Health plans have been investing in care management, wellness, and population health programs for decades. While the size and scope of health-management programs vary by company, many face the same challenge: The ability to accurately determine a return on investment (ROI).
The definition of health management is broad. It can include everything that helps a member effectively manage a chronic condition. It could incorporate case management for patients who were recently released from a hospital. Health management also might include broader programs aimed at promoting preventive health and wellbeing for a health plan’s entire membership. It could also involve addressing the “social determinants of health” – or social needs. These are factors that influence health but are traditionally outside the health care system – such as housing, food insecurity, transportation, social support system, to name a few.
I have spent the last 20 years developing and implementing standards to measure the impact health-management programs have on financials and health outcomes. Despite the intellectual horsepower the health care industry and academics have put into measuring the impact of such programs, determining an ROI often remains elusive.
An employer survey conducted by the Deloitte Center for Health Solutions found that nearly all employers that offer incentive programs agree, or somewhat agree, that wellness programs help to control medical costs. However, fewer than half of them are measuring the program’s ROI.
There are several reasons it can be difficult to measure the financial impact of such programs:
- Health care spending is highly variable: Multiple factors influence why, how much, and the type of health care people seek.
- Health costs change unpredictably: Health care utilization and price trends are ever-changing. As a result, measuring year-over-year change is typically insufficient.
- Randomized trials are not feasible: In the real world, it is difficult to identify fair and valid comparison groups for health-management programs.
Health-management programs can be costly and disjointed. A large health plan, for example, might have six departments that are each trying to motivate a group of members to take actions that will improve their health. But employees in various departments might be oblivious to what their colleagues are doing. One department, for example, might be sending a monthly letter to chronically ill members, reminding them to get key tests. At the same time, another department might be calling this same group of members to encourage them to use prescribed medicines. Imagine a member who receives three mailings and four phone calls in the same week. Then imagine that happening to 3,000 people each week.
While motivating members to stay healthy is important, it likely won’t work if everyone reaches out to the same patient. Sometimes leaders within the organization need to take a step back and look at the companywide outreach efforts. This type of uncoordinated activity can further complicate the ability to precisely measure program outcomes because it becomes impossible to determine which activities succeeded.
Methodologies might be part of the problem
People who run health-management programs want to know if the programs reduce medical costs and produce positive health outcomes. However, typical measurement approaches make it nearly impossible to pinpoint the impact. There are too many factors that can’t easily be controlled influencing the results. Furthermore, the programs alone can’t reduce medical costs of an entire population because of the other forces. As a result, findings are often fiercely debated regardless of how rigorously they are measured.
It might be time to consider design-based outcomes measurement. Using this strategy, the health plan builds a consistent and transparent platform for measuring the impact of programs based on explicitly quantified actions that individuals take. For example, a health management program that intends to improve medication adherence should measure refills. The financial impact of improving refill rates can then be projected. The health plan can then build an approach to collect information about those actions, and project their value.
The virtues of this approach can be large. First, the costs are typically much lower than more traditional measurements. Secondly, debates about methodology are largely eliminated. Further, with this type of straightforward approach, small-scale tests (e.g., testing whether text messages are more effective than letters in driving an action) can produce clear results that are connected to value, and that can drive rapid improvements. Value—and decisions about what should and shouldn’t be done—can be evaluated using a common yardstick across the full range of health-management programs.
Some health plans have invested heavily in trying to address the challenges inherent in precisely measuring program impact. Some health plan leaders have given up on the idea of measuring outcomes and are instead basing decisions on gut instinct. These organizations might recognize that outcomes measurement is important, but they can be daunted by the expense and the unknown value of the analysis. I think it is possible to make a solid assessment of a program’s value without stepping into complex methodologies that are not exact anyway.
A straightforward evaluation process could give health plan executives a clear picture of whether their health-management programs are working as intended. Let’s leave behind our valiant efforts to require precision in areas where we can’t always be precise.