Potentially inappropriate medications with older people in intensive care and associated factors: a historic cohort study.

Background: The epidemiology of potentially inappropriate medications (PIMs) in critical care units remains limited, especially in terms of the factors associated with their use.

Objective: To estimate the incidence and factors associated with PIMs use in intensive care units.

Design And Setting: Historical cohort study was conducted in a high-complexity hospital in Brazil.

Methods: A retrospective chart review was conducted on 314 patients aged ≥ 60 years who were admitted to intensive care units (ICUs) at a high-complexity hospital in Brazil. The dates were extracted from a "Patient Safety Project" database. A Chi-square test, Student's t-test, and multivariable logistic regression analyses were performed to assess which factors were associated with PIMs. The statistical significance was set at 5%.

Results: According to Beers' criteria, 12.8% of the identified drugs were considered inappropriate for the elderly population. The incidence rate of PIMs use was 45.8%. The most frequently used PIMs were metoclopramide, insulin, antipsychotics, non-steroidal anti-inflammatory drugs, and benzodiazepines. Factors associated with PIMs use were the number of medications (odds ratio [OR] = 1.17), length of hospital stay (OR = 1.07), and excessive potential drug interactions (OR = 2.43).

Conclusions: Approximately half of the older adults in ICUs received PIM. Patients taking PIMs had a longer length of stay in the ICU, higher numbers of medications, and higher numbers of potential drug interactions. In ICUs, the use of explicit methods combined with clinical judgment can contribute to the safety and quality of medication prescriptions.