- A study of near 33,000 workers by medical researchers from Harvard and the University of Chicago finds that use of an employee wellness program does not lead to better health outcomes or lower health-care spending.
- Workers at wholesale retailer BJ's Wholesale Club did exhibit improved health behaviors, such as weight management and more regular exercise, over an 18-month period.
- More than 80% of large companies and more than half of small employers offer a wellness program, with an estimated 50 million American workers covered. The study authors conclude that firms may need to temper expectations.
For the tens of millions of American workers now enrolled in workplace wellness programs offered by four out of five large employers, it may be time to slow the treadmill.
A major new study suggests that as more rigorous scientific data becomes available, wellness is not living up to its early promise. The workers in the study did not experience improved health outcomes or better job performance, and neither employees nor the companies saw lower health-care costs.
The study, published Tuesday in The Journal of the American Medical Association, is one of the first large-scale studies of a corporate wellness program designed as a randomized trial. Researchers from Harvard and the University of Chicago tracked employees at a broad cross-section of BJ's Wholesale Club worksites over a period of 18 months.
"If employers are launching a wellness program with hopes of a short-term or quick savings in health expenditures or absenteeism, this study should give them pause," said Katherine Baicker, co-author of the study and dean at the University of Chicago's Harris School of Public Policy.
The wellness program did change health behaviors and awareness, but it did not result in measurable improvements in important physical health outcomes. "You can read that optimistically. The first step has to be paying more attention and having the information, capacity and ability to pay more attention to health behaviors," Baicker said. "But is that sufficient to reduce blood pressure or improve diabetes? We don't have evidence that it does."
High cholesterol levels and hypertension did not differ between the test and control groups, the study found.
Randomizing the study was key to distinguishing this data from past work on wellness, she said.
"The studies that suggested improvements almost all suffered from not being able to tell what would happen in the absence of a program," Baicker explained. "People who participate in wellness are the type of people already paying more attention to nutrition and exercise, and if you don't have a good control group, you will think it has a stronger association with positive health outcomes."
The study of close to 33,000 BJ's workers was conducted at roughly 160 worksites.
The study is the second recent randomized one to come away with sobering conclusions about wellness programs, following the January publication of a National Bureau of Economic Research study conducted at the University of Illinois Urbana-Champaign, which tracked more than 12,000 employees, though only at one worksite, the university campus.
Julian Reif, assistant professor of finance and economics at the University of Illinois and a faculty research fellow at the National Bureau of Economic Research who co-authored the Illinois study, said it looked at 40 different outcomes over a year, going beyond medical spending and absenteeism, but found only a few positive effects — employees having annual health screenings, and employees saying they felt their health and safety was valued by their employer. It also found that a large portion of the workers who do well in these programs are the ones already the most motivated to pay attention to their health.
"This is an important industry to understand, because it has really increased in size, to 50 million workers in the U.S. and $8 billion," Reif said.
The BJ's study found that employee absenteeism, performance reviews and tenure did not improve as a result of participation in the wellness program.
Baicker said the study at BJ's is continuing, with results for a 36-month period planned for future publication. "We will look again and continue to follow this cohort to see if longer-term effects are different, but 18 months is not so short," Baicker said.
Reif said his team continues to study the Illinois group as well, but he said the longer a study has to go to find proof, the less value the proof will have to employers. Many employers have what Reif referred to as a "decent" annual turnover rate in employees (10%), so it is important for these programs to pay off in the short-term for companies when considering the investment.
Baicker's co-author on the study, Zirui Song, assistant professor of health-care policy and medicine at Harvard Medical School, said the study is an important piece of counterevidence to earlier literature that suggested a "sizable" early return on investment.
Both Song and Baicker had worked on some of that earlier research and wanted to pursue a more rigorous, randomized trial because of the limitations they knew existed in the data.
"It is just one piece of evidence," but it does caution against relying on the earlier evidence. "Don't expect a quick or immediate return is achievable," Song said.
The differences in medical spending, pharmaceutical spending and health-care cost-sharing were not significant between the test and control groups, the study found.
In an editorial accompanying the study's publication in JAMA, Jean Marie Abraham of the University of Minnesota's Division of Health Policy and Management noted that four out of five large employers across the country now offer a wellness program, but "the growing evidence that demonstrates limited or no program effects" should encourage wellness companies and employers to critically assess the programs they are offering.
A BJ's spokeswoman wrote in an email, "We're committed to the health and wellness of our Team Members. We'll use the results of this study to help inform programs moving forward."
The company declined to make an official available for further comment.
Debra Wein, founder and CEO of Wellness Workdays, the third-party wellness provider for BJ's, said the fact that there were changes in employee health behavior over 18 months was a positive, especially since the employee base of BJ's is not a homogenous, high-income, white collar demographic, but a diverse workforce where education, income, access to technology and native languages spoken vary across the population.
"What do these programs do for more representative workplaces and workers? The answer seems to be, at least for the first year or so, that they can change behavior but not in a way that translate into health-care savings or reduced absenteeism," Baicker said.
Though she allowed for a more positive reading of her research. "These behavior changes are a necessary first step. ... If we had found no changes in behavior I would be fairly confident that even three years later we would not see any cost savings in health-care spending. ... Seeing some behavior changes is hopeful, or, more hopeful than not."
There are individual examples of companies where wellness has directly influenced the health-care spending trend positively, but Wein stressed that the conversation about wellness has been shifting away from a focus on return on investment even before these recent studies. "The current conversation around wellness has moved dramatically away from financial return on investment and more towards value on investment measures."
She noted that at BJ's, even lacking quantitative measures to make the wellness case, dieticians who worked with the employees witnessed "dramatic changes in culture and perspective," such as healthier food choices in break rooms and more employees drinking water over soft drinks. There were changes that involved both employees and management of the company playing a role in facilitating the healthier options.
"We are looking at quantitative benefits and measurements through the research but there are many stories across all of the BJ's clubs where we found individuals who were able to make positive changes ... physical activity, healthier eating, reduction in prescription medicine use and visits to doctors. Throughout the course of the program many registered dieticians came across people making great strides personally. That can't be captured in studies but to us as practitioners, that is encouraging," Wein said. "I'm not so surprised by the fact that there were no financial changes. Whether recruitment or retention there are lots of other potential benefits.
Public health experts, including Ron Goetzel, vice president of IBM Watson Health and senior scientist at the Johns Hopkins Bloomberg School of Public Health, have recently noted that employers need to play a broader part in public health policy and wellness return on investment is too narrow to deal with the magnitude of U.S. health care treatment and spending issues.
Abraham noted that while there has been a recent move to take a broader view of wellness, the fact is that employer investment in wellness programs is "driven largely" by the idea that they generate health and economic benefits.
"On average, healthier employees have lower medical-care spending, which affects employer-sponsored insurance premiums," she wrote, and concluded the JAMA editorial by saying continued research is important: "Given their broad diffusion, employer wellness programs that are able to demonstrate positive outcomes can provide a valuable complement to health system — and community-based approaches for reducing chronic disease prevalence and its economic effects on society."