Prevalence of drug-drug interactions in older people before and after hospital admission: analysis from the OPERAM trial

Zerah, Lorène [UCL] Henrard, Séverine [UCL] Wilting, Ingeborg O’Mahony, Denis Rodondi, Nicolas Dalleur, Olivia [UCL] Dalton, Kieran Knol, Wilma Haschke, Manuel Spinewine, Anne [UCL] et al.[show all]

Background: Drug-drug interactions (DDIs) are highly prevalent in older patients but little is known about prevalence of DDIs over time. Our main objective was to assess changes in the prevalence and characteristics of drug-drug interactions (DDIs) during a one-year period after hospital admission in older people, and associated risk factors. Methods: We conducted a sub-study of the European OPERAM trial (OPtimising thERapy to prevent Avoidable hospital admissions in Multimorbid older people), which assessed the efects of a structured medication review (experimental arm) compared to usual care (control arm) on reducing drug-related hospital readmissions. All OPERAM patients (≥70 years, with multimorbidity and polypharmacy, hospitalized in four centers in Bern, Brussels, Cork and Utrecht between December 2016 and October 2018, followed over 1 year) who were alive at hospital discharge and had full medication data during the index hospitalization (at baseline i.e., enrolment at admission, and at discharge) were included. DDIs were assessed using an international consensus list of potentially clinically signifcant DDIs in older people. The point-prevalence of DDIs was evaluated at baseline, discharge, and at 2, 6 and 12months after hospitalization. Logistic regression models were performed to assess independent variables associated with changes in DDIs 2months after baseline. Results: Of the 1950 patients (median age 79 years) included, 1045 (54%) had at least one potentially clinically signifcant DDI at baseline; point-prevalence rates were 58, 57, 56 and 57% at discharge, and 2, 6 and 12months, respectively. The prevalence increased signifcantly from baseline to discharge (P<.001 [signifcant only in the control group]), then remained stable over time (P for trend .31). The fve most common DDIs –all pharmacodynamic in nature– accounted for 80% of all DDIs and involved drugs that afect potassium concentrations, centrally-acting drugs and antithrombotics. At 2months, DDIs had increased in 459 (27%) patients and decreased in 331 (19%). The main factor predictive of a change in the prevalence of DDIs was hyperpolypharmacy (≥10 medications). Conclusions: DDIs were very common; their prevalence increased during hospitalization and tended to remain stable thereafter. Medication review may help control this increase and minimize the risk of adverse drug events.