Page 27 - Contributed Paper Session (CPS) - Volume 3

CPS1934 Atikur R. K. et al.

Prediction of daily respiratory tract infection
episodes from prescription data
4
3
2
1
Atikur R. Khan , M. Towhidul Islam , Tabin Hasan , Saleheen Khan
1 Gulf University for Science and Technology, Kuwai
2 Independent University of Bangladesh, Dhaka
3 American International University - Bangladesh, Dhaka
4 Minnesota State University, USA

Abstract
Changing weather pattern may directly or indirectly affect the incidence and
severity of respiratory tract infections causing huge economic burden for
healthcare services. Early warning for severity and extent of this infection may
help healthcare service providers to prepare for any epidemic well before in
time. Our aim in this paper is to explore the relationship between respiratory
tract infection episodes and climatic factors and to predict the number of daily
episodes in different weather zones defined by the coverage areas of active
weather stations in Bangladesh. Prescription data collected from clinics are
integrated with climatic factors of the nearest weather stations, and this
integrated dataset is used to predict the daily respiratory tract infection
episodes in response to climatic factors. We apply panel generalized linear
models and show that the number of episodes increases in a greater extent
for increasing magnitude of rolling standard deviation of relative humidity and
rolling mean of wind speed. A 7-day ahead forecast of number of episodes
based on rolling window models of regression tree and random forest are
produced to know the severity of epidemic for healthcare planning, and a
further 1-day ahead confirmation forecast is produced to assess the necessity
of healthcare service plan adopted based on a 7-day ahead forecast. Root
mean squared forecast error computed both for 7-day ahead and 1-day ahead
forecasts both from regression tree and random forest provide qualitatively
similar results, except for three weather stations where unusually high number
of episodes are observed because of climatic extremes and high level of air
pollution.

Keywords
Panel generalized linear model; prescription data; respiratory tract infection;
regression tree; random forest

1. Introduction
Respiratory tract infections (RTI) are the most common infections
worldwide causing a considerable economic burden to healthcare services.
There are substantial evidences on the seasonal variation of respiratory

16 | I S I W S C 2 0 1 9