Take Your Pills: The Benefits of Adherence

Persistent use of maintenance medications for individuals with chronic health conditions has been shown to contribute to good health. Individuals suffering from cholesterol imbalance are at risk of cardiovascular disease that can manifest in heart attacks and strokes. Heart attacks and strokes are costly to treat: in 2010, cardiovascular disease alone accounted for approximately one fourth of the total cost of inpatient hospital care in the United States. Statin medication maintains cholesterol balance but must be taken daily and for the most part indefinitely. By 2014, more than 28 million Americans suffered from cholesterol imbalance and regulated their cholesterol through the use of statins. Despite clear treatment guidelines, poor compliance is prevalent: an average individual in my sample only purchased 84 percent of recommended dosage annually.

In this paper, I estimate the effects of adherence to statin medication on medical costs using MarketScan claims data on 400,062 commercially insured Americans suffering from cholesterol imbalance from 2007 to 2012. Because statin adherence can be correlated with other unobserved behavioral factors that also cause variation in medical costs, I employ an instrumental variable approach to control for endogeneity of statin adherence. Given that undesirable side effects lead individuals to be less likely to consume statins, I instrument for adherence using variation in statin side effect prevalence within demographically similar enrollees.

Results show that a 30-day supply of statins induces a 1 percent decrease in both total and out-of-pocket medical costs in the following year. Statin adherence is also associated with an 11 percent reduction in heart attack incidence and a 3 percent reduction in stroke, although the effect on stroke incidence is not statistically significant. I also find that statin out-of-pocket prices are negatively correlated with adherence, suggesting that the consumer’s price is an important determinant of adherence. Finally, by calibrating the model using price elasticity estimates in the literature, I show that insurers can profit by reducing out-of-pocket costs of statin medication, inducing better adherence and thereby reducing future medical costs even after accounting for the increased expenditure on statins.