Healthcare outcomes associated with Beers' criteria: A systematic review

OBJECTIVE: To examine healthcare outcomes associated with Beers' criteria of inappropriate medication use based on a literature review. DATA SOURCES: A search of MEDLINE, International Pharmaceutical Abstracts, and Cumulative Index to Nursing and Allied Health Literature was conducted to identify articles published from October 1991 to October 2006. The following key words were used: Beers, inappropriate, medication, drug, use, prescribing, and elderly. A manual search was also conducted using the references listed in the articles identified through the database search. STUDY SELECTION AND DATA EXTRACTION: Studies that examined the impact or outcomes of Beers' criteria of inappropriate medication use (1991, 1997, and 2003 critera) were selected. Each article was examined for study setting, data source, study sample, study design, criteria, analysis and covariates, type of healthcare outcome, and study findings. Of 235 articles retrieved, 18 presented studies that examined healthcare outcomes associated with inappropriate medication use based on Beers' criteria. Specifically, selfing-specific evidence as well as overall evidence was examined from the selected studies. The review considered evidence of association if more than 50% of the findings were statistically significant. DATA SYNTHESIS: Most of the 18 studies evaluated were retrospective cohort studies involving patents 65 years of age or older from diverse healthcare settings. In community settings, there was no evidence of association with respect to mortality and other healthcare use, and evidence regarding quality of life and costs was inconclusive. However, inappropriate medication use was associated with hospitalization measures in community elderly. In nursing homes, there was no evidence of association with mortality and the association with hospitalization measures was inconclusive. In hospitals, there was inconclusive evidence to make any generalizations. Across healthcare settings, inappropriate medication use was associated with adverse drug reactions and costs but not with other outcome measures. CONCLUSIONS: There is evidence that Beers' criteria of inappropriate medication use is associated with adverse healthcare impact in the community-dwelling elderly. With increasing use of Beers' criteria as quality-of-care measures, there is a need to strengthen the predictive validity of these criteria in all healthcare settings.