Adherence to statins in primary prevention: yearly adherence changes and outcomes

Adherence to statins in real-world practice settings is known to be suboptimal. However, less is known about how adherence changes over time and whether changes in adherence are associated with adverse cardiovascular (CV) outcomes.

To (a) characterize yearly changes in adherence among initially adherent patients taking statins for primary prevention and (b) assess the association between changes in statin adherence with subsequent risk of CV events.

A 10% random sample of the IMS LifeLink Health Plan Claims Database covering the time period from July 1, 1997, to December 31, 2008, was used to identify a cohort of primary prevention statin users. Adherence was estimated in yearly segments beginning with the index statin prescription using proportion of days covered (PDC). PDC was categorized into 3 levels: PDC ≥ 0.80, 0.20 ≤ PDC < 0.80, PDC < 0.20. Patients were excluded if they experienced CV events or had PDC < 0.80 in their first year of statin exposure. Descriptive statistics were used to explore proportions of the cohort in each PDC category during each year. Cox-proportional hazards models were used to estimate the 5-year CV event risk associated with yearly adherence transitions.

Of the 11,126 patients beginning at the highest level of adherence (PDC ≥ 0.80) in year 1, 70% remained at this level in year 2. Of those in this level during year 2, 73% remained at this level in year 3. 828 (7.44%) experienced a CV event during their observable follow-up time. It was found that those who transitioned from the highest to the lowest level of adherence in year 2 (PDC < 0.20) experienced 2.26 greater CV event hazard (P < 0.0001). Adjusting for year 2 adherence, patients at the lowest level in year 3 experienced a 271% increase in CV hazard (P < 0.0001), as compared with the highest level of adherence.

This study found that patients' adherence levels tend to decline over time, and a transition to levels of adherence lower than a PDC of 80% was associated with increased risk of CV events. These results are useful in the context of targeting interventions that aim to improve patients' adherence.