Benchmarking Wellness Programs: How Does Your Program Measure Up?
What is benchmarking?
Benchmarking is a widely used business term describing research that compares one company's business processes to those of other similar or leading companies in the industry. For health management programs, benchmarking may be used to compare program implementation, operations, and outcome metrics. Benchmarking promotes learning from the experiences of others and can help you identify potential areas of concern in your own workforce.
According to Wikipedia the most desirable benchmarking practices involve comparisons to a recognized industry leader. Leading visionary corporate health management approaches are those taken by Bank of America, Dow Chemical Company, Johnson & Johnson, Union Pacific Railroad, and the California Public Employees Retirement System.
Their experiences have been described in several studies. More broadly speaking, benchmark information may be found in the 2004 National Worksite Health Promotion Survey data. Programs evolve rapidly over time and information that is more than just a few years old may have less value than information obtained more recently. We use an ever-evolving dataset for such comparisons. Metrics obtained from these data may help you focus on changes needed to improve your program early on, saving time and resources.
Which programs are most amenable to benchmarking?
One of the reasons benchmarking is such a common and useful practice is that virtually any business process or performance metric is amenable to benchmarking. If leading companies or existing survey data are less compelling for your particular use, benchmarking with reference to similar companies can help to reduce the number of potentially confounding variables that might influence health or productivity outcomes of interest.
Those confounding variables might include industry type, size of the organization, geographic location of the employees, demographic characteristics, and so on. By holding these relatively constant, comparisons of program participation rates, changes in health risks and health care expenditures over time, and comparisons of productivity-related metrics can be made with a fair degree of confidence.
Some of the more common health management program components amenable to benchmarking include health risk appraisal (HRA) survey programs and telephone-based lifestyle coaching and disease management programs. These programs are typically contracted to external vendors who may serve many of your benchmarking peers.
Benchmarking your experiences to those of other companies can help determine if your communication processes, program participation rates, health risk patterns, and health and productivity outcomes seem in line with your peers.
Where can you find benchmarks?
Recent information about what to expect from health management programs is not always readily available. One may query program vendors for information about their books of business, but vendors have a strong financial incentive to provide information from the more positive side of their experiences. The scientific literature can provide another source of information about program designs and effectiveness.
For example, reviews of return on investment from wellness and disease management programs have been published by Chapman and Goetzel et al., respectively. Baicker et al. offer a more recent review of corporate wellness programs. These reviews are likely to represent the best experiences observed in the industry, however, and the authors have acknowledged that less positive findings are less likely to be published.
Exemplary programs may show returns of three or more dollars saved per dollar spent on health management. If there is a strong desire to learn which returns are typical rather than exemplary, published studies may be of less interest. Third-party evaluators may have access to book-of-business benchmarks that show a broader range of experiences. We have analyzed a variety of health management programs for ten employer clients in the last two years.
These clients work in the energy, health, technology, transportation, and financial services industries. They range in size from roughly 9,000 to 120,000 employees, with a median of about 21,000 workers. These clients range from 24% to 85% male employees, with a median age of about 42 years. Five of these clients implemented their programs for the first time in 2005, and 89% of the programs we studied were offered to employees and their spouses.
Our clients also vary according to the number of risk assessment, coaching, and disease management program components they offer. Most include HRA, lifestyle coaching, and disease management programs, but only a few have onsite fitness centers or medical clinics.
With their permission, we present information on their health management program participation rates, changes in health risks over time, changes in health care expenditures associated with changes in risk, and return on investment in their wellness and disease management programs.
Participation rates and incentives
Information about program participation rates provides an early indicator of how a health management program is functioning. Participation rates depend on marketing and communication strategies, ease of access to program services, and the use of incentives to motivate participation. Incentives can be structured as cash or other payouts, free or reduced prices for merchandise, prizes for competitions designed to increase participation, reductions in health care premiums or other out of pocket requirements, or in other ways.
Incentives can help increase participation in health management programs if sustained year over year. If incentives are withdrawn, however, participation rates may drop precipitously. Companies differ in their thoughts about using incentives. Some use incentives frequently, while others do not use them at all. When used, incentives most often target programs that every employee can use (for example, the HRA survey that is designed to find candidates for other wellness, coaching, or disease management programs).
Among our ten clients, incentives range from $50 to $100, payable upon documented participation. Participation rates for their HRA programs varied from 12% to 81% of eligible employees or spouses. When installed for the first time, incentives were associated with participation rates that grew from about 10% to 20% to about 50% to 80%.
In contrast to HRA programs which are designed for entire working populations, lifestyle coaching programs typically target high risk individuals such as smokers, heavy drinkers, poor exercisers, those with nutrition problems, and those who are overweight or obese. Disease management programs are typically limited to those with selected chronic conditions such as diabetes, coronary artery disease, congestive heart failure, depression, or musculoskeletal (back) problems.
Among our clients, incentives to encourage enrollment or to complete programs were used less often for these programs. Participation rates among those eligible for lifestyle coaching and disease management programs range from 8% to 81%, with medians of 41% for lifestyle coaching and 30% for disease management programs.
What about health outcomes?
Health outcomes are typically measured via the health risk appraisal survey and/or biometric testing of blood pressure, blood sugar levels, cholesterol levels, and body mass index (BMI). Different HRAs may target different sets of health risks, but health risks known to be disease predictors are measured on almost all HRAs. These include physical inactivity, stress, blood pressure, cholesterol, smoking, and BMI.
Other risks that are commonly measured relate to poor eating habits, depression, safety belt use, and measures of job or life satisfaction. The scientific literature shows a strong positive relationship between the number of risks that are observed and medical and pharmaceutical expenditures. Examples include studies by Musich and Edington.[5,6] Our clients vary on how they characterize their populations in terms of these risks.
For example, some consider those with 0 or 1 risk to be at low overall risk, while others consider those with 0 to 2 risks to be at low overall risk. In our benchmarking database, roughly one-fifth of the HRA participants are at low overall risk, about two-fifths are at medium overall risk (i.e., they had 2, 3, or 4 individual risk factors), and the remaining two-fifths are considered to be high overall risks because they have more than 4 or 5 individual health risks.
Changes in the prevalence of individual health risks and health status can be documented using repeat HRA participation. In our experience, about half of those who complete a baseline HRA opt to complete another one a year or so later. Stronger incentive structures can drive repeat percentages up to 60% to 80%. Higher participation in baseline and repeat HRAs provides more information on the target population.
Within one year, there should be evidence that selected health risks have begun to improve. Poor nutrition and physical inactivity are examples of health risks that can demonstrate significant improvements in a relatively short time, especially with risk-specific programming and communication strategies. The ultimate goal of health management programs is to enable participants to maintain or improve their health and productivity.
Low risk individuals typically have the lowest medical and productivity-related costs. They also have the lowest probability of developing chronic diseases and enjoy the highest quality of life. Unfortunately, achieving decreases in risk over time is difficult and more people tend to increase rather than reduce health risks as they age. The programs we analyzed showed minimal long-term risk improvement. However, for many people even maintaining risk status over time can be considered positive.
Cost outcomes or return on investment?
The scientific literature indicates that changes in medical and productivity costs can follow changes in health risks: costs are observed to decrease when risks decrease and to increase when risks increase.[6,7], Maximizing cost savings thus requires risk reduction strategies (to reduce costs as risk are reduced) and risk maintenance strategies (to prevent cost increases by preventing risk increases).
Comprehensive health management programs start with the HRA and biometric screening to assess risks, then offer easy access to fitness centers or other exercise opportunities and lifestyle coaching to manage health risks. Comprehensive programs will also offer disease management programs for selected chronic conditions.
In our experience applying rigorous statistical analyses to estimate return on investment (ROI) for these programs, ROI estimates have remained remarkably consistent for well-managed programs. Such programs yielded savings of about $2 for every dollar invested. Generally, however, ROI estimates were not positive in the first program year and sometimes not even in the second year.
HRA surveys are relatively inexpensive to apply, so efforts devoted to these tend to break even more quickly. Lifestyle coaching and disease management programs require more sophisticated expertise and are more costly to apply. Among our clients, lifestyle coaching and disease management programs generally did not offer a positive ROI until at least the second or third year; others have found this as well.
A key challenge is to make programming attractive enough and to maintain motivation long enough to keep people enrolled. So far sustaining enrollment has been difficult; very few people remained in these programs for longer than a year or two.
A review of key benchmarks can be used by employers to set expectations for their health management programs. The earlier in the process this is done, the better-informed program managers will be. To ensure the most comprehensive and valid comparisons, key metrics to be tracked include participation rates, health risks, changes in health risks over time, medical and productivity-related expenditures, and return on investment.
Knowing these will make it easier to present the business case for health management to senior executives year after year. The business case can be greatly facilitated by incorporating benchmarks from leading companies or others one wishes to emulate. The search for benchmarks is a fluid process. The scientific literature is helpful in establishing benchmarks from visionary employers who have used sophisticated design, development, marketing, communication, implementation, and evaluation strategies to apply leading programs.
The information to be obtained from the published literature is not typical though, because there is little incentive to publish poor results. The ten-client benchmarks noted herein are admittedly a small slice of a big industry pie, but they are timely and probably more typical than can easily be found from other publicly available sources. To help others evaluate their wellness programs, we encourage you to make your findings public.
As employers increasingly disseminate and discuss their findings, the knowledge required to improve program outcomes will grow for everyone, and the health and productivity of the U.S. workforce will improve along the way.
About The Authors
Shirley Musich, PhD
Dr. Musich is a Senior Researcher in the Health Care Innovation and Information Group at Ingenix, where she is responsible for providing decision support to lead employers and other Ingenix clients through health evaluation, strategy design, intervention, measurement and evaluation processes. Dr. Musich has published numerous research articles in her research area of interest which focuses on the associations between participation in health promotion, risk reduction, and disease management programs, and their effects on health status (health risks) and medical and productivity cost outcome measures. Dr. Musich received her Ph.D. in Kinesiology from the University of Michigan in 1998.
Ron Ozminkowski, PhD
Dr. Ozminkowski is Vice President, Research and Development, in the Health Care Innovation and Information Group at Ingenix, and he also serves as Vice President, Research and Policy for UnitedHealth Group Alliances. Dr. Ozminkowski has been conducting health services research and evaluation projects since 1983. He is internationally recognized as an expert in the evaluation of corporate wellness and disease management programs, and has published widely on these and related issues. Dr. Ozminkowski received his Ph.D. in 1989 from the University of Michigan, School of Public Health, with an emphasis in health economics.
Frank Bottone, Jr., PhD
Dr. Bottone is a Senior Publications Medical Writer at Ingenix. Dr. Bottone received his Ph.D. degree from the interdepartmental program in Nutrition, at North Carolina State University in Raleigh, NC, where his research focused on the chemo-preventive effects of dietary and other compounds. In addition to publishing numerous scientific articles in peer-reviewed journals, Dr. Bottone, Jr. is the author of two books and over a dozen magazine and newspaper articles.
 Linnan L, Bowling M, Childress J, et al. Results of the 2004 National Worksite Health Promotion Survey. American Journal of Public Health. 2008;98(8):1503-1509.
 Chapman L. Meta-evaluation of worksite health promotion economic return studies: 2005 update. American Journal of Health Promotion. 2005;19(6):1-11.
 Goetzel RZ, Ozminkowski RJ, Villagra V, Duffy J. Return on investment in disease management: A review. Health Care Financing Review. 2005;26(4):1-19.
 Baicker K, Cutler D, Song Z. Workplace wellness programs can generate savings. Health Affairs. 2010;29(2):1-8.
 Musich S, McDonald TM, Hirschland D, Edington DW. Excess healthcare costs associated with excess health risks in diseased and non-diseased health risk appraisal participants. Disease Management and Health Outcomes. 2002;10:251-258.
 Edington DW. Emerging research: a view from one research center. American Journal of Health Promotion. 2001;15:341-349.
 Musich SA, Adams L, Edington DW. Effectiveness of health promotion programs in moderating medical costs in the USA. Health Promotion International. 2000;15:5-15.