This is a response to the DiNardo, Horwitz, and Kelly Health Affairs Blog post in which the authors replied to my previous commentary, “Structuring Legal, Ethical, and Practical Workplace Health Incentives: A Reply to Horwitz, Kelly, And DiNardo.”  In my prior post, I highlighted my disagreements with many of the points made by Horwitz et al. in their Health Affairs article entitled “Wellness Incentives in the Workplace:  Cost Savings through Cost Shifting to Unhealthy Workers.”  This post continues that dialogue.

I begin this commentary with some hesitation.  I want to be clear that my intent in posting these blogs is not to “dig in my heels.”  In fact, I fully understand, appreciate, and empathize with DiNardo et al.’s positions.  They are rightfully concerned about protecting poor, minority, and disenfranchised workers whose rights may be threatened by unscrupulous employers who wish to place the onus on employees to “become healthy” or “else” — the “else” meaning paying a higher health insurance premium than their “healthy” counterparts.  I appreciate that DiNardo et al. are protecting the interests of workers who, through no fault of their own, have become ill and are now faced with the prospect of paying more for health care coverage because of their illness.

Let me unequivocally state that paying more for health insurance because you are ill or have certain health risk factors is not the goal of workplace health promotion (wellness) advocates.  Quite the opposite is true.  Our intent is to keep workers healthy for as long as possible so that they can be spared the human and financial burden of paying for health care services that might otherwise have been avoided.  The point of workplace wellness programs is to inspire people to improve their health behaviors and biometric measures so that they do not suffer from illnesses that are to a large degree attributable to lifestyle practices — e.g., lung cancer, type-2 diabetes, chronic obstructive pulmonary disease (COPD), and coronary heart disease (CHD).

In my previous post, I highlighted ways to structure incentive programs so that they are fair and contain provisions to guard against abuse.  Here, I address additional challenges to workplace health promotion programs posed by DiNardo et al.

The New Wellness Incentives Regulations

Since the last interchange with the authors, new regulations have been issued by the Departments of the Treasury, Labor, and Health and Human Services entitled “Incentives for Nondiscriminatory Wellness Programs in Group Health Plans.”  These regulations are clear about how workplace wellness programs should be structured with particular attention paid to protecting the rights of workers.  Within these regulations, distinctions are made between wellness program incentives that are based on “participation only,” “activity only,” and “outcomes.”

In terms of the regulations for outcomes-based incentive programs, the following rules apply:

  1. the employer must give individuals an opportunity to qualify for the full reward at least once a year;
  2. the reward for all health-contingent programs with respect to a given individual may not exceed 30 percent of premium (with the exception of smoking where insurance differentials may be as high as 50 percent);
  3. the health promotion program must be reasonably designed to promote health or prevent disease;
  4. the reward must be available to all similarly situated employees, and
  5. the employer must provide adequate notice of the availability of a reasonable alternative standard regardless of whether the health outcome is unreasonably difficult or medically inadvisable to achieve.

The last provision, the reasonable alternative standard, is especially important because it allows the employee to meet the requirements of an outcomes-based incentive program without actually achieving a predetermined health outcome.  Further, the employee does not need to provide physician verification for a waiver to be issued.  The employee is not required to undertake an “unreasonable” time commitment for participating in the alternative program, nor is the individual to bear the cost of the program.

In short, the new rules provide several protections for workers affected by the Affordable Care Act regulations regarding wellness incentives.  Quite simply, an employee is exempt from any health-contingent wellness reward or penalty by either participating in a health improvement program or obtaining a waiver from a physician noting that achieving a specified health outcome is inadvisable, unachievable, or potentially harmful.  As an example, a smoker can still receive an incentive for not smoking by attending a free smoking cessation program, and there is no limit to the number of times the smoker can participate in the program.

General Principles    

A central issue in this discussion of workplace incentives is whether an organization should be involved in influencing personal health matters of its workers.  If the health of employees had no effect on the business, then the answer would certainly be “no” — an organization has no business interfering in the private lives of employees.  However, if the organization is paying a large portion of the health insurance coverage for its workers, then it does have an interest in maintaining the health, well-being, and overall performance of its workforce.  That is, after all, one of the main reasons why companies introduced health insurance programs in the first place, to keep workers healthy and productive.

Further, organizations and their workers gain economically when they avoid paying for unnecessary health care services.  After all, money not spent on hospitalizations, procedures, and medications can fund higher wages and greater investment in the organization’s future.

Before I begin to respond to the various arguments made by DiNardo et al., I want to reiterate an important point.  DiNardo and others have focused specifically on incentive-based workplace wellness programs.  As noted previously, the science of incentives as a method of driving behavior change is still evolving and requires additional research.  That is why incentives should play a small role in the overall structure of workplace health promotion programs.

Readers are reminded that workplace wellness programs that “work” are comprehensive, evidence based, theory-grounded, well implemented, and fully resourced, with many individual and environmental components built into them.  The literature reviews on the topic of workplace health promotion conducted by the Centers for Disease Control and Prevention and Harvard University economists examined a variety of workplace programs where outcome-based incentives played a minor or non-existent role.  In most workplace programs, incentives have been used to boost participation in the program, not to change behavior.  It is only recently that outcomes-based programs are becoming more popular.

My colleagues and I agree that a workplace program that relies exclusively on outcomes-based incentives is likely to fail.  Good programs increase awareness and knowledge about healthy lifestyles; provide tools and resources to change harmful health behaviors such as smoking; provide social and environmental support for achieving good health habits; and are continuously monitored to make sure they are effective in improving population health.

Enough said about general issues.  Below, I address specific points made by DiNardo et al.  To start, they highlight two examples where the “conventional wisdom” may be in conflict with scientific evidence: practices aimed at reducing sodium consumption and glucose levels.  I examine each of these issues separately.

Sodium and Healthy Eating

DiNardo et al. cite a randomized clinical trial conducted by researchers in Palermo, Italy, that evaluated the differential effects of a normal sodium and low sodium diet, and I quote from the article, “in combination with high dose furosemide and severe fluid restriction in patients compensated after recently decompensated CHF (congestive heart failure), on readmission for worsening CHF.”  The patients studied were previously hospitalized and unresponsive to treatments.  I would argue that this is a medical intervention directed at quite sick patients, not one directed at a generally healthy, working-age population that participates in workplace health promotion programs.  The Italian researchers cite other studies that report higher risks of heart attacks, strokes, CHF, and death for people consuming more than 7,000 or fewer than 3,000 of mg of sodium a day.

So, too much or too little salt consumption is bad for people.  It also seems clear that the amount of sodium in the diet plays a role in disease treatment, but the ideal level of sodium consumption is yet to be determined.

This study does highlight the need for more medical research on sodium consumption.  However, I am not aware of any workplace health promotion program that monitors salt intake or penalizes workers for consuming too much salt.  In reality, healthy eating education programs cover such topics as total caloric intake, fruit and vegetable consumption, frequency and timing of meals, the pluses and minuses of various diets, good vs. bad fat, eating out, and so forth.

Nutrition education is complex and the science is still evolving.  For example, there is mounting evidence that adhering to a Mediterranean diet that includes ample amounts of fruits and vegetables, nuts and grains, “good fats,” seafood, and wine may be health beneficial.  The existence of conflicting results from a medical study on sodium intake that involved very sick patients does not negate the need to educate employees about healthy eating habits.  It does not make sense to hold back health education related to nutrition simply because the science on nutrition is evolving and recommendations regarding healthy eating shift as new evidence is gathered.

Again, no one is recommending penalizing workers for poor diets.  But, helping people gain knowledge and skills related to preparing healthy meals is a mainstay of workplace health promotion programs; it is critical to helping employees manage their weight and reduce their risk for the development or progression of chronic disease.

Glucose and Diabetes Prevention and Treatment

On the subject of managing weight and avoiding obesity-related illnesses like type-2 diabetes, DiNardo et al. question the value of glucose lowering treatments.  First, it should be noted that workplace wellness programs do not recommend specific medical treatments for diseases such as type-2 diabetes.  Those decisions are made by patients and their health care providers who, we expect, are current with the medical literature related to the benefits and harms associated with different treatments.  As noted in the new regulations cited above, if the individual’s physician thinks it is inadvisable to reduce the patient’s blood glucose levels to a certain predetermined value, then a waiver is granted to the employee, who is then exempt from any penalties related to not achieving a given blood glucose reading.

DiNardo et al. cite the 2012 Position Statement of the European Association for the Study of Diabetes (EASD) and the American Diabetes Association (ADA), which expressed “mounting concerns about [the] potential adverse effects and new uncertainties regarding the benefits of intensive glycemic control.”  However, if you read the report, you find the following recommendations: “…interventions designed to impact an individual’s physical activity levels and food intake are critical parts of type-2 diabetes management….all patients should receive standardized general diabetes education …with a specific focus on dietary interventions and the importance of increasing physical activity.”  The report goes on to say that “… establishing a goal of weight reduction, or at least weight maintenance, is recommended” … “patients should be advised to eat healthy foods … high in fiber (such as vegetables, fruits, whole grains, and legumes), low fat dairy products, and fresh fish…” and “… as much physical activity as possible should be promoted, ideally aiming for at least 150/min week of moderate activity…”

In reading these recommendations, it seems that both EASD and ADA are strong health promotion advocates.  In terms of glycemic control, they suggest individual targets for glycemic control should be established through conversations between the doctor and patient with the added note that “…healthier patients should be given lower glycemic targets (e.g., HbA1c <6.5-7.0 percent) and tighter control of body weight, blood pressure, and circulating lipids…” to preserve the patient’s quality of life.  This is very instructive to health promotion practitioners, but certainly not prescriptive.

Another study cited by DiNardo et al. is presented as evidence that Diabetes Prevention Program (DPP) is not effective because it was stopped prematurely.  The authors imply that the study was stopped because of poor long-term results or, even worse, that the intervention may have been harmful to patients.

To refresh readers’ memories, the DPP was a clinical trial that randomly assigned overweight or obese patients to an “intensive” lifestyle intervention that promoted weight loss through decreased caloric intake and increased physical activity.  Intervention group patients were compared to control group patients who only received “diabetes support and education programs.”

It should be noted that the control group received three group sessions annually focused on diet, exercise, and social support, over a four-year period.  In subsequent years, that level of “support” for controls was limited to once a year.  In my opinion, the control group received a substantial intervention, although not as intensive as the intervention group.  Regardless, the reason the study was stopped was there were no significant differences in cardiovascular morbidity and mortality after 10 years.  Death rates were about the same (1-3 percent per 100 person years) for both treatment and control group subjects.

However, there were noteworthy differences between the treatment and control groups.  Treatment subjects with type-2 diabetes were able to lose weight and maintain modest weight loss over 10 years (with an average weight loss of 6 percent at the end of the trial for the treatment group vs. a 3.5 percent weight loss for the control group).  Other positive findings included improvements in glycated hemoglobin levels, meaning that fewer treatment patients needed insulin.  Treatment subjects experienced a partial remission of diabetes during the first four years of the trial and benefited from reductions in urinary incontinence, sleep apnea, and depression, coupled with improvements in quality of life, physical functioning, and mobility.

For employers, these results should be compelling.  Employers, for obvious reasons, are concerned about functional abilities, productivity, and quality of life.  The DPP showed that overweight and obese individuals’ lives can be positively influenced by intensive programs aimed to increase physical activity, healthy eating, and weight management.  Whether the DPP can be scaled so that it can be delivered to large numbers of employees at an affordable price is yet to be determined.


DiNardo et al. correctly point to a randomized controlled trial demonstrating that smoking causes poor health outcomes, or more precisely that providing anti-smoking advice to middle-aged men resulted in fewer deaths from heart disease and fewer cancers for individuals exposed to a smoking cessation programs.  These results, consistent with observational studies, were reported in a 20-year follow up study by Rose and Colwell.  That study provides strong evidence for the value of smoking cessation programs that are central to most workplace health promotion efforts and supported by the new ACA regulations.

In fact, continuing to smoke after being offered free anti-smoking advice, and declining to participate in smoking cessation programs, does result in a 50 percent penalty for smokers.  Most people would agree that this is similar to smoking bans imposed on airplanes, restaurants, and workplaces.  Although smokers are adversely affected by such bans and their freedoms are impinged upon, the public at large benefits.

Hopefully, a 20-year prospective trial focused on a single risk factor is not the minimum requirement for providing incentives for behavior change.  It would be ideal if we could conduct these decades-long studies examining all common risk factors addressed in workplace health promotion programs to determine whether they indeed result in reductions in costs and improvements in productivity.  However, that is an unrealistic goal.  Those studies will not take place, at least not in our lifetimes, for several reasons: the difficulty of tracking health care utilization and productivity cost trends over long time periods; the lack of funding for such research; and, quite honestly, the lack of interest by public agencies in conducting applied research studies in workplace settings.

However, there is some emerging research showing that changing the health risk profile of workers can have an effect on disease incidence.  A study we recently published focused on Vanderbilt University employees; over an 8-year study period, employees who changed their BMI from ≥30 to < 30 had significantly lower diabetes incidence when compared to workers who became obese during that period.  Further, employees who became physically active after being sedentary also significantly decreased their diabetes risk.  Admittedly, similar long-term research performed in an applied setting (workplace) is still quite rare and certainly subject to multiple internal and external validity problems.

Back to smoking, the Rose and Colwell study proves that getting people to quit smoking will result in fewer heart-related deaths and cancers.  So, why is it not in the employer, employee, and society’s interest to provide financial incentives to do so?  The employee will be healthier, the employer will benefit from increased productivity (e.g., fewer smoking breaks on company time), and society will not spend as much in the form of Medicare payments for the treatment of avoidable cases of heart disease and lung cancers.  I still hold that this is a “win-win-win” for all involved.


DiNardo et al. revisit the issue of ethics — whether it is fair that the most vulnerable employees, those from lower socioeconomic strata with the most health risks, bear a greater cost that, in effect, subsidizes their colleagues with fewer health risks.  They refer to a study by Harold Schmidt in which 1,000 U.S. residents were surveyed about their opinions on three types of benefit plan incentive designs: “pure carrots” – rewards; “pure sticks” – penalties; and “false carrots” – where rewards are made possible by increasing insurance costs for everyone and then giving compliant individuals an opportunity to “buy back” or negate the additional premium amount.  In the study, respondents were randomly assigned to one of the three conditions in which the carrots or penalties were tied to achieving a BMI target.

Schmidt found that 60 percent of respondents (about two thirds of whom were overweight or obese) agreed that it was reasonable to align some form of penalty or reward with meeting BMI targets.  Higher-weight respondents set lower levels of penalties (or rewards) for not meeting targets, but even overweight and obese individuals agreed to a $150 differential in premium for those with high BMI values compared to those at normal weight.  As would be expected, lower income people felt the penalty or reward should be set at a lower amount compared to higher income people, but in the carrot model, that differential in incentive amounts was only $50, with higher income people recommending a $200 reward for a normal BMI and low-income people suggesting a $150 reward.  Schmidt concludes, and I agree, “…care is required to ensure that employees do not perceive any form of incentive program merely as unfair cost shifting and reject the approach as a whole.”


Finally, DiNardo et al. point to the difficulties involved in changing health behaviors, especially achieving a healthy weight, as an argument against setting goals and financial rewards related to weight loss.  There are references to books and articles illustrating how multiple approaches to getting Americans to lose weight have been abysmal failures.  They highlight a book by J. Eric Oliver, Fat Politics, that claims that “a handful of doctors, government bureaucrats, and health researchers with financial backing from the drug and weight-loss industries, have campaigned to create standards that mislead the public.”  I did not read the book, but it sounds rather conspiratorial.  I doubt that the CDC and other federal agencies are being paid off by the drug and weight-loss industries to sound an alarm about the epidemic rise of obesity rates among adults and children.  Until I am convinced otherwise by good scientific evidence, I will continue to believe that obesity, along with physical inactivity and poor diet, are genuine health risks that exert a substantial cost on society both health-wise and dollar-wise.

DiNardo et al. also highlight the results of the Multiple Risk Factor Intervention Trial (MRFIT) study as an example of why “exhortations to individuals to lose weight may do little to improve health, and in some cases may actually make matters worse.”  That certainly is a bold statement.  In reviewing the main findings from the MRFIT study, I noted that MRFIT researchers underscored “disturbances in human culture” as producing “epidemic” increases in risk factors associated with coronary heart disease (CHD) that include adverse dietary patterns, cigarette smoking, and sedentary lifestyle; these in turn account for adverse levels of serum cholesterol, blood pressure, and other metabolic risk factors.

As for DiNardo’s statement that “exhorting” people to lose weight is not very effective – I agree.  Exhorting anyone to do anything that he or she does not want to do is usually futile.  However, today’s employers are adopting policy and environmental strategies to support workers in their efforts to lead healthier lives — potentially more powerful than simply “telling” workers what to do.  What are these strategies?  They include providing healthy food choices in cafeterias and vending machines; making healthy foods the easy and cheaper choices; labeling “healthy” items; only allowing healthy food at company-sponsored events; selling half portions in the cafeteria; giving employees smaller plates; providing free water; making people wait for unhealthy food to be prepared; subsidizing fruits and vegetables; providing healthy cupboards; paying for microwaves and refrigerators; educating workers about how to prepare healthy meals; supporting introducing soda taxes in the community; building bike and walking trails on campus; opening stairwells; slowing down the elevator to encourage the use of stairs; promoting public transport; subsidizing gym membership; working with local schools to increase physical activity programs and healthy cafeterias; promoting farmers’ markets; and provide educational seminars, workshops, or classes on nutrition.

The above list is purposely long to highlight the many ways employers can encourage healthy habits for the workers without “exhorting” them to do so.  In fact, the reason that Americans are becoming more obese is largely due to factors other than their reduced will power.  Americans are driving more and doing less walking; eating out more often; increasing their screen time in front of televisions, computers, and video games; performing more sedentary work; and taking advantage of innovations such as moving sidewalks, escalators, automatic doors, and elevators, not to mention dish and clothing washing machines and electric powered windows in cars.  If we, as a society, devote more time and attention to the social and ecological factors causing us to become more obese, we would need less “convincing” and exhortation.

Bottom Line

I believe this dialogue between supporters and opponents of workplace health promotion programs has been helpful.  The discussion surfaces the legal, ethical, and practical issues related to organizational efforts to improve the health and well-being of workers.  It has also unearthed actual and suspected threats to individual rights and has shaped public policy so that it is balanced and supported by evidence.

Unfortunately, all too often, health policy debates are driven by emotion and ideology instead of science.  Ongoing challenges to “conventional wisdom” should be encouraged — and if the evidence points us in new directions, so be it.  A sober and unemotional review of the evidence will lead us to good public policy.  After all, we all want to achieve a common goal: improving the health and well-being of Americans, and doing so in a fair and equitable way.

Thank you again to DiNardo et al. for elevating this discussion to new heights.