An Apple a Day: Easier said than done

Written by Prof. Fernanda Márquez-Padilla

A few months ago, I pulled a muscle doing yoga and started going to physical therapy on a weekly basis soon after. I was supposed to do a 5-minute routine every day, and my discipline at doing so was mediocre at best. It wasn’t particularly hard, or painful, but still: it was so much easier to not do it.

At the same time, I was starting a research project on hypertensive patients’ behavior with respect to taking their medications as prescribed by their doctors (known in the medical literature as medication adherence), and had been reading about how people tend to be bad at doing so (with non-adherence considered “a worldwide problem of striking magnitude” by the WHO). “It doesn’t make much sense”, I remember thinking. Proper adherence to heart medication has been found to increase life expectancy, and significantly reduce the probability of negative health outcomes such as heart attacks, strokes, and other cardiovascular hospitalizations. And it’s “just” taking pills. Why don’t patients adhere? Then it hit me. I’m one of them: I’m terrible at adhering.

An important issue for health economics focuses on how to modify patients’ behavior. How can we motivate patients to engage in healthy conducts? Patient behavior has been found to be key for keeping individuals healthy. Improving patients’ medication adherence has great potential to reduce the costs of healthcare—especially for chronic patients who must often take specific medications for extended periods in order to manage their condition. However, modifying individuals’ behavior has been proven to be a challenging task, despite its positive implications for health outcomes and cost reductions.

A recent policy in Mexico undertaken by its largest public health provider, the Mexican Institute for Social Security (IMSS), created an interesting setup that unintentionally incentivized patients to improve their health behaviors—in this case, their medication adherence. The Receta Resurtible policy decreased the frequency with which hypertensive patients (i.e., high blood pressure) needed to see their physician and renew prescriptions, as long as their blood pressure remained stable and they were not late for renewing their prescriptions. In the new regime, patients could see their doctor every 90 days (as opposed to every 30). The policy’s main goal was to increase efficiency by eliminating arguably unnecessary check-ups from relatively stable chronic patients in order to free up clinic space and physicians’ time.

Waiting room at an IMSS Hospital. Source:

Now, why would this be an incentive for people to improve their health behavior? The key insight is that while consuming healthcare is a benefit for patients, it can also be time consuming and costly. Therefore, allowing chronic patients—who must be checked-up constantly—to go less often to see their doctor could actually be a type of “reward” that may be used to improve patient behavior. We may think of this as children being incentivized to study harder in order to avoid summer school.

In my research, I find that patients on the 90-day regime improved their medication taking behavior considerably. The number of days that they are out of medication between prescription fillings fell by 2.6 days in response to the policy (from a baseline of around 7.5 days). This is an improvement of 35%, comparable to the effects of other interventions for improving medication adherence, such as educational interventions or sending reminders to patients. My estimates suggest that patients improve their adherence as the total cost of getting their medication, which includes the non-monetary cost of actually renewing a prescription, falls. More interestingly, they further improve their behavior to be allowed to remain on the 90-day regime since they value its convenience. I was able to empirically test this thanks to great data from IMSS administrative records and a unique policy design.

Additionally, I find that patients’ health remained stable in spite of meeting with their physician less frequently. This point is particularly interesting for health policy, where the allocation of scarce medical resources should be done as efficiently as possible. Much debate has revolved around some prominent policies that seek to reallocate inputs for the production of health, such as reducing the frequency of certain procedures (i.e., consider the ongoing debate about the recommended frequency of mammograms) or allowing nurse practitioners to prescribe controlled medications. The value of these policies lies in the extent to which they can reduce the costs of providing healthcare, while not generating additional costs in terms of patients’ health or general wellbeing. In this sense, the Receta Resurtible policy appears to have increased efficiency by reducing how often patients should attend doctor’s appointments without harming their health.

I draw several general lessons on how to affect patients’ behavior from studying IMSS’s change in the frequency of prescription renewals. First, it is important to acknowledge that patients have a hard time adhering, and that sticking to a treatment is generally costly. Second, that in order to design the correct interventions to improve medication adherence, it is important to understand all the costs and benefits that patients face for engaging in any type of health behavior, and that these costs and benefits can be both monetary and non-monetary (such as the time and effort required to renew a prescription). Third, that incentives can come in the form of “getting out of something”—in this case, getting out of 8 check-ups per year. In a way, the policy created an additional benefit for improving medication adherence: the possibility of staying on the 90-day regime. This type of policy instrument may be useful to modify individuals’ behavior in other settings, and its design is particularly interesting as this type of incentive can be cost efficient and welfare improving: in this case, providing less healthcare is not only more efficient but it makes patients behave better as well, while keeping their health stable.

Perhaps next time I’ll be better at following my doctor’s suggested treatment!


Fernanda Márquez-Padilla holds a Ph.D. in Economics from Princeton University and is Assistant Professor at CIDE in Mexico City. Her research interests lie in the intersection of health and development economics, and is particularly interested in understanding patient behavior. She has worked as a consultant for the World Bank and RAND Corporation, worked for the Mexican Ministry of Finance, and has conducted research at Banco de México.


One thought on “An Apple a Day: Easier said than done

Leave a Reply

Your email address will not be published. Required fields are marked *