Abstract
Traditional methods to standardize exposures in pharmacoepidemiologic studies, like defined daily-doses, may be inadequate to capture drug class effects when there are many in-class medications, formulations, and administration routes. Antipsychotic medications are one example of a drug class with these complexities. Direct dose conversion methods are pharmacologically-based but often overlooked, potentially for lack of real-world guidance and examples of their implementation. The purpose of this article is to describe a method to implement dose conversion, using an example study that quantifies antipsychotic use among a cohort of older adults with dementia. We identified 45,442 older adults (aged ≥66 years) with dementia initiating antipsychotic therapy between January 1, 2009 and December 31, 2012 in Ontario, Canada using linked administrative healthcare databases. We developed and applied a data cleaning and dose conversion algorithm to quantify antipsychotic exposure in chlorpromazine dose equivalents at initiation, month 6, and month 12 of therapy. Results were stratified by route of administration. At initiation, 14% of patients received multiple antipsychotic prescriptions simultaneously. Patients initiating regular injectable and multiple administration routes received the highest median chlorpromazine equivalent daily-doses. Data cleaning changed 3, 16, 36, and 42% of total equivalent daily-doses in patients initiating oral, regular injectable, long-acting injectable, and multiple administration routes, respectively. Dose conversion of prescription claims data was a feasible method to quantify and present antipsychotic drug exposures. Dose conversion methods can be considered for drug effects studies of antipsychotic therapies and other medication classes with complex use.