Ask About Adherence is a blog series featuring Q&A’s with experts in medication adherence. In this post, we speak with Bruce Stuart, PhD, executive director of the Peter Lamy Center for Drug Therapy and Aging and professor in pharmaceutical health services research about his recently published article “Does good medication adherence really save payers money?” Dr. Stuart set out to determine to what degree medicines reduce medical spending by comparing Medicare spending among Part D enrollees with diabetes who were adherent to their medications to those who were not. His analysis found that there is a substantial benefit even when taking the healthy adherer effect into consideration – up to $6400 a year per person.
Stay tuned for the next Q&A and be sure to share your thoughts in the comments section below. We’d love to hear from you on ways to improve medication adherence!
SAMANTHA DOUGHERTY: Why focus your research on the benefits of adherence among Medicare Part D enrollees?
BRUCE STUART, PhD: There are two reasons we choose to focus on Medicare beneficiaries. Foremost, we had a Medicare dataset (MCBS) that contained both claims data and survey-based variables that captured aspects of patient knowledge and behavior known to be correlated with good medication adherence. This permitted us to perform a direct test of the magnitude of the healthy adherer effect.
Second, while our subjects were Medicare beneficiaries, the plans providing Part D drug benefits are all commercial insurance carriers. What makes these private plans interesting from a theoretical perspective is that capitated Medicare Advantage plans (MAPDs) have an economic incentive to promote medication utilization behaviors that return savings in medical and hospital costs whereas stand-alone prescription drug plans (PDPs) do not. And indeed, we found that beneficiaries enrolled in PDPs had much higher adjusted expenditures on traditional Part A and Part B services compared to MAPDs.
DOUGHERTY: What challenges and opportunities did your study present?
STUART: The biggest challenge we faced was to construct a convincing test of the healthy adherer effect. So many factors influence patient medication taking decisions that selecting the “right” variables took considerable forethought. In the end we selected variables in five domains: socioeconomic characteristics, self-reported health, use of preventive services, diabetes knowledge and behaviors, and adherence to other chronic medications. One potential consequence of including a large set of predictive factors is that we may have inadvertently overfit the true relationship. Had our sample been larger we could have used split-half validation procedures to test for overfitting. We leave that task to future researchers.
DOUGHERTY: What is the broader significance of your research?
STUART: We believe that our two main conclusions have broad generalizability; namely, that while it is likely that naïve models purporting to demonstrate cost offsets from medication adherence may overstate the true level of savings, there are clearly circumstances where significant savings can be had. Diabetes treatment is one such instance, but there are many others. We encourage future researchers to replicate our study in different chronic disease cohorts to help build the economic case for improving medication adherence.
DOUGHERTY: Are there implications for the payer community?
STUART: Yes, most decidedly. Payers are naturally skeptical about claims that spending money on drugs can actually reduce total medical outlays. The growth in value-based insurance designs suggests that payers are beginning to soften that stance, but more work needs to be done to identify where the greatest savings can be obtained.