wellness Association Mar 3, 2010 (www.healthandwellnessassociation.com/news)]

Rarely, for example, do “before and after” pictures of weight loss participants reflect the experience of the average participant, and even more rarely do they show the average participant after one or two years post completion of the program. Typically, something like 95% of participants gain weight back, often ending up weighing more than when they started. This yields virtually a permanent market for weight loss products and services, but little value for prevention investors.

While it is simpler, and far more persuasive to over-simplify the prevention issue, it is also a great disservice. Preventive efforts should be analyzed in terms of their full costs, including costs of achieving widespread adoption and any necessary repetition/continuation thereof. And they should be analyzed in terms of the true effects, recognizing their relapse rates, and other lifetime factors that may diminish their value over time.

Essential Elements of PHI: Predictions

2010-03-18T09:45:00.000-07:00

Predictions are among the most essential elements of PHI investment decisions, planning and evaluation. Yet they are the least likely elements to be present in the typical application. This may be due to general skepticism over the accuracy and reliability of predictions, but the effects of failure to generate needed predictions can be worse than the limitations of the predictions themselves.

Perhaps the most important prediction needed in PHI planning and evaluation is the depiction of the future without intervention. While the most common approach seems to be to rely on the most recent year’s experience as baseline for planning and evaluation, the best approach is to use a prediction for the next year if no PHI intervention is applied. When dealing with health, diseases, and risks, it is the normal situation for next year to be worse than this year, absent any intervention.

Predicting the next year realities of the metrics to be used in evaluating PHI is far better than relying on after vs. before comparisons. Moreover, when applied to disease management, after vs. before comparisons are likely to ignore the natural tendency of healthcare costs, for example, to “regress to the mean”, i.e. people with outlier levels of high expenditures in the baseline year are far more likely to have lower expenditures in the following year as a matter of course, whether or not any PHI intervention is applied.

Failure to take into account this regression to the mean tendency will result in significant over-counting of reductions in expenditures as consequences of interventions, rather than recognizing them as natural phenomena. A simple after vs. before comparison will ignore a host of factors that will tend to drive expenditures higher, such as inflation, effects of population aging, and worsening of risks without interventions.

But the most serious and most common problem is the failure to predict the likelihood of success, for individuals and populations. Literally thousands of studies have demonstrated that only a portion of populations enroll in PHI efforts, make the desired health behavior changes, achieve desired health status improvements and increases in productivity or performance. The probability and accuracy of predictions of such effects determines the success of PHI investments, so are essential in making good investment decisions.

Yet the vast majority of predictions rely on predicted costs of inaction, alone, rather than the probable savings of action. If the predicted costs without intervention are as high as $5000 per employee or plan member, for example, but only 30% of the population will enroll, only 50% of those will make the desired behavior change, 50% of those achieve the desired health status improvement, and 50% of those demonstrate improved productivity/performance, at an average gain in value of 5%, then the predicted effect would be only $5000 x .30 x .50 x .50 x .50 x .05 = $9.375 per member of the population.

If the costs of intervention amount to as little as $10 per population member, the results would be a net financial loss, though 7.5% of the population would gain improved health (30% enrollment x 50% making health behavior change x 50% achieving improved health = 7.5%). Without significantly higher levels of success at some point(s) in the cascade of value, there would be little reason to invest.

By contrast, an accurate prediction of the probability of success can add greatly to both the overall investment decision and the selection of targets for intervention, as well as the matching of interventions to individuals. If the predicted value of individuals or segments can be made and is reasonably accurate, interventions can be assigned to each based on the combination of potential (e.g. predicted healthcare costs and productivity/performance levels without intervention) x probability of realizing it.

This combination would at least enable the “scaling” of interventions to the predicted value of individuals or segments. Sponsors could use a predetermined level of ROI to make such matches, e.g. if they insist on a 2:1 ROI, they would set the limit for the cost of interventions to ½ the predicted combination of value. They could even use the same process to empower participants to select any combination of intervention costs and incentives they can reward themselves with for achieving the predicted results.

In any case, creating and continuously improving predictions of potential value gains times the probability of achieving them -- for populations, segments, or individuals -- should greatly improve the strategic and tactical investment decisions that sponsors can make. By contrast, relying solely on baseline measures and potential gains is as likely to turn out badly as well, depending on what the probability of success really is, and how well matches of investment are made.

There has long been an unfortunate practice in quoting an old saying as “Ignorance is bliss; ‘tis folly to be wise.” The accurate quotation is “If ignorance is bliss…etc.”. In the case of planning and making PHI investment decisions, ignorance of the potential/probability and value of success is not merely unwise, it could be disastrous.

Call Centers as Natural Experiments for PHI

2010-03-18T08:31:00.000-07:00

Operating call centers is an essential function for a wide range of organizations, though the function is all too often outsourced, even off-shored to other organizations. One advantage to making and keeping it an internal function is the opportunity such centers offer as natural experiment possibilities with respect to virtually all elements of planned and adopted PHI strategies and initiatives.

Call centers tend to employ lower-paid personnel, though some specialized function centers, including those in the PHI business, itself, often employ highly-paid professionals, such as nurse coaches for disease management initiatives. But turnover in such centers may be high, given the nature of the work, but that means that the importance and value of promoting both the health and engagement/loyalty of call center staff will be that much greater.

By applying PHI strategies and initiatives to call center staff, perhaps using call centers for pilot testing of basic policies and procedures, along with other efforts aimed at improving employee performance and value, employers and insurers can learn a lot about how well PHI elements work. Perhaps most valuable is the opportunity call centers offer for testing whatever method will be used for measuring productivity and performance.

It has long been a general practice in validating the various methods available for estimating productivity and performance levels through self-report surveys to compare such estimates to actual measured performance in call centers. One such validation, for example, found that self-reported productivity impairment due to health problems was 2.5 times greater than actual measured output. [G. Pransky, et al. “Performance Decrements Resulting from Illness in the Workplace” JOEM 47:1 Jan 2005 34-40]

Because of the normally high turnover and absence rates in call centers, due to the pressures of the work, such centers also are logical locations for pilot testing of efforts aimed at reducing both. [K. Carson “Tested Methods for Reducing Absenteeism in the Call Center” CustomerManagementIQ.com Mar 16, 2010] While health, and particularly problems related to stress, are likely to be major sources of impairment, there may be other causes of absence and turnover that need to be addressed, and these can be tested as well.

Because pay-for-performance (P4P) systems are often used with call center staff, such centers also represent logical testing grounds for such systems. In addition to testing the effects of P4P on call center staff performance, such systems also tend to promote “internal marketing” efforts by call center staff to encourage or demand back-office support improvements. [B. Hall, E. Lazear, et al. “Performance Pay at Safelite Auto Glass” Harvard Business School Dec 6, 2001 (Case Studies 9-800-291 & 292)]

Call centers may or may not, however, be good places for gauging the “multiplier” effect. This is the impact that the absence of particular employees with particular job skills and responsibilities, have on the performance of others. This effect has been measured at greater than 10x the individual’s lost performance, in the case of construction engineers, at 1.85x for medical assistants, but virtually nothing for easily replaceable short-order cooks, for example. [T. Parry “Capturing the Elusive: How Absence Impacts Lost Productivity” IBI Research Insights Aug 2006 (www.ibiweb.org)]

While it seems likely that the absence of one staff member in a call center could impact the performance of those on the job, the actual impact would likely vary according to the number of other agents available to take up the slack, whether “floaters” could fill in, and whether the call center is used mainly to handle incoming vs. outgoing calls. In any case, it would likely not be the best model for all other employees in the organization.

Whether call centers are the best model for testing PHI interventions, or even the accuracy of productivity and performance impairment estimates is not yet known. But at least the general applicability and impact of PHI can be tested therein for employers who are unsure of the results they might get from PHI investments. As more experience with call center validations of measurement and interventions emerge, we should gain a better idea of the value of such natural experiments.

Social Networking: PHI Applications

2010-03-18T07:40:00.000-07:00

Social Networking is one of the latest “can’t miss” ideas being touted and adopted in marketing to consumers, and even to business prospects and clients. Encouraging and enabling customers to share their experiences and impressions with their peers is one of the cheapest and most effective means available to influence people’s awareness, opinions, and actions, as social networking keeps showing in the political arena.

It is also being touted as a means of enhancing employee “engagement”, usually defined in terms of high levels of commitment/motivation and loyalty to their employers and jobs. But it is equally well suited to promoting enrollment, engagement, and success in PHI efforts aimed at health, performance, and value contributions that arise from improvement therein.

There are some basic “rules”, or at least logical things to keep in mind when planning and implementing social networking strategies – in PHI, as well as for promoting employee engagement in general. For example, a recent blog discussion on general applications suggests six things to keep in mind. [M. Finney “How to Use Social Media to Enhance Engagement” Smart Blog on Workforce Mar 16, 2009 (smartblogs.com/workforce)]

For PHI purposes, one of the key elements is deciding exactly how social networking will be used in overall strategy and particular initiatives. It can be applied to general promotion of the idea of health improvement and its performance consequences, to increase enrollment in risk assessment/screening, in particular initiatives, to improve behavior change efforts and success, and similar “marketing” aims.

The purpose selected will automatically determine the desired membership of the segment of the population intended to participate in social networking, as well as the behaviors that are to be promoted and sustained. The social networking “experience” should then be designed accordingly, though its members will have a lot to do with what the actual experience becomes.

Enlisting “opinion leaders”, and encouraging more successful PHI participants to join in the social networking effort can be a task assigned to PHI vendors, since employers are likely to be prohibited from identifying specific participants. They may also be self-selected through general invitations to PHI prospects and participants who consider themselves to be leaders and successful, though self-perceptions may be deluded.

Employers should monitor social networking activity and content, and may even set rules for the use of such networks, and monitor adherence thereto. But there should be no attempt to censor content that is merely critical of the PHI process or employer, merely content that is demeaning, rude, offensive, etc. to its participants.

The employer and vendor may participate in the networking, but should be clearly identified as such when they do so. Both should avoid any actions that will diminish participant trust in the network or PHI sponsors. Ideally, participants themselves should be empowered to set and enforce their own rules, in addition to any set by sponsors, so that they are confident in the network and its content.

All participants should be encouraged to contribute content, and about any subject of concern or value to them. The greater the participation level, the more value the network is likely to have, and participant ideas for increasing participation in the network, as well as for increasing the success of the PHI effort should be included in the general empowerment effort.

As is the case with social networks used in promoting purchases and use of products and services, PHI networks should promote participation and success at minimal cost to sponsors. The effects of networks should be continuously monitored to check on how well they are working, and to capture suggestions and feedback to help the network as well as the PHI effort in general to improve its effectiveness and efficiency.

Measuring PHI Results: Correlations, Conversions & Causes

2010-03-01T10:21:00.000-08:00

Measuring productivity and performance is usually either very simple, or extremely difficult. When workers are already being monitored, evaluated, and paid based on some sort of “piecework” arrangement, their output is already being measured. Translating that into “performance” can be no more than adding a few technical quality and customer satisfaction metrics, to be sure quantity is not being delivered at the expense of quality.

But when there is no system for objectively measuring productivity and performance, the task is likely to be highly complex, expensive, and subject to loads of doubt as to the validity and reliability of metrics. The most common approaches are either self-reported estimates by workers themselves, or subjective appraisals by their supervisors. Both of these are subject to understandable and well-documented biases.

Self-reporting of output and performance are subject to being exaggerated by employees, either through self-delusion or self-protection. Low performers are known to tend toward claiming at least average levels, if not higher. High performers may under-report, for fear they be thought of as “bragging”, or self-congratulatory. When it comes to self-reporting of the effects of health factors on productivity or performance, exaggeration may apply when people feel really bad, or under-stating impairment may be used lest the overall worth of the individual be undervalued.

In one example, where call center agents’ objective productivity was routinely measured, workers affected by health problems overstated their impairment by 2½ times, with actual output only off by 8%, compared to an average self-reported reduction of 20%. [G. Pransky, et al. “Performance Decrements Resulting from Illness in the Workplace” JOEM 47:1 Jan 2005 34-40]

There are at least half a dozen widely accepted and used self-reporting survey options, each of which is likely to vary measurably in results from the other, in addition to varying in difficulty and cost of application. Purveyors of such methods are likely to report high levels of correlation between objective measures and their results, but such a correlation can result from a consistent bias as well as consistent accuracy.

Mathematically, if a self-reported method delivered results that were consistently 2.5 times greater than objective measures, the correlation between the self-reports and reality could be quite high, approaching 1.00, the highest level of correlation available. But that only shows that the survey method varies in a consistent way with actual productivity or performance, not that it comes close to reflecting true values. It would be considered reliable, because of its consistency, but not valid because of equally consistent inaccuracy.

What is needed for validity in such situations is a consistent conversion factor that translates self-reported data into a far more accurate estimate. This is only possible when there is high correlation, showing that workers are consistent in the degree of over- or under-reporting. With such consistency, e.g. if they mainly come very close to overstating their impairment or output by a factor of 2.5x actuality, this factor can be used by convert self-reported data to reality by simply multiplying self-reports by 40%.

Even when productivity/performance is accurately measured, there arises the question of whether changes that are noted therein should be attributed to whatever causes have been initiated to bring them about. While improvements in health have often been found to yield apparent improvements in productivity/performance, there are many other causes that may have contributed, as well.

As is frequently noted in Health as Human Capital (hhcf.blospot.com), a wide range of factors contribute to both normal absences and longer disability as causes of lost productivity and performance. How much workers get paid compared to what they earn when present is one of the biggest causes of differences therein. Improvements in worker commitment to their work or employer, training and development, new technologies and support systems can all cause significant improvements.

Changes in compensation levels and systems can also be powerful causes. When Safelite switched from an hourly pay to a pay-for-performance system, productivity increased by 44% in the very first year the new system was implemented. [E. Lazear “Performance Pay and Productivity” American Economic Review 190:5 Dec 2000 1346-1361]

Increasing employee autonomy and flexibility can have similar impact. Empowering workers to perform their tasks whenever and wherever they pleased enabled Best Buy to improved corporate staff processing of orders by 36% in the first year of the new system, in addition to cutting staff turnover from 16.7% to zero. [M. Conlin “Smashing the Clock” Business Week Dec 11, 2006 (www.businessweek.com)]

If employers choose to offer incentives as part of their PHI initiatives that are partially or wholly based on employee productivity or performance, then the incentives, themselves, may be responsible for improving employee outcomes, perhaps as much as improving their health. The only way to systematically separate the effects of different possible causes is to use control vs. experimental groups, which also reduces the overall impact of the initiative being experimented with. Employers are understandably reluctant to reduce such impact, when they have faith that the initiative will work.

Correlations, conversion, and causes are all important concerns when measuring problems and the effects of solutions in PHI. All can be handled if sufficient care and planning is included in applying ways of improving productivity and performance. Such care is essential to ensure that “results” discovered are real and correctly attributable to the PHI intervention applied.

Broadening the Use of Marketing Tools in PHI

2010-02-26T11:41:00.000-08:00

Among the many applications of marketing concepts and tools to PHI, the use of predictive research to identify who is how likely to respond well to coaching, and who is how likely to engage in an “ambassador” role are both likely to be useful. While traditional marketing research creates a database and analyzes it around consumer purchasing patterns, it takes only minor adjustments to translate this application to consumer health behavior patterns. {"Meaningful Online Engagement Model" Immersionactive.com Download Feb 25, 2010]

First is the issue of whether or not people are likely to respond to the basic appeal, the “call to action” of the PHI communications. The action called for may be as simple as completing a health risk assessment survey, joining in a biometric/risk screening program, or whatever initial intake method is used to begin the PHI process. Or it may be to enroll in a particular PHI intervention, to persist therein, to make an actual health behavior change.

In such cases, learning about “audience” reactions to either proposed or actual communications aimed at achieving specific levels of action is an essential part of PHI “advertising”. Communications campaigns may be conducted by the PHI provider, by the employer or insurer sponsor thereof, or by both. In any case, determining up-front and post-exposure reactions to communications content and channels is sure to help.

This can include what audiences learn, how they feel about the content, whether they have the desired emotional response thereto, how well they feel the theme of the communications is conveyed, whether they believe/trust the content and its sources, etc. Advertising and market research firms, or the sponsors’ own advertising/market research staff should be able to devise survey methods and questions to yield this kind of information – from samples or the entire population targeted for PHI.

In addition, surveys can be used to identify the most likely “ambassadors” and their potential value to the PHI effort, in terms of social “buzz” communications with their peers. Learning which members of the population have the widest “reference networks” with which they regularly interact, how often they share information with them, over how many channels, etc. can identify the most likely to be useful participants in terms of “raving, referring, and recommending” the PHI program to others.

Such research should be systematically followed up by determining actual behaviors of targeted participants – how many enroll, make behavior changes, achieve health improvements, risk reductions, improved disease self-management, etc. The actual success data can then be used to continuously update and improve the predictive power of the attitude and perception surveys used to select the best targets.

The same follow-up should be used for research aimed at identifying and engaging “ambassadors”. Do those identified as best prospects engage in the desired role? Do they have the desired effects? How well do ambassadors succeed in the PHI dimensions, as well as in their role. In theory, at least, they should be more persistent and successful in their PHI goals, thanks to “cognitive dissonance” effects.

Such marketing research can complement the kind of analysis used to identify which members of the population are the best prospects for specific PHI interventions, what pattern of reinforcement is likely to be optimal for them, etc. as described in an earlier post. [“Optimizing Incentives’ Benefit/Cost in PHI” Feb 23, 2010]

By identifying and analyzing who actually adopts a desired ambassador role, in both increasing the number of PHI participants, and enabling peers to succeed in their goals, marketing research can end up paying for itself. Because word-of-mouth can be the most effective, as well as least expensive form or advertising, using marketing tools to identify, engage, and evaluate the social networking efforts of participants can significantly promote both PHI success rates and its ROI.

The Value of Turnover Reduction in PHI

2010-02-26T10:44:00.000-08:00

It is generally understood and accepted that employee turnover is a problem, though promoting “churn” among poor performers may yield more benefits than costs. Estimates of the cost to employers per employee who quits, retires, or is fired and has to be replaced range from a modest percentage of the annual compensation to more than the total compensation of each, combining lost productivity and replacement costs.

Added to such costs may be the “multiplier effects” of losing specific employees or categories thereof. Based on research reports, the effects of mere absence of people in some job categories may be far greater than the value each contributes alone. Effects on other members of the team, unit, or organization may be considerably greater than the value of individual contributions.

One study, for example, cited construction engineers’ absence as causing losses equal to as many as ten employees, since these engineers supply the instructions and directions for entire worksite teams. The average multiplier for all workers examined was 1.33, meaning that the loss of one worker includes the equivalent of one-third more of a worker in negative impact. [“Multiplier Effect: The Financial Consequences of Worker Absences” Knowledge&Wharton Dec 14, 2005 (knowledge.wharton.upenn.edu)]

Turnover also has deleterious impact on the ROI from PHI and other proactive health or disease management efforts. The positive impact of improved health in the workforce only applies if those whose health has improved remain as employees. Otherwise, the healthier employees and their value devolve on their next employer, not the one that invested in their health improvement.

To illustrate, consider the findings in a study by GlaxoSmithKline of a cohort of over 6000 of its employees over a four-year period. Average value gained per employee through better health was found to be $233 in the first year of their participation, $375 in the second, then $944 in the third, and $950 in the fourth. [G. Stave, et al. “Quantifiable Impact of the Contract for Health and Wellness” JOEM 45:2 2003 109-117]

If such a pattern prevailed in a workforce where turnover were high, the employer would not gain nearly as much as would be expected if turnover were low. At 100% turnover, a not uncommon level in some industries and organizations, the average annual savings from PHI investments would never go above the $233 first-year level, since by the end of the first year, all participants would have left for other employment.

If turnover were 50% per year, then as soon as the first year’s experience is over, half of participants would have left, and if an equal number of replacements enrolled in the PHI program, there would be half of participants yielding the first-year $233, while half yield the second year $375. The average for all participants would be ($233 + $375 = $608 divided by 2 = $304. Every year thereafter, assuming the 50% turnover rate persisted, would yield precisely and only that annual added value.

If the turnover rate were 33.33%, then after three years, there would be one-third of PHI participants delivering first-year $233, one-third yielding second-year $375, and one-third yielding $944. The average value gained for the third and subsequent years would be ($233 + $375 + $944 =) $1552 divided by 3 = $517.33 each. If the turnover rate were 25%, by the fourth year, the total participant group would yield an average value of ($233 + $375 + $944 + $950 =) $2502 divided by 4 = $625.50 each.
Only if turnover rates are below 25% per year will there be more participants yielding the higher $944 and $950 value gains than there are yielding only $233 to $375 each. And as the turnover rate gets closer to 0%, the average value yield will get closer to $950 each. Given the obvious advantages of gaining closer to $950 per participant per year than gaining closer to $233, it makes sense on this account alone to work on reducing turnover, in addition to traditional turnover costs.

Fortunately, investing in workforce health, whether through environmental and corporate culture improvements, or direct PHI interventions, tends to reduce turnover, and even make the attraction of higher performers when replacement is necessary easier. Given that a pattern of increasing value from PHI participants over time is the common pattern, considering as well as investing in efforts to reduce turnover makes even more sense when the overall added value of PHI impact is included.

Physicians vs. Retail Clinics in PHI

2010-02-26T10:03:00.001-08:00

The widespread movement among both retail and worksite clinics, staffed in most cases by nurse practitioners, has greatly expanded the options available to employees and their dependents, as well as to the general public. Since nurses carry the vast majority of coaching roles in phone-based disease management already, it seems a natural development, as well as a popular one.

But the American Academy of Family Physicians has taken a position opposing this expansion in the scope of practice for such clinics and practitioners. Physicians have long promoted the idea that they, not other professions, should be in charge of patients’ health, and that empowering others to deliver care independently fragments the healthcare “system”, though it is clearly already fragmented. [A. Robeznieks “AAFP Alters Stance on Retail Clinics” ModernHealthcare.com Feb 25, 2010.

With over 1000 retail clinics, and a rapidly growing number of worksite clinics already in operation, locking them out of the disease or health management domain would further strain problems that consumers already have in accessing proactive health services. Most employers and advocates of health reform seem to laud the growth in more convenient, health-focused sources of care.

Physicians are generally uneducated, untrained, and inexperienced in proactive health, and are clearly the most expensive sources for same. Moreover, with growing shortages of primary physicians, exacerbated by the recently reported 4-hours-per-week reduction in the time the average physician spends in practice, placing greater demands on physicians would seem an illogical choice.

It seems clear enough that health care reform will not happen, or not succeed until and unless it includes dramatic reductions in the incidence and prevalence of chronic and preventable acute conditions, this does not appear a good time to reduce available sources of proactive health and disease management. Physicians’ efforts to retain dominant positions in patient care make economic sense for them, but surely the idea that only physician practices should provide such care makes no sense.

With electronic medical records, more common in retail and worksite clinics than they are in physician practices, coordination of care is possible, even simplified, rather than dependent on the presence of a physician in charge. As primary physicians develop medical home and patient-centered primary care programs including proactive health and disease management, they should be able to first test the hypothesis that they can do it better than retail and worksite clinics.

After all, physicians are supposed to be scientists, not politicians. It would seem better for all concerned if physicians treated the issue of whether proactive health and disease management is best performed by themselves, or equally well, perhaps better by nurse practitioners and others in retail and worksite settings as a researchable question. Clearly rigorous research and analysis could answer the question better than political posturing and influence.

Cascading Effects of Pricing in PHI

2010-02-25T12:04:00.000-08:00

Perhaps because PHI or population health management in general are relatively new “disruptive innovations”, they tend to include wide variations in how they are priced, as well as in whether, which, and how expensive are the incentives adopted in their applications. But these variations can cause, while masking, vast differences in what PHI clients will get for their money.

The last time I checked, for example, PHI providers base their fees on a number of different factors related to the populations to which their services apply. Perhaps the oldest, and certainly the simplest, is the use of a charges per person per month (pmpm) pricing system which fits perfectly with insurance plan pricing and costing practices. This has the advantage of closely mirroring the number of people who actually belong in the population at risk as it changes throughout the year.

But it also has the disadvantage of creating costs for every person in that population, whether or not they will ever yield any benefit to the client insurer or employer. By contrast, other pricing systems may be based on how many members of the population at least participate in health risk assessments, screening tests, etc. which are normally used to determine which are eligible for or will be targeted for PHI “engagement”.

Others pricing systems may be based on the numbers of identified eligible members or targeted individuals actually enroll in a PHI intervention. In theory, prices could be set based on any one of a number of stages in the process of health behavior change where benefits for the client are expected to occur, perhaps on some minimal participation in the PHI intervention, completion of a particular program, making a desired behavior change, even achieving a particular health status improvement.

Pricing based on outcomes would have the advantage of gearing the costs that clients bear to the benefits they desire. Rather than paying for the PHI provider’s effort, clients may well prefer paying for results. And the closer they come to paying for results, the more likely they are to find that they get their money’s worth.

To illustrate, if we consider only options such as pricing on a pmpm or population size basis, per eligible, per participant, per completer, or per “successful” participant, the difference in what clients might have to pay could be enormous. If, for example, there is an “attrition” rate of 50% from each stage to the next in the “value chain” from population to success, then the effective costs to the client would double from each stage to the next.

If charges are set as $10 per person per month in the population, and only 50% of members are eligible, then each eligible person will generate costs of $10/.5 = $20. If only 50% of eligibles actually enroll in a PHI intervention, then the effective costs per enrollee will be $20/.5 = $40. If only 50% of enrollees complete the intervention, the effective costs per completer will be $40/.5 = $80. And if only half the completers actually succeed in improving their health behavior, for example, the costs per successful participant will be $80/.5 = $160.

If, perhaps, only half of those who succeed in improving their behavior end up achieving better health, the effective costs per healthier participant would become $160/.5 = $320. If only half of those whose health status improved yielded any measurable reduction in health care costs, in productivity or performance, then the effective costs per more valuable participant would become $320/.5 = $640.

Depending on how long the value chain is, and what are the conversion ratios from one stage in the chain to the other, the effective costs of PHI efforts could easily become overwhelming. For clients, paying fees based on any stage where no discernible benefit is likely would seem dangerous, if not foolish, though suppliers may prefer getting paid as early in the process as possible. If benefit does not actually emerge until late stages, but the probability of conversion to value can be accurately predicted, earlier payment to suppliers can more easily be justified.

In practice, it is more likely that PHI providers and clients would be able to agree on an interim basis, so that suppliers generate revenue to cover their costs as efforts are implemented and “consumed” by participants. Then, at the end of an agreed-upon period, adjustments can be made to reflect the actual conversion to value that the population yields. Already, some providers offer guarantees or risk/reward contracts that ensure some specified level of success, such as breakeven by the end of the first year, and specific ROI ratios at the end of subsequent years.

In the long run, it makes sense for both PHI providers and clients to develop value-based pricing so that providers achieve and clients are happy to pay for proven value, rather than let either no evaluation or sloppy evaluation continue. Since markets depend on fair exchanges of value, the more specifically value can be measured and used as a basis for pricing, the better all around.

Excessive Market Power in PHI?

2010-02-25T11:19:00.000-08:00

There has long been concern about “excessive market power” in health care. Most is directed at insurers, who have merged and acquired their way to dominant positions in a wide range of markets, while enjoying exemption from the controls of anti-trust regulation. [J. Cantlupe “AMA Bashes Health Insurer Monopolies in Many Markets” HealthLeadersMedia.com Feb 24, 2010] In at least 24 of 43 states reporting, two insurers owned market shares of 70% or more, qualifying as at least oligopolies, and this is a growing trend.

But recently, concern has also been expressed about growing market power in many markets by provider organizations, combinations of hospitals and physicians, or large multi-hospital systems. [R. Berenson, et al. “Unchecked Provider Clout in California Foreshadows Challenges to Health Reform” Health Affairs 29:4 April 2010] In some ways, this might seem to be a natural development responding to the growth of market power among payers, but some worry about what it means for health reform.

Moreover, the movement among providers in general is toward greater and greater accumulation, with small physician practices and independent hospitals declining in all but isolated rural areas, though California, for example, prohibits hospitals or other corporations from owning physician practices. And with growing interest in patient-centered medical homes that work best in large organizations, and a push toward Accountable Care Organizations, it seems likely that concentrating market power among providers will continue.

This can be a problem, as can concentrated market power among insurers, to the extent that competition helps to control prices, while concentration tends to increase prices. But in PHI, whether applied by employers for their workforces, or by commercial or government insurers for their members, different market dynamics apply than is the case with sickness care.

Sickness care, which constitutes at least 95% of all “healthcare” expenditures, develops according to its effectiveness. Competition based on quality considers only what works best, not what works most cost-effectively, though recent calls for a national program of cost-effectiveness research may mitigate this in future. By contrast, PHI is at least subject to, though not yet fully responsive to, comparative cost-effectiveness.

Where sickness care providers can claim superior levels of quality as justification for their high costs, PHI providers have to justify their costs in terms of what they enable their employer or insurer clients to save (in medical/hospital care, workers compensation, disability, and overall labor costs) or gain (in improved workforce performance and value). The worth of PHI is at least measured in precisely the same currency as its costs, where sickness care compares benefit
“apples” to dollar-cost “oranges”.

The full benefits of competition in PHI are by no means widely enjoyed, at least not yet. Only a portion, though at least a growing portion of payors even calculate the benefit/cost ratios or ROI on their investment. PHI providers may count only their fees as costs when reporting ROI, while ignoring what may be significant costs that clients incur in promoting and supporting PHI efforts internally.

In addition, PHI providers have historically been guilty of what is at least sloppy measurement of outcomes, often ignoring major sources of exaggerated findings such as self-selection bias and regression to the mean. Academics usually err on the side of excessive caution, by only considering examples where random control trial rigor has been applied. So we have the strange situation where the federal government keeps concluding that PHI (though normally limited to disease management and sickness care savings) doesn’t work, while employers and commercial insurers regularly conclude that it does.

In PHI, if anything, we have too little market concentration. While many of its specialized providers are included in the usual merger & acquisition mania (including insurance plans that are acquiring PHI providers while at the same time operating their own PHI programs), there are still hundreds, if not thousands of providers, including many, perhaps most hospitals and integrated health systems. And retail clinics have lately joined in, at least for some kind of disease management services, adding hundreds more.

It seems likely that the evaluation of PHI efforts, and more systematic and rigorous evaluations of competing providers’ success rates, true benefit/cost ratios, and ROI ratios/amounts, will enable the kind of market concentration that will actually help. In the PHI market, survival of the fittest will only work if there is an effective way of identifying which providers truly are the most cost-effective for their clients.

Optimizing Incentives' Benefit/Cost in PHI

2010-02-23T12:16:00.000-08:00

There has long been, and no doubt there will continue to be disagreement about the use of incentives in managing people’s behavior – in pretty near all domains where attempts are made to modify such behavior for the better. Much like the present extremes of polarization in politics, there tend to be two diametrically opposed views on the subject – either incentives are wonderful and the most effective ways to modify behavior, or they are terrible and should not be used.

The truth is almost sure to lie somewhere between these extremes, closer to Aristotle’s “Golden Mean” than to either pole of the disagreement. There is simply too much evidence supporting contentions that incentives can work well for some people in some situations for some time to ignore. There is also the unavoidable reality that incentives automatically add to the costs of behavior change efforts, and do not work well for all people in all situations for all times.

The challenge should clearly be to identify and employ incentives optimally, combining intrinsic and extrinsic, since intrinsic cost so much less, and applying them in ways that deliver the greatest benefit at the least cost to whoever sponsors behavior change efforts. Why should they use sub-optimal approaches, or even worse, ineffective approaches.

One example of a significant effort aimed at optimizing the benefit/cost of incentives is that developed by HealthHonors, called Dynamic Intermittent Reinforcement (DIR). Its application so far seems to have focused primarily on achieving higher levels of adherence to medications, which enjoys wide support among providers, pharmaceutical firms, insurers, employers, and consumer health advocates, at least where such medications are proven to be competitively cost-effective. (www.healthhonors.com)

But the principles and logic of DIR clearly applies to virtually any behavior change plus sustainment challenge significant to health. Its aim is to identify and apply, and more importantly, to adjust over time, the most cost-effective use of incentives of all kinds possible. It does so through continuous customization and response-based modification of incentives in intermittent use. [M. Kalayoglu, et al. "An Intermittent Reinforcement Platform to Increase Adherence to Medications" American Journal of Pharmacy Benefits 1:2 2009 91-94]

B.F. Skinner was one of many behavioral scientists going back to Pavlov who favored intermittent vs. constant reward as both more effective and more efficient. People who are constantly rewarded for the same behavior will tend to “extinguish” that behavior significantly sooner than those who are rewarded only intermittently, once the optimal intervals or schedule is determined. And by definition, intermittent rewards need be used far less frequently than constant rewards, even when the level of the reward is the same, in order to achieve initial and continuing impact.

The biggest challenge in DIR is that it calls for total and continuous customization of rewards, where in the employment context, one-size fits all forever is by far more common. Even PHI providers who offer customized communications for participants typically fail to continuously update such communications based on how participants respond to them. Continuous customization of rewards can enable reduction in both the size and frequency of reward over time, and thereby significantly reduce incentive costs, while maintaining comparable effects.

To create a workable solution, DIR is adjusted according to continuous updating of the individual participant’s history of incentive use and behavior change response. In effect, each participant teaches the DIR system what is the optimal amount and schedule of rewards, adjusting both over time in order to optimize the cost-effectiveness of their use.

Financial incentives can be customized as to form, in terms of cash payments, health spending account contributions, premium or deductible/co-pay/co-insurance reductions, for example. Over time, they can be shifted from one to another form, in order to keep the impact of incentives from getting stale, rather than relying on increasing the size of the incentive to achieve the same effect.

DIR is equally capable of including and optimizing the use of intrinsic rewards, rather than depending entirely on extrinsic “points” leading to financial rewards. Individual participants’ behavior change can also be promoted through methods that increase their self-confidence, convenience, competitive self-esteem, sense of belonging, and similar intrinsic values, and these can be varied over time in similar fashion to financial or other extrinsic rewards.

Unlike most other behavior change methods, which rely on rare monitoring of effects, as rare as once a year when health risk assessments, screenings, or claims analysis reveals what has changed, DIR depends on constant monitoring, based on indicators such as self-reports, observation, or objective metrics that can be gauged efficiently. Such monitoring is the basis for the “Dynamic” nature of DIR, since it permits continuous adjustment along with customization of incentive usage.

This is unlikely to be the only option for finding a middle ground between total reliance on standardized incentives and total avoidance of extrinsic rewards, but the principles and logic of DIR at least represent a far more reasoned approach to coping with the issue than does arguing about it from polar positions that never change and are insensitive to evidence.

[Note: I have no connection with nor financial interest in HealthHonors – merely a strong interest in what they are learning and doing. For anyone interested enough in the excruciating details of DIR, there is a record of its patent application (“Behavior Modification with Intermittent Reward”) available online at: www.faqs.org/patents/app/20100015584]

Specialization vs. Generalization in Health Management

2010-02-08T11:20:00.002-08:00

The history of medicine, along with that of most professions, is one of movement from the general to the specific. Physicians and lawyers were once all generalists, and mainly trained through an apprentice system under already practicing counterparts. Eventually professional schools replaced mentoring, and gradually specialization has overtaken “general practice”. Among physicians, those once deemed “generalists” now prefer being recognized as “specialists” in primary medicine, and have their own domains with respect to pediatrics, geriatrics, vs. “family” medicine.

When medical organizations, including hospitals, take on proactive health management (PHM), they tend to reflect the already well-established preference for specialization. While the most proactive elements of health management deal with holistic challenges related to a host of “health behaviors”, medical health management solutions tend to focus on specific diseases and narrowly specialized approaches thereto.

A recent example is “Sereno. The Center for Snoring Solutions”, opened in January at the Crocker Galleria shopping mall in downtown San Francisco. It offers a package of services – initial evaluation, customized medical interventions and follow-up – for $3600, which is normally not covered by insurance. [C. Rauber “Sereno Opens Snoring Clinic at Crocker Galleria” San Francisco Business Times Feb 5, 2010 (sanfrancisco.bizjournals.com)]

The physician in charge is an ENT specialist who focuses on snoring and sleep apnea problems. There are clearly a lot of advantages in specialization, since it enables physicians and other practitioners to focus their learning more efficiently, as well as maintain their proficiency by repeating the diagnosis and treatment of similar, if not identical problems far more often than would be possible in a generalized practice.

But when it comes to PHM, specialization may suffer from being overly focused on single narrow problems, where people tend to have multiple, broad health “issues”. For example, the Baptist Health Foundation, Jacksonville FL, is using a $15,800 grant to study a 90-block area of downtown to identify the reasons for its population having higher levels of heart disease, HIV/AIDs and homicide than anywhere else in the city, along with highest rates of STDs, diabetes, and teen birth.

Its focus is definitely on the general problem, looking for ways to identify and correct obstacles to access and gaps in care, as well as better ways to prevent the worst chronic diseases. Among the hopes for the study and its subsequent responses are that overuse of ERs and delays in obtaining care can be reduced, while the health of the community is improved. [K. Morrison “Baptist Studies New Town Area’s Chronic Health Issues” Jacksonville Business Journal Feb 5, 2010 (Jacksonville.bizjournals.com)]

The Baptist study will employ quite a different approach to that of the Sereno center, using community surveys, focus groups and volunteers in the tradition of health planning that goes back to the 1960s to identify problems and enlist the community in solving them, as contrasted to a provider and market driven solution. This may or may not include medical care interventions, since the solutions may depend more on economic, environmental, and social changes, rather than from medical care.

The built-in advantages of taking the community approach include those of “social networking” in general. Members of the community may know more about their problems than any experts in specialized solutions, and their participation may well increase the acceptance of resulting solutions by the community.

In any case, the most efficient approaches to reducing the kinds of chronic disease problems being addressed, including endemic violence, relate primarily to altering behavior patterns in the population. Medicine is far too narrow in its focus, and not always the best avenue for changing behavior. And until a wholly new approach to compensating physicians for their effective efforts in changing patients’ health behaviors is devised and implemented, specialized medical solutions are unlikely to be the best approach to improving the health of the community

How Far Will Hospitals Go Toward Value-Based Operation?

2010-02-04T08:01:00.000-08:00

Having worked with and for hospitals most of my career, I am familiar with how slowly they change – in any direction. I enjoyed my opportunity to be present at and perhaps influence their adoption of marketing concepts and techniques beginning in the late 1970s, and to help a number of them move into proactive health management as a new service line added to their traditional reactive sickness treatment mission.

One of the changes many hospitals, along with a host of healthcare gurus are calling for is the move to “value-based payment”, one of a growing number of “value-based” (V-B) approaches, such as V-B benefit design, V-B prescription drug benefits, and V-B medical care. Unfortunately, the traditional definition of “value” that typically forms the foundation for hospital V-B efforts is that of quality for the cost. [D. Cutler “Move Toward a Value-Based System” HFMA Wants You to Know Feb 3, 2010 (www.hfma.org)]

Even including cost along with quality, i.e. moving toward cost-effectiveness rather than just effectiveness, and eventually comparative cost-effectiveness rather than championing all treatments that are cost-effective, is a major change for hospitals and most medical care. But a shift toward value should include a lot more than this. It should include choices between emphasizing prevention and proactive management of disease and injury, not just treatment thereof, in a comparative cost-effectiveness context.

There are really only two major areas in which hospitals have espoused and adopted proactive management, and both are decidedly limited in their potential. These are in the prevention of readmissions (estimated to happen in roughly 20% of all Medicare admissions, for example) – and in the management of chronic disease (which seems to continuously fail to meet government expectations as a cost-savings device.)

The recent move toward “medical homes”, for example, is decidedly first a medical shift in the treatment of disease, despite originating in the specialty of pediatrics, which is unique in its emphasis on prevention, patient health and development, rather than merely disease treatment. Medical practices devoted to health as much as to disease have developed mainly among often criticized “concierge medicine” practices, while medical homes focus mainly on managing diseases that already exist.

Hospitals have generally tended toward solving problems one at a time – the one or two problems that a given patient “presents” at the time of admission or an ER visit, for example, of the set of problems its medical staff, facilities and equipment are capable of dealing with. Even when moving into the proactive realm, such as in disease management, which is intended to reduce the incidence of crises, complications and worsening of disease, rather than merely treat them, hospitals have tended to approach their efforts one disease at a time.

By contrast, taking a holistic approach to patients’ health and development over time is rarely mentioned as a way hospitals can truly move into value-based operation. A number of hospitals have taken somewhat holistic approaches to managing health and diseases among special populations, such as frail elderly who would represent money losers if they have to be treated in the hospital. Many have adopted employee health management programs to improve their relationships with employers and increase their market share of admissions among their workforces.

But hospitals truly interested in value-based operation would consider proactive health management (PHM) alongside reactive sickness treatment in a comparative cost-effectiveness context. Where PHM is predicted and found to be more cost-effective than reactive treatment, hospitals would re-focus their efforts and investments toward reducing the incidence and prevalence of disease and injury in populations, and treat un-avoided sickness as a failure of the most value-based alternative.

While there may be many ways to “tweak” the current payment system and current sickness care focus of hospitals and medicine to save money. But unless it includes continuous consideration of proactive health alternatives to reactive sickness as a focus for hospitals and the entire medical care system, value-based operation will never be achieved by hospitals or the “healthcare” system as a whole, and our constant battle to control its costs will continue to be unsuccessful.

Improving Performance via Environmental Factors

2010-02-03T09:12:00.000-08:00

For me, at least, the challenge of improving workforce performance is best addressed through focusing on a combination of three factors: motivation, competence, and environment. In practice, however, most attention seems to be devoted to motivation, with competence commonly in second place, and environment more often overlooked.

The use of incentives of many kinds, including pay-for-performance (P4P), has long been suggested as the key to motivating improved performance. The principle that “The things that get rewarded get done” has been touted as The Greatest Management Principle in the World (by Michael LeBoeuf, G.P. Putnam 1985). Rewarding and punishing have always been major sources of power for managers of all kinds.

Education, training, and self-development have an equally long history of success in improving performance. When people feel that they have gained new levels of competence, and have confidence in their abilities, higher levels of “self-efficacy” are often sufficient to prompt improved performance by themselves. If motivation is increased, but competencies left as they were, workers may become frustrated, rather than more productive.

Managing the work environment in order to improve performance can be far less expensive than increasing motivation or competence. Simple reminders built in to the environment, such as pill bottles that make noises or light up when it is time for people to take their medications, have proven highly effective in promoting prescription drug compliance, and thereby the control of chronic diseases.

The work environment can directly promote or interfere with worker performance. Poor light, high noise levels, distractions of many kinds, noxious odors, overly high or low temperatures, etc. can significantly reduce output, as well as worker retention. Ergonomic engineering of workstations can reduce worker injury and promote performance.

Empowering workers to enjoy greater control over their own environment, particularly over the level and timing of demand for productivity and performance can yield major reductions in stress and its negative effects on health as well as employee value contributions. Personal autonomy is likely to be valued by most workers, and can promote both better performance and higher retention levels.

Multi-sensory environmental management has been shown to have significant impact on customer behavior and satisfaction, and thereby with sales, revenue and profit measures of customer value. It seems likely that it can have similar impact on employees, particularly if they are granted opportunities to participate in decisions regarding specific changes to be made.

Compared to the costs of increased incentives or competence improvements, environmental changes include a vast array of low-cost options, particularly those that employees may devise for themselves. While rewards are typically the first options considered for improving performance, their relatively high and continuing costs should prompt considerations of alternative steps that can make the work environment a more effective influence in the promotion of better health and performance.

Who Gets Paid for the Value They Deliver?

2010-02-02T11:45:00.000-08:00

One of the key challenges for organizations of all kinds is identifying, evaluating, acquiring, retaining, and getting the most out of their “talent”. Ideally, what talent means is the value that individuals can, do, and might deliver to the organization. And a key element in talent management is how much they are paid.

In most cases, talent is defined and assessed in terms of a combination of performance and potential, much as the value of customers is defined in terms of actual and potential revenue and profit contributions over the “customer lifetime” of each. A combination of capabilities, behaviors and values or attitudes typically determines whether the potential of talent is realized, and challenges organizations to figure out how to influence all three. [A. Orlowska “3…2…1… Talent Management” HumanResourcesIQ.com Jan 26, 2010]

In many businesses, sales people are the highest paid employees, often surpassing managers and even executives in compensation. The value of sales people is far easier to measure than is true for most other categories of employees, and the effects of their compensation packages on the organization’s performance are often promptly and clearly demonstrated. [T. Tolan “Why Sales People Are the Highest Paid Employees” FistfulofTalent.com Feb 1, 2010]

If anyone doubts the high value of the sales force, they can try becoming, or simply trying out the sales role, themselves, and see if they can deliver anything close to similar results. The business can test this value by putting a given sales person on leave and watching what happens to revenue from the territory thus abandoned.

But attributing all the revenue or profit that a given sales person delivers to that sales person is simplistic and unreliable. Consider the value that is delivered by workers who determine the quality and cost of the products and services being sold – do they not contribute as much value as the persons who only sell? Clearly, managers and executives will argue that they deliver great value, themselves, often many hundreds of times the value delivered by workers, or at least argue that they should be paid that much.

Is there not a significant difference in the value added by sales people who sell competitively inferior, even unreliable and poor-performing products and services compared to those who have superior, reliable, excellent options? On the other hand, sales people who sell by over-promising compared to what the business can deliver can create more harm than good in terms of performance over time. Is not the sales job a lot easier, and does it not add somewhat less value, when the product or service virtually “sells itself” through proven performance, guarantees, etc.?

I recall an example in my own career when a marketing campaign undertaken by my unit was shown to have brought in roughly $1.5 million in added contribution margin, not merely revenue. We were fortunate in having five of the twelve staff members in the unit eligible for year-end bonuses, but the raises we got that year were nothing close to reflecting this proven value.

By contrast, the staff of the new chest pain emergency center where the higher volume of patients were treated were not eligible for any bonus payments at all, in spite of their full responsibility for delivering the product so well marketed. And our “sales force” of representatives who solicited referrals and admissions from physicians were paid directly according to the numbers of patients and revenue they were credited with adding, despite the fact that virtually all the added chest pain patients came through either their own, their family members’ or their emergency medical technicians’ choices, not their physicians’.

If I were a sales person, though I am temperamentally as unsuited for such a job as anyone I can imagine, I am sure I would want to take credit for any and all sales I could. But wouldn’t it be a lot easier to sell a product or services whose performance I could guarantee, for example? And since I would have nothing to do with delivering what was guaranteed, how should my compensation be adjusted accordingly?

When the Safelite windshield repair/replacement firm initiated a pay-for-performance system for its installers, it achieved a 44% increase in productivity and revenue with only a 10% increase in compensation for the installers. Who shared in the other 34% gains? Did the “back-office” staff who supported the installers by making sure they had the right replacement windshields and equipment to install them share in terms of higher compensation?

Paying someone more to do something is about as simple a method for getting more output as is reducing the price for generating more sales. Should the managers or executives who came up with such an obvious ploy get the lion’s share of the gains that result? What if lowering the price yields a reduction in profits along with an increase in sales? Should the sales force be rewarded as if profits had increased? Should the managers and executives gain as well?

It is always easiest to give the greatest credit to those who seem to have the most direct connection to results achieved. But “direct” doesn’t necessarily mean “greatest”. Paying for performance, while clearly an effective way of improving performance to some degree in many situations, is by no means a sure-fire tactic, particularly in the long run. Paying for value added is clearly a better idea, though it requires far more understanding of both what value differences are made by whom, and how much of the desired impact will be achieved through sharing how much of the differences with those who made them.

The Need for Systems Thinking in PHI

2010-02-02T10:38:00.000-08:00

I had the opportunity to attend a week-long workshop on “systems dynamics” at the Massachusetts Institute of Technology in 1973, and it was probably the most useful educational experience I ever had. It addressed the complex dynamics of “systems” -- not vacuum cleaners and soaps labeled as “cleaning systems”, but real interdependent, interacting elements where the results of change are often different, even opposite what was intended.

The impact of PHI should be examined, predicted, and understood in terms of the system in which it applies, both the macro economic system that affects all organizations, and the micro organizational system in which efforts to improve health are applied. Otherwise, efforts to improve health, productivity and performance in the workplace, for example, may prove disastrous for both the employer and its employees.

This is perhaps best illustrated through the specific systems dynamics related to productivity and health. It is clear from countless published studies that health affects productivity, in negative ways when health is poor, and in positive ways when it is improved or good to excellent. But productivity can cause problems, not merely yield financial gains.

One of the built-in challenges in any effort aimed at increasing productivity is the risk that it will lead to layoffs among existing employees. If they “improve” their output significantly, say by 30 to 40%, as has been achieved by employers including Best Buy and Safelite Corporation, there is a significant risk that a corresponding number of employees will be laid off, since they will become superfluous, assuming that the market will yield a translation of the increased output into increased sales.

The mathematics of this risk is relatively simple. If output per employee increases by 40%, then sales will have to increase by 40% in order to absorb the increased output. If there is no increase in sales, then the workforce should be decreased to (100% divided by 140%) = 71.53%, or by 28.47% in order to match the workforce to sales. This could be accomplished by reducing everyone’s hours by 28.47% in order to maintain full employment, or by laying off 28.47% of the workforce, or by any combination of the two approaches that achieves the necessary match.

At Lincoln Electric Company, in East Cleveland, Ohio, systems dynamics have been used to maintain full employment while “absorbing” dramatically higher productivity gains over the years than the industry norm. This has been possible through passing on the cost savings achieved through improved productivity to customers, and thereby increasing market share and overall sales enough to ensure no layoffs are needed. (Hours have been cut there in the recent economic downturn, but “no layoffs” is still the company policy)

Lincoln guarantees that its solutions will save customers money, or it will pay the difference, itself. (www.lincolnelectric.com/knowledge/custsolutions/gcr/asp) By offering and delivering such savings, it has been able to increase its market share enough to ensure that its workers keep their jobs. But one reason it is successful at this is that it rewards employees for their overall performance, not just their output.

The performance criteria include both the quality of the products employees produce, but the continuous improvements in quality and efficiency achieved through employees’ “citizenship” contributions. The suggestions and feedback by employees, as well as by managers who have been trained to do the same work as employees, helps ensure the necessary market share to avoid layoffs. And since employees also share in the profits achieved by the company, they benefit in that way, in addition to their performance incentives. [“Working/Careers at Lincoln®” (www.lincolnelectric.com/corporate/career/openings.asp)]

In some situations, increased productivity can achieve even greater results through promoting the departure of employees. But this happens when the pay-for-performance system used to promote productivity also motivates low performers to leave, since they typically do worse under the P4P system than they used to, and worse than they can do for other employers who pay by the hour. Safelite, for example, found that turnover among low performers increased when it switched from hourly to performance pay, while turnover among high performers decreased. [E. Lazear “Performance Pay and Productivity” American Economic Review 190:5 Dec 2000 1346-1361]

The dynamics affecting employee health and performance are as complex as in most systems. A variety of factors, including employee health, development, compensation, and management all affect performance, and work in complex ways when having their effects. PHI must compete with other factors, and be applied in the context of these other factors in order to optimize its overall impact on the success of the organizations that use it, and on the individuals and populations that participate in it.

A Retail Health (R)Evolution in Health Management?

2010-02-01T13:16:00.000-08:00

When I began my graduate education forty years ago, the shift from “hospital” to “healthcare” and “health” administration was just beginning. Since the program where I pursued my studies was then located in the university’s school of public health (it has since shifted to the business school), there was a lot of emphasis on both the idea of health vs. sickness care, and on public, along with private efforts related to both.

But the students in the master’s program in hospital/health administration almost all pursued careers in hospitals, and at upper echelons of management in such institutions. And despite shifts in the naming of institutions from “Hospitals” to “Healthcare” or “Medical Care”, an effort to make them more broadly focused, their careers have almost invariably focused on sickness care, rather than on proactive or preventive health activities.

The same was true for me, as I wended my way first through academic pursuits, mainly teaching and consulting vs. research, then to “corporate” sickness care at two multi-hospital systems. But the call of proactive health returned in my most recent position, when my efforts to create strong working relationships with large employers in the system’s market led to the discovery that many of them were interested in improving the health of their workforces, not merely treating their sickness.

At the same time, there was clearly another shift in the market, from a focus on large “edifices” in downtown areas where sickness care was focused toward “retail” locations for services aimed at making them more convenient for patients to use. These were first primarily of the “urgent care” variety, with physicians setting up shop in locations convenient to where people worked, initially, and later to where they lived. In recent years, the emergence of nurse practitioners and physician assistants as staff in retail clinics has grown even more common than physician urgent care.

Another development that occurred at roughly the same time has been the emergence of proactive health along with reactive sickness care in retail locations. Concierge medical practices, numbering somewhere in the thousands so far, have separated proactive health services as a special focus and major justification for annual retainers of one or two thousand dollars, though combined with some mix of patient- and insurer-paid sickness care in almost all cases.

And gradually, retail clinics, usually those staffed by non-physician practitioners, have added proactive health services to their routine sickness care menus. Some offer only basic prevention, such as flu shots or school physical exams, while others include wellness programs aimed at preventing or managing common chronic diseases such as diabetes, obesity, etc. In general, there seems to be a move toward expanding the proactive health list of services, perhaps to increase overall volume and revenue.

At least one example has recently emerged where a retail clinic focuses exclusively on health vs. sickness care. Two “WellnessMart MD” locations are available in California, operating as “stores”, with a wide range of health-focused services, products and information at each, along with referrals to selected, low-cost providers of supportive diagnostic testing and scanning. [“A Retail Clinic for Healthy People” Wall Street Journal Health Blog Apr 4, 2008 (blogs.wsj.com/health)]

The other major movement in “retail health” is clearly the growth in a new form of worksite medical clinic that includes comprehensive wellness, risk and disease management services along with traditional sickness care. These clinics have expanded greatly the kinds of services they offer, as well as the people they serve, in many cases offering care to dependents of workers, employees of related and even simply nearby employers, as well as to their own workers.

Worksite clinics have to walk the thin line of what is acceptable to both employees and governments in terms of what the employer can know about individuals’ health. But employers are also in the best position to measure and evaluate what they gain from their investments, since they have or can access metrics reflecting the full range of value that improvements in employee health can yield. In at least one case, that of Quad/Graphics in Wisconsin, employers can even find new business opportunities as providers of worksite clinics to other employers, as does QuadMed.

Employers will continue to face strategic choices related to whether they invest in employee health benefits, including proactive health management. So far, most seem to be holding the line in continuing to invest, despite the economic downturn and uncertainty over health reform. In any case, it seems likely that retail health management, overall, will continue to grow in one form or another.

Social Networking to Promote PHI Participation

2009-12-24T09:39:00.000-08:00

Marketers are often being told that they should at least consider “social marketing” as an element of the communications tactics they can use with prospects and customers. Human resources professionals get similar advice with respect to their recruitment and retention goals. And social marketing would seem fit well with PHI goals and tactics, as long as they are planned and executed well.

While much of the emphasis in social marketing advice seems to be on hiring someone to help its prospective users, there are some simple, common sense ideas that apply, as well. One of the most important is the same idea that is essential to PHI, in general. Be careful in selection of prospects for social marketing efforts, in the same way that such care is essential in selecting prospects for PHI interventions.

The normal history of PHI interventions is one of gradual improvement, among both those who participate therein, and across the total population that can deliver the desired value to sponsors, as well as to themselves. Rarely do all members of that population join in PHI efforts from the outset – laggards are commonly large in number, as is the case with any “innovation adoption” pattern.

Selection for PHI often deliberately looks for those most likely to deliver success. This combines their risk/reward potential, because of how many or which chronic diseases or health risks each has, with their probability of succeeding in the PHI intervention intended for or expected of them. But in social marketing, both this reward x probability potential and the “contagion” impact of each participant are important.

Clearly, the best prospects for the first year of a PHI intervention would be people who are most likely to succeed, so that they will have success stories they can share with their peers. And those who have the most peers in their reference network, as well as the greatest likelihood of telling and being listened to by them, would be the best prospects among those most likely to succeed.

Those who are “early adopters”, in terms of the pattern of innovation actually tend to be opinion leaders, though some may be “isolates” who simply strive to distinguish themselves from the crowd, and would prefer that the crowd does not follow their example, since that would reduce their distinction. Understanding the social network that applies among workers can help significantly in making social networking more effective and efficient.

The science of sociometrics aims specifically to identify which individuals in a population have what role with how many and which others. Learning which play a “leadership” role, usually in the informal sense, is equally important. But such leaders cannot be accurately identified by asking prospects if they are such a leader; only asking peers to name others as leaders will reveal who has the most influence. [“Who’s the Leader?” Knowledge@Wharton 2009 (knowledge.wharton.upenn.edu/article.cfm?articleid=2170)]

Once the opinion leaders have been identified, those who are successful in their personal PHI pursuits can be encouraged to engage in social marketing. Merely asking them to act as sources of positive “rave/recommend/refer” comments to their peers has been found to dramatically increase their activities in this regard. Suggesting ways by which they can do so, including chat rooms, e-mail, website listings as “references”, as well as social media networks such as FaceBook, MySpace, Twitter, etc. may help them be more active.

By the same token, inviting prospects and members of the population at risk in general, to ask those who they respect can provide a “pull” accompaniment to the “push” from opinion leaders. Most who succeed will be proud of their accomplishments, and happy to brag about them, though some may be embarrassed or otherwise reticent about them.

While incentives are commonly used to promote participation in PHI efforts, offering incentives for people to engage in “buzz marketing” may be counter-productive. It may make potential opinion leaders uncomfortable about becoming “shills” for PHI, particularly where employers and insurers gain more from its success than do employees. It may also have negative impacts if it gets out that such leaders have become “paid agents” for the PHI provider or sponsor.

Selectively identifying and enlisting opinion leaders to help promote PHI participation will have to carefully avoid violating any applicable labor laws, privacy rules, and individual sensitivities, of course. But engaging the right people the right way in social networking can be highly influential in promoting participation. It can also help increase success rates as participants

Managers Can Add to or Subtract from PHI Success

2009-12-24T09:02:00.000-08:00

In most cases -- based on my experience, at least – managers are not directly and overtly opposed to PHI, worksite wellness, and similar efforts to improve the health of their workers, and thereby their performance. But they often seem to act in ways that are inimical to the ultimate aims of such efforts, though probably unaware of how their actions effect results.

One of my favorite examples involves managers who still rely on authority as the major, if not sole foundation for getting workers to do what is needed. When faced with a problem, they simply tell their employee or employees: “Fix it, just fix it, and right away!” Of course, they may have been given a similar order by their boss, so it may seem natural to simply pass it on.

But in many cases, the problem is not caused by the employees addressed, nor is the solution under their control. In fact, in many cases, the manager, or the manager’s bosses may have caused the problem, and the employee is powerless to solve it. But managers seem to think they add something to the process if they gruffly order their employees to find a solution and implement it.

By contrast, there are managers who will at least ask the employees to whom responsibility for the solution is being delegated: “What can I do to help you solve this problem?”. Reflecting a view that it is at least a joint problem can greatly reduce the stress that managers delegate along with their orders. And actually joining in to develop the solution will further reduce stress, and the negative impact this often has on employees’ health and performance.

Managers who are proud that “I don’t suffer from stress, I give stress” may not even know how much they are detracting from performance, rather than ensuring it. Until they recognize that they may well be a significant part of the PHI problem, rather than part of the solution, it can be expected that PHI results will be significantly less than optimal.

Another way that managers, or “management” in general, can undermine the potential of PHI and other methods being used to improve employee, and thereby organizational performance, is by rigidly, or selectively enforcing rules and policies, rather than treat minor transgressions as problems to solve. In a recent example, Fidelity Investments in Texas fired four employees for violating “company policy” with respect to gambling, for engaging in a Fantasy Football game. [“Fidelity and Fantasy Football – Here’s 10 More Stupid Reasons to Fire Employees” Great Leadership Dec 20, 2009 (www.greatleadershipbydan.com)]

Having a policy that prohibits employee gambling may well make sense, for an investment firm in particular, but firing employees with no warning or second chance seems a bit Draconian – for them as well as for the organization. It will presumably have to replace them and wait for the replacements to reach the same level of value as those who were fired, incurring significant productivity and performance losses during the interim, as well as dampening morale among “surviving” staff.

Allegedly, nine other participants in the game were simply given warnings, and kept their jobs, including a manager. While one of the fired employees was the “commissioner” of the “league”, i.e. had a “leadership role” therein, there seemed to be no logical basis for selectively firing the four who did not “survive”. And legal definitions of “gambling” apparently exclude fantasy football games.

It might even be that PHI incentives that recognize winners of bonus challenges, such as for losing the most weight, are “gambling” in a similar sense that the football game enables. It might also be the case that players of the football game take time away from productive work in their pursuits. But so does firing them. Other employers have recognized the added social and working relationship value of employees engaging in e-mail and other communications unrelated to work, however.

It is certainly a 19th century idea that anyone who fails to obey company rules and policies can and should be fired. In the old “command and control” model of management, this was one of the major sources of managerial power and influence. But this is the 21st century, and this old model has been largely discredited, though clearly not abandoned.

If managers see such actions, or similar activities, as demonstrating or causing a significant problem, it would probably make more sense for them to identify that problem and present it to the “offenders” as one needing solving. Adopting the familiar negotiation tactic of having managers and employees (either literally or figuratively) sit on the same side of the table and work out a solution to such a modest problem would seem to be a far better approach.

At a minimum, it might be wise for managers to ask themselves: “Will how I approach this problem add to as opposed to reduce or eliminate the damage involved and value possible through a solution?” And if there is a clear risk that damage may be caused or value added by the solution, perhaps inviting employees to help solve it would be a better political approach, and one that would add value to the manager/employee relationship, as well as to the organization’s performance.

Measuring Productivity, Performance and Value: Political Issues

2009-12-22T10:18:00.000-08:00

There is a vast amount of information available in government reports, academic research, articles and books on the subject of measuring productivity and performance. Both have been major challenges in business at least throughout the modern industrial era. And most discussions of this challenge focus on the technical aspects of gauging the output and value that employees contribute to their organizations in return for their wages and salaries.

But, as I still recall being drilled into my head in graduate school courses in both public and health administration, the effectiveness of such measurement depend as much on their political aspects as on their “scientific validity and reliability”. Both labor and management, or all parties in organizations where there is little distinction between the two, must accept, and ideally support and be motivated by the way in which worker value is measured.

Measuring workforce value varies widely in terms of its technical difficulty. People who have both clear individual or team outputs, or measured results (e.g. sales revenue, widgets produced) and simple measures of the value thereof (e.g. revenue vs. costs) face less difficulty than those involved in “knowledge work”, complex services, and other efforts which tend to represent the majority of cases.

Given the political nature of measuring employee value, it makes sense to apply a political process thereto. The W.L. Gore firm, for example, which operates in a horizontal “latticework” model, rather than a hierarchy, gathers relevant peers together annually for a discussion of the value that each has contributed in the past year. The consensus reached becomes the basis for compensation, though the operational performance of the entire organization will naturally dictate what is available to spend thereon.

SEMCO in Sao Paulo, Brazil uses a negotiation process, where individuals propose how much they will accomplish during the year, and how much they expect to be paid for it. Negotiations among peers eventually reach an agreement, which may vary widely and frequently across different work assignments (both tasks and workstations change frequently).

The Transtec Group uses a more complex process, though it has simplified its performance rating system from an original list of seven dimensions to just three. It relies on its knowledge workers to not only manage but innovate new approaches to highway and runway construction. Workers are rated on a simple scale where “C” = “needs improvement”, “B” = “okay”, “A” = “good”, and “A+” = “Exceeds expectations. Ratings get translated into bonus payments based on 25% of the company’s profits.

The more that the work people do fails to fit traditional notions of productivity (e.g. “piecework”) or even performance where quality, customer satisfaction, and other measurable dimensions can be applied, the more the political dimensions of measurement apply. The use of consensus approaches such as 360o rating discussions and negotiations can handle most challenges, including those where objective measurements contribute major inputs into the process.

Employees at the Lincoln Electric Company in Cleveland have a more objective approach, with each worker’s output measured as machines that pass customer evaluations. But “performance” also includes “citizenship” measures such as contributions to efficiency and cost reduction, since these pay off in terms of more customers and business volume. Fully half of workers’ annual income normally comes from performance bonuses.

There have been many attempts to calculate or at least estimate the value contributed by employees, usually applying a far more stringent and objective approach than is used for managers and executives. I have read sources that claim the ratio of workers’ value to their annual compensation as at least 2:1. One author asserted a range of 3:1 to 5:1. {J. Leutzinger “Return on Investment: How to Determine an ROI and How to Get There – Healthy Habits at WIC” Health Improvement Solutions Oct 5, 2008 (www.calwic.org/docs/wellness/leutzinger.pdf)

Regardless of what the average ratio may be, employers and employees alike need to operate within both the technical and political realities that affect workforce value measurement. Consensus methods offer an application of “the wisdom of crowds” as well as a politically acceptable approach to measuring productivity, performance, and value that will probably work far better than the host of arbitrary and onerous performance appraisal methods commonly in use.

Personal Health Budgets in PHI?

2009-12-22T09:23:00.000-08:00

While it is common practice in the U.S. to criticize the National Health Service in the U.K., there is one intriguing Idea that the NHS is testing that might fit well with payer-sponsored population health management strategies, including the PHI approach. It is the development and use of “personal health budgets” (PHBs) which NHS patients are being permitted to at least partially control in their efforts to manage their healthcare needs.

The concept is based on the notion that consumers, as patients, have their own interests and preferences when it comes to spending for their health and illness problems. If they are empowered to manage expenditures related to illness, their experiences may be more prudent, effective, and satisfying than when either the government or care providers control expenditures.

In the NHS approach, PHBs are set based on a combination of inputs, from Primary Care Trusts to which NHS funds are assigned, the providers that will be involved in caring for patients, and the patients, themselves, as long as they are not demonstrably unqualified to do so. Pilot tests of this concept. [“Personal Health Budgets” NHS Update December 2009 (www.dh.gov.uk)]

One subset of pilots is testing the concept of enabling patients to pay directly for their care, including care rendered by alternative care providers who are not members of the NHS system. Patients would have special bank accounts into which NHS money would be deposited, and from which patients could disperse funds to pay for services that are included in an approved care plan. [“Direct Payments for Health Care” A Consultation on Proposals for Regulations and Guidelines” NHS Dec 2009 (www.dh.gov.uk)

While these pilots will test the efficacy and efficiency of sickness care managed through PHBs and direct payment accounts, the idea has clear applicability to PHI. They could be significant steps in promoting personal responsibility for one’s health, in addition to that for sickness care. The challenge, of course, would be to ensure that the amounts allocated to each PHB, and the manner in which the funds therein are allocated, make optimal sense.

Because PHI efforts have almost invariably been sponsored by insurers or employers, these payers naturally have chosen to control them completely. Some employers that I am aware of, and perhaps some insurers that I am not aware of, may offer participants the “right” to select their own health goals, or at least the health risk or disease they hope to manage, but then control the types and amounts of resources and support that will be allocated to help them.

By contrast, a true personal health management budget could be logically developed in ways similar to those in the NHS example, with payers, providers, and participants all having some input. Payers would want to be sure that the amounts allocated are well within the range or the risk/reward potential and probability of success estimated for each participant. Modern predictive models are clearly capable of producing such estimates via automated computer programs.

Ideally, predictions would be adjusted based on whatever “solutions” are selected by participants plus whatever providers of PHI services are involved. After all, the estimated probability and value of success would vary according to the solution applied, as well as the ROI predicted therefrom, since the costs of different solutions would vary widely.

Dispersals from these budgets could be handled entirely by the sponsors, i.e. payers involved, or they could be deposited in a limited use account, as is the option in the NHS version. If the theory is correct, participants who control more of the resources involved will be more likely to manage both their health and the expenditures well, more often succeeding than in cases where sponsors have all the control.

As is true for the NHS, it makes sense to at least test the idea, in order to learn if it behaves as well as it sounds. A process for developing PHM budgets would have to be developed and tested, along with the processes through which participants would exert some control over their expenditures. The federal government has already tested this idea in the management of social and support services, and applying it to PHM would, in most cases, require far less money being at risk.

Will HCOs Take "The Road Less Traveled By"?

2009-12-21T10:02:00.001-08:00

Since we are in the last year of those beginning with 200-, (though not the end of the decade, which is at the end of 2010) there have been lots of articles looking forward to what health care organizations will see and do in the immediate future. While dozens of different ideas have been put forth, there have been few suggestions that HCOs truly and vigorously invest in “health”; virtually all relate to “sickness”.

Having been born and raised in New England, I recall a lot that Robert Frost wrote, including his poem “The Road Not Taken”, written in 1920. He described how he chose the “…one (road) less traveled by, and that has made all the difference.” HCOs have a similar option today, though not as clearly and irrevocably distinct as Frost’s options.

My guess (because I have found no report on the actual number) is that hundreds, perhaps thousands of hospitals and integrated health systems have invested in “proactive health management” (PHM)– for their own employees, employer markets, or both. Major systems, such as Cleveland and Mayo Clinics have done both, and Mayo, for example, serves an international employer market.

It is an inevitable reality that “reactive sickness care”, the other side of the PHM coin, has been the dominant business for HCOs since Benjamin Franklin helped create the first US hospital in Philadelphia (about 1751 as I recall, though I wasn’t there at the time). And an accompanying reality is that PHM competes with sickness care for attention and resources, and conflicts with it as a market, since PHM deliberately aims to reduce the incidence and prevalence of sickness (including injury and illness).

Hospitals and systems are already investing in efforts to reduce “never events”, for which they will not likely be paid, as in their best interests. Many are investing in preventing re-hospitalizations, which have been declining for some time, since these will likely not be paid for as well. [J. Simmons “Potentially Preventable Hospitalizations Declining for Older Adults” HealthLeaders Media News Dec 18, 2009 (www.healthleadersmedia.com)]

If we consider the label “potentially preventable hospitalizations” in its full meaning, it includes much more than re-hospitalizations of recently discharged patients. Many hospitals have already invested in home care programs aimed at keeping frail elderly and severely chronically ill patients out of their facilities. Such prevention typically has the advantage of avoiding hospitalizations that are not profitable for the hospital, when DRG payments are too low, or Medicare sticks to its parsimonious payments.

But other “preventable” hospitalizations include a growing number that payers are deciding could have been handled on an outpatient basis. [“Hospitals Face Admissions Denials for Stents, But May Be Able to Prove Medical Necessity” AIS’s Health Business Daily Dec 21, 2009 (www.aishealth.com)] While payers’ decisions may lean toward the arbitrary and capricious regarding which are potentially ambulatory care cases, it is certain that payers will stretch the definition of “potentially ambulatory” care as far as they can, as they will for “potentially preventable”.

Already, there is a label in use for admissions that could have been prevented through good primary care, though the prevention clearly depends on patients doing their parts in such are, such as actually going to a physician for it, not merely what physicians and hospitals might do. Payers seem likely to adopt a similarly stringent payment policy toward such admissions, if they have not already.

But the ultimate gain for payers will surely come from admissions, even ambulatory care episodes, that “could have been prevented” through a combination of wellness, health risk, and chronic disease management. Estimates have been suggested that over half of all hospital admissions might fall into this category. What if they became “never events” that payers would simply deny payment for?

CMS has already calculated that “Prevention and Health Promotion Could Save Medicare $1.4 Trillion Over 10 Years” (L. Masterson in HealthLeaders News July 30, 2009 (healthplans.hcpro.com)] Similar savings are clearly available through worksite wellness and employee health management efforts, along with gains in productivity and performance that employers and their suppliers/insurers are eager to achieve.

If hospitals and health systems devoted as much time, energy, thought, and resources to proactive health as they do to reactive sickness, they could no doubt make significant inroads in the direction of reducing our sickness care burden in this century. And since we clearly cannot, or at least will not, be willing and able to pay for all the sickness predicted glowingly for the rest of this century, it might be in HCOs’ best interests to join the movement toward reducing it.

Doctors are known to be the most powerful source of motivation and understanding of how and why to adopt healthier behaviors among their patients. Primary physicians are already moving in that direction through “Medical Home” demonstrations, and as they gain increased payment for keeping patients out of hospitals, they should be helpful allies in such efforts. Hospitals are already linking or sponsoring retail clinics, which are increasingly adding proactive health services to their value propositions. What they learn and can demonstrate through their own proactive health efforts with their own employees should serve them well in expanding their role.

Perhaps the time has come for HCOs to consider the paths they might take for the rest of the century – to persist in holding on to as much of the sick care market as they can, or to join with pretty much all other stakeholders in the “health system” to reduce the size and costs of that market. Most have already taken tentative steps, and it is certain that committing to that less traveled road will make all the difference -- to themselves, and to everybody else.

New Sites and Sources for Training in PHI

2009-12-19T09:43:00.000-08:00

It seems finally to have dawned on the clinical training community that there are new locations increasingly important for training during and after clinical education. Given the already significant and growing shortages of clinicians who are trained specifically in primary care, particularly in prevention and proactive health management (PHM), new sites and support are clearly needed.

Current clinical training, particularly for physicians, is almost completely limited to hospitals, particularly large, urban medical centers that qualify as academic medical centers or teaching hospitals. But as primary care becomes devoted increasingly to ambulatory sickness care and health plus chronic disease management, in medical and health “homes”, hospitals are increasingly too narrow as sources of essential training experiences.

Community health centers have been specifically suggested as site for primary care medical residencies. [F. Lawry “Training Primary Care Physicians in Community Health Centers May Help Solve Workforce Crisis” Medscape Medical News Dec 18, 2009 (www.medscape.com/viewarticle/714066)] A program based on community health centers would help them care for the growing number of patients who depend on them for care, as well as help increase the number of physicians trained in primary care vs. other specialties.

But physicians are likely to be effective mainly as leaders of health management teams, rather than as sole providers of wellness, risk and disease management. They are over-educated in sickness care, and usually prefer to maintain a major focus on “primary” levels of sickness, along with wellness efforts. They are also too highly compensated to enable health management to be efficient in most cases.
One exception is the large and growing number (some have estimated 5000) of “concierge physicians”. While these are typically limited to patients willing and able to pay at least $1000-2000 as an annual retainer, to cover proactive health services, as well as special amenities, there are many that offer wellness services for significantly less. Procter & Gamble recently bought the largest organization of concierge practices, MDVIP (www.mdvip.com) to complement its already huge health & wellness business line.

New physicians’ huge debt after medical school may well motivate many to join the National Health Service Corps, which will pay up to $50,000 of their debt if they spend two years in underserved areas. This may increase the numbers willing to pursue primary care careers, as well as work in Community Health Centers. Other clinicians, including nurses and dieticians/nutritionists, as well as physical therapists, professional trainers, and others with strong wellness credentials, may use the same sites for training.

But PHM is adding to the numbers of places where it is being practiced. Retail clinics staffed by nurse practitioners and physician assistants are increasing the range of proactive services they offer, and making PHM a more important part of the retail clinic service line. The RediClinic chain, for example, combines “Get Healthy” routine sickness care services with “Live Healthy” PHM services (www.rediclinic.com)

The best opportunities for primary care clinicians may be in the growing number of worksite medical clinics that offer a similar mix of sickness and health services. These typically create a two-pronged source of savings for both employers and employees by offering lower-cost (than urgent care, emergency room, or traditional physician practices) care, as well as saving lost time from work in pursuit of such care.

When they function as PHM sources as well, worksite clinics can save employers even more, and serve as valued sources of positive ROI through reducing health-related productivity/performance impairment. They typically improve employee recruitment and retention success, and save expenses that was as well. And many have been able to yield “positive presenteeism” effects through more healthy, happy, energetic and engaged workforces as well.

The more that clinicians of all kinds, as well as non-clinical health professionals, get at least part of their training in clinics and practices that feature, perhaps even focus on PHM services rather than limited sickness care, the better the entire “health system” will be prepared to achieve the only result that will ultimately ensure its survival. Unless the incidence and prevalence of sickness and injury are significantly, ideally dramatically reduced, we will simply not be able to afford paying for it, regardless of how much sickness care is “reformed”.

Moreover, while sickness care is a cost center for employers, and a major, incessantly growing one at that, as well as for government, the stress will invariably be on reducing what is paid to providers of it. By contrast, PHM is a cost-saving center, and has even been shown to be a revenue-enhancing center as happier/healthier employees improve customer satisfaction/loyalty and new business. This can enable providers of PHM success to share in what employers gain through proactive investments, rather than groveling for what payers are willing to spare for sickness care.

Productivity, Performance, or Value in PHI?

2009-12-14T11:31:00.000-08:00

We have come a long way since the two separate and unequal domains of worksite wellness and chronic disease management began 30 of 40 years ago. We have at least begun, and perhaps made some significant progress toward integrating the two domains, and combining with other domains as well to move toward an integrated approach that goes beyond health alone. Another of the major shifts has been the expansion of what kinds of results are being sought and measured – sickness care costs and perhaps some worker morale in the early days, contrasted to a wide range of added value parameters today.

Originally, it made logical, if slightly lazy sense, to focus on measures that were already being monitored and reported: hospital/medical, workers compensation and disability insurance expenses. These were then, and are even more so today, major operating expenses, and subject to some degree of “management”, at least. But when expanded metrics were sought, serious problems with measuring them appeared immediately.

While measuring productivity was once far more common and simpler, based on recorded “piece work” output of craft and manufacturing workers, measuring the productivity of “knowledge workers” is acknowledged to be problematic. When workers operate in teams and on different shifts, as in hospitals, for example, it is a challenge to determine who exactly “produced” whatever outcomes are gauged.

Moreover, measured productivity has always been a two-edged sword. It enables “pay for output” motivation schemes, but may also anger workers if quotas are set too high before bonus levels are reached, or if no pay is given for sick and absent workers, for example. Professionals such as doctors and nurses have tended to prefer being paid for the quality of their work over its quantity.

Productivity can also become the wrong thing to pay for, if it motivates workers to ignore or give short shrift to quality and customer satisfaction. When the Safelite windshield repair chain adopted a pay for output scheme roughly ten years ago, it found that the number of disgruntled customers and amount of sloppy work increased. Only by ensuring those responsible for sloppy work had to repeat the job themselves, with no added pay, and that customer satisfaction standards were enforced, with bonuses for high levels thereof, did the company overcome this problem.

Performance is, in some ways, far simpler to measure than is productivity, but it is also far easier to measure it inaccurately. Annual “performance reviews” are notoriously unpopular – with both managers and managees. They rarely provide anything like a credible and timely measure that can be used in pay-for-performance systems to good effect. Given the importance of matching incentives to performance and rewarding workers as close to the time they achieve such performance, annual reviews are all but useless in PHI.

Normally, performance requires many separate measures, such as customer satisfaction, work quality, output, etc. – and these can compete or even conflict with each other. Moreover, while the value of goods produced or calls handled, sales delivered, and similar productivity measures is usually relatively easy to gauge, the combined worth of diverse performance measures is rarely anywhere near as simple to translate into a single metric.

Performance is likely to be a more popular measure to use than is productivity, simply because it incorporates quality, customer satisfaction, perhaps value delivered to customers, which can translate into meaningful “making a difference” measures for workers. But combining multiple measures into a single performance rating, and having that rating be accepted by both manager and worker, as well as yielding results that the organization prizes, can be an enormous challenge.

Focusing on value, the worth that individual employees, teams, units, etc. contribute to the organization would be the ideal approach to evaluating PHI results, along with almost any other Human Resources efforts. But it could also be the most dangerous. If it were possible to calculate the true “value added” by workers, their “lifetime worker value” in the same way that “customer lifetime value” is estimated and calculated, the information might create a revolution.

There is a well-known and common problem with respect to individuals' assessments of their worth and objective measures thereof. Higher-performing individuals tend to rate their performance somewhat lower, on average, than objective measures, while low-performers rate theirs higher than "reality". And managers, of course, may range all over the map in terms of how well they appreciate the worth of individual workers, in addition to being biased in terms of their ratings of workers' vs. managers' and executives' value.

Some organizations seem to assess worker value better than others. SEMCO, the innovative manufacturing firm in Sao Paulo Brazil, empowers its employees to set their own standards for output and performance, as well as to set a proposed value on achieving such standards. This value is their proposed compensation package, and is negotiated individually for each worker each year.

Negotiation is one way to reach an agreed-upon value for individual workers, or teams, for that matter. Teams could then be empowered to agree on their own approach for dividing their team-based compensation, and they might use productivity or performance in doing so. It has the advantage of enabling acceptance through the negotiation process, particularly if an a “Getting to Yes” approach is used.

But agreement does not necessarily mean that the value amount negotiated will truly satisfy either management or labor, nor that it will fit well with the organization’s overall performance goals and shareholder expectations. On the other hand, it seems to work well at SEMCO, and it has the advantage of being one of the most empowering approaches to deciding how and how much to pay workers.

As such, it offers a value of its own to workers – one that may be more powerful in terms of recruiting and retaining talent than financial compensation alone. (SEMCO has very high retention.) In any case, any one of the three options – productivity, performance, value – may be the focus for PHI strategies and interventions. As we develop our models for PHI, and particularly as we engage in comparative cost-effectiveness evaluations of how we perform it, we should also gradually learn which of the three works best in terms of continuously improving the productivity, performance and value of workforces and their organizations.

Inflation in PHI Results Reporting

2009-12-14T10:50:00.000-08:00

Just as “grade inflation” has tended to shift more and more grades in school toward the “A” level, without reflecting any real improvement in student learning, so “results inflation” is a common problem and practice in PHI. Both PHI providers and their clients may engage in this practice, consciously or unconsciously. Both have an interest in believing and persuading others that they have succeeded, and both may use what is called “motivated reasoning” to select and interpret result information in order to ensure a positive result will be reported.

In one example, I found a brief report in Employee Assistance and Wellness News (www.healthrespubs.com/News/wellness.html) with the astounding headline: “New Wellness Study: Employees with Balanced Diets Have 10 Times More Energy” (July 6, 2009).

It argued that “Energy is a key factor in productivity.” – and reported results of a survey of more than 1000 employees working for a wide range of employers in varied industries. With regard to the headline of the article, it reported that: “… only 5 percent of employees with unbalanced diets had high levels of energy, while 50 percent of workers with balanced energy had high energy.”

For one thing, this finding is based on employee self-reported energy levels, not on any objective measure of their actual energy. For another, there is no automatic translation possible from high energy reports to high productivity or performance value to their employers. Workers might spend all that added energy in wasteful pursuits, rather than productive work.

But even more questionable is the translation of the apparent fact that 50% of workers who reported balanced diets also reported high energy vs. only 5% of those with unbalanced diets. For one thing, the “balanced/unbalanced” diet measures clearly indicates a continuum in fact, though it seems to have been measured and reported as a binomial or yes/no parameter.

Some of the workers who reported an “unbalanced” diet most likely followed a diet very similar to others who reported a “balanced one”. There are a host of metrics in health that use what is often an arbitrary point on a continuum, such as blood pressure, cholesterol or sugar, BMI, waist measurement, etc. as the “boundary” between healthy vs. unhealthy, despite the tiny differences between the two when people come out close to the boundary.

There is no indication that the diet followed by those reporting theirs is “unhealthy” is twice, five or ten times as unhealthy as the diet followed by those reporting theirs as “healthy”. No degree of difference can be assumes, only some reported their diets as healthy, while others chose unhealthy – and that these same people reported their energy levels as high or “not high”, at far different rates.

Moreover, the actual findings were that: “Ten times as many workers reporting their diets as “healthy” also reported their energy as “high” as did those who reported their diets as “unhealthy”. This is entirely different from a finding that employees with healthy diets had “ten times as much energy”. The amount by which the “high energy” workers exceeded the energy levels of the “not high energy” workers is completely unknown.

While it may be understandable that PHI providers or clients report results in ways that exaggerate the extent and value of their interventions, it is not a good idea to let such reports be accepted at face value. An editor with a decent understanding of mathematics and a sense of duty to honest reporting might well have revised the headline to state only what really was found.

In such a case, the headline would have read that “Employees Reporting Balanced Diets Ten Times More Likely to Report Having High Energy”. This headline only applies to this particular survey, of course, and may imply a far more general reality, but it would at least reflect what was actually measured.

Retail Clinics Becoming Major Force in PHI

2009-12-07T20:39:00.000-08:00

The field of population health management was originally dominated first by employers, then came specialized provider organizations that focused typically on chronic disease management, later insurance plans and traditional healthcare providers, including hospitals, integrated systems, and physician practices joined in.

Meanwhile, retail clinics featuring nurse practitioners or physician assistants staffing emerged most recently, and have, themselves, climbed on board the PHI “bandwagon”. Most of these started out offering only basic preventive services, such as flu shots and school physicals. But recently, more are adding in disease management and wellness coaching, joining with the growing number of worksite medical clinics as a new source of PHI services.

In addition, many are joining with insurance plans and traditional healthcare organizations (HCOs) in such efforts, as well as interfacing with them for routine sickness care. They are creating interfaced electronic medical record (EMR) systems, for example, so that patients may move seamlessly among these different providers.

CVS Caremark‘s Minute Clinics, for example, have aligned with Allina Hospitals and Clinics of Minnesota and Western Wisconsin. As hospitals face demands that they reduce re-admissions, lest they become unpaid “never events”, and take on greater responsibilities for episode of care management, such alliances make good sense. Retail clinics need back up for complex cases they may see first, as well as gain added visits for ongoing disease and risk condition management. [“CVS, Allina Align Clinics to Get More from a Minute” DrugStoreNews.com Nov 16, 2009]

Hospitals are seriously handicapped by their ages old “edifice complex”, where their locations are typically few and concentrated in urban centers, while people looking for care may prefer closer and more convenient locations, such as the pharmacies, grocery and super stores where free parking and other places to shop make them attractive.

Insurance plans and employers often prefer that their employees to retail clinics, where they can get care faster and return earlier to work, and where costs are typically significantly lower than physician offices, and dramatically lower than emergency rooms. Integrated or at least coordinated providers can prove a win-win all around.

Minute Clinic and Humana are collaborating to improve the insurance plan members’ access to care, including management of routine illnesses and risk factors for chronic diseases. Humana will also open a clinic in its own headquarters facility in Louisville, KY, which will be managed by Minute Clinic and Humana’s health and well-being subsidiary LifeSynch. [“Minute Clinic, Humana Expand Partnership’s Focus on Chronic Illness Management” DrugStoreNews.com Dec 2, 2009]

Minute Clinics will become sources of prevention and wellness services for Humana members at roughly 500 locations nationwide. They have been official in-network providers for Humana members since 2006. The expanded relationship will make them eligible for health risk screening, while taking advantage of Humana’s health coaching services offered by LifeSynch.

As with Allina, Humana and Minute Clinic will interface EMR systems, and have indicated they will explore the development of new services designed go further enhance member health and wellness. Retail clinics in other chains and locations are also moving in a similar direction, as are worksite medical clinics.
Retail clinics already overlap with worksite medical clinics, and serve effectively as worksite clinics for workers at large pharmacies, grocery stores and super stores where they are located. The combination of retail and worksite clinics should go far to make up for the already serious shortage of primary physicians, as well as adding to the number and convenient locations of sources of wellness, risk and disease management services needed for effective health care reform

A Non-Health Approach to Improving Performance

2009-12-07T20:04:00.000-08:00

While lots of people in the wellness, disease and health risk management professions have argued for years or decades what is the best route to take, there is at least one voice that advocates taking an indirect approach. Instead of addressing health, explicitly, why not use easily altered business practices and financial incentives so that employees, themselves, manage their own health?

Wendy Lynch, PhD has posted a series of fascinating discussions, many with statistical backing, advocating the use of incentives alone to improve employee health, or at least to achieve the same results as such an improvement would yield. This could save a lot of money otherwise being invested in attempts to improve employee health behaviors and status, and thereby save employers money.

Her most recent posting (“Reduce Absence, Improve Productivity: Aligned Incentives Are a Simple Formula” Health as Human Capital Dec 6, 2009 (hhcf.blogspot.com) is the latest example. It describes an effort by a manager newly challenged to improve the performance of the Accounts Receivable (AR) department, where the only "health" problem was a high rate of worker absences.

The performance problems seemed far more related to how employees were managed than how healthy they were. There were no incentives for good or even adequate performance, everyone got the same pay, and usually nothing but negative, critical feedback. The total amounts collected by the AR staff were well below industry standards and expectations, and absence rates in the department were higher than normal.

The new manager set standards for acceptable, high, and exceptional performance, with bonuses of 12-15% for doing better than acceptable. Feedback on performance was measured and reported daily, and bonuses were paid at the end of every month. Performance improved virtually from the first day of the new system.

Staff attitudes and energy improved, and two low-performing members of the department left for other jobs. Collection amounts increased by 35%, while the rate of noncollectable accounts decreased from 12% to 3%. The AR department gained positive recognition from corporate leadership in addition to bonuses.

Moreover, where the average absences per month for the department had been 19 days per month (1.5 days per person), it dropped to only 6 days (0.5 per person), without any changes in absence policies and practices, or any investment in employee health improvement. Of course, there is no way of telling, in most cases, how many “sick days” reflect true disease or injury vs. low motivation, for example, so it cannot be concluded that the new management approach actually improved their health.

There have certainly been many other examples of equally dramatic improvements in performance as the result of better management. Safelite windshield repair increased its productivity by 44% in the first year of operating under a pay-for-performance system. Best Buy corporate office achieved a 36% increase through its “Results-Oriented Work Environment, enabling workers to choose when and where they worked, along with ending worker turnover during its first year.

SEMCO in Sao Paulo, Brazil achieved wonders in worker performance and low turnover by initiating an employer-managed work environment, where workers choose their own performance standards and compensation combination, change workstations and assignments regularly, and even “retire early” by taking one day off a week to practice for what they will do in retirement.

Clearly, there are indirect routes to achievement of the kinds of work performance results that PHI seeks to achieve through health improvement. On the other hand, these indirect routes are not in conflict, only in competition with PHI investments. The healthier workers are, the more they should be able to enjoy their lives while they work, as well as their retirement, when improved health yields improved performance and financial rewards.

There should be no argument over which is the best approach, in general. Each situation represents a unique challenge for testing and choosing the best option or combination of the two. As long as better management of personnel as well as their health can both yield positive ROI, employers should be willing to do both.

Random Variation in PHI

2009-12-07T19:27:00.000-08:00

My statistics professor in graduate school used to have a sign in his office advocating: “Help Stamp Out Random Variation”. This tongue-in-cheek message reminded all of us students of the problem that plagued efforts to identify useful trends, causes of phenomena, solutions to problems, etc. based on statistics.

Stamping out random variation, or at least its effects, has long been a holy grail for clinical studies. The “gold standard” of random control trials are intended to eliminate as much of chance events as can be controlled, so that results deserve and achieve high levels of confidence in their reliability and validity.

When it comes to PHI, or any other set of three initials denoting the management of health in populations, random variation is anything but stamped out. It is ever present, an essentially permanent and unvarying problem with just about any study of a particular solution, or of multiple solutions to the same essential problem – finding cost-effective ways of controlling population health, and thereby gaining financial benefits for sponsors, and health/life quality benefits for populations.

Randomness in populations

The first unavoidable challenge in stamping out randomness in PHI is the randomness in populations, the fact that they vary greatly from one to another. When I began investigating the potential for PHI in Denver in the 90s, I became immediately aware that the workforce at Coors Brewing Company was vastly different from that at Celestial Seasonings, even though both were in the “beverage” business.

Coors employees were far more likely to be older, male, smokers, overweight, and serious health challenges, to say nothing of their natural tendency to drink beer. Celestial Seasonings were more likely to be female, young, “health nuts”, with far less potential for reductions in their healthcare costs or health-related performance impairment. Testing the same intervention in both populations would invite vastly different results.

Populations often vary as well in dimensions such as the extent of their trust in their employer and insurance plans. They may believe in medical interventions, or prefer any number of alternative, complementary, and “new age” options. Health risk assessments that are the foundation for most PHI interventions may be based on sickness care claims, on biometric screenings, or on surveys of self-reported health status, perceptions, attitudes, productivity, performance, etc. – or any combination of all three.

Randomness in Problems

Partly because of the randomness in populations, the problems that are to be addressed in PHI interventions can vary widely. Coors should have a lot more overweight and obesity problems, more heart disease, diabetes, and a host of other problems more common among older men than would Celestial Seasonings. Moreover, the problems may be far more tractable at Celestial Seasonings, if its workforce is generally health-focused and feels responsible for its health.

Even if the problem “map” for two populations were pretty similar, the problems chosen as targets for attention may be different, simply because they are chosen differently. Employers may differ widely in what results they wish to pursue, from sickness care to disability to workers compensation expenses to productivity and performance impairment, positive presenteeism and revenue gains.

Depending on what results each is after, employers with the same problems may choose vastly different problems to address, more chronic disease or occupational safety issues on the one hand, or more lifestyle problems, lack of sleep, stress and job/life satisfaction issues on the other. Since different problems call for different interventions, the “solutions” chosen by different employers will tend to be different from each other, and different from the problems that insurance plans select.

Randomness in Solutions

While the standard in random clinical trials calls for precisely the same medication or other intervention to be used with all patients in the “intervention group”, and a placebo in the “control group”, such a practice is decidedly rare in PHI practice. There are hundreds, perhaps thousands of different solutions, offered by specialized PHI providers, retail clinics, onsite clinics, employers, healthcare organizations, and even concierge physicians.

It is statistically impossible, even ridiculous, to even attempt to reach a general conclusion about whether or not and how successful PHI is as a solution, precisely because it is not “A” solution, it is hundreds of same. Providers of different solutions typically maintain a proprietary interest in keeping some aspects of their solutions a secret, rather than sharing what they learn with their competitors. There is no FDA that decides which solutions are permitted to be sold, and no published data on comparative-cost effectiveness in most cases.

As a consequence of the random variation that cannot be stamped out in PHI, we are beset by advertising claims from hundreds of sources on the results they get, with little basis for confidence that such claims will pass any test of “scientific rigor”. The early years of Disease Management (DM) for example, were replete with “optimistic” or exaggerated claims, even guarantees of results, later found to be heavily contaminated by “regression to the mean” and “self-selection bias” effects.

While there is great furor over the idea of a federal program for comparative cost-effectiveness testing and publishing the results thereof, it might make sense to create a comparable effort, public or private, in PHI. Current efforts at health reform are all doomed if they do not include the most cost-effective solutions possible to the challenge of reducing the incidence and prevalence of disease and injury in our populations.

At the moment, there is no real hope that we will ever be able to identify which these are, to move PHI in the direction that evidence-based medicine has moved – identifying and promoting the widespread adoption of the best practices we can possibly identify, in spite of the permanent challenge of random variation. We can certainly do a lot better than we have so far.

Evaluating Early Indications of PHI Intervention Results

2009-12-04T11:05:00.000-08:00

There seems to be a widespread belief that strategies and particular interventions aimed at improving populations’ health and performance will take a while. Many providers of PHI programs predict success only in two or three years, though others, such as HealthMedia, Inc. (www.healthmedia.com) consistently report measurable results in as little as 90 days.

Fortunately, when PHI is seen as ultimately a marketing challenge, in part or in whole, that means that there are some immediate effects that can at least be looked for, and evaluated when found. Marketing rests on the belief that changing people’s behavior requires or at least greatly benefits from changing their perceptions, beliefs, and attitudes – and these changes can be gauged as early as they seem likely to occur.

Perhaps the most common behavior change model in use for wellness, risk and disease management programs is the “Stages of Change” or “Transtheoretical” model developed by James Prochaska, PhD and others. It postulates that people tend to move through a series of stages as they change behavior, from “pre-contemplation”, to “contemplation”, to “planning” to “action” to “maintenance”, though they occasionally relapse backward from one stage to a previous one, and ultimately may reach the point where no particular effort on their part or anyone else’s is needed.

These stages are easily measured through surveys that can be inexpensively administered via online communications at home or work, with results complied automatically by computer. Once payors learn what kinds of patterns characterize their populations, including dependents eligible to participate in PHI efforts, they can apply these patterns to predict what will happen with new participants.
Market research should be able to determine how many people tend to move from each stage in behavior change to the next, how long after they begin participation in the PHI initiative, how many relapse and how long it takes for them to recover and resume progress. These rates and timing for changes can then be used to predict when a desired level of success should be reached.

As they use marketing tactics, incentives, or any combination of the two to move more people faster through the stages of change, payors can track how much either or both approaches adds to the rate and timing of success, and calculate the internal ROI of such options. Since all findings of such tracking are likely to be unique to each population and payor combination, extrapolation from past histories of other populations may not prove reliable, though they may offer good enough estimates for use in the early stages of PHI implementation.

When PHI initiatives enlist family members or other supporters in particular efforts, and this often proves significant in influencing participants, early results can also be tracked through surveys. Participants can be asked about the extent to which they are getting support from family, friends, support groups, and other “social media” sources, as well as how they perceive such support.

Comparisons can easily be made between participants who get vs. don’t get social support, and even between those who have positive vs. negative attitudes toward that support, or different opinions depending on its source. This can help payors to evaluate the difference that such support makes to their desired achievements, and adjust their interventions accordingly.

Where early results of attitudes and self-reported efforts indicate the need for adjustments in the intervention strategy, they can also be used to select appropriate changes. Such early interventions should prove far more timely and effective than those that have to wait for end-of-year repetitions of health assessments, productivity or performance data.

Marketing May be an Alternative to Cash in PHI

2009-12-04T10:32:00.001-08:00

As a career-long devotee of marketing as a method of influencing people’s behavior, I have long advocated its use in workforce and other population health management efforts. Beside the fact that it fits, logically, marketing can, if successful, reduce dependence on what is otherwise the automatic first choice in most such effort, namely paying people to behave healthier.

There are significant problems with paying for behavior, though also significant potential benefits. Pay-for-performance, for example, has long been advocated, and when applied, often seems to work well. The immediate productivity increase of 44% achieved by the Safelite windshield repair company illustrates what can happen when current productivity levels are hampered by low worker motivation.

Of course, paying people to do something that the “system” is the primary reason for them not doing so already can be a total waste. Pay often has to be significant in order to work, and can end up costing more than the changed behavior delivers in financial gains. It easily becomes perceived as an “entitlement”, and then mainly serves to dissatisfy and de-motivate employees when it is reduced.

Marketing, by contrast, can work continuously and for a long time, when it is done well. As described this morning in a webinar on “Moving the Needle by Maximizing Participation” (www.wellsteps.com), even basic marketing concepts fit well with the challenge of communicating to people in ways that promote their engagement in wellness initiatives.

In its “Marketing 101” discussion, it cited the three simple challenges that must be successfully met:
• Identifying and understanding what people want, what problems they want solutions to, what they wish or at least would welcome as a difference in their lives
• Devising compelling value propositions linking what they want to the wellness solutions available, and communicating the “offer” in a compelling manner
• Calling for a specific action on their part to such communications, so that they are moved to act in a measurable and significant way toward engagement in the wellness initiative

The ideas presented during the webinar were backed up by a client of WellSteps, and by Judy Garrett of Syngenta (www.syngenta.com) who described the way it has implemented marketing concepts. If has devised familiar, “down home” labels for specific interventions, such as “Oh, my aching back” for back pain efforts, and makes the idea of participation more compellingly convenient by using a mobile van to reach employees wherever they work. It engages spouses, not only as possible participants in wellness efforts, but as sources of motivation and support for its worker participants, as well.

Marketing should at least reduce employers’ reliance on, and greatly reduce the size of any incentives offered to induce participation. Marketing research, for example, can help identify precisely what forms of incentives (recognition, gifts, “points” that may be “spent” as participants wish, etc.) are likely to be, and end up being most cost-effective.

By communicating to prospects and participants what they will be able to add to their lives, and enlisting their spouses, friends and family in efforts to enable them to do so, gains in personal life quality may prove more powerful than mere cash as motivators. And by reminding those who progress and succeed in their wellness efforts of what gains they have actually made, levels of engagement in wellness, and even gratitude and other positive attitudes toward their employer may be increased far more than is the case for cash incentives.

Of course, the only way to find out how well marketing works is to try it and see. It is likely to take more time and effort to design and implement a marketing campaign than to merely “bribe” workers into participation. But the long-term effects may well prove far more positive, in terms of both financial ROI and employee morale, retention, etc., than can be achieved through incentives alone.

Some Major Barriers to PHI Success

2009-12-02T12:06:00.000-08:00

The fact that people behave “irrationally”, that emotions, sub- and un-conscious motivations often drive behaviors, has long complicated efforts to induce healthy behaviors in populations at risk. But even when “rational” or “cognitive” attitudes and beliefs play a significant role, they are subject to a major challenge – people are often erroneous in their convictions, but absolutely certain of them.

The incidence and prevalence of “erroneous certainty” has recently been covered in a highly readable book [Philip Hansten Premature Factulation: The Ignorance of Certainty and the Ghost of Montaigne Philiponus Press 2009]. It covers the wide range of sources for erroneous certainty, and offers a wide range of methods for preventing it in oneself or correcting it in others. The fact that these sources and methods require an entire book to cover them indicates how significant a problem erroneous certainty can be.

A recent article also discussed the challenge [L. Wallace “Why Fact Can’t Compete with Belief” The Atlantic.com Oct 9, 2009. It suggests that one major reason for the triumph of erroneous beliefs over verifiable facts is what is called “motivated reasoning”. Because any instance of being forced to alter one’s beliefs or opinions is, to some degree at least, an attack on one’s sense of self and worth, it often motivates us to find ways to dismiss or ignore information that doesn’t fit our preconceptions.

This typically includes what may be complicated re-interpretation of facts, dismissing their sources, selectively listening, reading, or remembering only those facts that represent no conflict. Often, people confronted with antagonistic information will actually intensify the beliefs that the information puts in doubt.

If any of the ideas put forth in the book and article sound familiar, it may be because of the recent fiascoes at “town meetings” regarding health reform proposals, or plans to send more troops to Afghanistan. Scientists have shown themselves to be just as resistant to new ideas as are the rest of us, witness decades of objections to Darwin’s theories among his peers.

It is often necessary, or at least useful, to change people’s minds about something related to their health and behavior/lifestyle in order to induce them to adopt and persist in healthier alternatives. When intractable but erroneous beliefs hinder such changes, they can be just as difficult to deal with as are sub-/un-conscious thoughts and emotions.

Fortunately, both brain science and “the wisdom of crowds” can help to overcome such hindrances. Brain science offers a number of methods for framing information, for example, that will add to the pure content impact thereof. Peer pressure and support can often overcome the resistance of individuals based on erroneous certainty.

One specific approach is to alter behavior by altering the environment in which the behavior takes place. A recent study in Connecticut middle schools, for example, demonstrated that when “junk” foods and drinks were removed from school vending machines, the students significantly reduced their use of both, more so in the second year of the experiment than in the first. [C. Bruderlin-Nelson “Kids Eat Less Junk Food When Middle Schools Stop Providing It” Health Behavior News Service Dec 1, 2009 (www.cfah.org)]

Employees, themselves, seem to support the idea of eliminating, reducing, or charging more for unhealthy food and drink in worksite cafeterias and vending machines, for example. In the middle school example, the children did not make up for the lack of junk at school by consuming more of it at home, so there was a net decrease, at least. Unfortunately, schools generate revenue by making such foods and drinks available in vending machines, so not all schools may consider it a worthwhile investment.

By learning more about how brain science can be adapted to influencing health behaviors, and experimenting with methods for overcoming erroneous certainty, PHI initiatives may be able to increase their effectiveness well beyond what traditional “management” or even “marketing” methods have achieved. It is certainly worth some effort to improve the effectiveness and efficiency of PHI, given the worldwide challenges of too much disease and injury for too little money available to treat them, as well as the need to continuously improve workplace performance.

Is the U.S. Ahead or Behind the World in PHI?

2009-11-20T13:44:00.000-08:00

While we in the US like to think we are at the “cutting edge” of most things, it has always been clear that this is true only in the narrow domain of healthcare technology – and even there, in the sickness care side of that. When it comes to proactive health management in general, and even to its application to the workforce, we may be different from the rest of the world, but we are by no means ahead.

In a recent survey of over 1100 organizations employing over 10 million people in 45 countries, Buck Consultants found that: “Improving Productivity Driving Wellness Programs Globally, But Not in U.S.” Employee Benefit News Nov 17, 2009 (http://ebn.benefitnews.com). Wellness programs are only slightly more common among employers in the U.S. than in other countries, and are not really more well-developed.

In at least one area, U.S. employers are slightly ahead of the rest of the world. Globally, only 22% of the organizations surveyed measure the impact of their programs. In the U.S., where only 13% did so a few years ago, 42% were measuring ROI in this survey. The economic downturn has affected employers globally, with 24% reporting having to reduce investments, though 19% reported increasing them.

Priorities differ across the world. The highest priority in Africa, Australia, Europe, and Latin America, and even in Canada, was given to improving productivity, mainly through reducing presenteeism. Asian and U.S. employers ranked this as second priority, though Asia ranked improving employee morale and engagement first, while the U.S. ranked reducing health insurance costs at the top.

This is probably understandable, in that healthcare costs in the U.S. are significantly higher than anywhere else in the world, and employers pay a large portion thereof, with more of their revenue going to cover such costs than are gained as profits. Moreover, most other countries have a government-supported or managed healthcare system, where employers do not pay directly for their own workforce’s healthcare.

The drawback to U.S. employers focusing more on healthcare costs is that this automatically cants their investments more toward chronic diseases, for example, than is the case with other countries. This often means ignoring or giving short shrift to other health and personal problems, such as sleep deprivation, family issues, stress, etc. that are among the most severe causes of lost productivity.

Employers in Africa, Asia, Australia, Canada and Europe, for example, list “Stress” as their top health problem, while Latin America lists it as third. U.S. employers did not list it among their top three, which were “Physical Activity/Exercise”, “Nutrition/Healthy Eating”, and Chronic Diseases”. Given that U.S. employees are among the most obese in the world, the emphasis on nutrition and exercise may be understandable, but stress is also both a major health risk and a major hindrance to productivity and performance.

The methods used to address health priorities vary a bit in terms of commonality, but the lists are all but identical across the world. Healthy food options at the worksite, online information and coaching, improving the psychosocial work environment, and personal health records are among the top five almost everywhere. There seems to have been no questions asked about the extent to which employers in different countries have integrated their wellness efforts with other strategies and programs to improve productivity and performance, though these must be major priorities almost everywhere in the current downturn.

Wellness programs were found in 77% of employers in North America, though the worldwide average is 64%, up from 49% in 2007. Among global firms, 48% have globally centralized responsibility for wellness, up from 22% just a year ago. Where U.S. employers may have enjoyed a significant lead in promoting worksite health a decade ago, it seems that the rest of the world is definitely catching up. It may even be passing U.S. employers in tying productivity improvement efforts to health, rather than getting bogged down in what is often the far less rewarding domain of healthcare costs.

[The full report “Working Well: A Global Survey of Health Promotion and Workplace Wellness Strategies” Buck Consultants 2009 is available for purchase (executive summary was free when is accessed it) at www.bucksurveys.com.]

The Fifth Life Asset in PHI

2009-11-20T09:34:00.000-08:00

I have found that a “life asset” construct can be very useful in planning and executing population health management/PHI strategies and programs. The first four of these assets represent elements that people can have or lack enough of, assess their status, apply with some discretion, and balance their pursuit and protection of. These assets are:
• Competence/talent – useful knowledge, skills, and attitudes/traits that can either promote or hinder their performance on the job and in their lives
• Health – overall “wellness”, energy levels, specific behaviors and conditions that represent risks of future acute/chronic diseases or disabilities, or impair productivity and performance at work or in their lives
• Time – total amount of time available at work and discretionary time available for life pursuits, how well it is managed in order to enable the demonstration of their competencies, pursuit of health, etc.
• Wealth – cash and other assets that people possess and can dispose of, how much is “discretionary” vs. committed to satisfy obligations and meet the necessities of life

These four assets are at least somewhat under the control of, or at least ability to be influenced by individuals. In the realm of personal performance, they can promote or hinder what people are able to. They interact with and affect each other – wealth can buy better health, and health can exhaust wealth. Time can limit or enable the pursuit of better health, or the maintenance of good health. Competence may include or omit a wide range of abilities that enable or hamper the pursuit and protection of health, time and wealth assets.

The fifth life asset is power. While “power” is defined as the potential to control or influence what happens in one’s life and what other people do, it also includes the ability to resist the efforts of others to influence or control one’s own life. The concerns and ambitions one has with respect to this life asset may include widely ranging amounts or degrees of dominance over others, or autonomy vs. dominance by others.

Time is a unique life asset in that it cannot be stored or accumulated for future use – it only exists during each day, week, month, or year, to be “spent” on some combination of necessities and discretionary uses. Power is a unique asset in that it only exists because of the assent of others, and can vary widely across different contexts, such as work, family, community. It can also vary by situation, particular moments in time when one may be granted or denied power or autonomy by others or by one’s competency, health, time, or wealth assets.

All five assets share the ability to be used by others or one’s self as incentives and rewards to promote PHI-relevant behaviors. While wealth is by far the most popular asset category used in promoting both health and performance behaviors, at least in the United States, it is by no means the only one. Moreover, its use is severely circumscribed by the tendency for wealth asset rewards to be PHI expenses for employer, insurer or government sponsors.

Competency assets are often deliberately promoted by employers to both satisfy workers who wish to develop these assets as part of their career or job, and such development often improves employee performance, engagement, and retention. Time assets can be used in the form of paid time that can be used in PHI pursuits, extra days off as a reward for PHI accomplishments, including work performance as well as for health improvements.

Power, in the form of autonomy and personal control over one’s life can be used in promoting health, such as enabling participants in specific programs to choose of formulate their own goals, even to devise their own methods for achieving such goals. The Duke (University) Prospective Health Program, for example, enables participants to formulate their own health “mission” statements and goals, though it exercises its own power relative to the types and costs of support for which participants are eligible. (www.dukeprospectivehealth.org)

PHI participants may also be afforded a modest degree of power through social networks for group support, where individuals may become “mentors” or other influentials in support groups, for example. “Wealth” incentives may be offered in ways that empower participants to make choices among different rewards, rather than merely one specified by the PHI sponsor.

Of course, power may also be used to improve employee performance directly, through tele-work and similar arrangements which enable workers to perform tasks when and where they wish. Best Buy’s “Results-Oriented Work Environment” program at its corporate headquarters, for example, gave its workers full autonomy with respect to where and when, and enjoyed an immediate 35% increase in worker productivity, with a reduction in annual turnover from 16.7% to zero. [M. Conlin “Smashing the Clock” Business Week Dec 11, 2006 (www.businessweek.com)]

Power can also be used to improve health. With stress, one of the major causes of health problems, understood to be a function of the demands that life and work put on us together with our ability to control such demands and how we respond thereto, enhancing workers’ ability to control their own work can be a significant health promotion tool.

When workers have the power to control their own performance, and are rewarded directly for increases therein, via pay-for-performance (P4P) systems, for example, they often significantly improve and sustain such performance. Employees at the Safelite windshield repair company increased their output and their employer’s revenue by 44% in the first year of a P4P scheme that replaced an hourly pay system that only rewarded workers by requiring more time expenditures.

Empowering workers or health plan members to have more choices about the goals they pursue, the methods they use, and the rewards they can gain may be a very effective combination in PHI. While PHI sponsors understandably wish to control at least the costs they incur, there are choices across the various life assets besides wealth that may prove to have equal impact at far lower costs.

Marketing Applications in PHI -- Lead Development

2009-11-12T09:12:00.000-08:00

One of the elements that marketing can add to traditional management approaches to health management is the application of the concepts and techniques of “lead development”, a major part of marketing and sales. Where management would generally select prospects for PHI based on assessments of problems, marketing would look far beyond that to identify and qualify “leads” representing the people most likely to become “valuable customers”.

This means looking at both which members of the population at risk combine both high potential for contributing value to the insurance plan or employer sponsor of PHI efforts, and high probability of doing so. The mathematical product of potential times probability represents the predicted benefit that each individual can contribute, which can then be compared to different costs of gaining that benefit to both select prospects and determine which interventions each would fit best.

Unfortunately, this combination is complicated by the fact that the probability of realizing valuable contributions – in reduced sickness care, workers compensation, and disability expenditures along with improvements in productivity and performance – is affected by which interventions are selected to realize such contributions. Lower-potential members of the populations would logically be steered to, persuaded to select, or limited to lower-cost interventions, in order to promote positive returns on the investment (ROI). And lower-cost interventions may have significantly lower “success rates” in terms of bringing about the desired behavior changes and financial value contributions.

Management approaches would tend to lean toward either selecting the “right” intervention for each member, based on each’s potential x probability predicted value, or if the member is allowed to select each’s own preferred health goal, to limit what kinds and intensity of support each will be eligible for in order to achieve the right match. Ideally, the success rate for members with given known characteristics and attitudes would be predicted with some confidence for each of the interventions available, so that a logical match could be made.

While by far the most common management approach is to identify the best match, from the sponsor’s perspective, then seek to enroll the member in that intervention, using incentives, education and persuasive communications in order to achieve the prospect’s participation. The Duke Prospective Health Program, however, (www.dukeprospectivehealth.org) is one example of “empowering” employees and dependents who are members of its self-insured plans by getting them to write their own mission/vision statements and formulate their own goals. Duke can then control the interventions each participant is eligible for in order to match each’s predicted costs to each’s predicted contribution.

In effect, the predictive modeling used to predict both the contribution potential and the realization probability of each member is a way of “qualifying” leads. The same information can be used to control the kinds and costs of the efforts and methods used to seek the participation of each prospect, not merely the costs of the intervention that fits each best. After all, the cost of enrolling and engaging participants is likely to be one of the major costs of the entire intervention.

Marketing concepts and techniques are designed specifically for the full “campaign” that will be used to promote enrollment (“sales conversion”) active engagement, and success of as many participants as possible. The targeting of the right people for each communication, media, frequency, and timing of each, for both achieving the desired enrollment/engagement results and the desired health and performance behavior changes that represent success, fits well with marketing capabilities.

Employers and insurance plans will usually have their own marketing staff, or outsource marketing functions to external agencies. The same people involved in promoting sales of their products and services may well have capabilities that can be applied to PHI challenges. Or since PHI is most often outsourced, due to privacy/confidentiality and trust considerations that make it risky for sponsors to control or even access PHI data, the PHI provider should have access to marketing support.

Given that a key element of PHI will necessarily be evaluating the impact of its strategy and particular interventions on traditional “customer” metrics (satisfaction, loyalty, “customer lifetime value”, ROI, etc.) it makes sense for marketing to be involved in, if not in charge of this element, as well. The contributions that marketing can make, through its traditional expertise in market research, offer design and modification, and “advertising/sales” should add significant value, as well as awareness of that value, to most PHI efforts.

Retail Clinics and Proactive vs. Reactive Health Care

2009-11-02T08:56:00.000-08:00

There has always been the potential for retail clinics, particularly those staffed primarily by nurse practitioners, to become health, not merely sickness clinics. Most were certainly started with a main focus on treating routine illness and injury, offering convenience of time and place, and the ability for consumers to use any wait time required to do shopping in the same location, as well as obtain prescription filling and refill services there.

Most offered basic preventive health services, such as flu shots, vaccinations for vacation trips or to meet requirements for school, but rarely included anything comprehensive or even extensive in the way of personal health management services. One exception has been the RediClinic chain of clinics, primarily in the Austin and Houston, Texas markets, though it recently opened a clinic in Missouri.

It offers what it calls a combination of “Get Well” services, typical for retail clinics, but also an extensive set of “Live Well” services, including primary, secondary, and tertiary preventive services for common risks (e.g. smoking cessation) and chronic conditions (e.g. allergies). It cites seven different ways by which it can reduce employee healthcare costs by 5% or more:
• Offering lower costs for routine sickness care than urgent care clinics and Emergency Rooms charge
• Reducing absenteeism and increasing productivity
• Offering lower cost health screenings, immunizations, exams
• Providing commonly needed preventive services at low cost
• Reducing the travel time for most employees
• Minimizing time spent obtaining services
• Offering services seven day a week and extended hours per day
(www.rediclinic.com/resources/value.asp0

RediClinic, like many other retail clinics, has developed working relationships with healthcare systems, in its case, with St. David’s HealthCare in Austin, which has 24 sites of its own in Central Texas. (www.StDavids.com) Such relationships usually combine physician oversight for the retail clinics, and places to which the clinics can refer patients, often with mutually accessible electronic health records that can be shared among linked providers.

Minute Clinics, one of the earliest chains and one of the largest, with 560 locations in 24 states, recently adopted the use of Keas care plans for five acute conditions: flu, sinusitis, bronchitis, pharyngitis, and acute (strep) laryngitis. It the same services are also being offered at 2200 Quest Diagnostics service centers in the US. These and other health-focused care plans are intended to enable patients to develop and follow personalized health action plans based on shared medical records. Minute Clinics and Keas have working partnerships with a number of major integrated healthcare systems, including Allina in Minnesota and Partners HealtCare in Massachusetts. [B. Dolan Pedometer Plan: Keas Partners with Partners HealthCare” Mobihealthnews.com Oct 22. 2009]

Any combination of personal health monitors, that can be used by consumers to monitor and report specific health indicators, along with the ready availability of retail clinics in thousands of convenient retail stores, can make it far easier for consumers to manage their own health. Whether they will invest enough of their own time and effort in doing so, and how extensively retail clinics and personal physicians will join in such efforts is open to question, though the potential is clearly substantial.

Ezekiel Emanuel, Special Advisor on Health Policy at the Office of Management and Budget, has expressed doubt that: “…patients are willing to give up much of their time to monitor their health and become empowered.”. He believes that physicians will retail the role they have always had in such pursuits. Others, however, feel that both patients’ and physicians’ roles are changing already. B. Dolan “TEDMED: OMB, Qualcomm, Intel and More” MobiHealthNews.com Oct 29, 2009]

While retail clinics, themselves, have been forced to close or limit their growth since they began a resurgence in the past decade, it is clear that they could play a significant role, along with remote monitoring, medical homes, and other developments already under way. They will probably need the acceptance and support of traditional healthcare providers, insurers, and consumers, as well as governments, in order to achieve their full potential in proactive health as well as reactive sickness care, however, and only time will tell if they will get it.

Shouldn't Most Things Be Value-Driven?

2009-10-29T10:03:00.000-07:00

This decade, however it may be labeled, has been replete with examples and exhortations of “value-based” (VB) behaviors. They range from benefit design (VBBD) to purchasing (VBP) to human resources (VBHR) strategy (VBS) and overall management as well as policy making (VBPM). Even calls for cost-effectiveness policies are, in effect, demands that health care, prescription drugs, and health expenditures be value-based (VBCE?).

The basic idea is that all kinds of behaviors, policies, practices, and investments of time, effort, and resources, should be designed and applied in ways that optimize the value gained therefrom. This value may include the benefit vs. cost felt by an organization, an individual, a population, an industry, society as a whole, or almost any number of different entities and people. And while the idea of “value” seems to be generally understood, it is still unclear, for example, whether the relationship between benefit and cost is calculated as benefit minus cost (net gain) or benefit divided by cost (ROI ratio).

In any case, the most common examples of value-based practices seem to be uncertain or unclear about how many that are participants in or affected by such practices should be included in whatever value calculations are to be applied. It is at least generally the case that PHI at least considers the value to be gained and investments made by both the sponsors and providers thereof, as well as the individuals and populations who participate in and are affected by PHI strategies and initiatives.

Of course, when medical insurance plans and employers apply value-based PHI, they focus most of their attention on their own value. But since the prospects and realities of prospects and participants are major factors affecting the number of participants, degree of commitment, and amount of success achieved, the value gained by them is at least a significant, pragmatic concern.

What often gets lost in PHI efforts, however, is the extent to which the definitions, dimensions, and expectations of value among participants differ from those of sponsors, and vary across the populations involved. Insurers and employers, as sponsors of PHI efforts who usually pay the costs of such efforts, and even offer incentives to participants for their cooperation therein, tend to feel that their definitions and dimensions should have priority.

Despite such an understandable feeling, it has until recently been the case that sponsors of PHI were slow and rudimentary in their evaluation efforts. Only in the last year or two have even a majority of sponsors applied a formal evaluation process to their investments, and the vast majority still do so myopically. Employee health has been shown by many and argued by most to affect not only insurance (including medical, workers compensation and disability) expenses, but absenteeism, presenteeism, turnover, morale, productivity and performance as well, yet few sponsors evaluate even half of the affected dimensions.

As long as evaluations yield positive results, of course, it might be argued that they need not go any further than “satisficing” PHI sponsors, i.e. showing that they have gained enough to show that their investments have paid off and should be continued. But such an approach can also reduce the value gained, by leading to both under-investment in PHI, and mid-directed investment, missing many opportunities to achieve even greater net gains.

For example, to the extent that sponsors look narrowly and for short-term effects alone, they can easily miss many of the effects that generate participants’ enthusiasm, engagement, and success. Failure to include workers’ own gains in terms of increased productivity and improved performance in evaluations can reduce both the financial rewards they obtain, and a personal sense of accomplishment, achievement, and self-realization.

Insurers that fail to look beyond the direct medical, workers compensation, or disability expense savings that PHI produces can miss opportunities to not only delight more of their employer clients, but to develop a revenue-generating business focused on not only current insurance clients, but potential PHI clients who are not insurance clients. Both CIGNA and Aetna have developed such a business, for example, though they are in the minority.

Employers that fail to look beyond these same direct and easily measured effects of PHI investments, or look only at immediate outcomes, may miss the potential for long-term wellness-focused efforts involving their entire workforce, and limit themselves to disease management among the minority of workers with high-risk/cost chronic conditions, even though these are often the most difficult to improve. They may also miss important health-related factors that impair workers’ productivity and performance but generate little medical care expense, such as poor fitness, diet, and sleep habits.

But the most serious flaws for PHI providers, insurers and employer clients alike, may be that their narrow definitions and dimensions of value are coupled with narrow definitions and dimensions of their “value-based” PHI investments, per se. Even many PHI advocates and practitioners often forget that wellness, risk and disease management are only one of many avenues to improving worker productivity and performance, and the overall value of their contributions to their organizations.

It may be that the expansion of the value-based investment and evaluation processes to the point of recognizing and integrating all the employer efforts at improving productivity, performance, and value of employees will occur only many years, even decades into the future. But the sooner it happens, and the more completely employers and PHI providers, as well as insurers, recognize the potential of integrating PHI with other human capital asset efforts such as training and development, compensation and benefits, work/life balance and flexible work arrangements, etc. the greater the value they figure to gain, as long as they measure it well enough to plan, manage, and evaluate it well.

Genetic Predispositions and PHI Interventions

2009-10-29T08:57:00.000-07:00

There has long been a debate as to whether certain health risks, both conditions and behaviors, are “genetically programmed” into some people. If so, the argument usually goes, it would be unfair to penalize such people for their born proclivities, just as it would be to penalize people who are left-handed or short. At its most severe interpretation, health risks become diseases that medical insurance plans should cover as they would any other illness or injury.

Smoking, for example, has been deemed an addiction by many, and not a voluntary choice, at least as far as quitting the nasty habit is concerned. Overweight/obesity, it has been argued, are diseases, just as are anorexia and bulimia, even though all are clearly behavioral problems as well. They often have behavioral solutions, as well, including many where "cognitive behavioral therapy", "motivational interviewing", and other behavioral approaches have been found to be the most cost-effective options. [A. Gardner "Psychotherapy Beats Light Treatment for SAD" HealthDay.com Oct 29, 2009]
A recent study found a gene variant that appears to predispose its owners to being bad drivers. [“Blame Genetics for Bad Driving, Study Finds” CNN.com Oct 29, 2009 @ www.cnn.com/2009/TECH/science/10/29/bad.driver.gene/index.html]

If genetics indeed predispose some people, but not others, at least not as much, toward health-risky behaviors and conditions, would it be their constitutional right to adopt and persist in such behaviors, while the rest of us could be punished for it? Predisposition can be statistically significant to scientists while not making it even close to impossible to resist, after all. Predisposition does make it more difficult for people to resist certain temptations, or more prone to become addicted to certain behaviors, but rarely make them entirely irresistible.

If anything, it would seem that an argument could and perhaps should be made that people with predispositions toward a given behavior or condition (overweight/obesity tends to run in my family, for example) should be offered higher incentives to avoid or reform the risk involved, rather than forgiven entirely. If a given predisposition makes individuals twice as likely, or 25% more likely to have a given risk, each could be offered and paid twice as much or 25% more, respectively. Disincentives could be adjusted downward, to half as much or 25% less, respectively.

Of course, such a policy and practice would be complicated by the fact that the Genetic Information Nondiscrimination Act of 2008 (GINA) appears to make it illegal for anyone to even collect information on genetic predispositions, including family histories, much less to use incentives or punishments based on such genetic factors. [L. Masterson “Genetics Rule Could Hamper Wellness Efforts” Health Plan Insider Oct 29, 2009 (http://healthplans.hcpro.com)]

It would be unusually silly, even for our Congress, to insist that any health risk with any genetic proclivity involved should be protected against either identification or “incentivization”. This would handicap wellness, risk and disease management to the point of absurdity. The GINA regulations, as promulgated by Health & Human Services, Labor and Treasury departments, appear to go well beyond the original wording of the legislation, which focused solely on the use of genetic information in insurance underwriting and employment decisions.

In the long run at least, it seems likely that almost every health risk behavior and condition will be found to have some kind of genetic predisposition for some people. If this means prohibiting the discovery of the behaviors and conditions involved, it will not merely “hamper” wellness, risk and disease management efforts, it would probably make them impossible.

Of course, there are ways around most legislation, as this country has ably demonstrated in such examples as the alcohol Prohibition of 1919-1933. One way would be to make it entirely voluntary on the part of insurance plan members and employees as to whether or not they report a given wellness concern, risk behavior, condition, or chronic disease. Or perhaps PHI providers who are insulated from insurers or employers could be allowed to analyze or collect information on risks as long as they don’t share it with their clients.

As for incentives, the same relatively safe options of offering incentives for participation, rather than for proven risk reduction, or in the case of employees, for improved productivity and performance, whether it results from improved health or any other cause, could be available. In any case, except in the rare cases where a genetic predisposition is truly a necessary and sufficient cause of a given risk or disease, it would seem illogical in the extreme to prohibit any attention being given to them whatsoever, though lawyers and cause advocates may disagree.

Adjusting Measurements -- Team to Individuals

2009-10-28T09:57:00.001-07:00

When it is difficult, expensive, or otherwise too much trouble to objectively measure the performance of individuals, the usual option selected is to measure the performance of teams, departments, or units, instead. This can enable at least a pay-for-performance (P4P) or performance appraisal system (or both) based on objective measurements rather than either managers’ or employees’ own subjective estimates.

When P4P systems based on groups suffice to achieve the desired degree of motivation, performance, retention, and other measures of effective management, this may be all that is needed. But in many cases, individuals will still desire, often prefer, that their individual productivity/performance also be measured and rewarded, either along with group rewards, or instead of them. In such cases, it is at least possible to adjust the group data to account for individual member differences.

Since the most important attribute of any adjustment is that it should be accepted wholeheartedly by the members of the group, particularly by its true high-performers, the adjustment approach should probably be devised, or at least ratified, by the members of the group, itself. They may be assigned the responsibility to come up with their own system, with management merely having the right to approve or not, or management may present its own approach to be considered by the group, or participate in the group process.

One approach, for example, could be to have each group member propose their own individual productivity or performance rating, then obtain 360o feedback from the other members of the group to reach a consensus or at least average rating. The measured productivity or performance value of the group would then be used, in comparison to the total rating of its combined members, to adjust the individual ratings. If the consensus or average ratings of the combined group are less than the measured value, the ratings would be adjusted upward based on the difference; if group ratings are more, adjustment would be downward.

If supervisors and other managers are involved, they may add their own ratings to the consensus or average ratings calculations, with similar adjustments made based on the measured value of the group’s performance. Ideally, there should be only small differences between managers’ and peers’ ratings, though if there are large differences, another approach may be needed.

In my own experience with employees’ own ratings of their performance, for example, the pattern I have noticed most often is one of minimizing the absolute differences among them, while generally accurately reflecting the relative differences, or ranking of each other. Workers who operate in a true team environment or system usually know which among them is contributing better than average, and which less than average value, at least in terms of observable performance and output.

In cases where managers agree with the ranking, but feel the gaps between workers are greater than peer ratings indicate, adjustments can be made – perhaps “splitting the difference” between what managers feel the gaps are between members and the smaller gaps that the peer consensus numbers indicate. After all, numerous studies have indicated that low-performing individuals tend to over-estimate their performance, while high-performers tend to underestimate it. This will tend to “crunch” the variation to something like the Lake Wobegon phenomenon of everyone being above average.

Of course, the ideal situation in any team would probably be one where everyone really is above average. The tendency of P4P systems, themselves, is to drive low-performers toward higher turnover, while retaining more high-performers because of their greater rewards. In the Safelite auto windshield repair case, turnover among high performers decreased, while it increased among lower performers. If such unbalanced attrition persists, it could easily achieve the Lake Wobegon ideal over time.

This is a major reason for avoiding any “forced normal” distribution in performance ratings. In my last position as Corporate VP for Strategy and Marketing, I was lucky enough to have four direct reports who were definitely above average, and all four secured gain-sharing bonuses reflecting the organization’s improved performance accordingly. Forcing a normal distribution would have been not merely unfair, but self-defeating in such a situation.

There should always be some basis for determining the extent to which workers as a group, and ideally, as individuals, are above, below, or roughly average in their productivity and performance. At Lincoln Electric, for example, its workforce productivity increased by more than three times as much as did its industry as a whole. Its workers clearly deserved, and thanks to finding markets for its increased productivity, have enjoyed compensation combining wages and bonuses that are twice the industry average. [“The Lincoln Electric Company” Harvard Business School Case Study # 9-376-028 and “Archive for Feb 27, 2002 FuturePositive.synearth.net]

The less cohesive and co-located team members are, the more difficult it is likely to be to achieve a P4P rating and adjustment system that is equally and sufficiently acceptable to all. Creating smaller teams for rating purposes should help make solutions more acceptable, but it may take some years to reach the point where the system is both acceptable and effective, as far as the organization is concerned. Objective measurement, such as that used by Lincoln Electric, makes everything easier, but at least when it can be used at the team or unit level, it can achieve more than traditional management performance reviews are likely to.

By adjusting objective measures of productivity and performance, or even subjective ratings based on both measured value and consensus or average peer and manager ratings, most of the benefits of P4P systems should be achievable. And considering examples such as the Safelite case, where productivity increased by 44% with only a 10% increase in wages, these benefits are definitely worth pursuing. 173. E. Lazear “Performance Pay and Productivity” American Economic Review 190:5 Dec 2000 1346-1361

Adjusting Productivity and Performance Estimates in PHI

2009-10-28T09:07:00.000-07:00

There is no question about the difference between objectively measuring and subjectively estimating employees’ productivity and performance. Objective measurement is enormously more useful and valuable, for both employers and employees. Subjective estimation, by either employees or managers, is so far back as to not deserve being in second place. It it is possible and inexpensive enough to do, productivity and performance should both be objectively measured, rather than estimated by any known method.

On the other hand, if objective measurement were easy and inexpensive, it probably would be far more widely adopted. Of course, some of the resistance to measurement may come from managers. Objective measurement can significantly reduce the power and importance of managers, particularly immediate supervisors. In organizations where objective measurement applies to most or all employees, there tend to be far fewer managers, since employees can more easily manage themselves.

Lincoln Electric, in Cleveland, Ohio, is one of the more well-known examples. Its employees are measured on their productivity, quality, and “citizenship” contributions to improving organizational performance. Its managers are trained in doing the work of employees, as well as in management. And it gets by with roughly 1/14 as many managers as is common for firms in its industry, while paying employees more and getting far greater productivity and performance from them. [“The Lincoln Electric Company” Harvard Business School Case Study # 9-376-028 and “Archive for Feb 27, 2002 FuturePositive.synearth.net]

While managers’ “performance reviews” are apparently the most common method used, and the closest most organizations come to measuring performance, there is little support for this as the best, most accurate or most effective approach. Employee self-reporting is the second most common method used, though it can be overly generous when employees see either reward or punishment potential in how their reports are used.

For example, when experts compared the self-reported productivity impairment among agents in a large call center operation, where productivity was also objectively measured, they found that agents over-reported the extent of their impairment by 2.5 times the objective measure on average. [G. Pransky, et al. “Performance Decrements Resulting from Illness in the Workplace” JOEM 47:1 Jan 2005 34-40]

When self-reported estimates can be compared to objective measurements, the adjustment needed can be made easily based on simple arithmetic. If employees over-report their impairment, improvement, or current levels of productivity or performance by 2.5 times, their reports can be multiplied by 40% (1 divided by 2.5) to yield a more accurate adjusted level. Of course, ideally, the extent of over-reporting by individuals should be adjusted by their individual degree, rather than the average for all employees.

If the call center employees are found to over-report their productivity or performance by some degree, such as the 2.5 x exaggeration in the above example, the same degree of over-reporting could be used to adjust the self-reporting of other employees, where their efforts are not objectively measured. But the other employees may not exaggerate to the same extent at all, may be far more or less accurate, may even under-report, for example.

In any case where objective measurement is not feasible or simply not practiced, as well as all cases where such measurement is applied, the measures should be adjusted based on objectively measured value that should be similarly affected by measured or reported productivity or performance improvement. If workers are to be paid more because of their productivity/performance, the additional payment should accurately reflect the actual value their efforts produce, not simply the measured improvement therein.

For example, if productivity measures show distinct improvements via self-reporting or objective measurement of output per se, there may or may not be a corresponding increase in value yielded, since all that is produced may not be sold. It could be simply added to inventory, or result in such an increase in stock on hand that the employer has to reduce prices in order to sell it. In such cases, it only makes sense to adjust the increased productivity in terms of the reduced revenue per sale.

If improved productivity enables reductions in the workforce, then the estimated or measured increase in productivity should logically be adjusted to reflect decreases in labor costs that result. In the current economic slide, for example, productivity has often increased by necessity, as workers work longer and harder to make up for those who have been laid off. In such cases, the remaining, harder-working employees should be paid more, based on their productivity and labor cost reductions, so that they will not decide they have been ripped off and leave at the first opportunity when the job market improves.

When performance improves, as reflected in objective measures such as new business, improved customer satisfaction and loyalty, reduced injuries, errors, etc., the value of such improvements should be used to translate performance improvements among workers into value added. This value added should then be used to calculate pay-for-performance bonuses or raises, or both, depending on how the P4P system works.

The measured value of improved productivity and performance is what most closely and accurately aligns the value that the organization derives from improved productivity and performance, rather than the improvement in productivity/performance per se. By adjusting measured productivity/performance based on value contributions directly linked thereto, both employers and employees can be sure that they are using the most appropriate and aligned P4P system, and most accurately share in the gains enjoyed by the organization.

Outcomes vs. Value for PHI Investments

2009-10-26T08:52:00.000-07:00

Whether labeled “outcomes”, “results”, “effects”, “consequences”, or any synonym, it is increasingly a requirement for investments in time, effort, and resources, particularly financial ones, to report what happened after they are made. But there is often a great gap between the outcomes reported and the value that they contribute – to the organizations that make them, as well as the individuals who deliver them.

In a recent report I read, McGhee Productivity Solutions cited the following “client results” it had contributed to achieving:
• 11 hours saved weekly per employee
• 19% saved time from meetings
• 32% reduction in time spent doing e-mail
• 33% time saved in looking for information
• 60% time saved from interruptions
• 78% reduction in number of e-mails in inbox
• 6 more hours/week spent in goal-directed activities
• 30% improved ratings in work/life balance
• 30% improved ratings in perceived sense of control
[R. Pierce comment to posting “Healthy Workplace, Employee Wellbeing” on Wellness Is a Business Strategy Group at Linkedin.com Oct 11, 2009]

It was not clear to me whether these results represented a collection of effects noted by different clients or the average consequences found across all clients, but the list clearly reflects a wide range of outcomes that most clients would welcome. But few of them can readily be translated into value, either for the employer or the employees affected.

The numbers of hours saved weekly per employee represents a potential value for employees and employer, but only if they are translated into activities that yield value for either or both. The value may be stated in terms of improved productivity or performance, or any other balanced scorecard as far as the employer is concerned. It may also be translated into opportunities afforded to employees to reduce the time they spend at work and increase time spent with family or engaged in other preferred activities.

The percentage reductions in time spent on particular activities cannot be translated into any value contribution without knowing what the percentage amounts to in actual time. The reported 19%, 30% or 60% reductions in time, if the baseline time spent on particular activities is measured in minutes may represent very modest value, and only if the time saved is spent doing something more worthwhile.

The most valuable-looking effect noted was the six more hours spent in goal-directed activities, but even there the value may be gained by the employer or the employee, depending on what the goals involved are. The 30% improvements in employee perceptions of work/life balance and perceived control should translate into improved morale, perhaps engagement, retention, and other valuable metrics, but in and of themselves, are tough to put a value upon.

Of course, it may well be that McGhee Productivity Solutions reported only the metrics it collected in assessing its performance, whereas its employer clients measured a variety of value dimensions, which varied by client, so could not be aggregated. In any case, these results illustrate one of the key challenges for all efforts and investments made to improve productivity or performance. Value perceived by employees is vital to long-range value for employers, and value for employers should almost always be translatable into dollar gains that can be compared to dollar investments.

Many other outcomes can help to justify investments, even when they fail to pay off in dollar terms. But unless the dollar payoff is known, along with the dollar costs of the effort, there is no way for employers to objectively evaluate what value they are getting for their money.
Translating results into measures such as dollars saved in labor costs, improved productivity, increased customer satisfaction, new business, etc. – in business value -- is likely to be far more persuasive than percentage changes in time spent or other process indicators.

Health Reform without Prevention and Wellness?

2009-10-23T08:34:00.001-07:00

With the population of the US already enjoying high ranking in world statistics for poor health, you'd think health reform would be replete with investments in programs aimed at reducing health problems by reducing the incidence and prevalence of risky behaviors, risk conditions, diseases and injuries in the first place. But with CBO and other criticisms of the cost-effectiveness of "prevention and wellness", along with predictions that it would take years for any significant effects to be achieved, it is perhaps no surprise that reform is scheduled to proceed with little or not such investments. {T. Theobald "Health-Care Reform Plans Neglect Wellness Prevention" The Tennessean Oct 23, 2009 (www.tennessean.com)]

We already know that "voluntary" health behaviors represent as much as 60-70% of the causes, or at least risk factors for sickness care costs, and that health risk and health-related factors are major causes of productivity and performance impairment, hence significant sources of competitive challenges in the world economy. But we appear to have little confidence that such behaviors can be changed in time to do much good, or that the good to be done will take to long to realize, in practice.

This lack of confidence exists despite the fact that employers around the world are already investing billions in maintaining and improving the health of their workforces. With most of the countries that compete with the US in the global economy already "blessed" with some form of national health insurance, employers in these countries are increasing their investments in employee health based on expectations and experience of improvements in productivity, performance, and retention of employees.

One of the major problems that academic experts and the government analysts who rely upon their advice face in evaluating workforce wellness efforts is that they rarely, if ever, use random control trials in evaluation of such efforts. Employers are far too practical to waste half of their population in control groups to ensure their evaluations meet academic standards for scientific rigor. I recall a major study undertaken by the Wisconsin State Education department more than ten years ago that used a large number of school districts in the state as controls when studying wellness investments in other districts. While the study called for multiple years of comparisons, the results achieved in the first six months were so positive and so dramatic that it dropped the control group idea and enrolled all school districts in the wellness effort. Anything less would have seriously reduced the cost savings it needed to achieve.

It may be that we lack confidence in our ability to "manage" consumers well enough to achieve significant savings and net ROI. After all, it seems to be universally agreed that the best way to engage large portions of the populations at risk is to bribe them with incentives to participate, and these incentives often become the highest source of costs for wellness/prevention investments, automatically threatening ROI. But recent research in the domain of "brain science", relative to both marketing and management, indicate that money is one of the weakest sources of motivation in the long run, in addition to being the most expensive approach in most cases. {C. Mattie "Money as a De-Motivator" Human Resources IQ June 11, 2008 (www.humanresourcesiq.com)]

Impatience is clearly another challenge in discussions of PHI's potential. Most of the data I have reviewed over the past fifteen years of my involvement in PHI have indicates significantly higher financial returns in the second year of programs than in the first, and in the third compared to the second. If costs are not controlled well enough, first-year results may well be disappointing, and investors simply give up trying. GlaxoSmithKline, one of the few employers to report results in the same employee cohort for as long as four years, calculated first-year savings of only $233 per participant among the 6000 or so employees followed. But savings in the second year were $375, and then $944 and $950 in the third and fourth years. [316. G. Stave, et al. “Quantifiable Impact of the Contract for Health and Wellness” JOEM 45:2 2003 109-117]

Unless we are successful in reducing the incidence and prevalence of illness and injury in our population, we have little or no chance of significantly reducing our sickness care system costs, nor of retaining our ability to insure our population against such costs. There are surely some reforms in how sickness care is delivered that could save significant amounts, but most likely only in the form of "bending the curve" of the trend of increasing costs that has plagued us for as long as any can remember. If we want truly significant savings, we need to apply, or if we truly lack cost-effective methods, to develop some, that will significantly, preferably dramatically, reduce sickness, itself.

How "health reform" can be achieved without achieving such savings is totally beyond my power of comprehension or confidence. It should probably be beyond the imagination of reformers as well, though in politics it is usually the most persuasive lobbies with the strongest pull that make the difference. If there has ever been a situation where persuasion, i.e. sophistry and politics is not the best approach to solving the problem, it surely is the sickness care crisis. As it is now, the chances of success for the reforms thus far being considered in Congress represent little or no hope for success in solving the problem, even if they succeed in being enacted into legislation.

WHat Will the Newest Consumer Do for PHI?

2009-10-23T08:03:00.000-07:00

The idea of a “new consumer” has been around for a decade or two already. These consumers are more informed about choices available in the market, more demanding, tougher to satisfy, and more insisting on control of how others market to them. Some have even described the change in terms of a “buyer-centric market”, where consumers will rely on their own devices or professional buyers’ advocates and agents to control who they deal with and what information they will accept from sellers.

But there is yet a newer consumer on the way, if not already here. What distinguishes this consumer is a direct consequence of the recent economic downturn, which may never recover to the former level of glory in terms of employment and consumer spending levels. [L. McPheters “Consumer Spending and the ‘New Normal’ Economy” Knowledge@W. P. Carey (School of Business, Arizona State University) Oct 14, 2009 (http://knowledge.wpcarey.asu.edu/article.cfm?articleid=1820)]

The newest consumer will be significantly more frugal, buying less and saving more, according to a growing consensus among economics and business experts. Instead of an economy based on growing credit and consumer debt, spending will be relatively stagnant. Experts disagree about how severe the effects will be, and how long this lower spending pattern will last, but already see the signs in consumption levels reported during the past two years.

This amounts to a “correction” in a trend that has been followed since 1960, during which consumer spending as a percent of the nation’s GDP had increased from 60% to 70%. This higher level is significantly greater than in most other developed countries, where levels tend to range from 55% to 65%, and far greater than in the fastest developing economy in China, where it was only 36% in 2008.

Whether this percentage declines to former levels, or drops even further to the levels prevailing in countries such as Germany (56%) or Japan (58%) is open to debate, but few predict the level of spending will continue the past trend, at least not for a while. The slow recovery of employment levels compared to stock market prices tends to reinforce more pessimistic forecasts.

But when it comes to PHI, the “new frugality” may tend to support consumers’ willingness to invest their time and effort, though not, perhaps, much of their money) in maintaining or improving their health. Since health is at least one of the significant factors that determine how well consumers produce and perform in their jobs, it will clearly have impact on their careers and income. With increased competition for fewer jobs, consumers in the best of health may well do better in the job market, despite regulatory proscriptions against discrimination based on health.

There may be a second, indirect effect as well. Consumers have already demonstrated a noticeable reduction in their use of health services, along with the increase in the number of non-paying patients. Both will threaten the financial security of health care organizations (HCOs), and even health reform may not create the level of payments needed for HCOs to thrive.

Already, the majority of HCOs offer their own employees some sort of PHI programs. It would not take much for the number of HCOs that offer the same kind of programs to other employers in their markets to increase. If HCOs learn how to deliver cost-effective services and competitive levels of ROI to employers, or even to health insurance plans, they may find PHI a significant new source of revenue in a declining or stagnant sickness care market.

After reaching the absurd level of 130% of personal income in 2008, consumer spending is clearly heading toward a lower level. As consumers increasingly care more for their “wealth assets”, it only makes sense for them to do the same for their “health assets”, as well. Compared to the notorious stinginess that government and commercial insurance plans have shown with respect to sickness care, PHI programs, that deliver cost savings to insurers, as well as performance improvements to employers, figure to deliver significantly higher levels of “generosity” based on their clear financial returns.

In Praise of Discrimination in PHI

2009-10-13T09:55:00.000-07:00

We have long held diverse views on the subject of discrimination. Being a person of discriminating tastes is supposed to be a good thing, though it usually requires far more than an average income to indulge such tastes. Being able to discriminate between right and wrong, justice and injustice, safe and dangerous is essential to being a good citizen and person.

But discrimination against some, favoring some at the expense of others is generally held to be un-American, and illegal or at least unethical and immoral. Discrimination is definitely a naughty thing in the context of using incentives in PHI, for example, where a host of laws and regulation prohibit discriminating against people who are “disabled” by obesity, “addicted” to smoking, alcohol, or drugs, and are deemed unable to control their behaviors.

But the financial realities that attend PHI and the entire range of health, wellness, and disease management call out for discrimination. If people have widely varying risks of being sick or otherwise impaired with respect to their performance at work, when said risks are voluntary behaviors or the results thereof, does it not make sense to favor those who make the healthiest choices instead of the unhealthiest?

When people vary widely in terms of both the value of their potential health risk and disease reform, as well is in the probability that such potential will be realized, does it not make sense to invest differently in their efforts? At least one employer of which I am aware empowers its employees and their dependents who are members of its self-insured health plan to choose their own health mission and goals. But it then reserves the right to decide how intensive and expensive will be the support if offers such members in their pursuit of such goals.

Another recognizes that different kinds of health risks and disease conditions represent different levels of difficulties for those who commit to managing them. Because of this, it offers greater incentives where the difficulty levels are higher, so that it avoids discrimination in terms of the prospects and payoffs for success among its workforce members. [“Vitality – Changing Behavior through Incentives” The Vitality Group 2009 (www.powerofvitality.com)]

When incentives are limited to 20% of healthcare premium costs, as is the case today, there is not that much room to discriminate rewards compared to the variation in potential x probability value across the workforce or insurance plan membership. But if proposals in health reform that would raise that limit to 50% are included in reform laws, there will be far greater opportunity for appropriate discrimination, though the laws may also make it impossible or difficult to take advantage of the opportunity.

When dealing with a population of workers who are paid at widely varying levels, it is normally the case that lower-paid workers will be influenced by lower levels of incentive payments than are higher-paid workers. When higher-paid workers are also more valuable to the employer, and why not when they are paid more, does it not make sense to discriminate among workers based on their pay levels? Incentives that offer the same percentage increase in their pay foe achieving a given wellness, risk or disease management goal would automatically be discriminatory as to amount. Whether such offers would violate laws and regulations is something that lawyers can advise on.

It makes eminently good sense to discriminate among prospective participants in wellness, risk, and disease management efforts in terms of the extent to which individual employees have problems or potential for improvement in such arenas. Yet at least one legal “expert” has warned against doing so in terms of marketing programs aimed at such programs. He asserts that “An employer has to be careful that the program is not advertised in such a way as it appears to be targeting particular individuals, or individuals with disabilities.” [C. Wooten “Wellness Warnings: Programs Must Not Be Seen as Mandatory or Discriminatory, Expert Says” Houston Business Journal Oct 2, 2009 (http://houston.bizjournals.com)]

The same expert warns against offering wellness programs and activities at the worksite, lest any accident or other injury that affects workers be deemed “work-related” and have to be paid for by the employer. Incentives that relate to behaviors away from the worksite, such as smoking, and perhaps even over-eating, not exercising, not sleeping enough, etc. – which are known to impair work performance but are not “work” behaviors – may be charged as invasions of privacy.

Fortunately, there is at least one domain in which discrimination has always been legal, and to degrees that often boggle the mind – in how much we pay different workers, particularly executives vs. everybody else. Since improved health, reduced risks and productivity/performance impairment, improved self-management of chronic conditions are all linked to financial gains for the employer, the incentives and rewards therefore could all be embedded in pay-for-performance systems.

The federal government is, itself, engaged in a discriminatory incentive program that pays hospitals, physicians and patients alike for differential levels of success in its Acute Care Episode (ACE) demonstration project covering Medicare beneficiaries. Five hospital systems are currently participating in this project, which offers to share cost savings the government gains with those who contribute thereto.

While the project is equally focused on improving quality, it is the cost savings resulting that determine what amounts will be shared. There are 22 quality measures to be monitored and reported to ensure that cost savings are not achieved at the expense of high quality. Hospitals receive the cost savings incentive payments from CMS, then can share up to 25% of the physician fees involved with physicians, and up to $1157, the Medicare part B premium, with patients. Medicare gains a discount of 1 - 6% on the fees it pays the hospital, while the hospital gets whatever is left after sharing with physicians and patients. [D. Finley "Providers Nationwide Watch Medicare Experiment Here" MySA MySanAntonio.com/news/64040247.html)

Of course, in this arrangement, there is no direct discrimination by Medicare, though since savings will vary widely across patients, as well as physicians and hospitals and the “cases” they take responsibility for, there will necessarily be discriminatory levels of incentives paid. Discriminating among people based on different levels of value they contribute to payors is an essential element of the free market, and is permitted among both buyers and sellers therein. Why should it not be not merely included, but recommended, when properly applied, in PHI?

Positive and Negative Effects of the Economy on PHI

2009-10-12T09:56:00.000-07:00

While the economic downturn has abated significantly, with respect to consumer buying behavior and confidence, as well as stock market averages, it continues to have significant effects on performance health management efforts. While some employers have continued the trend of reducing their health investments, many others have increased or at least continued their pursuit of the widespread and significant financial gains available through PHI.

As for employees and consumers in general, one might have thought the economic uncertainty might have made them more likely to become more prudent managers of their health. But while it has made them less enthusiastic purchasers of health care, they have reduced their use of preventive services and proactive health management care when it costs them money as well.

And if anything, they have become less careful about their personal self-management. The threat of or reality of job loss and home mortgage foreclosure has added to the already high level of stress at work. Moreover, the same employer prudence that makes them slow to hire back previously laid off employees even when business improves is increasing the stress on workers who have retained their jobs because of higher workloads and productivity expectations.

The tough economy has led significant numbers of employees to cut back on their healthy food purchases, and made them more susceptible to the appeal of unhealthy “comfort” and fast-foods. Many have cut back on exercise due to longer work hours, and desire to save money that gym and fitness center membership or use add to their expenses. Many have also cut back on vacations, in order to augment their income through unused paid time off repayment, adding to their stress.

In the UK, for example, a survey of employees revealed that 15% of respondents were working longer hours, 7% were drinking more alcohol, 6% smoking more, 19% sleeping less, 57% buying less healthy food, with 15% reducing their fruit and vegetable purchases, and 21% using exercise facilities less. [“Health and Well-Being News” Health & Performance Management (UK) 1:8 2008 p.22 (www.vielife.com)]

One positive result of the downturn has been a reduction in the expected number of physicians retiring, as their retail investments are often in as bad shape as are those of other workers. This could make it easier for workers and other consumers to obtain health and disease management services from their personal physicians, and should support the growing development of medical homes which typically include at least some such services. [“More Docs Put Off Retirement as Financial Crisis Lingers” Modern Physician Oct 12, 2009 (www.modernhealthcare.com)]

Overall, it seems likely that the downturn has had generally negative impact on employee and consumer health, in general, and with unemployment one of the slowest elements of the economy to recover, this seems likely to continue. Employers are still evidencing a healthy commitment to PHI, though whether this will survive current movement toward health reform remains to be seen. The experience of Europe would suggest that PHI will generally deliver value to employers willing to invest in it, even when medical/hospital insurance is no longer their concern.

The movement of physicians toward medical homes may prove a positive development that is continuing during the downturn. How much they and their “homes” will emphasize general health vs. narrow disease management, and how much they will do toward employee health and performance remains to be seen, and these domains may remain beyond the interests and capabilities of the medical profession, though nurses and particularly nurse practitioners may take it up more enthusiastically.

Specificity vs. Generality in PHI: Risks

2009-09-18T11:39:00.000-07:00

The risks involved in PHI strategies and interventions are among the least well understood elements thereof. The nature of risks, and their significance are far more complex than are traditional “problems”, or “goals”, because they involve statistics and their interpretation. One difficulty of dealing with risks is the labeling and interpretation of the different kinds of statistical realities they indicate.

For example, an article I read today on the subject of prostate cancer asserted the rather disturbing prediction that “Every sixth man will contract prostate cancer this year…”. [J. Ramsey "Free Prostate Cancer Screening"TOPNEWS Sep 18, 2009 (http://topnews.us.content/2718-free-prostate-cancer-screening) As a man, this naturally disturbed me quite a lot, until I thought about it for a minute and realized it was probably a case of mixing up lifetime risk with annual risk. It seems perfectly plausible to me that roughly 16% of all men will contract prostate cancer during their lifetime. The thought that 16% of all men will contract it each year would scare most men and all health insurance firms a lot, and indicate that I am well past due, since I have already lived a large number of six-year periods.

It helped that later on in the article, it was mentioned that: “Every year 185,000 men are diagnosed with prostate cancer”. Since 185,000 men is by no means one-sixth of all adult males in the US, this clearly denotes that it is lifetime risk we are dealing with, not annual risk. But risk is dependent for its meaning on the time period it covers, and in PHI, at least, it is not the statistical risk number as much as the risk (or productivity/performance) impairment factor that is of greatest concern. And normally, it is not just the risk of something happening, but the effect of its already being in place for some significant number of people that makes it important.

Risk behaviors (smoking, drug use, physical inactivity, poor diet, poor sleep habits, etc.) and risk conditions have at least two types of meaning in PHI. For insured populations, the risks are mathematical expressions of the likelihood that individuals identified as having each risk will end up with a chronic or acute disease or injury, and cost a lot of insurance expense, in medical/hospital care, disability or workers comp. This may extend to risk of turnover, if health condition precludes continued employment in the same job.

The statistical risk of contracting a given disease or injury can be predicted based on reported behavior and biometric screening indicators, or the existence of a given “risk” can be linked to measured impairment of productivity or performance. In the latter case, the risk is not a statistical prediction but a measure of present reality – people with the risk factor identified will be, on average, less productive and valuable workers than those without it.

The great difficulty that arises with risks that are related to productivity/performance impairment, whether behaviors or conditions, is that they rarely occur alone. In the typical population, with as few as a dozen or so of the common health risks measured or reported, there will usually be less then 5%, and often less than 1% of the population who have no risks at all, and not many more who have only one. Studies have shown, for example, that fewer than 5% of most populations meet four standards for healthy weight, not smoking, eating desired numbers of fruits and vegetables and getting enough exercise.

Because risks tend to occur together, with the average person having as many as half of the number being counted, it is difficult, often impossible, to determine the impairment effect of any individual risk. In theory, those members of the population who have only one risk would be the preferred basis for estimating this specific effect, but unless the population is very large, there may be far too few in a given population who have only one risk, while covering all the risks of concern.

In practice, all members of the population are counted with all their risks measured or reported, and the measured or self-reported impairment of each member is recorded for each. But that also means that the degree of impairment for each individual is counted for each risk each has, up to as many as the total number of risks being counted. In one example, the pattern of risks and impairment levels of over 200,000 people, of whom only 1.4% had none, and less than 5% had just one, ended up multi-counting the impairment level of the entire population by over 300%.

Fortunately, this over-counting can easily be overcome. The total level of impairment can be counted by summing up the population where risks are reported in ways that count each individual only once. This happens, for example, when the population impairment level is reported across the number of risks each member has. Since each person will belong to only one category of number of risks, each will only be counted once, and the total impairment will be accurate for the entire population, assuming measured or self-reported data from each is accurate. Or, of course, the total impairment attributed to all the separate risk factors can be divided by the degree of over-counting, and a decent estimate generated that way, since this directly accounts for the over-counting.

But the best way for overcoming over-counting of risk effects is to count the impact of PHI interventions. Whether members of the population are enlisted in pre-designed intervention programs determined by the employer or provider – or are empowered to choose their own program, the results of that program should only count the results for each participant once.

An exception arises if members can enroll in more than one program at a time, or during the same evaluation period. When this happens, then there is a risk that the overall effects of the PHI program will include over-counting, just as the overall impairment levels are over-counted. In most cases, this does not arise in practice, and in any case, as long as the improvement in productivity/performance and reduction in insurance expenses are each counted only once for each participant, there will be no over-counting in the total.

The only flaw in the situation will be that where population members can and do enroll in more than one intervention during the same evaluation period, the effects of individual interventions may be over-counted. Since the total PHI impact will not be, however, this can be adjusted by any agreed-upon approach, such as dividing the improvement achieved for each individual by the number of different programs each participated in.

The greater issue related to specificity vs. generality of risks arises in cases where members of the population are labeled according not to the specific risks or impairment factors each has, but the number of such risks each has. This has been a common practice in reporting results of initial population assessments and evaluating impacts of PHI interventions. Populations are typically segmented by risk numbers into “low” (e.g. 0-1 risk), “medium” (2-3 or 4), and “high” (4+ or 5+) risk segments.

Those in the low-risk segment may be assigned to minimal effort/cost interventions, perhaps website or online coaching based on automated communications. Those in the medium-risk may be assigned the same plus additional efforts, perhaps mailed materials, support groups or onsite group sessions. Those at high-risk may be assigned to personal health coaches either onsite, where the employer has an onsite medical clinic, for example, or by phone, if not. Success would be tracked in terms of the reduced percentage of the population that remains in the high or medium risk, compared to those who improve to the medium or low categories, along with reductions in insurance spending related to each. (This approach is not usually used with productivity/performance measures)

Such a general vs. specific approach challenges the provider of interventions, since the people enrolled in each level of intervention share mainly the same number of risks, rather than the same risks. There may be many situations where the essential behavior change objective is the same, e.g. adherence to medications, but when people have different risks, there may be difficulties making communications with all of them relevant and effective. This is not a problem with phone coaching, of course, since coaches would know what risks each participant has, and could be sure to address each differently for each participant.

It may be possible to use a fairly general intervention for most disease management interventions, where the major focus is on adherence to medications and lifestyle regimens. Automated computer analysis and generation of customized coaching content to be delivered by mail or online is simple and inexpensive enough to use even with low-risk participants. Participants can customize their own website interactions, and work with coaches to ensure their unique needs and preferences are respected.

In general, however, it is clear that the more that coaching and support services provided to the individual can be customized to each’s unique problems, challenges, and preferences, the better results tend to be. This same reality has long been recognized in consumer marketing, and PHI shares a lot of similarities with marketing in terms of both striving to influence the behavior of diverse populations.

Specificity vs. Generality in PHI: Part 2 Behavior Changes

2009-09-18T10:29:00.000-07:00

Part 1 of this discussion pointed out the PHI challenges related to specificity vs. generality of therapies, or the interventions applied to the population at risk, whether employees or insurance plan members. Care must be taken to avoid being either too generous or too narrow in identifying, predicting, and evaluating PHI results – to be sure that the “effects” of interventions are truly attributable to such “causes”.

The same admonition applies to the direct objective of PHI interventions, namely to altering the health-related and performance-related behaviors of workforce and insured populations. While specific behavior changes may be set as goals for individuals and populations, these changes may have far-reaching results that go well beyond the original reason for promoting and enabling them.

The example of smoking cessation has already been cited in Part 1 of this discussion – when smokers overcome their addiction and no longer have to take smoke breaks away from their workstation, they will often improve their productivity and performance due to that particular behavior change, not because of any health improvement. Behaviors often have far- and wide-reaching effects, extending well beyond the original reason for promoting or extinguishing them.

Exercise is one of the best examples. While only a minority of the population “adheres” to recommendations for ongoing physical activity, such activity is one of the healthiest things people can do. It may be recommended as part of a weight loss regimen, for example, but its effects are likely to be far greater on other health and life quality dimensions. Dietary intake is far more important in weight management than is exercise, for example, though exercise has as great a set of extended positive effects.

Exercise has been credited with having significant life-extending and life-quality enhancing effects, at any age, though the earlier good exercise habits are initiated, and the longer they are followed, the better. Exercise is a significant help in the management of depression, as well, and depression is one of the more significant life-quality as well as productivity and performance impairment factors, as well as a far too common problem in most populations.

And since depression is a common co-morbidity in a number of chronic diseases, including heart disease, stroke, cancer, and diabetes, including exercise recommendations and support in DM programs aimed at these conditions, as well as for those whose physical indolence is merely a risk factor, makes good sense. And physical fitness, which is a desired effect of exercise, is also helpful in reducing work accidents and injuries, thought exercise can cause injuries, particularly among those not yet physically fit.

Enabling members of a population to master the skill of controlling their own behavior -- whether this involves overcoming addictions to tobacco, alcohol, or drugs or adopting new healthy habits such as exercise, stress management, or healthy eating – can help them with other life challenges as well. It can enhance their self-esteem and respect from peers, and give them confidence relative to other health and life challenges they may face.

One of the elements of the PHI participation experience may well be that of joining a “healthy community” of peers who are pursuing the same or similar health behavior goals. This can not only help participants make better and faster progress toward their goals – it can also improve their overall relations with fellow workers, and between managers and workers. This can help in team efforts and in overall corporate culture and climate improvements, while costing employers virtually nothing beyond the built-in costs of PHI programs.

Empowering employees to do better at coping with or reducing worksite stress or work-life imbalances can also greatly improve their productivity and performance, in addition to reducing health, disability and workers compensation insurance/expenses. This may require, or at least benefit greatly from changes in work processes and policies, such as enabling workers to have greater control over work demands and their own efforts, in addition to enabling them to gain skills. But since high levels of stress is a major cause of both productivity/performance impairment and turnover, enabling workers to learn or adopt this healthy behavior can have widespread positive results.

Enabling workers to learn or adopt healthier sleeping habits can be another significant accomplishment for them, and a major source of cost reduction and performance improvement for employers. Both too little and too much sleep are separate but significant causes of safety, health, and performance problems, as well as of reduction in life quality for workers. And whenever employers enable their workers to improve their own life quality, that is likely to pay off in terms of worker morale and retention.

As was the case with PHI interventions, themselves, the value of positive changes in employee behaviors, both health and life-quality related changes, can have impact far beyond the immediate objective thereof. The full effects of behavior changes should be identified – to both employer and employee, in order to both promote and evaluate the total benefits of such interventions, while guarding against being too generous with cause attribution and ignoring the effects of other programs and interventions that may also be affecting value achieved.

With employees, themselves, tracking and reminding PHI participants of the personal as well as organizational benefits they are contributing and gaining should be a major element of PHI programs. To the extent that employees anticipate and are aware of significant personal benefits they gain from PHI, dependence on extrinsic incentives for motivating their participation should be reduced, making PHI investments that much more likely to generate positive returns.

Specificity vs. Generality in PHI: Part 1 -- Therapeutic Interventions

2009-09-18T09:38:00.000-07:00

There is a “rule” or at least strongly recommended convention in predicting and evaluating the effects of “treatments” or therapies, called “therapeutic specificity”. It holds that only the effects that evidence has shown are truly caused by a given treatment should be counted as the differences it makes to those treated.

In a classic example of failure to consider and apply this rule, the “number needed to decrease” the use of treatment was employed to calculate the prospects for disease management (DM). [A. Linden “What Will It Take for DM to Demonstrate an ROI?” Disease Management Colloquium May 2006 (www.ehcca.com)]

When applied to diabetes patients, it was determined that number of diabetes-specific hospitalizations would have to be decreased by at least 74% in order to make up for the costs of the DM effort. While such an effect could result, in theory, it might be so unlikely as to cause those considering that specific DM investment to reconsider.

It might also cause them to look more favorably at other DM investments where the number needed to decrease utilization and costs would be more likely. [A. Linden & T. Biuso “In Search of Financial Savings from Disease Management: Applying the Number Needed to Decrease (NND) Analysis to a Diabetic Population” Oregon Health & Science University, Portland, OR (undated working paper)]

Of course, care must be used when applying the principle of therapeutic specificity that the specificity requirement should focus on the therapy, not just the disease. With diabetes, for example, the “treatment” used in DM often includes therapies and lifestyle change recommendations aimed at common co-morbidities of diabetes – e.g. depression, obesity, high blood pressure and cholesterol.

Treatment for such co-morbidities would usually be expected to help reduce risks of heart disease and stroke, in addition to the common complications of diabetes such as eye and kidney disease, circulatory problems that might lead to amputation of limbs, for example. In cases where the treatment induces effects beyond those limited to diabetes, the full effects that are logically attributable to that treatment should be counted, or at least considered as likely outcomes.

When treatment for diabetes or other chronic diseases includes therapies that have effects well beyond those normally attributed to the condition, itself, it makes sense to also consider the impacts of such treatment on outcomes such as employee productivity, performance, and turnover, as well as medical care costs. With diabetes, for example, treatment for its common co-morbidity depression could have dramatic impact in reducing productivity and performance impairment, as well as turnover.

In fact, because the effects of depression are so great on productivity and performance, it is possible, and often likely, that the full scope of DM interventions aimed at employees with diabetes, perhaps even including DM that is used by dependents of employees, will achieve economic benefits in the presenteeism and absenteeism domains. Dow Chemical Co., for example, found depression to be one of the most expensive causes of lost productivity, and its impact on absenteeism and presenteeism were calculated at almost twelve times as great as medical care costs alone. J. Collins, et al. “The Assessment of Chronic Health Conditions on Work Performance, Absence and Total Economic Impact for Employers” Journal of Occupational and Environmental Medicine June 2005 547-557

On the other hand, of course, it is certainly possible that the “co-morbidity” of depression in some, perhaps many cases, was not caused by diabetes at all, but had preceded it in time, and was not affected thereby. Or it may have already existed, but been only exacerbated by the addition of diabetes. In any case, it is the therapy, involved, i.e. the full scope and total elements of the DM intervention, whose effects are to be determined – not merely those that are directly related and attributed to treatment of diabetes.

This same logic applies to all DM and PHI interventions. Regardless of what the medical diagnosis, risk behavior/condition, or wellness dimension is the target of the intervention, it is the total set of outcomes attributable to the intervention, not merely those directly related to the problem, that should be counted. In any workforce application, DM or other PHI interventions may be only part of the “treatment” aimed at improving worker performance, but the full effects of the intervention, not merely the targeted condition, per se, should guide what outcomes are attributed thereto.

For example, when the PHI problem is smoking, one of the most significant and widespread health risks, the effects of smoking cessation interventions may include effects that have nothing to do with the employee’s health. Productivity and performance may improve among smokers because they cease to take “smoke breaks”, which often have to be taken outside the work environment, or away from the workstation, and automatically reduce productivity and performance.

In such cases, the cessation of smoking would logically be identified as a probable cause of improved productivity and performance, well before and independent of any predictable health improvement. Offering pay-for-performance (P4P) bonuses for such improvement may well be a contributing factor, but the most immediate cause would often be smoking cessation and the particular PHI intervention that made that possible, possibly including nicotine-replacement therapy – unless, of course, the P4P system were adopted at roughly the same time as the PHI intervention began, and was a major cause of smokers attempting to quit.

The degree of specificity for evaluating traditional medical therapies has long included concern over the “placebo effect”, where patients’ belief in and anticipation of positive effects has a separate and often dramatic impact on patients’ condition, separate from the effects of the treatment. The equivalent effect may be present in PHI interventions, when the added attention that the organization gives to workers may induce what has become known as the “Hawthorne Effect”, also independent of the intervention.

In general, it is always wise to consider, and to guard against or adjust for causes of desired outcomes -- both reductions in health, disability and workers compensation insurance/expense and improvements in worker productivity, performance and retention – as potentially arising from multiple causes. But the reason for doing so is nowhere near as narrow as those effects known to occur solely because of the PHI intervention’s targeted problem. Both the full impact of the “therapies” applied in the intervention, and the “placebo” effect of the organization’s showing interest in workers, can logically be considered results of the PHI investment, and counted when evaluating results.

Is Value-Based PHI a Dream?

2009-09-14T07:36:00.000-07:00

Last week, on a Linked-in online group discussion site, the question was asked: “Are Biometric Screenings Worth the Investment?” [http://www.linkedin.com/groupAnswers?viewQuestionAndAnswers&discussionID=7025700&gid=89892&commentID=6482979&trk=view_disc] I immediately contributed my two-cents-worth, insisting that such a general question simply could not be answered. To be worth the investment, the circumstances have to be right = the value of the information gained through biometric screening would have to be worth the costs added to wellness efforts thereby.

Clearly, my opinion did not sway my colleagues who contributed their own arguments. Essentially all of them insisted that the need for objective, biometric data was so great that no wellness effort should be without it. Relying on far less costly health risk assessment (HRA) surveys, or on analysis of past healthcare claims, or even a combination of the two, apparently falls far short of the value of biometric metrics when assessing and monitoring the health of populations.

While I have no argument with the assertion that biometrics offer objective and current data of significant importance, even superior to other data in many cases, the original question still cannot be answered in the abstract. If the argument goes that biometrics, or any other element of a wellness program for that matter, delivers such great value, the validity of this assertion can only be discovered after the fact (a posteriori), when the investment made on the data analysis based on biometric data proves to yield an ROI that makes the investment worthwhile. Where such value can be predicted in advance for any situation where wellness programs are to be used, it has to be predicted specifically for that situation, not insisted on as a matter of principle.

While discussions of “value-based” investments in general are almost universally positive, the idea is also one that cannot be championed in the abstract without recognizing that value is the relationship between desired benefits and the costs of achieving them. It is not a relationship that can easily be predicted in advance, and it is one that can never be predicted with such certainty as to permit insisting on a particular element of wellness, such as "biometrics or nothing" up front (a priori).

In a healthy population -- where its members are already largely adhering to healthy lifestyles, and have healthcare expenditures and health-related productivity/performance impairment levels on the low side -- the costs of biometric screening of the entire population, including the costs of incentives needed to ensure the vast majority of members are measured, can easily exceed the potential gains from available wellness programs. If the insurer or employer sponsor only considers health care cost reduction as the financial benefits in the value relationship, or even if the wages paid to workers are on the low side, there may well not be enough achievable reductions in healthcare expense plus productivity/performance impairment to cover the costs of biometric screening plus planned wellness interventions.

Having spent almost my entire professional career in health care organizations and teaching about health management, I have long been familiar with the strong tendency in healthcare to insist that quality is king. If some people cannot afford quality health care, we should tax ourselves to make it possible. If the country can’t afford to deliver high quality care to everyone, so the healthcare reform argument goes, it should still strive to do so, hang the expense.

Well, we have already learned the serious flaws in any “hang the expense” position, even when applied strictly to sickness care. That is what got us into the mess we're in with sickness care. We insist that healthcare should not be rationed even as we recognize that paying for our quality-first healthcare system is unsustainable without serious and significant reductions in its costs. While town meetings feature angry shouting matches around “death panels” and other imaginary examples of using comparative cost-effectiveness in allotting health resources, the unpopularity of reducing healthcare expenditures is clear.

But at least in proactive health management, the need for comparative cost-effectiveness should be eminently clear. If a proposed wellness program, or any single element thereof, is not expected to yield or lead to a positive ROI, why should the investment be made? If it will have such significant positive impact on the health of the population involved, why should the members of that population not at least pay for part of its costs, if that would be necessary to yield a positive ROI for sponsors?

There is simply no single element of PHI, nor any particular approach to any element, such as biometric screening for gathering baseline data, or phone coaching for wellness participants, that is so essential as to be imposed on every wellness investment before analyzing the prospective benefits and costs thereof. While employers have often invested in PHI in the past without credible predictions of positive ROI, even without any plans to measure ROI, this is a practice that is rapidly disappearing. If biometric screening cannot be justified in terms of positive impact on predicted financial benefits that are at least equal to its costs, it makes no sense to insist on it.

Even if there are serious flaws and limitations in self-reported risks and claims data, these may well be the only sources of data that can be afforded in particular cases. Insisting on biometric screenings without even considering the ROI impacts of using this method is simply replicating and continuing the hang-the-expense mentality that has led to sickness care being unaffordable. And since wellness and PHI are supposed to be part of the cure for our unsustainable sickness care system, it makes no sense to handicap the cure with any necessity insisted on a priori.

By definition, and because of the definition of “value”, there can be no value-based solution to healthcare reform, sickness care crisis, or any other major challenge we face without considering the benefit/cost relationships of the solution in the specific situation to which it will be applied. Biometric screening may well prove to deliver such significant improvements in the value gained through wellness investments that it should be included in the vast majority of cases. But this does not mean it should always be included in all cases – not until it is clearly justified by its predicted benefit/cost relationship in the specific situation for which it is suggested. The use of biometric screening should always be decided based on the evidence, not insisted on when there is no situation-specific evidence that justifies it.

Self-Management a Major Option in PHI

2009-09-11T09:02:00.000-07:00

Proactive health management efforts, including PHI, began with sponsors -- such as employers pushing worksite wellness or employers pushing disease management – and a continuously growing array of providers selling their particular solutions. But there has always been a much cheaper alternative – “empowering” employees and health plan members to manage their own health and their own efforts to achieve health improvements.

While the most popular belief relative to consumers’ managing their own health is that they should do so as part of accepting responsibility for being prudent users of health care as well as controllers of their own health – as an obligation they accept, perhaps in return for their employer or government-paid health benefits. But while responsibility may be a good subject for philosophical argument, it is relatively weak as a basis for actually getting people to behave in more healthy ways.

The most common approach to enlisting/enrolling consumers in managing their own health is to "bribe" them – offering incentives and rewards of various kinds in return for any number of specific behaviors. These may include completing a health risk assessment (HRA) survey or biometric screening program, enrolling in a coaching effort, making or maintaining specific health behaviors, or attaining/sustaining specific health status measures.

Of course, there tend to be many members, often a large majority of employee or plan member populations, who have such low levels of health care use or productivity/performance impairment due to their health that the costs of providers’ programs and incentives added for their participation exceed the potential savings. So the fastest growing sector of PHI seems to be in lower-cost approaches, such as e-mail and wireless coaching and web-site self-service methods.

PHI has already moved past a narrow focus on health care costs to add health-related and even comprehensive efforts to improve productivity and performance, as well as reduce talent turnover. It is also moving past the “selling” of one-size-fits-all “solutions” to empowering people to make their own choices among a wide range of community as well as PHI provider options. It has long been understood in marketing that enabling consumers to choose among options is far more likely to work than insisting that they buy the same one as everyone else.

In addition, PHI has moved beyond traditional “educational” and informational means for “correcting” consumer health behavior to modern marketing methods and the growing domain of brain science. As we learn more about the real ways people make choices and adopt behaviors, there are many more options available to both PHI sponsors and participants.

One of these involves the personal “bet” approach, combining the making of a public commitment to adopting and persisting in a given behavior or reaching a given goal, with an agreement to penalize one’s self if one’s behavior fails to match the commitment or goal. This approach is based on the well-established reality that people tend to give far greater weight to the risk of losing something they already have and value than to the potential for gaining something they do not but would like to have.

There is a website available, for example (www.stickk.com) that offers consumers a way of making a personal bet, selecting a “referee” who will monitor their behavior to see if they keep their commitment, plus joining a “social media” network of friends or peers who are pursuing a similar goal. The consumer makes all the choices involved, so is fully “empowered” in contrast to traditional sponsored efforts to enroll each in a particular program.

Because the bets that people make with themselves, where losses are often paid to charities or other worthy causes, or to “unworthy” causes such as a political party or campaign totally opposite to one’s inclinations. This latter approach adds even more negative consequences to “losing” the bet. People can make as many bets as they feel they can handle, and bet as much as they think they can afford while ensuring enough “pain” of loss to motivate them to fulfill their commitments or achieve their goals.

This betting approach has been used throughout the world, though only for a few years, it seems, and has had admirable success rates with a wide variety of health behaviors and goals: quitting smoking, losing weight, exercising regularly, controlling calories or portion sizes, reducing alcohol intake, getting enough sleep, even achieving specific improvements in productivity or performance at work. Success rates as high as 85% have been reported. [“Bet Dieting Takes Hold in the UK” BBC News May 23, 2009 (http://newsvote.bbc.co.uk)]

Bets can cover behaviors in weekly or monthly patterns, making it possible to increase the bet by 52 or 12 times. They can be made competitive, where a group of participants put up a similar sum each, with only the “winners” who achieve the goal sharing the total sum involved. Bets can also be paid off weekly or monthly, to add the impact of immediacy and frequent reminders, rather than waiting a long time till some ultimate goal is reached.

Bets can also help to prevent or at least reduce relapses, since past winnings would have to be returned, repeating the impact of lost value. For goals such as weight loss and tobacco use cessation, a continuing “threat” of loss will often be essential, because the typical long-term relapse rate is so high. Social networks can also continue to play a role relative to relapses, with continued monitoring of behavior or status metrics.

Since individuals can set their own bet levels, and therefore risks of loss at whatever level they choose, they can ensure their risks are at least affordable, rather than catastrophic, while still being significant enough to provide significant and lasting motivation. In one example, an “Ian Ayres” bet $500 per week if he failed to achieve a 20-pound weight loss and maintain the lower weight.

Over the year he reported in his “case study”, he lost more then his 20-pound goal and in half the time he had set for his bet, and kept his weight below his limit fluctuating only a few pounds from a 25-pound loss for the entire year. As a result, though he had put himself at risk for losing $26,000 during that year, he actually lost nothing and has continued the bet. (www.stickk.com/story/founders.php)

Dean Karlan, one of the founders of the stickk.com website has maintained a $1000 per week bet with fellow M.I.T. graduate Ian Ayres, who once had to pay $15,000 for failing to keep his weight off.
Their bet has been renewed successfully since.

Dr. Kevin Volpp, a physician at the Philadelphia, PA Veterans Affairs Medical Center and professor at the University of Pennsylvania has done studies of the use of a lottery approach for incentives, with chances of winning up to $100 a day vs. a “deposit” group who bet their own money with the chance of doubling or tripling it. The lottery group lost an average of 13 pounds, while the betting group lost 14 pounds, versus a control group whose average loss was only 4 pounds. [A Gorman "Wagering on Weight Loss" ABC HealthCheck Apr 1, 2009 (http://abclocal.go.com)]

Both the size of the potential gain/loss and the frequency of its payoff or penalty make big differences to the effectiveness of incentive programs in general. But with self-bet penalties, not only the effort involved, but the incentives, themselves, can be entirely in the control of participants, with essentially no cost whatever being borne by sponsors. While no single approach will work for all participants, and probably no one approach for even a majority thereof, the potential for self-managed/self-paid incentives offers a significant new alternative to sponsors, as well as to consumers.

An Expanding Number of Problems Face PHI

2009-09-09T08:59:00.000-07:00

When proactive health management of populations first emerged a few decades ago, it comprised mainly three problem domains: worksite wellness and safety for employers, and disease management for insurance plans. These three remain the most common areas for investment today, but there are many others that have joined them as the potential for proactive management vs. reactive treatment of health problems becomes the more favored option.

In addition to disease management, for example, there has emerged a growing emphasis on both acute and chronic disease prevention, as well as chronic disease reversal. Immunization for infectious diseases has grown, while promoting home as well as worksite safety has also emerged. The new campaign to prevent H1N1 flu is only the latest example.

The idea of reversing chronic conditions has joined efforts to control them. This means empowering patients with chronic conditions to reduce their risks of crises, complications, and worsening thereof, while keeping their condition under control through lifestyle methods alone, rather than relying on medications. Reversing both heart disease and diabetes have been shown to be possible, and efforts are under way to do the same for prostate cancer.

Preventing the onset of chronic conditions requires the management of both risk conditions, such as “pre-diabetes”, “pre-hypertension”, etc. and syndromes, such as metabolic syndrome, which is a risk for a number of conditions. The challenge is to identify the relative risk/reward potential of preventing vs. reversing risk conditions, the same as is true for chronic conditions.

But in addition to health behaviors and conditions that cause high sickness care expense, there has emerged a new set of behaviors and conditions that impair the value of workforces, as well as the personal life quality of insurance plan members. These include sleep deprivation, inadequate stress coping/management, poor diet and activity habits, which may cost far more through impairing workers’ performance than they do through adding sickness care costs.

And gradually, once employers and PHI providers target conditions and behaviors that reduce the value that workers deliver to their employers, PHI is open to including a wide range of non-health-related causes of value reduction or limitation. These include workers’ motivation, capabilities and support, not merely their health.

Motivation may be boosted or maintained through a wide range of investments, including those focused on reward/recognition systems, as well as those based mainly on workers’ intrinsic values and drives. Capability includes both measurable abilities and talent, and the self-confidence needed to ensure that workers will apply them when needed. Since self confidence is often lower among the more talented, relative to their true capabilities, while higher among the less talented, this is a key factor in not merely limiting performance, but promoting poor performance.

Support includes all the readily available information, tools, equipment, supplies and reminders that increase the likelihood of good performance, and make it more efficient. These can be left to workers to arrange for themselves, or at least to demand from their peers, or ensured by employers and the systems they maintain for the purpose, and often both are needed. This domain is often the weakest, since responsibility for it is often split between employer and worker, and across the workforce.

PHI should be, or at least be in the process of becoming, part of an integrated approach to enhancing the performance and value of workers, including managers and executives. The same tools for evaluating performance and rewarding it should apply to all three categories of employees, with full recognition that the best integrated approaches for employees will often reduce the need for managers and executives to “lead” them.

Consumer Responsiblity in Health Reform and PHI

2009-08-29T08:32:00.000-07:00

While the current debate over health reform focuses primarily on what people disagree about, often contentiously and even violently, there seems to be at least one thing that most of us agree upon. Consumers should be or at least behave responsibly with respect to their health and their health care use. If universal health care is to be a right, it must be accompanied by the corresponding responsibility to make it affordable for all.

The trouble is, of course, that we can all agree, in principle, that consumers (i.e. every person on earth) should be responsible. But what does it mean in practice? How do we actually get people to behave responsibly as health self-managers and health care consumers? Agreement on the principle may be easy to secure, but how do we at least increase the number of responsible consumers and improve the degree to which they behave responsibly?

There is at least a plurality of consumers, and of health reform gurus as well, who agree that it would be helpful if there were incentives built into the reformed healthcare system to promote responsible behavior. Employers and insurers alike seem willing to build incentives into promoting both preferred health and health care behaviors. Philosophers may argue that true responsibility means making commitments that are right and proper, without having to be paid for it. But since we’re willing to pay health care providers to do the right thing, perhaps it makes sense to pay health care consumers, as well.

One example I recently ran across is the position taken by the Health Alliance Plan (HAP), a major health insurance plan in Michigan (www.hap.org). It has posted “Your Rights & Responsibilities” on its website for all of its Health Engagement/Enhanced Plan members to read. (www.hap.org/info/rights.php) The rights listed relate to the kinds of services, information, participation in decisions, responses, treatment and voice each can expect in their dealings with the plan and care providers who are affiliated with it. One of them, for example, is the right to make recommendations regarding the rights and responsibilities policy.

The responsibilities are:
• To disclose your medical history and symptoms before and during the course of treatment.
• To follow the plans and instructions for care that you have agreed on with those providing your health care.
• To respect the rights of other patients and members.
• To notify your doctor of any unexpected changes in your health.
• To understand our procedures and use the plan appropriately.
• To notify us of any changes in pertinent membership information.
• To participate in understanding your health problems and developing mutually agreed upon goals.
(Copied and pasted from the HAP website on Aug 20, 2009)

While these responsibilities deal mainly with members’ use of health care services, they clearly apply to health self-management as well. “Disclosing medical history” would logically include answering health risk assessment (HRA) surveys, for example, as well as communicating openly and honestly with their physician, who is intended to be an active partner in health improvement efforts. “Following plans and instructions” would include participating in efforts to understand and work toward achieving “mutually agreed upon goals”. “Notifying your doctor” about changes in health should extend to notifying whoever members work with on health goals about changes in their health relevant to such goals.

HAP empowers members to carry out their responsibilities through a wide range health management program options. It offers and includes varying degrees of coverage for weight management programs, including: Weight Watchers® and its “Momentum” program: nutrition counseling; physician-supervised weight loss programs; prescription medications (for members who have an Rx rider with their health plan; HAP’s own “Weight Wise” program; HealthMedia Balance®; and GlobalFit Fitness Solutions. (www.hap.org/healthy-living/weightmanagement.php).

HAP also includes rewards for its Health Engagement/Enhanced Plan members who fulfill responsibilities relative to managing their own health. These take the form of lower out-of-pocket costs when using covered healthcare services. Those who do not meet enough targets can qualify for the same improved coverage by agreeing to specific behavior change plans or participating in particular programs.

The number of targets required to qualify for the Enhanced Plan vary depending on which ones they meet. In order to qualify, they must earn 80 of 100 available “points”, and since being a non-smoker or agreeing to a smoking cessation treatment plan earn 25 out of the 100, all must meet that particular target. Meeting targets for healthy weight or agreeing to participate in a weight management program, meeting targets for blood pressure or cholesterol, or agreeing to follow their physician’s treatment plan for these two targets, will earn 15 points each. Similar options for blood sugar, alcohol use or preventive tests will earn 10 points each. So only meeting the tobacco use target plus at least two of the 15-point targets plus at least one of the 10-point targets would qualify for the Enhanced plan.

While it is common for employers to sponsor most, if not all, their health management programs at no charge to their employees or dependents, HAP does not completely protect its members against some out-of-pocket costs. Physician visits to secure treatment plans for some risks, prescription drugs to control them or quit smoking, and private weight management programs may require members to pay some fees. But normally, these can be readily compensated for through lower out-of-pocket payment requirements when using other healthcare services.

The challenge which tends to be the major problem for most employers and insurers, namely weight management, is also the one with the greatest number of choices for members. And it has worked well for most who have participated. The average weight loss for participants stood at 9.3 pounds on HAP’s website “billboard”, based on loss of 215,284 pounds among nearly 23,200 participants, reported on Aug 20, 2009. Of course, it is often the case that employees and plan members have strong personal reasons for wishing to lose weight, and gain a number of personal benefits such as better appearance, pride in accomplishment, etc. in addition to any incentives offered.

We still have a long way to go before we have achieved what would amount to “evidence-based health management” (EBHM) as a counterpart to evidence-based medicine (EBM). As more insurance plans, employers, governments and PHI providers gain experience and report on what works, we should gain appreciably more understanding of what works in translating the abstract notion of “consumer responsibility” into cost-effective and efficient interventions. While the principle of consumers’ taking responsibility for their health may be readily agreed to by all concerned, it is the practice that will ultimately make desired differences.

Marketing Applications in PHI

2009-08-29T08:09:00.001-07:00

Marketing came late to health care – there were virtually no discussions of the possibility of applying marketing principles and practices to health care until the 1970s, and no persons seem to have occupied marketing positions in health care organizations until that time. But it caught on, though slowly, and I was lucky enough to enjoy a career as both a health care marketing practitioner and professor for more three decades.

PHI has attracted people from a wide range of disciplines – education, medicine, nursing, coaching, etc. as well as specialists in particular health challenges: nutrition, fitness, stress management, etc. It has always needed help from “marketing” to promote particular strategies to potential clients (employers, commercial and government insurance plans), as well as potential participants (employees, dependents, health plan members).

But it can also benefit from the application of marketing models to thinking about, planning and implementing behavior change interventions, themselves. Fundamentally, PHI and all other efforts to reduce the incidence and prevalence of personal health risks, acute and chronic disease and injury require consumer behavior changes. [L. Masterson “Health Reform Is Useless without Behavior Change” HealthLeadersMedia Aug 19, 2009 (http://healthplans.hcpro.com)]

Consumer behavior change can be seen and approached as a marketing challenge. It can also be seen as a challenge for education, management, regulation, and other models of behavior change, however, and there seems little agreement on which model applies and works best. While I don’t claim that the marketing model is the best, I strongly believe that it has much to offer, when it is applied as its definition dictates, rather than as the limited, advertising-focused and seller-driven communications efforts most frequently practiced.

Marketing as a behavior change model rests on the idea of engineering mutual exchanges of value. As a behavior engineering approach, it seeks to design and deliver value to customers for the value they deliver in return. Usually, this means product and service value to customers in return for their payment of prices and fees. But it can equally well apply to consumers changing behavior to benefit sponsors of PHI efforts, in return for both extrinsic and intrinsic personal benefits they gain.

The “marketing mix” of tools to use in bringing about behavior change relies on a familiar set of four “Ps”:
• Product = the benefits that consumers can gain and enjoy by adopting and maintaining a given behavior pattern
• Price = the prices consumers pay or other burdens they endure by adopting and maintaining that behavior pattern
• Place = how easy or difficult, convenient or inconvenient it is for them to adopt and maintain the behavior pattern
• Promotion = all communications aimed at promoting consumers’ confident expectations and perceptions of gains vs. costs should they adopt and maintain the behavior pattern

Sponsors of PHI initiatives tend to rely on a management model for changing health plan members’ or employees’ behaviors. The need to control people’s behavior is strong among employers, insurers, and governments, after all, so it is natural that almost all programs in the PHI domain use the word “management” in their labels: population health management, disease management, environmental management, etc. They also tend to rely on “authority” (of experts, physicians, managers, etc.) and on the use of extrinsic rewards or punishments to achieve their aims.

By contrast, a marketing model can apply ideas such as:
• Product – identifying the inherent/intrinsic benefits that consumers can gain through changing their behavior in the intended way
• Price – reducing the “costs” to consumers of making the intended behavior change
• Place – decreasing any difficulties or inconveniences associated with making such changes
• Promotion – reminding consumers of the intended behaviors to stimulate and maintain high levels of awareness of the best time, place, and manner to make changes, and of the benefits they are gaining thereby

For example, while consumers overcome addiction to tobacco will gain a host of health-related benefits, beginning soon after they quit, they will also gain financial benefits through avoiding the expense of buying tobacco products. To the extent that they are rewarded for their performance, they should also gain through being able to devote time formerly lost to “smoke breaks” to productive effort and performance. Workers who are able to move from obesity or overweight to healthy weight levels are likely to gain greater energy and self-confidence.

Reducing costs may include specific strategies that reduce what consumers have to pay for healthy products or services – subsidizing purchases of exercise equipment, healthy food, gym memberships, etc. Providing worksite shower facilities so workers can comfortably walk, run, or bike to work or exercise during breaks can eliminate concerns over the “price” of such efforts.

Making it easier for consumers to adopt and maintain healthier behaviors can include having fitness centers, walking/jogging/biking trails, education and services onsite, offering healthy food in cafeteria and vending machines, even offering healthy take-home meals. Consumers can be encouraged to make their own home environments more convenient for healthy behaviors and inconvenient for unhealthy ones, such as by not buying unhealthy foods to tempt them, buying enjoyable exercise products to have at home, such as video exercise games. Reducing the size of plates and silverware can reduce the amount of food we eat, for example. [P. Bregman “The Easiest Way to Change People’s Behavior” How We Work Mar 11, 2009 (http://blogs.harvardbusiness.org/bregman)

Reminding consumers to adopt, repeat, and persist in healthy behaviors can be something that sponsors do, or consumers can be empowered to devise their own reminders. Keeping healthy food, including snacks in the front of cabinets and refrigerators can remind us to select them by making them the first option we see. Reporting to them, or enabling them to readily access progress reports, records of goals achieved, incentive awards, and the significance of their health efforts and achievements as well as performance improvements, can reinforce positive attitudes toward PHI programs in which they participate.

By focusing on the full "participant experience" in PHI initiatives, sponsors and providers can achieve the same kinds of financial gains as are available with their customers. Value contributions from employees and health plan members can be optimized in the same ways that they are from consumers of other services and products. Marketing may well prove to be significantly more cost-effective than management as a model for influencing "consumer" behavior.

Behavioral Economics vs. Empowerment in PHI?

2009-08-26T09:32:00.000-07:00

For centuries we have widely accepted Adam Smith’s theory of free markets being the best way for economies to operate. We have also accepted the idea that both buyers and sellers make decisions largely based on rational analysis of the utility or value of competing options. But recent developments in “behavioral economics” and “brain science” have significantly undermined our faith in both ideas – they are based on assumptions and beliefs about how we make decisions that are demonstrably false. S. Stewart “Can Behavioral Economics Save Us from Ourselves?” University of Chicago Magazine Feb 2005 (http://magazine.uchicago.edu)]

Essentially, we do not make decisions rationally, even when we are acting for large businesses or other organizations, and certainly when we are acting as consumers. Our emotions tend to overpower and co-opt our rational capabilities. And what is perhaps the worst consequence of this reality is that sellers know about behavioral economics and can use it to manipulate our decisions and behaviors while we blithely assume we are rational and independent thinkers, deciders, and buyers.

One of the essential flaws in Adam Smith’s model of the free market is that it is based on roughly equal knowledge and capabilities in both buyers and sellers. This may have been true in the village-based economies of 1776 when he wrote The Wealth of Nations, but the industrial revolution changed all that. Even with the “new consumerism” and broad consumer access to information about market choices, sellers still know enormously more about the markets in which they operate and their offerings than consumers ever will.

Since we make most decisions emotionally, where we are open to manipulation by sellers who are willing and able to spend the resources needed to do so, behavioral economics is mainly used by sellers, rather than studied by consumers. As a result, PHI sponsors and providers have choices to make about whether they will apply the new realities of consumer behavior to manipulate it – after all, they’re only seeking better health, which is in consumers’ best interests – or stick with the rational model.

The “rational” model, for example, favors the idea of offering and paying incentives to consumers in order to promote their adopting particular behaviors – and that adds even greater benefit to those who respond favorably to such offers. Of course, if some behavioral economics guru discovered a way to achieve the same kinds of results without using incentives at all, or while using far less expensive ones, I imagine sponsors would be delighted to “rationally” choose the lower-cost option.

Because PHI relies on a continuous relationship for optimal results, however, its sponsors and providers run serious risks if they apply behavioral economics to elicit desired consumer behavior changes. The word might get out, and any trust that has existed between consumers and either sponsors or providers who are found to be cleverly manipulating consumers might be severely damaged, if not destroyed. While “rational” models of consumer behavior can be used openly, finding ways to promote emotion-based choices and behaviors is likely to be viewed as inappropriate by consumers.

On the other hand, because a relationship is involved, and time is available, PHI might also become unique in its use of behavioral economics by offering consumers the choice of rational vs. irrational appeals. They could be asked from the outset, once they have selected a health goal to work on, whether they would prefer the PHI intervention provider to use strictly rational or irrational, perhaps a mix of the two when seeking to persuade them to adopt or discontinue the behavior(s) in question. If consumers give their permission, then the use of irrational behavioral economics techniques might both work well and be accepted as appropriate by those who gave their okay.

Even after consumers indicate a preference for a rational approach, if they fail to succeed or make significant progress toward their goal(s), PHI participants could be offered the choice of either switching to irrational, or to a mix of rational and irrational approaches. Of course, it may prove to be the case that “permission-based manipulation” will not work as well, since consumers will recognize it for what it is. Or, since they have given their explicit okay for its use, such an approach may work as well or even better than surreptitious approaches. Permission-based e-mail has certainly proven to be more cost-effective than blanket spam, for example.

Ultimately, it may depend on how sponsors and providers view participants, and participants view them. If the PHI challenge is undertaken as one of “manager vs. managee”, behavioral economics may be selected as long as its "tricks" work better or more cheaply than other options. If the challenge is seen as one of empowering “partners” toward achievement of mutually equitable value, then being open and up front about the use of irrational approaches would make more sense. In any case, it would seem advisable for both PHI sponsors and providers, as well as participants, to make informed choices between the two, or relative to any mix of the two, in practice.

How Long Till PHI Results and ROI?

2009-08-19T08:01:00.000-07:00

There seems to be a widespread belief, or at least a contention, that it takes over a year before results and positive returns on investment (ROI) from performance health management efforts should be expected. Many PHI providers warn that this is the case, preparing clients for meager if not negative returns in the first year, then improvements in subsequent years. This should not be a serious problem for employers and health plans that are “true believers” in the value of PHI, though it might create hesitancy or under-investment among those that are not yet convinced.

Disease management demonstration projects undertaken by Medicare, for example, have tended to yield equivocal or mixed results in their first years, for example, perhaps reflecting the inherent difficulties in “teaching old dogs new tricks” within such a short time frame. But in at least one case, results by the third year were significantly better. Fully half of the ten physician group practices participating in CMS’ demonstration achieved quality benchmarks in at least 28 of 32 measures reported, and qualified for performance bonuses amounting to $25.3 million as their share of the $32.2 million saved by Medicare during that year. [J. Simmons “Physician Groups Earn Over $25 Million for Quality Improvements” Healthleadersmedia.com Aug 19, 2009]

Marathon Health, a PHI and onsite medical care provider, asserts that its clients “…have the potential to double your investment in our program over three years…” First come “hard dollar savings” from the lower costs of onsite medical care, lower use of hospitals, ERs, and specialists, improved compliance with evidence-based medicine, better referral management, and reduced disability/absenteeism. And second, “soft dollar savings” arise for employers can include improved productivity, employee morale/retention, and less time lost away from work when seeking medical care. (www.marathon-health.com/return-on-investment.html)

Marathon is willing to back up its predictions of savings by putting a portion of its fees at risk. “If your goals are not met, we’ll match the discrepancy with a percentage of your total cost.” (www.marathon-health.com/program-costs.html) Though only a modest number of case examples have reported any provider’s or client’s multi-year results, the examples that are available tend to reinforce expectations for increasing ROI over at least three or four years. [G. Stave, et al. “Quantifiable Impact of the Contract for Health and Wellness” JOEM 45:2 2003 109-117]

Of course, since some PHI investments look only for savings in the “hard” measures of health, disability, and workers compensation insurance or direct expenditures, it probably makes sense to be conservative when predicting or promising ROI. On the other hand, considering that absenteeism/presenteeism losses tend to yield two to five times as much in savings as medical care cost reductions alone, providers that measure and focus on health factors that cause absences and impaired productivity at work, may feel comfortable predicting or promising earlier achievement of positive ROI.

Marathon Health, for example, adds to its claims of potential doubling of clients’ investment over three years the prediction that there will be “…noticeable savings realized six months into the program.” HealthMedia.com has reported significant savings within 90 and 180 days from its PHI initiatives. When productivity/performance improvements are included in reported results, however, they are usually based on employee self-reported changes therein, rather than objective data, so real savings are far more challenging to measure credibly than is the case with insurance expense reductions.

There is often a built-in discrepancy between what PHI providers may count as results and what employers accept. Employers or their insurance plans will have to provide cost figures relative to health, disability, and workers compensation insurance expense reduction, for example, and they may not credit PHI interventions with the totals for reductions measured after PHI interventions have been implemented. They may feel that other employer-sponsored interventions were responsible for a share of such results.

It has been reported to me that it is “industry practice” for PHI providers to count as costs of an intervention program only the providers’ fees. Since employers, themselves, may incur substantial costs related to their own investments in workforce health improvement: promoting awareness through internal publications, memos, posters, etc. plus perhaps investments in incentives, subsidies for fitness club memberships, creation of walking/jogging/biking paths, showers for employees who exercise onsite or on the way to work, etc. – employers’ own investments might be as much or even more than the fees charged by PHI providers.

The history of PHI interventions is full of examples of “rigorless” evaluations. Simple “after” vs. “before” comparisons often ignore the effects of a predictable “regression to the mean” effect that may be responsible for a substantial of apparent cost reductions. Side-by-side comparisons of participants’ insurance costs or performance improvements and those of non-participants in PHI interventions may ignore differences between the two cohorts that already existed and account for a significant share of “results”.

The measurement and evaluation of PHI interventions need significant improvement and validation. Fortunately, employers, themselves, have dramatically increased their insistence on ROI calculation, rather than taking on faith the idea that PHI is a good thing to do. Where as few as 14% of employers actually measured their ROI in 2007, this proportion has risen to 71% in 2009. If they also increase the sophistication, completeness and accuracy of their ROI measurements, employers and providers alike may discover that more valuable results are being achieved and earlier than even providers have suggested.

Should PHI Focus on Numbers of Risks or Selected Risks?

2009-08-12T09:19:00.000-07:00

It is easier to measure and report the sheer number of people at specified levels of risk than to gauge the level of particular risks for multiple risks for the same people. This may be why there are more examples of both the costs of health risks and reductions therein related to changes in overall risk levels than in separate risks. There may be a dozen or more individual risks included in analysis, after all, making it difficult to summarize risk status across each one. It is far easier to create breaks between “at risk” and “not at risk”, or among “low”, “medium”, and “high” overall risk scores, than it is to characterize a population based on scores across a dozen or more separate risk factors.

Studies of the effects of sheer numbers/percentages of people at risk --or numbers at low, medium, and high risk levels – invariably show that when there are more people at risk, or more at high and medium vs. low and medium, the effects of such risks show clearly in terms of higher healthcare expenditures or absenteeism/presenteeism losses.
And when risk levels go down, such costs also decline, though there may not be a perfect correlation between risk reductions and cost reductions.

But it is also known that when individual risk factors are analyzed separately, they vary significantly in terms of how much higher health costs or productivity losses are for different factors. Chronic diseases such as congestive heart failure, chronic lung diseases, heart disease and diabetes tend to be major causes of high health care costs, while allergies, stress and emotional health factors are often the most significant factors in lost productivity. In theory, at least, significant reductions in some risk factors could lead to far greater cost savings than with other factors. But if only the number of factors affecting individuals is measured and monitored, the differential effects of different factors would be lost.

On the other hand, it is also the case that most people who have health risks have more than one. When HealthMedia®, Inc. of Ann Arbor, Michigan reported productivity impairment linked to fourteen health risks and chronic conditions, it found that only one or two percent of the entire population of workforces in its database had zero risks, and the majority had four or more. And since it can only report the degree of productivity impairment of each individual, the impairment may be caused by any one of many risk factors each has. There is no realistic way of determining how much each of multiple factors contributes to total impairment separately.

In such circumstances, it may be best to focus on the whole individual as a coaching participant, and to promote whatever changes in health risk behaviors each is willing and able to pursue, rather than assign each to one particular risk-focused intervention. If the employer or PHI provider selects the individual risk factor to be addressed, each may select one that the individual is least likely to change – e.g. tobacco addiction or weight loss, where long-term success are often most difficult to achieve. If all risks are treated equally, then reducing a more tractable risk behavior could prove far more successful, even if the potential benefits of another are greater, since it is the potential times success achievement rate that determines the benefits of particular risk-focused interventions.

Moreover, when individuals have multiple risks, as the vast majority of workforces and insured populations have been found to have, enabling them to choose their own risk reduction targets may promote higher levels of commitment to and engagement in the risk reduction intervention than is the case when employers or providers assign them to specific risk-focused interventions. To some degree at least, PHI interventions should increase individual participants’ knowledge and skills relative to self-management of risk behaviors in general, not merely the specific behavior(s) targeted for “reform” in specific interventions. By enabling participants to gain skills and self-confidence (= “self-efficacy”) in self-management in general, letting them start where success has its best chance can help when added challenges are selected.

As is true for all PHI interventions, the only way to find out whether risk-type or risk-number approaches lead to the greater success is to try both and see. Large employers could easily test both approaches in different locations, or even with participants randomly assigned to one approach vs. the other. Insurance plans and providers could easily test the two options with half of their total populations at risk randomly assigned to each option based on the group or client involved. Whether insurers or providers will share what they learn thereby with the general public, or choose to treat such learning as a proprietary competitive advantage will remain to be seen.

Both the success rates and the costs of the two kinds of interventions may well differ significantly, so it may be necessary for providers to continue to offer their clients choices as to which approach to apply. New clients, for example, may prefer to start with the lowest cost option, in order to minimize their financial risks, and only try a higher-cost but more promising option when they have already succeeded with their initial trials. In any case, it would certainly add much to the body of knowledge available regarding PHI if there were some scientifically sound comparative cost-effectiveness studies undertaken on the two different approaches.

The Full Costs of Turnover in PHI

2009-08-11T08:53:00.000-07:00

The costs of employee turnover have long been recognized as a major problem in many industries and individual businesses. Estimates have emerged that put such costs as equal to a full year’s salary for the worker that has to be replaced, though this may apply only to higher skill/value categories of workers. The kinds of costs discussed include both costs of replacement and losses in productivity/performance due to the time it takes to replace the worker, as well as to bring the replacement up to speed.

But there is another significant cost that is rarely mentioned – the impact that turnover has on PHI results and return on investment (ROI) levels. Depending on the pattern that describes how PHI investments affect the benefit of worker participation to their employer, turnover may easily cut PHI value in half or even worse. Given that the costs of PHI interventions tend to be the same for participants, whether or not they leave before results are achieved, the ROI potential will be significantly reduced.

The pattern of PHI results over time has not been studied or reported to the degree that immediate results are published. This is unfortunate already, since the pattern discovered in most cases seems to be one of improved results in the second year of a PHI program compared to its first year, then still better results in the third. And this is true even when turnover is already affecting results in the wrong direction.

Virtually all PHI results are reported for the set of participants who were engaged in particular PHI interventions during the year analyzed. But turnover, whether reflecting employees who leave their employer entirely, or simply those who drop out of PHI participation, will change the mix of participants every year. Results are rarely reported for a cohort of the same employees who participate in a PHI program for multiple years.

One of the few exceptions I have found involved a group of over 6000 employees at the GlaxoSmithKline pharmaceutical firm. It reported results for four full years compared to the baseline, pre-PHI year, where the intervention years were 2000 through 2003. And the pattern it reported for that cohort of participants showed savings of $233 per participant in the first intervention year, then $375 in the second, a jump to $944 in the third, then a leveling off to $950 in the fourth. [G. Stave, et al. “Quantifiable Impact of the Contract for Health and Wellness” JOEM 45:2 2003 109-117]

If this were a typical pattern for cohorts of participants continuously engaged in a PHI program for four years, then the turnover rate among such participants would have dramatic, potentially devastating impact on the ROI achieved by the employer. It could easily turn a potentially magnificent return level to a thoroughly disappointing one. To illustrate, consider another employer where the cohort-specific results follow the GlaxoSmithKline pattern.

100% turnover – if the annual turnover rate were 100% or more, i.e. if all participants left the company or dropped out of the PHI initiative during the year, the pattern of results would be the first-year level of savings repeated every year. Since none of the participants would reach their personal second year of continuous participation, none would achieve even the second-year level of results – returns would never get higher than $233 per participant, and would presumably be even lower than that since turnover would occur throughout the year and many would leave before achieving even the full-year level.

50% turnover – if only half of participants dropped out each year, the first year results would be much the same as with 100% turnover, but the second year would be better, since half the participants would continue till they achieved second year savings of $375 each. The second year savings would be .5 x $233 plus .5 x $375, or $116.50 + $187.50 = $304.00. Depending on the costs of the PHI program overall, this might well be enough to represent a positive, even a highly positive ROI.

33.33% turnover – if one-third of participants dropped out after the first intervention year, then another third did so after the second, and the last third quit after the third year, the first year would yield the same $233 savings as in the preceding examples, while the second and third years would be better. The second year would yield 1/3 x $233 plus 2/3 x $375, or $77.67 + $250 = $327.67. Then the third year would yield 1/3 x $233 plus 1/3 x $375 plus 1/3 x $944, or $77.67 + $125 + $314.67 = $517.34.

25% turnover – if only one-quarter of participants dropped out each year, there would be one quarter who stick it out till the fourth year, when savings would be $950 each. That would mean that the fourth year savings would be ¼ of $233 plus ¼ of $375 plus ¼ of $944 plus ¼ of $950, or $58.25 + 93.75 + $236.00 + 237.50 = $625.50. First year savings would be the same as in all the preceding scenarios, but second and third year savings would be a good deal greater, as well.

After first-year savings of $233 per participant, second year savings would be ¼ x $233 plus ¾ x $375, or $58.25 + $282.25 = 339.50 per participant. Third year savings would be ¼ x $233 + ¼ x $375 + ½ x $944, or 58.25 + $93.75 + $472 = $624. Savings for the four-year investment period would be $233 + $339.50 + $624 + $625.50 = $1822.00.

As turnover rates get lower each year, i.e. as retention rates get higher, the pattern of yearly savings would approach the cohort-based pattern for continuous participants, namely $233 + 375 + $944 + $950 = $2502. For example, at 10%, first year savings would be $233; second year -- $23.30 + $337.50 = 360.80; third year -- $23.30 + $37.50 + $755.20 = $816; and fourth year -- $23.30 + $37.50 = $94.40 + $665 = $820. The total four-year savings would be $233 + 360.80 + $816 + $820 = $2229.80.

As turnover rates approach zero/retention rates approach 100%, the value added by reducing turnover/increasing retention will diminish compared to the gains when high turnover decreases/low retention increases. The impact of turnover on PHI returns is only one of the reasons to promote retention, particularly among the higher performers in the workforce, but it can be significant, even dramatic, where turnover has been high in the past. And luckily, successful PHI programs tend to improve retention – they show workers that their employer values them, plus they yield personal benefits to participants, which tends to increase loyalty.

Measuring Productivity-Performance in PHI

2009-08-05T10:27:00.000-07:00

Virtually all the early attempts to “manage” employee or insurance plan member health focused on measures of healthcare use and expense. Some employee-focused efforts also measured workers compensation and disability expense, but those expenditures tended to be far less than healthcare costs because so few employees incurred them. Focusing on these easily measured costs makes sense in terms of convenience for planning and evaluation, but tends to miss the greatest savings potential for employers.

Some employers have a relatively easy time measuring productivity/performance (P/P) losses, because they already measure P/P as their basis for paying employees. Even if only some employees are paid on a pay-for-performance (P4P) basis, the results achieved with them can be extrapolated to those who are paid on a wages or salary basis alone, though this may stretch the bounds of “science” in cases where piecework employees are significantly different from the rest.

[Note: productivity is normally used to describe the sheer amount of output that workers account for, while performance extends to their total qualitative as well as quantitative value they contribute. If worker performance is already well measured through annual evaluations, it may be easier to skip output, already virtually impossible to measure with “knowledge” workers, and start with performance, instead, though determining the dollar value of performance improvement may be problematic.]

For those employers who cannot use P4P information to gauge the effects of health management efforts, including P/P in their planning and evaluation still makes sense, even if the measurement challenge is significant. Not only do P/P considerations add greatly to the value that health management delivers, they also significantly change the “health” problems that are selected for attention. Worker performance tends to be far more affected by non-diseases such as physical inactivity, sleep problems, stress, emotional problems, etc. It is also significantly affected by motivation, capability, and support systems unrelated to health.

For this reason, both PHI and other or integrated approaches to performance improvement are challenged to measure current degrees of “impairment”, as well as “sub-optimal” performance, in hopes that both health-focused and motivation/capability/support (MCS) investments will improve performance enough to justify the expense involved. Without performance measures or at least credible estimates, such investments are unlikely to pay off in terms of healthcare expense reduction.

But measuring P/P when there is no P4P or other system already in place is a real challenge. There are many different approaches available, and since they are applied mainly where there are no objective measures available, it becomes difficult to check their validity and accuracy. Most validations of productivity estimates are made with workers who are at least monitored or paid on the basis of measured performance, such as call center or sales staff.

One useful source of information and insights on P/P evaluation was published in 2007 – S. Mattke, et al. “A Review of Methods to Measure Health-Related Productivity Loss” American Journal of Managed Care 13:4 April 2007 211-217] It examined and reported on 17 different methods, though five of these dealt solely with a single condition (3 migraine, 1 allergy, 1 hepatitis). These methods varied greatly in term of how and what they measured, as well as the numbers of questions used in their questionnaire (from 3 to 44), and the recall period respondents were asked about (1-4 weeks).

Most of the questionnaires have been “validated” to some degree, usually by comparing their results to objectively measured P/P among workers paid or at least managed on a measured productivity or performance basis. But such a validation only deals with output or qualitative performance metrics, while employers are likely to focus on the costs of absence and impaired P/P, not just the amount of it. And costs represent a measurement challenge at least as great as that relative to P/P, itself.

Some experts have argued, for example, that the most common approach to estimating costs, i.e. multiplying the degree of impairment or absence times the effective hourly or daily wage costs of the affected workers. Given privacy concerns, it is often necessary or advisable to use average wages rather than look at each individual separately. But wages may or may not accurately reflect the value of what employers lose when workers are absent or impaired by health or other factors.

In many situations, the absence or impairment of one worker is made up for by peers taking on more than their usual workload. When this creates overtime work, there are costs involved, but when there is enough “flex” in terms of how hard and efficiently workers to their jobs, there may be virtually no cost impact at all.

On the other hand, some workers, such as a surgeon in a hospital, or a construction engineer on a job site, may cause a loss of P/P across an entire team of workers. The effects of particular workers on the P/P of their teams may be as high as ten times or more their own costs in terms of value to their employer. Of course, if workers are only paid when they are present and working, and on a P4P basis, their absence or impairment may create no costs at all for employers, merely a bad day or week for the worker.

The vast majority of research into health and productivity deals solely with the costs of lost output or value, not with the recovery thereof through health improvement efforts or interventions focused on workers’ motivation, capability, or support. In one study, for example, the healthcare costs of having a given number of risks were calculated, and analyzed relative to changes in risks achieved. Reducing the number of risks workers had yielded an average of $153 in healthcare cost savings. But preventing workers from gaining one risk yielded $350 in cost savings. [“University of Louisville: Delivering Positive ROI through Health Management” HealthFitness.com Case Study 2009]

One approach frequently taken by PHI providers or their clients has been to measure only healthcare costs, or perhaps add in disability and workers compensation costs as well, since all are routinely reported. If these domains yield enough financial benefit to demonstrate a positive and acceptable or admirable ROI, then P/P value gained can be treated as “icing on the cake”. If not, then evaluations can move forward incrementally, perhaps counting just absence reductions, since absence may be already tracked, leaving P/P measurement to be attempted only when necessary to make the case for persisting in programs and investments.

There is likely to be enough value achieved and measured through most PHI efforts to cover the investments required, particularly as providers increasingly offer low-cost approaches in addition to the higher-cost options originally most common. But learning what approach to P/P measurement works best, for both current impairment and achieved improvements, is likely to add so much more useful intelligence, that it will often be worthwhile even when current results measured make a good enough case for continuing efforts.

Where measurements have been made, the added value of P/P effects have ranged from two to five times as great as those from healthcare, disability and workers compensation expense reductions alone. Not measuring P/P effects to the full extent feasible is likely to result in employers missing out on the greater part of the potential value of PHI, even if they still achieve positive and acceptable ROI based on incomplete measures. Only complete measurement methods will enable complete PHI strategies and tactics.