Predictors of Hospital Length of Stay among Patients with Dementia: A Comparison of Ordinary Least Squares and Robust Regression
Article Sidebar
Main Article Content
Abstract
BACKGROUND: Dementia is a major public health concern in aging societies. Patients with dementia frequently experience prolonged and highly variable hospital stays, placing a significant burden on healthcare systems. Conventional statistical approaches, such as Ordinary Least Squares (OLS) regression, may yield biased estimates when applied to clinical data with outliers or heteroscedasticity. Robust Regression offers a more appropriate alternative; however, its comparative performance in psychiatric hospital settings in Thailand remains underexplored.
OBJECTIVES: This study aimed to investigate the factors associated with hospital length of stay (LOS) among patients with dementia and to compare the performance of Ordinary Least Squares (OLS) regression and Robust Regression in a psychiatric hospital setting in Thailand.
METHODS: A retrospective analytical study was conducted using the medical records of 120 patients with dementia admitted to Suansaranrom Hospital between June and October 2025. The variables examined included age, Charlson Comorbidity Index (CCI), Clinical Dementia Rating (CDR), Behavioral and Psychological Symptoms of Dementia (BPSD), and length of stay (LOS).
RESULTS: The mean LOS was 29.3±15.2 days, with 8 outlier cases (6.7%) having a LOS longer than 60 days. OLS regression did not reveal any statistically significant overall relationships among the variables. In contrast, Robust Regression identified age, CCI, and CDR as significant predictors of LOS, with CDR exhibiting the strongest influence. BPSD was not statistically significantly associated with LOS in either analytical method. Furthermore, patients aged 60 years and older were more likely to experience prolonged hospitalization.
CONCLUSIONS: Robust Regression is more appropriate than OLS for clinical datasets containing outliers and heteroscedasticity, providing more reliable estimates of factors associated with LOS among patients with dementia. These findings may support individualized care planning, systematic management of behavioral and psychological symptoms, multidisciplinary collaboration, and the development of community-based preventive strategies aimed at reducing prolonged hospitalization among high-risk elderly patients.
Thaiclinicaltrials.org number, TCTR20260329010
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
World Health Organization. Global status report on the public health response to dementia [Internet]. 2021 [cited 2025 Jan 15]. Available from: https://iris.who.int/server/api/core/bitstreams/9e8aab50-ca34-45e0-9fb0-26e675746f65/content
GBD 2019 Dementia Forecasting Collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the global burden of disease study 2019. Lancet Public Health [Internet]. 2022 [cited 2025 Jan 15];7(2):e105-e125. Available from: https://www.thelancet.com/action/showPdf?pii=S2468-2667%2821%2900249-8
Department of Older Persons. Statistics on elderly persons in Thailand [Internet]. 2024 [cited 2025 Jan 15]. Available from: https://www.dop.go.th/th/know/1
Srikirin P, Jirakittithavorn C. Dementia Patients’ Caregiver Association of Thailand. Situation of dementia in Thailand [Internet]. 2025 [cited 2025 Jan 15]. Available from: https://policywatch.thaipbs.or.th/article/life-233
Bayes CL, Bazán JL, Valdivieso L. A robust regression model for bounded count health data. Stat Methods Med Res 2024;33:1392-411.
Maronna RA, Martin RD, Yohai VJ, Salibian-Barrera M. Robust statistics: theory and methods (with R). 2nd ed. Hoboken, New Jersey: John Wiley & Sons; 2019.
Rousseeuw PJ, Leroy AM. Robust regression and outlier detection. New York: John Wiley & Sons; 1987.
Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale: Lawrence Erlbaum Associates; 1988.
Morris JC. The clinical dementia rating (CDR): current version and scoring rules. Neurology 1993;43:2412-4.
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373-83.
