Time series analysis of cholera cases in Bindura district (2014-2023)

Author

Magariro, Madeline

Title

Time series analysis of cholera cases in Bindura district (2014-2023)

Abstract

Cholera outbreaks raise a significant public health challenge in many developing countries, including Zimbabwe. Bindura District which is located in the province of Mashonaland Central in Zimbabwe have experienced severe outbreaks over the past decade. The aim of the study is to investigate the trends, seasonality of the cholera cases and cholera deaths in Bindura District from 2014 to 2023 and forecast the future cholera outbreaks. A quantitative research design was employed to analyze 40 quarterly observations of cholera cases and deaths, spanning a period of ten years. The data was obtained from the Ministry of Health and Child Care’s Bindura District office and the analysis was conducted using R Studio software. Initial analysis revealed that the data exhibited non-stationarity and autocorrelation. To address these issues and develop an effective model, auto ARIMA algorithm was employed to identify the optimal ARIMA specification, which accounted for the non-stationarity and autocorrelation in the data. The Akaike Information Criterion (AIC) was employed to select the optimal models for cholera cases and deaths, respectively. The optimal models were identified as ARIMA (2,1,0) (0,1,1) [4] for cholera cases and ARIMA (0,0,1) (0,1,0) [4] for deaths. The data was used for forecasting and the results showed that both cholera cases and deaths were expected to fluctuate until 2026. Mean squared error, mean absolute error and root mean square error metrics were utilized to analyze prediction accuracy of the model. The performance metrics were low indicating that the ARIMA models had good predictive accuracy and it demonstrates the effectiveness of ARIMA model in capturing the patterns and trends in the data. The forecasting results revealed a discordant trend between cholera cases and deaths. While cholera cases are predicted to increase, cholera deaths are not expected to follow a similar pattern, suggesting a decoupling of the cholera cases and deaths. Since the cholera cases and deaths are fluctuating, there is need to strengthen disease surveillance systems to detect cholera outbreaks early, increase public health measures during peak seasons and conduct regular public awareness campaigns to educate communities about cholera prevention, symptoms and treatment.

Date

June 2024

Publisher

BUSE

Keywords

Cholera

Time series analysis

Cholera cases

Bindura district

Supervisor

Dr T. W. Mapuwei