Association between polypharmacy and the persistence of delirium: a retrospective cohort study

Abstract

Abstract Background Although the association between polypharmacy and the occurrence of delirium has been well studied, the influence of polypharmacy on the persistence of delirium remains unclear. We aimed to explore the effect of polypharmacy on the persistence of delirium. Methods This retrospective cohort study was conducted at a tertiary hospital. The medical records of patients diagnosed with delirium who were referred to the Department of Psychosomatic Medicine were reviewed. Presentation with delirium on day 3 was set as the outcome in this study. We counted the number of drugs prescribed on the date of referral, excluding general infusion fluids, nutritional or electrolytic products, and psychotropics. To define polypharmacy, we developed a classification and regression tree (CART) model and drew a receiver operating characteristic (ROC) curve. The odds ratio (OR) of polypharmacy for the persistence of delirium on day 3 was calculated using a logistic regression model with the propensity score as a covariate. Results We reviewed the data of 113 patients. The CART model and ROC curve indicated an optimal polypharmacy cutoff of six drugs. Polypharmacy was significantly associated with the persistence of delirium both before [OR, 3.02; 95% confidence interval (CI), 1.39–6.81; P = 0.0062] and after (OR, 3.19; 95% CI, 1.32–8.03; P = 0.011) propensity score adjustment. Conclusion We discovered an association between polypharmacy and worsening courses of delirium and hypothesize that polypharmacy might be a prognostic factor for delirium.

Keywords

• Classification and regression tree
• Delirium
• Polypharmacy
• Propensity score
• Receiver operating characteristic curve