Early Rx Refills Predict Future Adherence Patterns


Medication dispensing within the first few months of prescribing chronic therapy can precisely predict future patient adherence.

Medication dispensing within the first few months of prescribing chronic therapy can precisely predict future patient adherence.

This finding from the CVS Health Research Institute and Brigham and Women’s Hospital could better identify patients who will struggle with long-term medication adherence and target interventions to them.

“With the increasing availability of rich patient data, we can better anticipate how the patients we manage will take their medications,” explained senior study author Niteesh Choudhry, MD, PhD, in a press release. “…By focusing on a patient’s initial, short-term medication filling behavior—are they or are they not refilling their prescription on time during the first few months of therapy—we can predict with great precision whether a patient will continue to take the medication as prescribed over the long term.”

The retrospective cohort study examined prescription drug claims from more than 77,000 statin initiators enrolled in a Medicare Part D plan from CVS Caremark, the pharmacy benefit manager (PBM) of CVS Health, over a 3-year period.

The researchers used group-based trajectory models to classify patients into 6 adherence trajectories ranging from nonadherence to near-perfect adherence.

For each group, the authors first estimated 12-month adherence trajectories based on only demographic characteristics such as age and baseline clinical characteristics such as type of statin, health services usage, and comorbidities likely to influence adherence.

Then, they estimated adherence trajectories that included baseline clinical predictors as well as statin fills observed during the first 2 to 4 months after treatment initiation.

Across the groups, the researchers found that these observed fills accurately predicted 1-year statin adherence, but baseline demographic and clinical characteristics did not.

“Physicians, pharmacy benefit managers, or other providers with timely access to patient refill data could easily implement a dynamic prediction system for adherence trajectories,” the study authors concluded. “Because both the trajectory model and the prediction model methodology are relatively simple and require little beyond pharmacy refill data, highly accurate predictions are possible for a wide spectrum of patients at providers with varying resources.”

Relatedly, CVS recently launched the Vulnerable Patient Index (VPI), which uses pharmacy claims data to identify CVS Caremark members who are most likely to generate high total health care costs resulting from poor medication adherence or unsafe use of complex treatment regimens.

“…Through better analytics, we can deliver the right intervention to the right patient at the right time,” stated study co-author and CVS Health senior vice president and chief scientific officer William Shrank, MD, MSHS.

The study titled “Predicting Adherence Trajectory Using Initial Patterns of Medication Filling” was published in The American Journal of Managed Care.

Related Videos
Rear view of Audience listening Speakers on the stage in the conference hall or seminar meeting, business and education about investment concept -  Image credit: THANANIT | stock.adobe.com
Pharmacist helping patient use glucose monitor -- Image credit: Kalyakan | stock.adobe.com
Pride flags during pride event -- Image credit: ink drop | stock.adobe.com
Female Pharmacist Holding Tablet PC - Image credit: Tyler Olson | stock.adobe.com
African American male pharmacist using digital tablet during inventory in pharmacy - Image credit: sofiko14 | stock.adobe.com
Young woman using smart phone,Social media concept. - Image credit: Urupong | stock.adobe.com
selling mental health medication to man at pharmacy | Image Credit: Syda Productions - stock.adobe.com
Medicine tablets on counting tray with counting spatula at pharmacy | Image Credit: sutlafk - stock.adobe.com
© 2024 MJH Life Sciences

All rights reserved.