Time series analysis of diabetes patients: A case study of Jigme Dorji Wangchuk National Referral Hospital in Bhutan

Abstract

© Published under licence by IOP Publishing Ltd. Diabetes has become a concern in Bhutan with its growing number of patients reported to the hospital. Increasing number of patient is likely to result in rising demand for the medical emergencies. Due to having only three referral hospitals in Bhutan, it is important to forecast the future incidences and prepare with proper resource planning. The monthly number of Diabetes patients obtained from Jigme Dorji Wangchuk National Referral Hospital (JDWNRH) is fitted by autoregressive integrated moving average (ARIMA) model. Such dataset starting from January 2006 to December 2016 is divided into two sub datasets; training dataset (January 2006 - December 2015) and validation dataset (January 2016 - December 2016). Using ARIMA, several models were evaluated based on the Bayesian Information Criterion (BIC) and Ljung-Box Q statistics. ARIMA(0, 1, 1) is the best model to describe and predict the future trends of Diabetes incidences on considering the simplest parsimonious lowest order model. Therefore, the proposed model will help to plan appropriately and allocate resources for emergencies.