Determinants of unit cost for hypertensive heart disease inpatients: the case of hospitals in Ilala district.

Abstract

There are limited studies on modeling cost estimates for specific disease conditions. The aim of this study was to identify determinants of unit cost and to identify a costing model of the unit cost for hypertensive heart disease inpatients for hospitals in Ilala district. The study involved 400 hypertensive heart disease patients admitted in 5 hospitals in the year 2017. Data were analysed by using R (R-3.5.1) software and Microsoft Excel. In the analysis, the study employed multiple linear regression analysis, Principal Components Analysis and Analysis of Variance. The findings showed that the key determinants of unit cost for hypertensive heart disease inpatients were surgical services fee, admission charges, consultation fee, medical procedure fee, hospital type, diagnostic examination fee, and age of the inpatients. Other key determinants were length of hospital stay and cost of prescription drugs. Three multiple linear regression models were compared to get the best model in predicting unit cost for hypertensive heart disease inpatients: first one was the regression model fitted by using original variables (with log transformed cost variables), second one was regression model fitted by using four Principal Components and the third one was regression model fitted by using five Principal Components as predictor variables. Performance of the models was compared using Akaike Information and Bayesian Information Criteria. The study concluded that, nine variables were key determinants of unit cost and multiple linear regression model fitted by using original variables (with log transformed cost variables) was the best model in predicting unit cost for hypertensive heart disease inpatients. The study recommended that before identifying appropriate costing model in healthcare, expertise opinions on modeling and clinical aspects should be sought at the design stage.