Consequences of ignoring patient diagnoses when using the 2015 Updated Beers Criteria

Abstract

Background: Beers Criteria are one of the best known explicit criteria to identify inappropriate medication in elderly that can be used in medication review. The access to patients’ medical records may be different among healthcare professionals and settings and, subsequently, the identification of patients’ diagnoses may be compromised. Objective: To assess the consequences of ignoring patient diagnoses when applying 2015 Beers Criteria to identify potentially inappropriate medication (PIM). Setting: Three nursing homes in Central Portugal. Method: Medical records of nursing home residents over 65 years old were appraised to identify medication profile and medical conditions. 2015 Beers Criteria were used with and without considering patients’ diagnoses. To compare the number of PIM and PIM-qualifying criteria complied in these two judgements, Wilcoxon signed-rank tests were performed. Main outcome measure: Number of PIMs and number of PIMqualifying criteria. Results: A total of 185 patients with a mean age of 86.7 years (SD = 7.8) with a majority of female (70.3%) were studied. When assessing the patients with full access to the diagnoses, median number of PIMs was 4 (IQR 0–10) and number of PIM-qualifying criteria was 5 (IQR 0–15). When evaluating only patient current medication, median number of PIMs was 4 (IQR 0–10) and PIM-qualifying criteria was 4 (IQR 0–12). Statistical difference was found in the number of PIM-qualifying criteria identified (p < 0.001), but not in the number of PIMs per patient (p = 0.090). In 171 patients (92.4%) PIMs identified were identical when using or ignoring their medical diagnoses. However, in 80 patients (43.2%) the PIMqualifying criteria complied were different with and without access to patient diagnoses. Conclusion: Although restricted access to patients’ diagnoses may limit the judgement of Beers PIM-qualifying criteria, this limitation had no effect on the number of PIM identified.