INTRO: While expectations for the return on investment for wellness programs should be high, employers need to temper those expectations with a dose of realism. LuAnn Heinen, vice president of the National Business Group on Health, outlines the challenges facing those who want to implement these programs and offers advice on how to plan for them.

Large companies with comprehensive wellness offerings often spend in the range of $200 to $300 per employee per year or around 2% of total claim costs on wellness programs. Increasingly, spouses and domestic partners are eligible for wellness program activities and incentives as well.

Evidence for a positive return on investment on wellness programs is growing in the peer-reviewed literature.

Key review articles summarizing high-quality published return-on-investment studies conclude that strong employer-sponsored wellness programs improve health, save money and should be encouraged as a strategy to slow health cost increases not just in the private sector but also in Medicare.

However, most chief financial officers don't read this literature. Even if they did, the data is old and refers to other, often unidentified, companies.

Instead, CFOs want to see a return on investment for the specific programs offered to their own employees. And they want it yesterday.

%%BREAK%%

Complexity, challenges

Unfortunately there is no roadmap for ROI analysis, and the lack of an industry standard for measuring financial impact means each attempt is custom-designed and may or may not produce valid results. Executing an ROI analysis requires careful answers to many (not entirely straightforward) questions such as:

Which program components will be included — health assessments/biometric testing, health coaching, corporate fitness challenges, Weight Watchers at Work, subsidized healthy lunches, use of on-site fitness facilities, stress management interventions, premium discounts for tobacco cessation, or other? Comprehensive programs by definition are multicomponent, and it is difficult to attribute behavior change or improved health to a single activity; typically evaluations cover many or most of these.

*What is the timeline for data collection? While employers can expect to see behavior change in year one, program startup costs are likely to be high while health care savings lag, resulting in a negative ROI. Reduced medical claims may appear as early as year two, and more commonly in year three, which is when most companies expect to see a positive ROI.

*What can change over the two- to three-year evaluation timeframe, and how might this impact the conclusions? Plan design changes, new vendors with different approaches and data definitions, poor program execution, external events that influence key outcome measures and other hard-to-predict factors need to be considered when interpreting the findings.

*How will employee/dependent participation in the wellness program be defined (e.g., health assessment/biometrics plus at least one additional activity)? While participating in a health assessment or biometric screening is a one-time event, more meaningful engagement is needed to produce outcomes for on-site fitness centers or weight management coaching. For example, participation in a physical activity program might be defined as two times per week for 10 of 12 continuous weeks, while coaching programs usually require at least four sessions to produce outcomes.

%%BREAK%%

*How long do employees need to stay with the company and participate in the program to be included in the analysis? In order to capture potential cost savings, the study population should be restricted to participants and nonparticipants who are continuously employed over the course of the evaluation.

*What results will be measured in addition to health cost impact? What additional bottom-line impacts are important — e.g., absenteeism, presenteeism, disability costs and workers compensation costs — and are the data to measure them available? Leaders in health and productivity analysis with access to integrated data have shown that these costs (and potential savings) appear to be even more significant than medical and pharmaceutical costs.

*What about non-financial effects such as change in health risks? We know today's health risks are tomorrow's claims, so it is desirable to include the impact of wellness programs on people who may not yet be generating medical claims.

*Is there complete data on a sample size that is large enough to detect differences between the comparison groups (e.g., participants and matched non-participant controls) for the chosen measures? This question must be answered by a statistician based on the research design; overall sample size is not usually the problem, but having an adequate sample to look at subsets (e.g. by gender, age, work location, job category or other segmentation) often is.

*Is there added value in measuring health care utilization differences and then attributing costs to these differences for physician visits, hospital/ER use or number of prescriptions? Yes, because medical claim costs can be distorted by network discounts and other confounders.

*How can selection bias be avoided? Simply comparing participants to non-participants will be flawed if employees who sign up for the wellness program are systematically different from non-participants (e.g. with respect to age, BMI, health risks/costs). Statistical techniques are commonly used to minimize this problem.

*Are wellness program costs fully captured? An incomplete “I” tally will make any “R” look bigger.

%%BREAK%%

Planning ahead

One of the most common pitfalls in designing an evaluation is failing to set evaluation parameters in advance. It is not appropriate to decide which employees will be included in the study (e.g., the Omaha office) or counted as participants once the wellness program is under way and relevant outcomes such as weight loss or increased physical activity become apparent. This means that the plan for ROI analysis must be developed simultaneously with the plan to launch a wellness initiative — and pretty clearly this is not the usual course of events.

Most employers will outsource ROI analysis because it is complex, and key choices will determine the validity of the final product. National employee benefit consultants and actuaries, data warehouses, and certain private and university-based research groups do valuable work in this area. Health plans and wellness suppliers also may have the capability and can typically provide estimated savings from their own “book of business” results. If following this path, employers should understand the supplier's approach to calculating ROI and may wish to retain a third party to review the method and results.

The cost of an evaluation depends on its scope and rigor, which are influenced by what key stakeholders want to know. Since budgets are typically limited, discussion about which outcomes are most important and feasible to measure should occur before the evaluation begins. The more outcomes to be tested, the more expensive the evaluation will be. A statistically rigorous comparison study (comparing program participants carefully matched to nonparticipants with similar demographics and risk factors) is likely to run $200,000 or more.

%%BREAK%%

Real-world approaches

Most companies don't have the time, budget or inclination to conduct a full-blown ROI analysis. Representative examples from the wide range of approaches used by large employers follow:

*Employer A calculates wellness ROI by comparing the change in population health risks for a cohort of wellness participants measured over two or more years compared with a reference population of nonparticipants. The per-person savings from reducing certain health risks — 200 fewer smokers, 300 moving from inactive to physically active — multiplied by the number of people amounts to the ROI. Per-person savings is based on costs associated with each health risk from published literature.

*At Employer B, ROI estimates are based on published research showing expected savings for the types of health and productivity programs implemented at this employer. Savings estimates factor in engagement rates as well as the full cost of the programs, incentives and communications.

*A third party expert was retained by Company C to define the method of ROI analysis and review the output. The vendor analysis compares participant to nonparticipant outcomes before and after program implementation. All claims except uncontrollable (e.g., maternity) are included for those eligible for the program and continuously employed.

*Company D uses a disability dashboard to show the number of disability, light duty and paid workers compensation cases, the number of days absent per 100 employees and the dollars spent per 100 employees on absence and disability claims. Success of health and welfare programs is based on this scorecard and on reduction in claim costs for preventable diseases.

*Company E internally calculates its ROI on key programs by taking total savings and dividing it by the total cost of providing the service. In addition, it calculates cost avoidance by comparing the health condition prevalence — diabetes, smoking, high blood pressure — for its population with national averages and applying published costs for each condition.

*Company F takes the total validated monthly savings from its vendor-administered wellness program divided by the total fees paid to the vendor to calculate ROI. An independent third party expert reviews the savings method to ensure accuracy.

%%BREAK%%

Perspective on expectations

ROI studies are difficult. And once a financial ROI is derived — even if it is 3:1 — it doesn't capture everything that is important. Employee wellness programs have a largely unmeasured impact on recruitment, retention and organizational culture. Employee trust in management, willingness to recommend the company as an employer, and voluntary turnover rates all can be impacted.

While employer expectations for an ROI analysis need to be realistic given the challenges and complexities described above, expectations for the wellness program itself should remain high. Reasonable expectations include increased employee/family awareness and engagement around their own responsibility, a better performing workforce through reduced absence and increased productivity at work, and financial benefits from both medical and productivity cost savings at least in excess of the dollars spent.

LuAnn Heinen is vice president of the National Business Group on Health in Washington. She can be reached at heinen@businessgrouphealth.org.