NEJM Catalyst: Knocking Down Barriers to Behavior Change

“’To help people change behavior, we can think about either pushing them harder or lowering barriers to make it easier,’ says Kevin Volpp, Patient Engagement Theme Leader for NEJM Catalyst and founding Director of Penn Medicine’s Leonard Davis Institute Center for Health Incentives and Behavioral Economics. ‘My philosophy is, we should start by lowering barriers whenever possible.’

Lowering barriers to enable behavior change is more sustainable and takes friction out of the system, and it can be done through a variety of complementary strategies. Volpp highlights seven of these:

• Changing defaults
• Simplifying complicated processes
• Reducing out-of-pocket costs
• Using technology in supportive ways
• Raising taxes on unhealthy items
• Using social support
• Making it fun

‘If we want these programs to be engaging, if we want participation rates to be high, it can’t feel onerous to participate,’ he says.

Volpp shares several real-world examples of lowering barriers. In the first, he describes the work of Penn colleague Mitesh Patel on generic medication prescribing. Generic prescribing rates at Penn Medicine were low. Patel’s team found that, inadvertently, brands had been set up as the default — so they changed that default.

‘What happened overnight was pretty dramatic,’ says Volpp. ‘Two and a half years later, these effects have been sustained; this has saved our health system about $32 million. The key was to make the path of least resistance the easiest path for providers, the one that aligned with the desired path.’ It’s not that providers were opposed to prescribing generics. They simply did not have the time to search for them on a long list of medications.

‘If we want these programs to be engaging, if we want participation rates to be high, it can’t feel onerous to participate.’

This and other choice architecture projects like it show real opportunity to impact health care delivery in a big way through small, systematic changes.

Another barrier is the health plan. These plans typically have so many incentives built in that they make it difficult for consumers to know what the cost of care will be and what the underlying incentives are. People also have trouble understanding coinsurance and deductibles; copayments are easier to understand. ‘We have a veritable alphabet soup where you can think about a guide to benefits that might be 70 or 80 pages long and all of these different incentives at play, which, of course, inherently leads to a lot of confusion,’ says Volpp.

A team led by economist and social and decision scientist George Loewenstein worked with Humana to design a new health plan called Humana Simplicity. Its design is simple in three ways:

1. Only copayments — no other cost-sharing mechanisms
2. A 2-page benefits guide
3. Everything collapsed into 1 of 7 price categories

American Journal of Health Promotion, 2016. Click To Enlarge.

Another opportunity to knock down barriers to behavior change is to lower the cost for people who participate health programs like Weight Watchers. In a landmark trial involving 23,000 employees of two large companies, researchers studied how much participation rates would change as employers lowered the cost of Weight Watchers to employees through subsidies.

‘The punchline here is the grand prize: If by the end of this 12-week contest you were at the highest levels of status, you would win the grand prize of a coffee mug.’

‘You might imagine that you get higher participation rates but people drop out at higher rates, or they lose less weight,’ says Volpp. However, providing 100% subsidy roughly doubled the rate of participation, and 80% subsidy increased it by about 60 to 70%. Most importantly, there was no trade-off between higher enrollment rates and weight loss.

Another health success is the reduction in U.S. smoking rates, where the primary driver — ‘probably the most important driver’ — has been raising taxes. Multiple studies show that a 10% increase in taxes leads to roughly a 4% reduction in smoking rates, notes Volpp. ‘These reductions are even larger among young people because they have less discretionary income, and that’s very important in terms of deterring people from starting smoking in the first place.’

In early April 2018, the Kentucky legislature passed a bill doubling the state’s tobacco tax. ‘That a state that historically was so important in terms of tobacco production chose to do this really highlights the health and economic benefits to the citizens of the state in raising tobacco taxes,’ says Volpp.

‘Another interesting and exciting opportunity comes with technology,’ says Volpp. A proliferation of new technologies makes it easier for people to monitor and engage with their health. ‘But it’s very important to recognize that although the capabilities of these technologies are considerable, many people, after given one of these devices, let it sit on a shelf. Once the novelty wears off, it’s no longer so exciting,’ Volpp reminds us. ‘We have to think about, how do we couple these technologies with social science engagement strategies that enable much higher participation rates on an ongoing basis as well improvements in outcomes?’

‘We have to think about, how do we couple these technologies with social science engagement strategies that enable much higher participation rates on an ongoing basis as well improvements in outcomes?’

One example is the first behavioral intervention study, led by Mitesh Patel, in the Framingham Heart cohort. ‘We took advantage of the multiple family members who were part of that cohort to design a social incentive system of gamification that utilized a number of behavioral economic insights,’ Volpp explains. The team provided each family with points at the beginning of the game and set easy-to-attain goals. They chose a member of each family at random every day, and if that person did not meet their step goals they would lose points — leveraging loss aversion and anticipated regret.

‘The punchline here is the grand prize: If by the end of this 12-week contest you were at the highest levels of status, you would win the grand prize of a coffee mug,’ he says. Despite the lack of monetary value, the people who met their daily step goals in the intervention group climbed to about 50% from 30% in the control group — an effect similar in magnitude to what financial incentive studies have shown.

‘These types of approaches can be used to make participation in health improvement initiative much more fun, more engaging, and thereby more successful,’ says Volpp.”

Read more at the NEJM Catalyst