July 16, 2015
Tips on Evaluating a Wellness Program
by Tom Emerick
Many companies evaluate wellness programs using employee surveys, but they are notoriously unreliable. There is a better way.
This is news you can use.
If you want to evaluate the cost/benefit ratio of a wellness program, the following is a list of costs that are almost always overlooked in wellness evaluations. These are not the only things that need to be evaluated, just the ones most commonly overlooked.
When the items in the following list are fully considered, wellness evaluations can look entirely different.
1. The cost of staff hired to manage the program. A rule of thumb is to multiply their salary times two to account for FICA, benefits, office space, training, workers comp, management, etc.
2. The cost of wages for workers while attending wellness events at work. One company I looked at was spending about $175 per employee per year on this, not a trivial sum.
3. The opportunity cost of the HR staff running the program.
4. The full cost of wellness communications. Sending wellness communications to people at work has a wage cost. See #2 above.
5. The total cost to evaluate the program periodically.
6. The cost of false positives, which come from sending employees to doctors when they’re not sick. This is especially pernicious if you’re paying for wellness exams for employees. At one company, the cost of the false positives, sometimes as high as $80,000 per event, nearly cost more than the physical exams themselves. You have to examine claims data to see this.
7. If you have a fitness center, you need to take into account sports injuries for users. (Understanding this also involves access to claims data.) I’ve evaluated the impact of fitness centers for three very large companies. Taking into account sports injuries, etc., you could not make the case for an ROI for any of the three of them. In one company, we examined claims data on a) moderate or occasional fitness center users, b) people who used the fitness center regularly, and c) nonusers. Nonusers had the lowest average medical costs. Moderate users had higher medical costs than nonusers and regular users had the highest medical costs, a perfect reverse correlation.
Surveys of employees are notoriously unreliable. They measure employee opinions, at best, and opinions are not facts. As we all know, sometimes in employee surveys people will say what they think the surveyor wants to hear.
Medical claims and sick pay data are about the most meaningful ways to measure wellness outcomes. Short- and long-term disability data can be useful, too, as can life claims experience when compared with norms. If you only use employee surveys and other surrogate data, too bad.
I met an actuary who spoke at a conference on this topic and used the measurements above to evaluate wellness programs. He said he’d never seen one that had a positive ROI, except ones that used payroll deduction penalties.