Workplace Wellness Programs to Improve Employee Health Behaviors in the United States

Researchers:
Zirui Song
Fieldwork by:
Location:
United States (Multiple Locations)
Sample:
200 BJ’s Wholesale Club sites
Timeline:
2014 - 2017
Target group:
  • Workers
Outcome of interest:
  • Enrollment and attendance
  • Health outcomes
Intervention type:
  • Preventive health
  • Monetary incentives
AEA RCT registration number:
AEARCTR-0000586
Partners:

In 2018, more than 4 in 5 large firms and half of small employers in the United States offered a wellness program. Though workplace wellness is an $8 billion industry, there is little rigorous evidence on the effects of these programs. Researchers conducted a randomized evaluation of a comprehensive, multi-site workplace wellness program to evaluate the program’s impact on self-reported and clinical health outcomes, health care spending and utilization, and employment outcomes. After 18 months, individuals who were offered the wellness program reported improved health behaviors for two out of the 29 self-reported outcomes (regular exercise and active weight management). The program did not have a measurable effect on any other self-reported or clinical measures of health, health care spending or utilization, or employment outcomes.

Policy issue

In the United States, workplace wellness programs have increasingly appealed to employers aiming to improve employee health and productivity and lower health care costs. Half of all small employers and over 80 percent of large firms offered a wellness program in 2018. The growth of this industry has been aided by public investment, such as the Affordable Care Act, which included funds to promote the development of workplace wellness programs. However, little rigorous evidence exists on the effects of such programs. Can an employee wellness program improve health and well-being, reduce employee absenteeism and turnover, and lower health care utilization and spending?

Context of the evaluation

BJ’s Wholesale Club (BJ’s) is a large employer that operates about 200 warehouse stores with about 26,000 employees across the Eastern United States. BJ’s has a diverse workforce, and a substantial share of BJ’s workers are lower income. About 55 percent of the firm’s employees are white, about 20 percent are black, and about 18 percent are Hispanic. The majority (about 84 percent) of the firm’s employees are paid hourly, with an average annual pay of about $25,000 for full-time workers (about 46 percent of employees), and about $10,000 for part-time workers (about 38 percent of employees).

Photo: Wikimedia Commons | Tdorante10

Details of the intervention

Researchers conducted a randomized evaluation to measure the impact of a wellness intervention on employee health and well-being, worker absenteeism and turnover, and health care spending and utilization. Researchers randomly selected 40 BJ’s sites and assigned 20 sites to the treatment group and 20 sites to the primary control group, with the remaining sites serving as secondary controls. Each site has about 100 employees at any given time.

Employees at sites assigned to the treatment group were invited to participate in a package of eight wellness program modules, including individual and team-based activities and challenges related to topics such as nutrition, physical activity, and stress reduction. Modules included modest incentives to participate, typically a $25 BJ’s gift card for completing a particular module; the average total potential incentives across the program was approximately $250. Wellness Workdays, an established wellness vendor, designed and implemented the program in partnership with researchers and BJ’s.

Researchers measured outcomes using survey and clinical data collected from the 20 treatment and 20 primary control sites, including: (1) surveys of health behaviors, physical and mental health, and overall well-being; and (2) on-site clinical health assessments, including cholesterol, diabetic blood sugar control, blood pressure, and body mass index. Researchers also measured outcomes using administrative claims and employment data collected from the 20 treatment and 140 primary and secondary control sites, including: (3) employment records, including absentee and sick days, job tenure, and work performance; and (4) insurance claims records for those insured through BJ’s, including physician visits, emergency department visits, hospitalizations, and prescription medication use.

Results and policy lessons

After 18 months, employees at worksites that offered the wellness program reported improved health behaviors for two out of the 29 self-reported outcomes (regular exercise and active weight management), but the program did not have a measurable effect on clinical measures of health, health care spending or utilization, or employment outcomes.

Employees working in treatment sites that offered the wellness program were 8.3 percentage points (13 percent) more likely to report that they engaged in regular exercise, relative to the control group’s average of 61.9 percent.1 They were also 13.6 percentage points (25 percent) more likely to report that they were actively managing their weight, relative to the control group’s average of 54.7 percent.

The workplace wellness program offered to employees in treatment sites did not have a detectable effect on any other self-reported health and behavioral outcomes, clinical measures of health (such as blood pressure or body mass index), health care spending and utilization, or employment outcomes (such as absenteeism or work performance) relative to the control group.

The findings from this study are largely consistent with those from another randomized evaluation of a large-scale workplace wellness program at the University of Illinois at Urbana-Champaign, which found that the program did not have an effect on any measured health or cost outcome in its first year. One key difference that may help explain the improvement in two self-reported measures of health behavior is that the wellness programming at BJ’s was implemented at the worksite (rather than individual) level. This may have facilitated changes in workplace culture and harnessed peer effects for behavior change.

The results from both studies contrast with much of the prior literature on workplace wellness programs, which tended to find positive and large returns on investment. Many of these prior studies could not rule out the effects of selection bias; for example, it is possible that the people who chose to participate in the wellness programs were already healthier than those who did not. This study eliminated concerns of selection bias by randomly assigning the offering of a wellness program to different worksites. The results suggest that the return on investment of wellness programs in the short term may not be as large as previous studies suggested.

1.
Since researchers made multiple comparisons between the treatment and control groups across a number of outcomes in this study, there was potential for the study to identify an effect even though there was no effect, due to chance. The researchers used a statistical technique to mitigate this; the effects reported took this statistical adjustment into account.