3050 The prevalence and impact of medication-related harm in people living with dementia—a systematic review

Abstract Introduction People living with dementia (PLWD) take five more medications on average than those without dementia. This can increase the risk of medication-related harm, defined as any negative outcome, harm or injury caused by taking a medication. The aim of this systematic review was to identify studies that reported the prevalence of medication-related harm in PLWD and to assess its impact by evaluating various outcomes. Methods Twelve databases were searched from date of inception to April 2023. Papers published in English, reporting on the prevalence and/or adverse outcomes of medication-related harm in PLWD using any study design were eligible for inclusion. Methodological quality was assessed using the Cochrane Risk of Bias 2 tool for randomised controlled trials (RCTs) or the Risk of Bias in Non-randomised Studies of Exposures for non-randomised studies. A meta-analysis was conducted to determine combined hazard ratios (HRs) and 95% confidence intervals (CIs) on studies with similar harm-related outcomes using Review Manager software. Results Ninety-seven studies were included in the review; 93 were non-randomised studies and four were RCTs. Quality assessments found all four RCTs and the majority of non-randomised studies (n = 58) to be at a low risk of bias. Adverse health outcomes, including hospitalisations and mortality, were most frequently reported (n = 45 studies), with psychoactive medications (such as antipsychotic medications) being the most implicated class of medicines (n = 54 studies). Analysis showed that the use of antipsychotics was associated with a significantly increased mortality risk in six studies (n = 25,715 participants; HR = 1.42; 95% CI 1.10–1.84; p = 0.008). Conclusion This systematic review is the first to report the impact of medication-related harm among PLWD, with evidence to suggest that antipsychotic medication use is associated with mortality. However, the included studies had high heterogeneity, which made it difficult to draw comparisons between studies.