Background/Purpose: Poor adherence with statin therapy is associated with increased mortality in the general population, but no corresponding data are available among patients with rheumatoid arthritis (RA). Since cardiovascular diseases (CVD) are the primary cause of excess mortality in RA, RA-specific data are highly relevant to RA clinical care. We evaluated the impact of statin adherence on risk of mortality in individuals with RA.

Methods: We conducted a cohort study of all incident statin users in a population-based cohort of RA identified from physician billing data, followed from May 1996 to March 2010 using administrative health data. Each individual’s observational time was divided into 90-day periods from the date of the first statin prescription until date of death or end of follow-up. For each period, we calculated the proportion days covered (PDC) of statin use and created three adherence categories: high adherence (PDC ≥ 0.80), poor adherence (0 < PDC < 0.80), and non-use (PDC = 0). Outcomes evaluated include deaths due to all causes and due to any CVD. We used Cox's proportional hazards analyses, modeling risk of death, with statin adherence as a time-dependent variable representing adherence in the current 90 day interval. We considered the following covariates: age, gender, comorbidities (e.g. acute myocardial infarction, cerebrovascular accidents, infections), use of diabetes, hypertension, and congestive heart failure medications, and use of medications known to influence cardiac risk (e.g. hormone replacement therapy), measured at baseline. Markers of RA severity and RA medication use that could influence mortality risk (DMARDs, glucocorticosteroids, methotrexate and rate of RA-related medical visits) were also included as time-dependent covariates. Covariates were included in the final model if they were significant predictors of death or if they had a confounding effect on the association between statin adherence and death. 

Results: The cohort of RA patients with incident statin use included 6,525 individuals with 36,097 person-years of follow-up (60% females; mean age 67 years). We documented 1,193 deaths overall with 386 due to CVD. Adjusted HRs for statin adherence status in the current period and mortality risk are summarized (Table).

Conclusion: These population-based data indicate that RA patients who are poorly adherent with statins have a higher risk of death from CVD and from all causes. Findings emphasize the importance of discussing adherence with statin therapy during health care professional encounters with RA patients.