Suicide Trend Analysis in Brazil: The Effect of the COVID-19 Pandemic
Abstract
Suicide is one of the leading causes of death worldwide. Linked to the mental health conditions of a population, stressful global events can affect trends. This study analyzed suicide trends in Brazil between 2003 and 2022, focusing on the COVID-19 pandemic period. We used data from the Mortality Information System (SIM) of DATASUS (218,707 suicide occurrences). Suicide mortality rates were predicted for the pandemic years and beyond (2020-2022) using time series models (Prophet, SARIMA, and LSTM) trained with data up to 2019. Here, we compare the actual results with those forecasted to evaluate a possible impact of the pandemic. Our main findings show that the suicide rate in Brazil nearly doubled over two decades (from 4.45 to 8.11 per 100,000 inhabitants), with a pronounced increase during the COVID-19 pandemic. The time series models tested here performed well prior to 2020 but underestimated suicide peaks during the pandemic, underestimating the impact of COVID-19 on forecasting complex phenomena such as suicide. Our analysis highlighted the importance of public policies focused on mental health and reinforces the need for preventive actions, especially in the context of crises such as COVID-19.
References
Araghinejad, S. (2014). Time Series Modeling. Springer.
Bosi, M. L. M. and Alves, E. D. (2023). Distanciamento social em contextos urbanos na pandemia de covid-19: desafios para o campo da saúde mental. Physis: Revista de Saúde Coletiva, 33:e33007.
Box, G. E., Jenkins, G. M., Reinsel, G. C., and Ljung, G. M. (2015). Time series analysis: forecasting and control. John Wiley & Sons.
Braun, B. F., Anjos, G. O., Fonseca, T. M. A., Trevisan, E. R., and de Souza Castro, S. (2023). Perfil epidemiológico dos casos de tentativa de suicídio: revisão integrativa. SMAD, Rev. Eletrônica Saúde Mental Álcool e Drogas (em Português), 19(1):112–122.
Caterini, A. L., Chang, D. E., Caterini, A. L., and Chang, D. E. (2018). Recurrent neural networks. Deep neural networks in a mathematical framework, pages 59–79.
de Lyra Reis, J. G. and Teixeira, K. H. (2023). Os determinantes socioeconômicos do suicídio nos municípios do brasil. In Anais do Encontro Nacional da Associação Brasileira de Estudos Regionais e Urbanos (ENABER).
de Vigilância em Saude, S. (2024). Boletim epidemiológico, volume 55, nº 04. Acessado em setembro de 2024.
Dubey, A. K., Kumar, A., García-Díaz, V., Sharma, A. K., and Kanhaiya, K. (2021). Study and analysis of sarima and lstm in forecasting time series data. Sustainable Energy Technologies and Assessments, 47:101474.
Feng, T., Zheng, Z., Xu, J., Liu, M., Li, M., Jia, H., and Yu, X. (2022). The comparative analysis of sarima, facebook prophet, and lstm for road traffic injury prediction in northeast china. Frontiers in public health, 10:946563.
Hochreiter, S. (1997). Long short-term memory. Neural Computation MIT-Press.
Jelodar, H., Wang, Y., Orji, R., and Huang, S. (2020). Deep sentiment classification and topic discovery on novel coronavirus or covid-19 online discussions: Nlp using lstm recurrent neural network approach. IEEE Journal of Biomedical and Health Informatics, 24(10):2733–2742.
Kaczmarczyk, P. (2017). Microeconometric analysis of telecommunication services market with the use of sarima models. Dynamic Econometric Models, 17:41–57.
Landi, F., Baraldi, L., Cornia, M., and Cucchiara, R. (2021). Working memory connections for lstm. Neural Networks, 144:334–341.
Lomnicki, Z. and Zaremba, S. (1957). On the estimation of autocorrelation in time series. The Annals of Mathematical Statistics, 28(1):140–158.
Lopes, L. O. R., Jesus, R. S. M., Souza, R. S. B., and Teodoro, M. L. M. (2023). Fatores de risco e associados ao comportamento suicida no brasil: uma revisão sistemática. Tempus Psicológico, 6(2).
Mohammadi Farsani, R. and Pazouki, E. (2020). A transformer self-attention model for time series forecasting. Journal of Electrical and Computer Engineering Innovations (JECEI), 9(1):1–10.
Nurmi, J. and Lohan, E. S. (2021). Systematic review on machine-learning algorithms used in wearable-based ehealth data analysis. IEEE Access, 9:112221–112235.
Rafferty, G. (2021). Forecasting Time Series Data with Facebook Prophet: Build, improve, and optimize time series forecasting models using the advanced forecasting tool. Packt Publishing Ltd.
Ryu, S., Nam, H. J., Jhon, M., Lee, J.-Y., Kim, J.-M., and Kim, S.-W. (2022). Trends in suicide deaths before and after the covid-19 outbreak in korea. PLoS One, 17(9):e0273637.
Silva Junior, A. P. d., Silva Júnior, F. J. G. d., Sales, J. C., Monteiro, C. F. d. S., Miranda, P. I. G., et al. (2023). Estratégias para prevenção e posvenção do suicídio em tempos de pandemia de covid-19. Interface-Comunicação, Saúde, Educação, 27:e230181.
Sirisha, U. M., Belavagi, M. C., and Attigeri, G. (2022). Profit prediction using arima, sarima and lstm models in time series forecasting: A comparison. IEEE Access, 10:124715–124727.
Sivaramakrishnan, S., Fernandez, T. F., Babukarthik, R., and Premalatha, S. (2022). Forecasting time series data using arima and facebook prophet models. In Big data management in Sensing, pages 47–59. River Publishers.
Soares, F. C., Stahnke, D. N., and Levandowski, M. L. (2023). Tendência de suicídio no brasil de 2011 a 2020: foco especial na pandemia de covid-19. Revista Panamericana de Salud Pública, 46:e212.
Tanaka, T. and Okamoto, S. (2021). Increase in suicide following an initial decline during the covid-19 pandemic in japan. Nature human behaviour, 5(2):229–238.
Taylor, S. J. and Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1):37–45.
Walczak, S. (2019). Artificial neural networks. In Advanced methodologies and technologies in artificial intelligence, computer simulation, and human-computer interaction, pages 40–53. IGI global.
Yu, Y., Si, X., Hu, C., and Zhang, J. (2019). A review of recurrent neural networks: Lstm cells and network architectures. Neural computation, 31(7):1235–1270.
Zhang, M. (2018). Time series: Autoregressive models ar, ma, arma, arima. University of Pittsburgh.
Bosi, M. L. M. and Alves, E. D. (2023). Distanciamento social em contextos urbanos na pandemia de covid-19: desafios para o campo da saúde mental. Physis: Revista de Saúde Coletiva, 33:e33007.
Box, G. E., Jenkins, G. M., Reinsel, G. C., and Ljung, G. M. (2015). Time series analysis: forecasting and control. John Wiley & Sons.
Braun, B. F., Anjos, G. O., Fonseca, T. M. A., Trevisan, E. R., and de Souza Castro, S. (2023). Perfil epidemiológico dos casos de tentativa de suicídio: revisão integrativa. SMAD, Rev. Eletrônica Saúde Mental Álcool e Drogas (em Português), 19(1):112–122.
Caterini, A. L., Chang, D. E., Caterini, A. L., and Chang, D. E. (2018). Recurrent neural networks. Deep neural networks in a mathematical framework, pages 59–79.
de Lyra Reis, J. G. and Teixeira, K. H. (2023). Os determinantes socioeconômicos do suicídio nos municípios do brasil. In Anais do Encontro Nacional da Associação Brasileira de Estudos Regionais e Urbanos (ENABER).
de Vigilância em Saude, S. (2024). Boletim epidemiológico, volume 55, nº 04. Acessado em setembro de 2024.
Dubey, A. K., Kumar, A., García-Díaz, V., Sharma, A. K., and Kanhaiya, K. (2021). Study and analysis of sarima and lstm in forecasting time series data. Sustainable Energy Technologies and Assessments, 47:101474.
Feng, T., Zheng, Z., Xu, J., Liu, M., Li, M., Jia, H., and Yu, X. (2022). The comparative analysis of sarima, facebook prophet, and lstm for road traffic injury prediction in northeast china. Frontiers in public health, 10:946563.
Hochreiter, S. (1997). Long short-term memory. Neural Computation MIT-Press.
Jelodar, H., Wang, Y., Orji, R., and Huang, S. (2020). Deep sentiment classification and topic discovery on novel coronavirus or covid-19 online discussions: Nlp using lstm recurrent neural network approach. IEEE Journal of Biomedical and Health Informatics, 24(10):2733–2742.
Kaczmarczyk, P. (2017). Microeconometric analysis of telecommunication services market with the use of sarima models. Dynamic Econometric Models, 17:41–57.
Landi, F., Baraldi, L., Cornia, M., and Cucchiara, R. (2021). Working memory connections for lstm. Neural Networks, 144:334–341.
Lomnicki, Z. and Zaremba, S. (1957). On the estimation of autocorrelation in time series. The Annals of Mathematical Statistics, 28(1):140–158.
Lopes, L. O. R., Jesus, R. S. M., Souza, R. S. B., and Teodoro, M. L. M. (2023). Fatores de risco e associados ao comportamento suicida no brasil: uma revisão sistemática. Tempus Psicológico, 6(2).
Mohammadi Farsani, R. and Pazouki, E. (2020). A transformer self-attention model for time series forecasting. Journal of Electrical and Computer Engineering Innovations (JECEI), 9(1):1–10.
Nurmi, J. and Lohan, E. S. (2021). Systematic review on machine-learning algorithms used in wearable-based ehealth data analysis. IEEE Access, 9:112221–112235.
Rafferty, G. (2021). Forecasting Time Series Data with Facebook Prophet: Build, improve, and optimize time series forecasting models using the advanced forecasting tool. Packt Publishing Ltd.
Ryu, S., Nam, H. J., Jhon, M., Lee, J.-Y., Kim, J.-M., and Kim, S.-W. (2022). Trends in suicide deaths before and after the covid-19 outbreak in korea. PLoS One, 17(9):e0273637.
Silva Junior, A. P. d., Silva Júnior, F. J. G. d., Sales, J. C., Monteiro, C. F. d. S., Miranda, P. I. G., et al. (2023). Estratégias para prevenção e posvenção do suicídio em tempos de pandemia de covid-19. Interface-Comunicação, Saúde, Educação, 27:e230181.
Sirisha, U. M., Belavagi, M. C., and Attigeri, G. (2022). Profit prediction using arima, sarima and lstm models in time series forecasting: A comparison. IEEE Access, 10:124715–124727.
Sivaramakrishnan, S., Fernandez, T. F., Babukarthik, R., and Premalatha, S. (2022). Forecasting time series data using arima and facebook prophet models. In Big data management in Sensing, pages 47–59. River Publishers.
Soares, F. C., Stahnke, D. N., and Levandowski, M. L. (2023). Tendência de suicídio no brasil de 2011 a 2020: foco especial na pandemia de covid-19. Revista Panamericana de Salud Pública, 46:e212.
Tanaka, T. and Okamoto, S. (2021). Increase in suicide following an initial decline during the covid-19 pandemic in japan. Nature human behaviour, 5(2):229–238.
Taylor, S. J. and Letham, B. (2018). Forecasting at scale. The American Statistician, 72(1):37–45.
Walczak, S. (2019). Artificial neural networks. In Advanced methodologies and technologies in artificial intelligence, computer simulation, and human-computer interaction, pages 40–53. IGI global.
Yu, Y., Si, X., Hu, C., and Zhang, J. (2019). A review of recurrent neural networks: Lstm cells and network architectures. Neural computation, 31(7):1235–1270.
Zhang, M. (2018). Time series: Autoregressive models ar, ma, arma, arima. University of Pittsburgh.
Published
2025-06-09
How to Cite
PRACHEDES, Luciana; BERNARDINO, Heder; MARTINS, Leonardo; SOUZA, Felipe; WENECHA, Ryanne; DIMENSTEIN, Magda.
Suicide Trend Analysis in Brazil: The Effect of the COVID-19 Pandemic. In: BRAZILIAN SYMPOSIUM ON COMPUTING APPLIED TO HEALTH (SBCAS), 25. , 2025, Porto Alegre/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 569-580.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2025.7580.
