Prevalence of potential drug–drug interactions in outpatients of a general hospital in China: a retrospective investigation

Background Potential drug–drug interactions are important factors resulting in adverse drug reactions or therapeutic failure. Therefore, potential drug–drug interactions need to be identified to prevent the related risk and improve drug safety. Objective This study was designed to determine the prevalence of potential drug–drug interactions and investigate the association of potential drug–drug interactions with characteristics in outpatient prescriptions. Setting A large-scale general university hospital in Jinshan District of Shanghai, China. Method The retrospective study was conducted on data obtained from prescriptions containing two or more drugs, written for outpatients older than 18 years. They were screened for potential drug–drug interactions using Lexi-Interact in UpToDate, Stockley’s Drug Interactions and Medicine Specification in the order of priority. Main outcome measure Drug–drug interactions with C, D, X risk rating and clinical parameters recorded at the prescriptions. Results 16,120 prescriptions were screened for the presence of potential drug–drug interactions and 4882 (30.29%) prescriptions containing 6667 potential drug–drug interactions were identified. Among 6667 potential drug–drug interactions, 90.81% (6054/6667), 8.49% (566/6667), 0.70% (47/6667) potential drug–drug interactions belonged to the risk category of C, D and X, respectively. Male, old age and polypharmacy increased the likelihood of potential drug–drug interactions. The most frequently prescribed drugs responsible for potential drug–drug interactions included pioglitazone, dihydrocodeine, thalidomide, sotalol, amiodarone and amlodipine. The predominant potential adverse outcome of potential drug–drug interactions was the increased central nervous system suppression function with the mechanism of reinforced pharmacological effects. Conclusion This study showed that potentially significant drug–drug interactions in outpatients were prevalent in real-world practice. Considering the risk of potential clinical consequences related to potential drug–drug interactions, it is necessary to implement the computerized surveillance and warning systems with drug–drug interactions databases as well as develop the clinical guidelines regarding the widespread potential drug–drug interactions. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (10.1007/s11096-020-01068-3) contains supplementary material, which is available to authorized users.