ROI of Wellness: How Good Is the Data?

By David Anderson February 22, 2013

As employers consider investing in employee wellness programs, many want to know there will be a positive return on investment (ROI), typically calculated as a ratio of health plan dollars saved per dollar invested.

A 2010 report by a team of Harvard University health economists, published in the journal Health Affairs, found that the average medical cost savings per dollar invested in wellness programs was $3.27. The report's finding was based on an analysis of more than 20 peer-reviewed ROI studies.

Few other health-related investments come close to having this much ROI support.

Other Views on ROI

A 2012 report by the not-for-profit International Foundation of Employee Benefit Plans, A Closer Look: Wellness ROI, similarly found that most North American employers that have analyzed the ROI of their wellness programs have found $1 to $3 decreases in their overall health care costs for every dollar spent. See the SHRM Online article “Study: Wellness Programs Saved $1 to $3 per Dollar Spent.”

Moreover, lowering health risk factors to their theoretical minimums, if this were possible, would reduce average annual health care costs per working-age adult by 18.4 percent, according to study results published in the January 2013 issue of the Journal of Occupational & Environmental Medicine.

Research Challenges

So, why is the ROI of wellness still controversial and why are many employers still hesitating to invest? Despite the favorable review, the 20 studies in the Harvard/Health Affairs report varied widely in methods and rigor, and none met all the criteria for a gold-standard randomized controlled trial that would definitively answer the key question of whether the lower health care costs for employees exposed to a wellness program were caused by the program. Unlike lab rats, employees can’t be randomly assigned to treatment and control groups in a sterile environment where even the white-coated lab technicians don’t know which group the rat is in.

Because a gold standard study isn’t feasible, researchers have had to resort to less definitive “quasi-experimental” approaches. Current industry evaluation experts recommend a pre-/post-measurement, participant vs. non-participant design with statistical adjustment to account for baseline differences across different groups.

This research jargon is referred to as a “differences-in-differences” approach and simply means researchers compare medical cost trends of employees participating in the wellness program to trends of employees who don’t participate, while using complicated statistics to try to overcome a problem of “selection bias” that may occur because employees themselves choose whether or not to be in the wellness program rather than being randomly assigned.

But researchers can only make these adjustments based on known differences, typically demographic and health factors. Motivational differences that may lead employees both to participate in the wellness program and make changes in their health—possibly with or without the program—are not accounted for or even fully understood.

Another challenge for researchers is that wellness is morphing into population health management that focuses on mobilizing entire populations, including those dealing with chronic disease, to engage in multiple aspects of the program.

Critics demanding scientific certainty will continue to point out that the significantly improved medical cost trends almost invariably found among participants in best-practice wellness programs may be partly or even completely due to selection bias rather than the wellness program. What can’t be argued is that the best practice research method is very costly to execute ($50,000 to $100,000 or more), is most suitable for evaluation of large populations (10,000 or more beneficiaries) and is increasingly complex because the growing use of incentives is eroding true nonparticipant comparison groups.

Where does that leave things? This imperfect evidence that wellness programs reduce medical costs is better than the data corporate executives have for most key business decisions. Even if it were the only evidence, a best-practice wellness investment would still be prudent. Fortunately, multiple studies by benefit consulting firms have also shown that employers who invest in employee health have substantially lowered medical cost trends and achieved better results on other performance measures important to their CFO and shareholders. This evidence also is far from perfect, but it’s based on a complementary approach and focuses on the key issue for many employers—slowing medical cost trends to a sustainable level.

The imperfect evidence that wellness programs
reduce medical costs is better than
the data corporate executives have for
most key business decisions.

How to move forward in this imperfect world? To begin with, investments must align with goals. If health plan costs are increasing 7 percent annually per employee and the goal is 2 percent, the average employer would have to reduce its trend by over $500 per employee plus the additional cost of the wellness program. If the Harvard health economists' ROI estimate of about 3:1 after about three years is correct, that would mean investing about $250 per employee annually on health and well-being. Our own research suggests that a 2:1 ROI by the third year of a program may be more realistic. This more conservative target increases the required investment to about $500 per employee annually to hit the medical-cost trend target.

Employers spending this kind of money need evidence to justify the value of their investment in population health management. Given the drawbacks to a costly medical claims-based ROI study, for most employers a straightforward measures of progress is advisable in five broad categories:

  • Engagement. The term is overused, but active leadership engagement drives culture change, and year-round employee engagement drives individual health improvement. If a metric doesn’t pass the smell test for real engagement likely to drive change, demand metrics that do.
  • Health trends. Consider if your company’s rates of smoking, obesity, elevated blood pressure and cholesterol, depression, diabetes, heart disease and other health measures are performing better than external benchmarks. In cases like obesity, not getting worse is an initial measure of success.
  • Health care use trends. A goal for most employers is to slow unnecessary use of health care services, which drives up costs. A leading indicator of success is a shift away from expensive in-patient hospital stays toward more physician visits for preventive exams, nonurgent care and medication adherence.
  • Medical cost trends. Relevant metrics include whether current medical cost increase trends are better than they were before the program, better than actuarial projections and better than outside benchmarks. Because employers do many things to slow their cost trend, it’s difficult to determine the exact cause. The bottom line, however, is whether the sum of the parts is achieving targets.
  • Performance trends. The questions here are similar to those for medical cost projections. They may be applied to clearly health-related metrics like absence and disability costs. They also may be applied to employee performance metrics and, possibly, business results. These are difficult to attribute to the wellness program per se, but the bottom line is success.

Researchers and program evaluators should continue building evidence on the value of wellness and what works best, using the best methods available to them. Perhaps someday large-scale, multi-employer research will provide more definitive evidence. In the meantime, employers need to say “good enough, move on” when it comes to promoting employee wellness, because inaction is simply not a sustainable option.

David Anderson, Ph.D., is senior vice president and chief health officer for StayWell Health Management, and the architect of StayWell’s health assessment model and predictive modeling tool.

Related Articles:

Keep up with the latest news. Sign up for SHRM’s free Compensation & Benefits e-newsletter



Hire the best HR talent or advance your own career.

Move your HR career forward.

Apply for the SHRM-CP or SHRM-SCP exam today! Applicants now have the option to test from home.

Apply for the SHRM-CP or SHRM-SCP exam today! Applicants now have the option to test from home.



HR Daily Newsletter

News, trends and analysis, as well as breaking news alerts, to help HR professionals do their jobs better each business day.