Time Series Forecasting and Projection of Diabetes Prevalence in India from 2023 to 2035: A Cross-sectional Study

Abstract

Introduction: Diabetes is a significant public health problem in India, with an estimated 74 million Indian adults suffering from the condition as of 2021, and the prevalence is continuing to grow. Aim: To investigate the prevalence and forecasted trends of Diabetes Mellitus (DM) in India through time series analysis and forecasting models. Materials and Methods: A cross-sectional study was conducted in the Department of Statistics at Manonmaniam Sundaranar University, Tirunelveli, Tamil Nadu, India, from January 2024 to March 2024. The prevalence of diabetes in India was forecasted from 2023 to 2035 using existing diabetes prevalence data from 2009 to 2021, with the help of Gretl software and the Autoregressive Intregated Moving Average (ARIMA) model. Results: Authors observed a consistent upward trajectory in DM prevalence, with rates steadily increasing from 7.1% in 2009 to 9.6% in 2021. Time series analysis reveals non stationarity in the data, necessitating the use of ARIMA models for forecasting. Among the models considered, ARIMA (2,1,2) emerges as the best fit, demonstrating strong explanatory power with an R-squared value of 0.80. Forecasting projections indicate a continued rise in DM prevalence, with rates projected to increase from 10.35% in 2023 to 13.46% by 2035, which translates to an increase from 97.5 million in 2023 to 139 million in 2035. Conclusion: These findings underscore the urgent need for comprehensive public health interventions to address the escalating burden of diabetes in India, emphasising the importance of prevention, early detection, and effective management strategies.

Keywords

• diabetes mellitus
• epidemiology
• population forecast
• prevention and control