Article Title

Impact of Employee Wellness Program on Health Outcome: A Propensity Score-Matched Analysis

Publication Date



health care financing, insurance, premiums, program evaluation, demographics, complex disease management, multiple chronic conditions, health promotion, prevention, screening, incentives in health care


Background: Since 2012, Geisinger Health Plan has redesigned its employee wellness program MyHealth Rewards (MHR) by requiring biometric screenings and goal achievements for blood pressure, body mass index, glucose and cholesterol levels to be eligible for premium discount in subsequent year. This study is designed to evaluate the impact of MHR on stroke and myocardial infarction.

Methods: Claims data from 2011 to 2015 were retrieved from Geisinger Health Plan, restricting to continuously enrolled members. Four mutually exclusive cohorts were identified: Geisinger Health System (GHS) employees who met goals (Group 1); GHS employees who met goals some years (Group 2); GHS employees who never joined MHR (Group 3); and non-GHS employees (Group 4). After one-to-one propensity score matching was used to balance the baseline characteristics (eg, sex, age, plan type, chronic condition, utilizations), 11,239 GHS employees (Group 1: 2,842; Group 2: 3,999; Group 3: 4,398) and matched non-GHS employees of equal numbers were included for analyses. Kaplan-Meier method and Cox proportional hazards models were used to estimate the difference in time to first stroke or myocardial infarction between GHS and non-GHS employees.

Results: After one-to-one propensity score matching, both GHS and non-GHS cohorts had similar baseline characteristics; yet, GHS employees in Groups 1 and 2 had a consistently higher probability of an event-free outcome at each time period since 2011 than the non-GHS employees, both in terms of stroke and myocardial infarction. However, the differences were not statistically significant. The estimated hazard ratios associated with the GHS employee status across all subgroups were 0.75 (P = 0.14) for stroke and 0.91 (P = 0.39) for myocardial infarction. In contrast, Group 3 (never enrolled) has higher risk of myocardial infarction (hazard ratio: 1.18, P = 0.30) but lower risk of stroke (hazard ratio: 0.68, P = 0.25).

Conclusion: Although the estimated MHR effects were not statistically significant, the results are consistent with the expectation that MHR may help prevent adverse health outcomes. This expectation is further supported by a separate analysis showing long-term cost of care savings associated with MHR.




June 27th, 2017


August 10th, 2017