Prevalence and correlates of cardiovascular medication use among nursing home residents with ischemic heart disease: results from the SHELTER study.

Objectives: Despite being the highest group of users of many medications, older individuals remain underrepresented in clinical trials. This leaves a gap in evidence to guide management of many conditions, such as ischemic heart disease (IHD), in this population. This study aimed to describe factors associated with IHD medication use among nursing home residents in 7 European countries and Israel to depict challenges facing disease management in this population.

Design: This study was a retrospective cohort analysis.

Setting And Participants: The sample included 4156 nursing home residents in the SHELTER study.

Measurement: All residents were assessed using the interRAI Long-Term Care Facility (LTCF) instrument. Use of angiotensin-converting enzyme inhibitor (ACEi) and/or angiotensin receptor blocker (ARB), beta-blocker (BB), antiaggregants (including acetylsalicylic acid [ASA]) and statins was analyzed. Based on the use of these medications, residents were classified into groups by medication use (as nonusers, 1-2 medications, or 3-4 medications). Generalized Estimation Equation modeling was used to explore predictors of medication use from items on the LTCF instrument as well as facility questionnaire.

Results: Of the 1050 residents with IHD, medication use was 77.7% overall, but only 16.9% were receiving 3 to 4 medications. Use of antiaggregants was highest at 51.7% and variations in medication use were observed by country (highest in France and lowest in Italy). Functional disability was the strongest predictor of medication use, reducing the likelihood of any or optimal management. Severe cognitive impairment also reduced the likelihood of optimal management, and comorbidity generally increased the likelihood of medication use. Polypharmacy reduced the likelihood of use of 3 to 4 medications for IHD.

Conclusion: Optimal management of IHD in nursing home residents was low and varied by country. Individual characteristics seemed to predict IHD medication use, suggesting prescribing bias and an effect of population differences from clinical trial cohorts.