Over the last decade or so, companies have been looking to “do well by doing good” with their health-care provisions: to cut the amount they spend on health insurance, and lose to absenteeism, by encouraging employees to take better care of themselves. A recent study by Rand found that in 2012, about half of employers with at least 50 employees -- and more than 90 percent of those with more than 50,000 -- had one of these programs in place.

They range from discounted gym memberships and wellness workshops to ultimatums; the Cleveland Clinic monitors its employees' blood levels for various conditions, and it won’t have sugared soda on its campuses. Moreover, the legendary institution refuses to hire smokers, and it forced those already working for them to quit -- smoking, or the job. There are also programs that focus on the behavior of people with expensive chronic conditions such as diabetes or asthma. Usually these involve a nurse regularly reaching out to these patients, cajoling them into taking their medication and monitoring their symptoms, and possibly catching potential problems early.

Such programs should reduce a company’s health-care spending. But until now, we haven’t had data showing whether they actually do.

Rand researchers have now released an article in Health Affairs that takes an in-depth look at PepsiCo Inc.’s Healthy Living program, which features both kinds of intervention: one aimed at encouraging all its employees to live healthier, and one that focuses on people who are managing chronic conditions. Unfortunately, it’s not a randomized controlled trial. But nonetheless, it does provide some information about what works, and what may not, in wellness interventions.

Credit: Health Affairs, "Managing Manifest Diseases, But Not Health Risks, Saved PepsiCo Money Over Seven Years"
Credit: Health Affairs, "Managing Manifest Diseases, But Not Health Risks, Saved PepsiCo Money Over Seven Years"

All the usual caveats apply: We don’t know that the people who participated in Pepsi’s wellness program aren’t somehow systematically different (like, more likely to test their blood sugar and take their blood pressure pills) than people who didn’t. You don’t want to run out and spend billions on disease management on the basis of these results.

What you might want to do is reassess your spending on general “wellness” -- at least, if what you’re trying to do is save money on health-care costs. These results make a certain amount of sense, after all. For people who don’t already have diabetes or hypertension or high cholesterol, exercise and eating better may provide benefits … but those benefits are likely to be decades away. At which point the employee may be working for someone else.

That doesn’t mean you should get rid of the gym discounts and the smoking-cessation help. Quitting smoking is good for people. So is exercising. Exhorting people to be healthier may not have all that much effect, but making it marginally easier to do so may really help people who are looking to make a change. Companies provide all sorts of benefits to their employees, and there’s no reason that this shouldn’t be one.

But understand that this benefit you’re providing your employee is for their benefit, like life insurance. If these results are correct, it’s probably not going to make the shareholders better off.