Potentially inappropriate drug prescribing and associated factors in nursing homes

Importance: Polymedication is frequent in nursing home (NH) residents. This increases the risk of potentially inappropriate drug prescribing (PIDP), which can lead to adverse drug events, such as falls and hospitalization.

Objective: To identify PIDP in NH residents and to investigate subject-related and NH structural and organizational factors associated with PIDP.

Design: Cross-sectional study.

Setting: A total of 175 NHs in Midi-Pyrénées region, South-Western France.

Participants: A total of 974 subjects randomly selected from the 6275 NH residents participating in the IQUARE study.

Exposure: Patients with PIDP.

Main outcomes and measures: PIDP was the main outcome measure. It was defined using a specific indicator, based on the Summary of Product Characteristics, on the Laroche list, and on residents' clinical data. PIDP was defined as the presence of at least 1 of the following criteria: (1) drug with an unfavorable benefit-to-risk ratio; (2) drug with questionable efficacy according to the Laroche list; (3) absolute contraindication; (4) significant drug-drug interaction. Associated factors were identified by using multivariable logistic regression models.

Results: Among the 974 residents included, 71% had PIDP. PIDP was more frequent in patients without dementia, with several comorbidities and taking multiple medications. In the multivariable analysis, age (odds ratio [OR] 1.02; 95% confidence interval [CI] 1.01-1.03) and Charlson Comorbidity Index (CCI; P = .003, CCI = 1 versus 0: OR1/0 1.22; 95% CI 0.85-1.74, CCI ≥ 2 versus 0: OR2/0 1.72; 95% CI 1.23-2.41) were associated with an increased likelihood of PIDP. By contrast, dementia was associated with a lower likelihood of PIDP (OR 0.70; 95% CI 0.53-0.94). Among NH structural and organizational characteristics, the access to psychiatric advice and/or to hospitalization in a psychiatric unit (OR 1.36; 95% CI 1.02-1.82) and the presence of a reevaluation of drug prescriptions (OR 1.45; 95% CI 1.07-1.96) were associated with an increased likelihood of PIDP.

Conclusions and relevance: Our work suggests that some NH characteristics are associated with an increased likelihood of PIDP. Gaining a better understanding of the factors influencing PIDP, especially structural and organizational NH factors, can help to determine the interventions that should be implemented.