Purpose
This study aimed to assess the association between antipsychotic doses and the risk of tardive dyskinesia (TD) in clinical practice using a Japanese claims database from 2010 to 2020.
Methods
The study population included patients 15 years or older with a diagnosis record of schizophrenia, depression, or bipolar disorder who were prescribed antipsychotics. Using a case-control design, we categorized patients newly diagnosed with TD as cases, with corresponding 1:10 matching in the control group. The primary endpoint was the relative risk of TD in the >median dose and ≤median dose groups, as determined using conditional logistic regression analysis adjusted for age.
Results
The analysis population included 58,452 patients, and the median daily antipsychotic dose was 75 mg/d of chlorpromazine equivalent (CPZE). Of these, 80 were identified as TD cases, and doses >75 mg/d were associated with a significantly increased risk of TD at the last prescription and the maximum dose, respectively, before the date of the first diagnosis of TD. Post-hoc analysis further showed a significant association between doses ≥300 mg/d and the risk of TD compared to doses ≤75 mg/d and doses >75 to <300 mg/d. Comparing ≥300 mg/d versus >75 to <300 mg/d, the odd ratios at the last prescription and maximum dose before the first diagnosis of TD were 3.40 and 3.50, respectively.
Conclusions
In the Japanese medical claims database of patients receiving relatively low doses of antipsychotics, doses >75 mg/d were associated with an increased risk of TD in a dose-dependent manner.