Trend of antidepressant sales and suicide records in Brazil: a time series analysis, 2014-2021

Antidepressants are the first-line of treatments for major depression, but adverse effects may include self-harm. We aimed to investigate the association between antidepressant sales and suicide rates in Brazil.

This time-series analysis used monthly records of antidepressants sales in Brazilian drugstores and suicide rates from 2014 to 2021. Sales were converted to defined daily dose per 1000 inhabitants per day (DID). Vector autoregressive models with one to six monthly lags were estimated. Model selection considered information criteria and the directionality of predictive relationships. Cumulative distributed-lag effects (β) and 95% confidence intervals (95% CI), impulse response, and forecast error variance decomposition were used to describe temporal dynamics.

A total of 101,346 suicides were recorded, increasing from 11,339 in 2014 to 16,248 in 2021. The incidence rose from 56.2 to 76.2 cases per million inhabitants, while antidepressant sales increased from 14.7 to 32.1 DID/month. The five-month lag model showed the best performance. Antidepressant sales were associated with suicide rates (β = 0.045; 95% CI, 0.023-0.066). A shock in antidepressant sales was associated with suicide rates in the fifth month (impulse response = 0.094; 95% CI, 0.053-0.134) and accounted for 22.0% of the variation in suicide rates by the tenth month.

The time-series design does not enable assessing causality, so it cannot be inferred that people who committed suicide had been exposed to antidepressants.

Suicide and sales of antidepressants increased in the period. Suicide rates were associated with prior antidepressant sales.