Modeling trends and forecasting future incidence of end stage renal disease in the U.S.

This study employs time series modeling to analyze and forecast the incidence of End Stage Renal Disease (ESRD) in the United States over the period from ۲۰۰۲ to ۲۰۲۲. Utilizing annual incidence data sourced from the reputable United States Renal Data System (USRDS), the research reveals an overall fluctuating yet upward trend in ESRD cases, peaking at ۱۳۶,۱۷۱ in ۲۰۲۱ before a slight decline in ۲۰۲۲. Through rigorous examination of the autocorrelation (ACF) and partial autocorrelation (PACF) functions, and guided by the corrected Akaike Information Criterion (AIC) as the model selection criterion, the ARIMA(۲,۰,۱) model was identified as the optimal fit. The stationarity of the data was confirmed by the Dickey-Fuller test, indicating no need for differencing. Forecasts for ۲۰۲۳ to ۲۰۲۷ suggest a gradual decline in ESRD incidence, potentially reflecting improvements in healthcare, earlier diagnosis, and demographic shifts. These findings underscore the critical importance of continuous surveillance and the implementation of targeted prevention and management strategies to sustain and enhance the encouraging trend toward reduced ESRD incidence.

Keywords:

Time series , End stage renal disease , Autoregressive integrated moving average , Autocorrelation function , Forecasting , Minitab