Determining the Length of Stay and Duration of Illness for Psychiatric Inpatients Using Multivariate Modelling


Mental and behavioural diseases account for a considerable portion of the worldwide disease burden. According to WHO predictions, by 2020, this burden will have increased by 15%, having a considerable impact on health as well as major social, human rights, and economic ramifications in all nations around the world. Using multivariate normal (MVN) and bivariate normal (BVN) normal distributions, this work presents a strategy for estimating hospital length of stay (LOS), total duration of sickness (TDI), and recent duration of illness (RDI). To do so, researchers used retrospective data from 146 patients with complete medical records and diagnoses. For the calendar year 2013-2014, data on people with mental and behavioural disorders (as defined by the APA’s DSM-V and the WHO’s ICD-10) was obtained from the Department of Psychiatry, Lady Hardinge Medical College & Smt. S.K. Hospital, New Delhi, India. The above-mentioned variables’ estimated values are determined to be compatible with the observed values. Finally, for patients for whom information on these variables is unknown, the MVN distribution is used to estimate the variables LOS and TDI. The model developed in this work will aid the medical community in not just advising patients on their expected duration of stay in the hospital at the time of admission, but also in developing and managing appropriate patient therapies.

Author (s) Details

Prof. Alka Sabharwal
Department of Statistics, Kirori Mal College, University of Delhi, Delhi, India.

Dr. Sakshi Kaushik
Biostatistics- Biometrics Department, Quartesian, Chennai, Tamil Nadu, India.

View Book :-

Leave A Comment