Impact of Covid-19 pandemic on tuberculosis mortality in Brazil: a time series analysis

To evaluate changes in the trend of tuberculosis mortality in Brazil over recent decades and assess the impact of the Covid-19 pandemic on this indicator.

We analyzed the national and regional time series of tuberculosis mortality from 2000 to 2023 using joinpoint regression. To assess the pandemic's impact, we applied two interrupted time series (ITS) approaches: segmented linear regression and the AutoRegressive Integrated Moving Average with eXogenous variables (ARIMAX model). We also used Autoregressive Integrated Moving Average (ARIMA) modeling to estimate excess tuberculosis deaths linked to the pandemic and to forecast mortality trends through 2030.

An increase in tuberculosis mortality was observed starting in 2021, reaching 2.4 deaths per 100 thousand people in both 2022 and 2023 - similar to rates observed in 2011. This represents a reversal in the declining trend seen throughout the 2000s and 2010s, affecting all macroregions. The annual percentage change from 2019 to 2023 was +6.5% (95% confidence interval - 95%CI 4.42 to 9.98), contrasting with an average decline of -1.93% (95%CI -2.19 to -1.69) over the full period. Both ITS models consistently demonstrated a detrimental long-term reversal of the mortality trend after the pandemic. While a precise level change was not apparent using traditional segmented regression, the ARIMAX-based analysis successfully isolated a significant acute and lagged effect (β = +0.211; p = 0.0029). We estimated 6,540 excess tuberculosis deaths in Brazil between 2020 and 2023 (95%CI 3,950 to 9,130). The forecasting model with the pandemic effect projected higher mortality rates from 2024 to 2030, while the counterfactual scenario showed a continued decline.

The Covid-19 pandemic had a substantial negative impact on tuberculosis mortality in Brazil, representing a setback in achieving national and global elimination targets.