Universal Journal of Public Health Vol. 11(1), pp. 177 - 184
DOI: 10.13189/ujph.2023.110119
Reprint (PDF) (319Kb)

Generalized Linear Model to Estimate Length of Stay in The Hospital due to Respiratory Diseases

Siti Wafiah Hanin Mohd Zulkifli , Humaida Banu Samsudin ^*, Noriza Majid
Department of Mathematical Science, Faculty of Science and Technology, National University of Malaysia, Malaysia

ABSTRACT

Several studies reported various factors associated with length of stay in the hospital due to respiratory diseases. This study aims to select the best Generalized Linear Models (GLM) for predicting LOS and identify the effects of clinical and demographic factors on LOS. The study was carried out using data on registered admission and discharge at the government hospitals for central region states in Malaysia. A total of 526,511 cases were classified under diseases of respiratory systems coded J00 to J99 and factors such as gender, age group, ethnicity, marital status, discharge condition and diagnoses of the patients were used in this study. Two regression models were used to predict LOS: GLM Zero Truncated Poisson regression (ZTP) and GLM Zero Truncated Negative Binomial (ZTNB). The best count fit model was chosen according to Akaike’s Information Criteria (AIC) and Bayesian Information Criterion (BIC). The median length of stay in the study was four days (Interquartile Range: three to six days). According to statistical comparisons, the best model for count data is GLM Zero Truncated Negative Binomial. The outcome of the model found that there were significant changes in log LOS for each predictor except for unknown marital status and patients diagnosed with other diseases of the respiratory system that were insignificant in ZTP but significant in ZTNB. ZTNB model helps to identify factors contributing to the log length of stay so that hospitals are well equipped with the facilities and prepared for the expenditures and resources.

KEYWORDS
Generalized Linear Model, Respiratory Diseases, Length of Stay, Zero Truncated Negative Binomial

Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] Siti Wafiah Hanin Mohd Zulkifli , Humaida Banu Samsudin , Noriza Majid , "Generalized Linear Model to Estimate Length of Stay in The Hospital due to Respiratory Diseases," Universal Journal of Public Health, Vol. 11, No. 1, pp. 177 - 184, 2023. DOI: 10.13189/ujph.2023.110119.

(b). APA Format:
Siti Wafiah Hanin Mohd Zulkifli , Humaida Banu Samsudin , Noriza Majid (2023). Generalized Linear Model to Estimate Length of Stay in The Hospital due to Respiratory Diseases. Universal Journal of Public Health, 11(1), 177 - 184. DOI: 10.13189/ujph.2023.110119.