A Nudge May Be Enough to Change Healthcare Behavior

Marcia Frellick

January 17, 2018

The nation's first research unit devoted to "healthcare nudges" has developed solutions that significantly improve outcomes and lower costs. For example, the group created methods to reduce opioid pill numbers and increase use of generics, but the team is also seeing some unintended consequences.

The Penn Medicine Nudge Unit, launched in 2016, is led by Mitesh S. Patel, MD, MBA, an assistant professor of medicine and health care management at the Perelman School of Medicine and the Wharton School at the University of Pennsylvania in Philadelphia.

Dr Patel and colleagues note in an article published online today in the New England Journal of Medicine that although theirs may be the first unit of its kind in healthcare, the use of nudges to motivate behavior is common in other industries. Airlines, for example, require that buyers actively decline or accept trip insurance to purchase a ticket, and Netflix, by default, immediately starts the next episode to encourage binge watching. And after the British Government established a nudge unit in 2010 to maximize efficiency through behavorial science, such units started appearing in other governments, including the United States.

Already, the Penn nudge unit has found that healthcare improvements can be easily made. The team starts by consulting with providers about gaps that might benefit from nudges, researches the best ways to implement them, and puts systems in place to evaluate the effects, often with changes to the electronic health record (EHR) system.

More Generics Used When That Is the Default Setting

For instance, in 2016, when the Penn researchers changed the EHR default to generics instead of brand names, making generics the path of least resistance, prescribing rates for generics went from 75% to 98% "essentially overnight," Dr Patel told Medscape Medical News.

"Penn was one of the lower-performing health systems in terms of generic prescribing before that intervention, and went to the best-performing health system in the region after that," he said. "It was a very simple change, and now we have every single clinic in every department across all the sites in Pennsylvania and New Jersey rolled onto the new platform."

Changing EHR Default Setting Cut Opioid Pill Numbers

In another example, the nudge unit researchers found that a simple change in the default setting in the EHR reduced the number of opioid pills prescribed per patient in two Penn emergency departments. With the new system, the default setting is 10 pills. For a clinician to prescribe more than that, they have to opt out of the default and select another number. By changing the default setting, the median number of pills prescribed dropped by about a third, Dr Patel said.

Specifically, the proportion of 10-pill prescriptions rose from 20.6% before the default change to 43.3% after, the team reported this week in an article published online in the Journal of General Internal Medicine. Meanwhile, prescriptions with 11 to 19 tablets, which had been the most common, declined from 33.5% to 20.1%.

Dr Patel said they recently received a grant to test the opioid nudges in the Sutter Health System on the West Coast, which has more than 50 emergency departments and urgent care centers.

Unintended Consequences

Amid the successes, however, can come unintended consequences. In the opioid example, researchers found a slight decrease in the number of prescriptions for fewer than 10 pills from 20.4% to 15.4%, which indicated that the nudge to 10 pills might have superseded the inclination to prescribe even fewer pills.

Authors write in the paper, "This suggests that future efforts to set default quantities should start with the lowest baseline prescription."

Researchers found another unintended consequence with a price transparency initiative. Dr Patel said when clinicians could see the cost of high-priced tests, the numbers of those tests ordered dropped. However, the cost savings was offset by greater numbers of less expensive tests ordered when clinicians found out they were cheaper than they thought.

"We design these things to be testable and we can always compare," he told Medscape Medical News. "This is an evolving science."

Vineet Aurora, MD, an academic hospitalist and director of Graduate Medical Education Clinical Learning Environment Innovation at the University of Chicago in Illinois, used electronic nudges in her work to reduce the number of times clinicians entered patients' hospital rooms overnight to take labs and administer medication so there were fewer interruptions to patients' sleep.

As reported by Medscape Medical News, Dr Aurora and colleagues found that the default request for vitals had been set to four times a day, and because the options on number of visits was not visible on the screen, physicians could not make a different choice.

The team also found that changing the default setting of heparin administration to every 12 hours from every 8 hours meant more clinicians chose the longer time, and there were fewer interruptions of sleep to treat patients for the prevention of venous thromboembolism.

Her results showed nighttime disruptions for vitals were more than three times less likely occur after the interventions than before (odds ratio, 3.35; P < .05), and disruptions for medication administration were more than four times less likely to occur (odds ratio, 4.08; P < .05).

She told Medscape Medical News that one key to making the changes is to examine all the ways a change can affect workflow. In her study, for instance, changing the times labs were taken would have meant a change in work hours and lifestyle for a team of phlebotomists, so they had to rethink staffing.

"If your nudge is creating work or changing workflow, you've got to think whose work has been changed," she said.

Upfront buy-in is also essential, so that clinicians see the changes as the means to improve outcomes and not as manipulation, she said.

"They need to see that the nudge is helping them implement something that they want to do already," she said.

Dr Patel said in one experiment at Penn, they tried to get primary care physicians to prescribe more statins, "but we didn't engage clinicians upfront in the design, and it didn't fit in their workflow as well as it could have."

As to whether groups resist the entire idea of nudges to change their behavior, Dr Patel says to front-line clinicians, "You're already being nudged," by whatever the current default is, and it is often not the direction the clinician would choose. He added that many of the ideas for change come from the clinicians themselves.

Why Change Is Slow

So why aren't all systems changing the EHR defaults if the improvements described in papers such as these are easy?

Dr Patel said in the last decade the focus has been primarily on just getting practices to use EHRs and meet regulatory deadlines.

"Ten years ago, only 1 in 10 doctors were using the electronic health record," he notes. "Healthcare often lags behind in [information technology] adoption in general."

Nudges represent the next level of optimization, he says.

Units Can Be Replicated Elsewhere

Dr Patel said nudge units can and should be replicated elsewhere, and the extra staff and time needed to run such a unit are relatively small compared with the larger investment on the part of the health system to say it is a priority.

There are thousands of settings in the EHR that could improve healthcare with a systematic approach, he notes.

"Having leadership on board invested in the idea goes a long way in terms of being able to make these interventions and testable experiments happen," he said.

Dr Aurora says the idea of a devoted nudge unit marries two key elements of advancing healthcare: "They are aligning the research expertise of those in behavioral economics with the clinicians who know what the problem is, and to me, that's the holy grail of healthcare delivery sciences."

Dr Patel reports personal and other fees from Catalyst Health, Healthmine Service, and Life.io outside the submitted work. Coauthors report personal and other fees from VAL Health and CVS Health and grants from Vitality Institute, Oscar Health Insurance, Humana, CVS, and Hawaii Medical Services Association outside the submitted work. In the Journal of General Internal Medicine study, lead author M. Kit Delgado, MD, an assistant professor of Emergency Medicine and Epidemiology at the Perelman School of Medicine at the University of Pennsylvania, reports receiving an honorarium for participating in an Expert Roundtable on Opioid Prescribing convened by United Health Group. Dr Aurora has disclosed no relevant financial relationships.

N Engl J Med. 2018;378:214-216.

J Gen Intern Med. Published online January 16, 2018. Article

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.