From Medication Refill Histories to Drug Exposure Trajectories: A State Sequence Analysis of Methimazole Use

Le résumé fourni par la source

Aims This study proposes the use of state sequence analysis (SSA) to investigate variability in the drug exposure trajectories. To illustrate this approach, the application of SSA to methimazole use in ASLTO4, an Italian Local Health Authority, is presented as a worked example. Methods Data on the dispensing of methimazole to adults in the ASLTO4 databases from 2019 to 2020 were used to reconstruct individual medication refill histories. For each treatment episode, the average daily dose (ADD) was calculated. The ADDs were stratified on a nominal five-level scale to obtain the ‘alphabet’ required for the SSA. Cluster analysis was performed on the trajectories to recognise subgroups of users with comparable methimazole exposure. Finally, for each cluster, it was calculated the proportion of individual with levothyroxine in addition to methimazole. Results Of the 1,277 methimazole users, 822 who filled ≥2 methimazole prescriptions were included in the SSA analysis. Women made up most of users (68.4%) and the median age was 66.0 years (51.0-77.0 years). The cluster analysis of the 822 sequences led to a typology of 4 clusters with distinct methimazole exposure. In addition, the proportion of patients receiving methimazole and levothyroxine varied across clusters, and, in comparison with the cluster with the lower methimazole exposure (cluster 3), was significantly higher [OR 4.2, (95%CI 2.1-8.2), P<0.001] in the cluster 1 with the higher methimazole exposure. Conclusions State sequence analysis enables the identification of medication use patterns based on longitudinal, individual-level data relating to the daily dose.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.