Boosting Workplace Wellness Programs With Financial Incentives

Alison Cuellar, PhD; Amelia M. Haviland, PhD; Seth Richards-Shubik, PhD; Anthony T. LoSasso, PhD; Alicia Atwood, MPH; Hilary Wolfendale, MA; Mona Shah, MS; and Kevin G. Volpp, MD, PhD

Several strategies have been suggested to promote use of preventive services, including expanded incentives and greater involvement of employers.1 Under the Affordable Care Act (ACA), new incentives were created to promote employer wellness programs and encourage employers and employees to support healthier workplaces. The law enables employers to offer financial rewards to employees who participate in wellness programs or meet certain health-related targets.2 Increasingly, employers offering wellness programs are incorporating these financial incentives with the belief that they will boost the impact of their programs; however, little is known about the effectiveness of these incentive programs. 

In 2015, 14% of all employers, half of employers with 200 or more employees, and two-thirds of employers with 1000 or more employees offered wellness programs.3,4 These programs include health screenings that collect self-reported health risk information and biometric data from an in-person health examination conducted by a medical professional.1 Many employers incorporate financial incentives into their wellness programs. Thirty-two percent of employers with biometric screening programs incorporate a financial incentive for employees who complete the biometric screens; among large employers (≥200 employees) with wellness programs, 56% do so.

The use of incentives in conjunction with wellness programs is largely driven by employers’ beliefs that programs with incentives are somewhat (64% of employers who use incentives) or very effective (27%). Incentives are disbursed in a variety of ways, including premium discounts, waivers of cost sharing, or additional covered benefits. In practice, typical incentives range from $150 to $500, while the average annual premium for single coverage is $6251.1

Despite their widespread use, systematic evidence about the effectiveness of wellness programs with financial incentives is lacking. One literature review concluded that incentives were effective at increasing participation in self-reported health assessments, but was unable to assess their impact on the completion of preventive screenings.5 In randomized controlled settings, financial rewards not paired with wellness programs have been effective at increasing participation in health risk assessment completion,6,7 smoking cessation,8 weight loss,9,10 and chronic disease management.11-13 Randomized trials provide significant proof of concept that targeted financial incentives can be effective; however, as with all randomized trials in which participation requires an opt-in consent, it is difficult to assess generalizability, as many such trials only enroll 10% to 15% of potentially eligible individuals and participants likely differ from those who do not volunteer to participate on unmeasured characteristics, such as motivation.

The current study takes advantage of a sizeable natural experiment in which 39 large employers within the United States initially offered wellness programs without financial incentives. Over time, 15 of them added financial incentives and the remaining 24 did not. Financial incentives were attached to specific health actions, including annual preventive visits, biometric screening, and selected screening services for diabetes, heart disease, and cancer. Employees received personalized scorecards, both online and mailed to their home, to help them track their progress toward receiving incentives, which were awarded as premium reductions, cash, or gift cards. To our knowledge, ours is the largest study of wellness programs with financial incentives to date and includes over 1.4 million insured enrollees. A key strength of our design is the ability to examine pre-post effects of the introduction of incentives with a contemporaneous wellness program control group that did not introduce incentives. 

In addition, we considered whether the financial incentive effect is greatest for nonregular users of preventive services or for users who had received the service in the prior year. The financial incentive program is designed to be broad-based rather than targeted. Nonetheless, a program is more economically efficient if it entices new or nonregular users of prevention services rather than rewarding individuals who likely would have received the services in any case. 

METHODS

The Wellness Program

The wellness program we studied allows enrollees to earn dollars or points for adopting better health behaviors. Enrollees are provided a personalized scorecard, which includes health actions that were completed, as well as those that were not, as an aid to maintain or improve their health behaviors. The points earned for various activities can be converted into cash rewards, premium reductions, or gift cards in a manner set by the employer. We examined clinical screening outcomes that can be measured via claims data. The incentive for any given clinical screening was the same for all covered individuals within an employer, but could vary across employers. Incentive amounts ranged from $0 to $80 for preventive visits, $0 to $100 for low-density lipoprotein cholesterol (LDL-C) tests, $0 to $100 for blood sugar ascertainment with glycated hemoglobin, and $0 to $75 for cancer screening tests. The maximum annual award that an individual could earn for receiving all of these services ranged from $250 to $900 across employers. By contrast, employers without financial incentives in their wellness program promoted and measured the same outcomes, but employees did not receive an explicit financial reward.

Study Setting and Employers

All employers used the same insurer to administer their wellness programs, allowing us to observe preventive and health promoting behaviors before and after financial incentives were implemented. The 15 treatment employers introduced financial incentives into their wellness programs between 2010 and 2012, providing variation in the start dates. Individuals’ outcomes were observed as long as they were covered by the insurer. In many cases, the insurer was the sole provider of coverage for the employers. The staggered implementation by employer is illustrated in the Figure. Our data span January 2009 through December 2013. 

Data and Study Variables

We used de-identified healthcare claims and health plan enrollment and wellness program data from the insurer. Outcome variables were obtained from claims data based on standard claims codes. These include utilization of annual preventive visits, LDL-C tests, fasting blood sugar tests, and breast, cervical, and colon cancer screens. Coding details are provided in eAppendix Table 1 [eAppendices available at ajmc.com]. We were able to determine chronic conditions from claims, but not body mass index (BMI). From enrollment data, we obtained individual age, gender, dates and type of coverage, and whether the insured member was an employee or adult dependent. Insurance enrollment data typically have limited demographic information. However, through a vendor, the insurer obtained imputed information on race, ethnicity, and education of covered members, all potentially important factors that could influence an individual’s propensity to seek preventive care. 

Population

Our sample was restricted to adult (aged 18-64 years) employees who were covered by the insurer administering the wellness programs for at least 12 continuous months (1 full plan year). Spouses and domestic partners were included if they, in addition to the employee, were offered the wellness program. All sample members were included in models that examined annual preventive visits, LDL-C screening, and fasting blood sugar screening. Different subgroups were examined for each preventive screening (ie, women aged 40-64 years for breast cancer, women aged 18-64 years for cervical cancer, and men and women aged 50-64 years for colorectal cancer). Individuals older than 64 years were excluded because they were eligible for Medicare coverage. eAppendix Figure 1 illustrates our sample construction. 

Statistical Approach 

Our study employed a panel-data difference-in-differences (DID) design in which 39 employers offered wellness programs without financial incentives at baseline, 15 employers added financial incentives and the remaining 24 comparison employers did not.14 This is also commonly referred to as a stepped-wedge design. Because all employers used the same national insurer to administer their wellness programs, we observed preventive and health promoting behaviors before and after financial incentives were implemented based on common data collection. 
The effect of the incentive is estimated for each preventive service outcome using variations on equation 1, which reflects a staggered DID or stepped-wedge regression model. Equation: 

PrevServit= β0 + β1Treatmentit + β2 Xi + β3 HealthPlanit + γf Employer + γtYear + εit

The observations are at the individual-year level, where i denotes the individual and t denotes the year. The dependent variable, PrevServit, is an indicator variable that equals 1 if the relevant preventive service was received by individual i in year t. X represents a vector of individual covariates, including age; gender; imputed race, ethnicity, and education; whether the covered individual was an employee or dependent; whether the individual had asthma, coronary artery disease, congestive heart failure, chronic obstructive pulmonary disease, diabetes, or hypertension; and whether the individual was offered a high-deductible health plan with a health savings account or health reimbursement arrangement. Year and employer indicators are also included. Including employer indicators allows us to control for the average differences across employers in any time-invariant observable or unobservable employer-level predictors. The error term is represented by ε

Our key independent variable is represented in equation 1 by “Treatment.” The variable Treatment is a binary variable that takes the value 1 in any year that the individual is offered a wellness plan with incentives and 0 in all other years. Controlling for individual and other plan characteristics, the coefficient on this variable is interpreted as the average change across enrollees in employers with financial incentives relative to the average change for those with standard wellness programs. Standard errors were clustered by employer, and all reported P values were for 2-tailed statistical tests.

We estimated multivariate linear probability models to isolate the degree to which offering a wellness program with financial incentives in a given year influenced the probability of an employee receiving a selected preventive visit, cancer screening test, or blood test in that year relative to employees of employers that offered wellness programs without financial incentives. We selected linear probability models because their coefficients can be interpreted as marginal effects of treatment and because our treatment variable was binary. 

In order to identify whether the program effect was greater for nonregular users of services versus users who had recently received the service, we repeated our models adding an indicator for receipt of the service in the prior year and its interaction with the incentive indicator. Although we used all years of data for each employer, we restricted the models to individuals present in the data for 24 continuous months. Introducing a lagged dependent variable (ie, prior receipt) could lead to bias if there is serial correlation in unobserved individual-level factors and it is greater or lesser at employers offering financial incentives.

We addressed observable differences between treatment and comparison employers with treatment-on-the-treated propensity score weighting. Our outcome models were weighted using inverse probability weights obtained from a series of propensity score models. The weighting models were estimated using boosted regression, as implemented in the “Toolkit for Weighting and Analysis of Nonequivalent Groups” package in R,15 which predicted the probability of being a treated observation in the year before those employers added financial incentives based on on age, gender, race and ethnicity, Census region, urban location, chronic conditions, and offer of high-deductible coverage with a health savings or health reimbursement account. Because the study used a DID comparison design, we weighted both treatment and control observations for each year to match the baseline year for the treatment group in order to balance the distribution of covariates both over time for each treatment group and between the treatment groups.16 This allowed us to control for any compositional changes over time in the treatment or comparison group as well as provide appropriate weights to observations in the control group. Separate weighting models were run for each treatment group by year combination. The high-deductible health plan variable was not balanced by weighting in all years and was removed from 6 of the 11 propensity score models, but was included in all outcome models. Weighting models were re-estimated for each of the cancer screening eligibility subgroups defined by age and gender. The propensity score weights were applied in calculating all reported results except sample sizes.

RESULTS

Study Population and Covariate Balance

The characteristics of individuals in the treated employers were substantively similar to those in the control employers in the baseline year after weighting (Table 1). Baseline rates for receiving a full biometric tests and breast cancer screening were 3 to 5 percentage points lower in the employers that did not introduce incentives. Thus, introducing financial incentives does not appear to be a selected response by employers to low baseline rates of targeted services. Covariate balance from propensity score weighting is typically measured by standardized mean differences (eAppendix Figure 2). The balance was successful; among 365 comparisons, all were below 0.12 and only 4 were greater than 0.10.

Preventive Visits and Blood Tests for Disease Screening

Baseline rates for preventive visits were an average 36.1% in the treatment group. In years when employers offered a wellness program with financial incentives, members were 7.7 percentage points more likely to have wellness visits (P <.05) (Table 2), a 21.3% increase. They were also 7.9 percentage points more likely to have cholesterol screenings and 7.1 percentage points more likely to have blood sugar tests for diabetes (P <.05). Results for the blood sugar and LDL-C tests were similar, and 95% of members who received the LDL-C test also received a blood sugar screening test. When considered as a set, the financial incentives resulted in 8.1% more individuals having all 3 biometric components (ie, a preventive visit and the 2 blood screenings) over a baseline rate of 19.0%, a 42.6% increase over baseline. Full regression results are shown in eAppendix Table 2.

Cancer Screening Tests

The wellness program with financial incentives was associated with smaller increases in cancer screening rates (Table 2). Financial incentives were associated with a 2.7 percentage-point increase in mammography (P <.05) and a 2.2 percentage-point increase in colorectal cancer screening (P <.01). Relative to baseline rates in the treatment group, these changes represent 5% to 7% improvements in screening rates. No differences were detected for cervical cancer screening rates over time in employers who offered incentives compared with those who did not.

Differences by Prior Service Use

Individuals who sought services in a given year were 17% to 30% more likely than others to receive them in the following year, controlling for other characteristics (Table 3). Estimates of the incentive program’s effect by prior receipt of service are mixed, but 2 results emerge. The program had a significantly greater impact on receiving the full biometric screen for those who had previously had one, by 6.1 percentage points (P <.05). In contrast, for the cervical and breast cancer screening tests, the program had a greater effect on individuals who did not receive screening in the prior year, meaning that here the incentives had a stronger effect on nonregular service users. The effect of the incentive on nonregular users was 3.1% (P <.01) for cervical cancer and 4.6% (P <.01) for breast cancer. We did not detect different impacts of incentives on colorectal cancer screening by prior year receipt. 

For all preventive tests, we found a strong association between having the test in the prior year and repeating it, independent of the incentive. Although federal screening recommendations do not support annual cancer screens for all individuals, we note that our data include a mix of individuals for whom an annual cancer screen would not be recommended and others who may have had positive results in the past and for whom annual exams are recommended.17-19

DISCUSSION

Within a single national insurer, workplace wellness programs paired with financial incentives were associated with increased utilization of targeted preventive services relative to worksite wellness programs without incentives. Increases ranged from 3% to 42% over baseline rates. The largest impacts were seen for receiving a full biometric screen (ie, preventive visit and 2 lab tests), as our study found that adding financial incentives to wellness programs nearly doubled the number of individuals receiving a full biometric screening exam. These results represent the first national data on the impact of adding financial incentives to wellness programs affecting all employees within a set of employers. Although data from randomized trials show significant impacts of financial incentives on health behaviors for self-selected populations of participants, these data speak to the impact across entire employer populations of implementing wellness programs with financial incentives geared toward increasing prevention and screening.

Efficiency and equity are potential challenges in wellness programs with financial incentives. Rewarding individuals who would receive services regardless of a financial reward is not an efficient use of resources, yet under the ACA, wellness programs implemented in employer settings are required to apply to all “similarly situated” employees and therefore do not allow programs to target just those individuals who would otherwise be unlikely to get a given program. Given this, it is interesting that we found that the incentive effect was similar for individuals who did or did not receive preventive visits, screening blood tests, and colorectal cancer screens in the past year and that it was more effective for individuals receiving prior-year services for the full biometric test. Thus, the financial incentives were not more systematically effective at bringing in “new” or infrequent users than “prior” or more frequent users for these services. This was not the case for cervical and breast cancer, where those who had not received screening in the prior year were more likely to be impacted by the reward.

Limitations

Our study has several limitations. First, we evaluated the incentive programs as they were implemented, which resulted in a limited range of incentive values. We cannot assess the impact of larger incentives. Second, we have taken several steps to mitigate potential selection bias on observed characteristics, yet we cannot rule out bias on unobserved characteristics that would occur if the trends in targeted service use systematically differed for the treatment versus comparison groups for reasons other than the addition of financial incentives. We were, however, able to rule out that employers with lower baseline use of targeted services were more likely to add financial incentives to their wellness programs. Third, because our analysis required at least 12 months of continuous enrollment, we lost members to attrition. If sicker employees were more likely to leave one set of employers than the other, this could bias our results. We addressed this issue partially through the use of propensity score weighting. Fourth, claims data have been found to underreport preventive services relative to medical records.20 Because our sample was composed of privately insured, continuously enrolled individuals, we believe this problem is attenuated. Moreover, any measurement error caused by our claims data should not differ systematically across the treatment and comparison employers or bias our results. Despite being the largest of its kind, the current study is limited to 15 intervention and 24 comparison employers, limiting the generalizability of the results. Finally, we were not able to assess impacts on health outcomes, as these were not included in the claims data.

CONCLUSIONS

We find that the addition of financial incentives to workplace wellness programs has a statistically significant impact on the receipt of targeted preventive care services. Although the magnitude of the incentives evaluated in this study were well below the federally determined maximum of 30% of premiums for outcomes-based incentives, the effectiveness of these programs signals that modest financial incentive programs in workplace settings can drive health behavior in desired directions. Because targeting financial incentives to particular groups, such as individuals who have not had preventive services in the past year, is challenging within the ACA framework, wellness programs may need to rely on other outreach efforts. Overall, our results highlight the potential promise of ACA-induced movement toward greater use of wellness programs in employment-based settings when they are paired with financial incentives.

Acknowledgments 
The authors thank Tanya Lewis-Walls for her descriptions of the UnitedHealthcare United Personal Rewards program. Preliminary results were presented at the 2014 American Society of Health Economists meeting in Los Angeles, CA, and at the 2015 National Institutes of Health, Economics of Prevention Workshop in Bethesda, MD. Dr Cuellar (principal investigator) had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Print | AJMC Printing...