Association of potentially inappropriate medications with rehospitalisation and death within three months in older patients: a systematic review and meta-analysis.

Introduction: Potentially inappropriate medications (PIMs) are medications with an unfavourable risk-benefit profile for patients aged ≥ 65 years. Currently, several screening tools are available and are used in clinical practice to identify PIMs.

Aim: Our objective was to systematically synthesize the available evidence on the association between potentially inappropriate medications (PIMs), as identified by various screening tools including STOPP/START and the Beers Criteria, and the outcomes of rehospitalization and mortality within three months in older patients.

Method: Adhering to Cochrane standards, we conducted a systematic review and meta-analysis to investigate the outcomes of patients aged 65 years and older, comparing those with at least one PIM identified by any explicit screening tool to those without PIMs. Primary outcomes were readmission and/or death within three months. The databases of Embase, MEDLINE, and CENTRAL were searched for retrospective as well as prospective observational studies published between 1991 and May 17 2024. The risk of bias was assessed for all included studies.

Results: In total, 1,642 studies were identified through the systematic search. Nine observational studies were included in this review. Our analysis revealed a statistically significant association between the overall presence of PIMs and rehospitalisation; odds ratio (OR) 1.47 [95% confidence interval (CI) 1.02 to 2.13, p = 0.045]. While stratification according to STOPP/START criteria yielded a statistically significant OR of 1.84 [CI 1.08 to 3.12, p = 0.024; n = 5], no statistically significant difference was seen according to Beers list [OR 1.25, CI 0.86 to 1.81, p = 0.235; n = 5]. Furthermore, no significant association was found between PIMs and mortality in either the overall analysis or the stratification according to the Beers and STOPP/START criteria. The risk of bias in the included studies was moderate to serious, and the certainty of evidence was very low.

Conclusion: The use of PIM screening tools during medication evaluations may reduce the risk of rehospitalisation and potentially lower mortality in older patients. However, further studies are warranted to confirm the association with adverse outcomes.