Objective: To compare the performance of the 2003 and 2012 Beers criteria (BC) to predict negative clinical outcomes associated with potentially inappropriate medications in hospitalized older adults.
Design: Retrospective cohort study.
Setting: Acute Care of Elders (ACE) unit in a community-based teaching hospital.
Participants: All patients admitted to an ACE unit who were older than 65 years of age and prescribed at least one medication upon hospital admission.
Main outcome measure(s): The primary outcome was hospital length of stay (LOS). Secondary outcomes included likelihood of experiencing adverse drug events (ADEs) and in-hospital mortality.
Results: A total of 340 patients were included in this study. Inpatients prescribed a BC drug at any time had a longer hospital LOS than those not prescribed a BC drug (2003 BC: adjusted geometric mean, 5.93 vs. 5.50 days, P = 0.003; 2012 BC: adjusted geometric mean, 5.87 vs. 4.21 days, P < 0.001). Patients prescribed a 2003 BC drug had an increased risk of experiencing an ADE compared with those not prescribed a BC drug (odds ratio [OR] = 1.86, 95% confidence interval [CI] 1.11-3.11); however, this outcome was not statistically significant after adjusting for confounders (OR = 1.51, 95% CI 0.870-2.63). There was no statistically significant difference in ADEs when using the 2012 BC (adjusted OR = 1.27, 95% CI 0.689-2.33). There was no difference in hospital mortality regardless of the BC version used.
Conclusion: Prescription of BC drugs in an acute care setting is associated with an increased hospital LOS; however, there is no difference in the risk of ADEs or in-hospital mortality.