Minnesota MHA e-Briefing
Jean M. Abraham, Ph.D. Assistant Professor Division of Health Policy and Management
MHA e-Briefing is an electronic publication from the Master of Healthcare Administration Program, University of Minnesota
January 2011
How Well Do Wellness Programs Work?
Each year, millions of individuals make resolutions to live healthier lifestyles, whether it is to exercise more, lose weight, or eat more nutritiously. For working Americans with health insurance, many now have additional resources available to them for pursuing these goals: employer-based wellness programs. In many organizations, wellness programs include a variety of components, such as health risk assessments, disease management for those with chronic conditions, lifestyle management and health coaching, weight management classes, smoking cessation, and fitness center discounts.
According to the Kaiser Family Foundation, in 2010 approximately 74% of all employers that offer health insurance as a fringe benefit offered at least one wellness program, up from 58% in 2009. Although wellness programs are still concentrated among larger employers, they are beginning to diffuse to medium and small employers. As a way to improve access to wellness programs by workers at smaller firms, provisions within the Patient Protection and Affordable Care Act provide grants for up to five years to small employers that establish these programs.
Why Do Employers Offer These Programs?
Employers provide health insurance to approximately 163 million non-elderly individuals in the United States. Between 1999 and 2009, employer-sponsored health insurance premiums more than doubled, with average annual premiums reaching $4,824 and $13,375 for single and family coverage in 2009 (Kaiser Family Foundation, 2009). Looking for long-term strategies to address rising costs, employers are implementing wellness programs that encourage employees to make positive lifestyle choices, including getting regular exercise and losing weight. In addition to having a strong incentive to control health care costs, employers provide a useful environment for promoting behavior change among the working-age population because individuals share a common purpose and culture at the workplace, according to Goetzel and Ozminkowski (2008).
The key issue for employers is whether investment in wellness programming yields a positive return on investment (ROI). That is, do these programs motivate employees enough to make sustained changes in behavior that they otherwise would not have made, such that they use less medical care and increase their productivity in amounts enough to offset program costs? A solid study design and careful statistical analysis are essential for obtaining unbiased estimates of a program’s effects on medical care consumption and productivity.
Research On Employer-based Wellness Programs
Dozens of evaluations have been published on employer wellness programs. The results are mixed. Studies that have been done rigorously have shown both positive and negative ROIs. Unfortunately, a large number of studies have not adequately addressed the thorny methodological issues associated with evaluating these programs. In particular, the two main methodological issues are selection bias and regression to the mean. Selection bias occurs when those who voluntarily enroll in a wellness program are different in unobserved ways from those who do not enroll, and these differences are correlated with the program’s effectiveness. For example, if those who participate in a disease management program have lower health care costs because they are better managers of their health conditions, then the program may appear to be more effective than it actually is.
Regression to the mean in the context of wellness program evaluation refers to the natural pattern of returning to an average level of health after an episode of illness. For example, if only those individuals with high health care expenditures are enrolled in a wellness program, comparing the expenditures before program participation to expenditures after program participation may show a decline, but it is not known whether it is due to the program or because the person simply recovered on their own. This issue can be eliminated by comparing medical care consumption or productivity of the participant group to those of a control group with the same pre-intervention characteristics as participants.
Research Being Done at the University of Minnesota
A team of faculty researchers in the Division of Health Policy and Management, including myself, Dr. John Nyman, Dr. Bryan Dowd, Dr. Roger Feldman, and doctoral student, Nathan Barleen, have been evaluating the University of Minnesota’s wellness initiatives over the past several years. Our published research appearing in the Journal of Occupational and Environmental Medicine in 2009 and 2010, as well as work that is in progress shows that both the disease management and exercise-incentive programs are in fact associated with independent reductions in medical care expenditures for the population of employees covered by the university’s health insurance. While our estimated ROI for these programs is not as optimistic as the $3.50 return for every $1 invested that is typically reported by the industry, we are finding that the savings generated by the programs are able to offset the aggregate costs of wellness programming sponsored by the university.
Additional Research Needed to Understand What Works and What Doesn’t
The first generation of studies evaluating employer wellness programs provides valuable insights regarding the potential aggregate impact of these programs on organizations’ bottom lines. However, much research is still needed to understand the determinants of participation, the specific effects of program designs, and employees’ perceptions and satisfaction with these programs. For example, do exercise-focused initiatives attract only those who are already exercising regularly or are programs helping less active employees to change their behavior? Which types of wellness initiatives generate the largest and smallest cost-savings for employers? How do medical care utilization patterns change among those enrolled in disease management? Do financial incentives make a difference in terms of attracting participants? If so, how large should they be and how should they be structured? Should employers use a “carrot” or “stick” approach in terms of tying wellness program participation and health insurance premium contributions?
The research team in the Division of Health Policy and Management is interested in pursuing new wellness program research in organizations outside of the University. If your organization may be interested in collaborating with us, please feel free to contact me at abrah042@umn.edu.
Thoughts, comments, or suggestions for future e-Briefing topics? Contact mrath@umn.edu.