Comparisons of potentially inappropriate medications and outcomes in older adults admitted to intensive care unit: A retrospective cohort study

Abstract

ObjectivesTo comparatively assess potentially inappropriate medication (PIM) use and subsequent impact on clinical outcomes among older adults admitted to the intensive care unit (ICU) by means of 3 different screening criteria for PIMs. DesignRetrospective cohort study. Setting and participantsDCH Regional Medical Center ICU. Patients 65 years of age and older admitted to the medical ICU in 2014 (n = 346). Main outcome measuresPIMs were identified with the use of the Beers criteria (2015 and 2012 versions) and the Screening Tool of Older People’s Potentially Inappropriate Prescriptions (STOPP). The proportions of PIM use at admission and discharge and proportions of in-hospital mortality and ICU and hospital readmission within 2014 among patients with PIM use were compared among the 3 criteria. Multivariable Poisson regression models assessed the associations of PIMs at admission with hospital and ICU length of stay (LOS). Statistical significance was considered to be indicated at P < 0.05. ResultsThe proportions of patients with at least 1 PIM identified through 3 different criteria (2015 Beers, 2012 Beers, and STOPP, respectively; at admission: 68.5%, 58.1%, and 44.5%; at discharge: 77.4%, 63.6%, and 42.9%) were significantly different from each other (2012/2015 Beers vs. STOPP: P < 0.01). PIM use at admission as determined by STOPP was significantly associated with longer ICU stay (relative risk [RR] 1.24, 95% CI 1.11–1.38) and hospital LOS (RR 1.24, 95% CI 1.16–1.33). However, PIMs identified through the Beers criteria (2015 and 2012 versions) were associated with shorter ICU and hospital LOS. No differences were found in proportions of in-hospital mortality and ICU and hospital readmission among patients with PIM use identified through the 3 different criteria. ConclusionAlthough the Beers criteria demonstrated the ability to identify PIMs more frequently in the ICU setting, PIM use identified by means of STOPP was associated with longer ICU and hospital LOS. Clinical interventions aiming to reduce PIMs identified by STOPP in inpatient or ICU settings may decrease patients’ inpatient or ICU LOS.