Evidence-Based Diabetes Management

Can Financial Incentives Improve Self-Management Behaviors?

Published Online: April 11, 2013
Kim Farina, PhD
For many health conditions and chronic diseases, positive outcomes are tightly linked to patient self-management behaviors. Yet, patients struggle to successfully selfmanage their conditions.

The field of behavioral economics has grown over the years and is increasingly being applied by health policy makers and providers to improve healthrelated behaviors and decision making. Behavioral economists are partnering with healthcare professionals to study the rationale behind behavioral decision making and to use this information to develop methods for reducing unhealthy behaviors (eg, smoking or poor adherence to medication). Presenters at the “Should We Pay People to Change Their Behavior? Behavioral Economics and Diabetes” symposium of the 72nd Scientific Sessions of the American Diabetes Association discussed how behavioral economics can enhance diabetes management.

Incentivizing Through Reinforcement

Nancy Petry, PhD, professor of medicine, University of Connecticut Health Center, shared lessons learned from the field of substance abuse treatment during her presentation, “Moving From Behavioral Economic Theory to Medical Decisions.” Incentive intervention programs rely on the promise of a tangible reinforcement reward for positive behavioral change. This technique has had proven success among substance abusers and accumulating evidence supports its use to encourage healthy behaviors (eg, weight loss and adherence to medication and exercise) among other populations. The US Department of Veterans Affairs (VA) is so convinced of its efficacy for substance abuse treatment that it has called for nationwide implementation of reinforcement interventions.

The concept of behavior modification through reinforcement is nothing new, Dr Petry explained. “Reinforcers are used to change behaviors in everyday settings, for example, employee salaries, commission, and awards.” Its underlying principles are: 1) frequent monitoring of a specific target behavior, 2) providing tangible positive reinforcement each time the target behavior occurs, and 3) withholding reinforcement if the target behavior does not occur.

“In terms of diabetes prevention and treatment,” stated Dr Petry, “there are a lot of potential applications of these techniques.” Improved attendance at clinic visits or group education sessions, self-monitoring of blood glucose, and adherence to medication and lifestyle changes are just some of the potential goals that can be supported.

“Reinforcement interventions can be powerfully effective in changing behaviors. The challenge is to effectively adapt and apply these techniques to improve patient outcomes in both the short and long term, while being cognizant of costs and patient and provider burden,” said Dr Petry.

During his presentation “Behavioral Economic Interventions in Obesity, Smoking Cessation, and Diabetes,” Kevin Volpp, MD, PhD, professor of medicine and healthcare management, Perelman School of Medicine and the Wharton School, University of Pennsylvania, and staff physician, Philadelphia VA Medical Center, explained that new technology is enabling incentive programs to be executed remotely. Dr Volpp’s group is exploring the usefulness of what he calls automated hovering.

Physicians don’t know what their patients are doing outside the clinics and do not have effective tools to influence the decisions patients make. “So, in a general sense, we need to figure out a more effective way of hovering over patients in a way that is well received,” remarked Dr Volpp. This may be achieved, he explained, through the application of a variety of devices and platforms. In one program, participants are given a scale that is linked to a server. The server automatically calculates an incentive, delivers information to the participant, and initiates an electronic fund transfer to the participant when appropriate.

Dr Volpp emphasized the importance of considering decision errors during the design of any health management improvement program. Humans tend to be wired for immediate gratification and this leads to decision errors. “It is very hard to say no to something that is right in front of you and very present as opposed to some future increased risk,” Dr Volpp said.

Ignoring the elements that underlie decision errors can lead to ineffective programs that may worsen outcomes and certainly increase costs. “There are a lot of mistakes people make and they are easy to make,” cautioned Dr Volpp. “You can spend millions of dollars on a program that doesn’t work.”

Redesigning Health Insurance to Change Behavior

High copays and increased patient cost sharing decrease the use of preventative medicine and worsen socioeconomic disparities in healthcare utilization. Value-based insurance design (VBID) is being purported as an innovative approach to healthcare financing that increases utilization of highvalue healthcare services while reducing that of low-value services. Allison Rosen, MD, MPH, ScD, associate professor, department of quantitative health sciences, University of Massachusetts Medical School, discussed the merits of VBID during her presentation “Redesigning Health Insurance to Change Behavior.”

The question that always comes up, according to Dr Rosen, is whether VBID saves money; however, the more appropriate question is whether VBID improves patient outcomes (or health returns) relative to every dollar spent. For VBID to save money, the cost of increased utilization of high-value services can be subsidized by medical cost offsets, enhanced productivity, decreased disability costs, and higher cost sharing for services of lower value. Dr Rosen suggests that return on investment can be maximized by strategic targeting. For example, return on investment for measures that prevent adverse events will be greater if they are targeted to those most likely to have an event rather than the general population.

Evidence shows that as barriers to quality healthcare decrease, utilization increases. Those increases are modest, admits Dr Rosen, ranging from 1% to 7%, and their clinical significance has yet to be determined. However, she noted, “If you are getting some increases, at least you are not spending the money on other services that are getting you nothing or perhaps decreasing health.” To date, most private-sector VBID programs have focused on removing barriers to evidence-based services for chronic disease. There is a growing recognition that improving value in healthcare will require increasing barriers to low value services as well.

A Primary Care Perspective

Michael Parchman, MD, MPH, director, MacColl Center for Health Care Innovation, talked about the building blocks required for high-performing primary care during his presentation “The Behavioral Economics of Improving Diabetes Management—A Primary Care Perspective.”

Interventions being developed to improve diabetes care delivery need to be integrated with primary care, where most diabetes care takes place, explained Dr Parchman. “Too often, these programs are completely isolated from primary care or are not designed to work in the context of primary care,” he stated. Practically speaking, implementation of many of the diabetes management improvement programs being developed would require a complete rework of primary care office flow at significant timecost effort for the practice.

Funding Source: None.

Author Disclosure: The author reports receiving payment for involvement in the preparation of this manuscript with no associated conflicts of interest.

Authorship Information: Concept and design; drafting of the manuscript; and critical revision of the manuscript for important intellectual content.
Issue: March/April
More on AJMC.COM