Analysis and Estimation of Length of In-Hospital Stay Using Demographic Data of COVID-19 Recovered Patients in Singapore

Singapore is one of the countries which has taken early, systematic, and rigorous nationwide responses to slowing the COVID-19 contagion down. By February 4, 2020, the government of Singapore restricted the mobility of the vulnerable age-groups. In the current study, we study the influence of the age-based vulnerability of the population with respect to COVID-19 conditions on the recovery of COVID-19 patients. We study 245 patients in Singapore recovered and discharged during January 23--April 01. We first study the descriptive statistics of the length of in-hospital stay (LOS) of the COVID-19 patients based on demographic variables, namely age, and gender. Then, we determine the distribution of LOS, using local and generalized linear regression models. We take the approach of periodization based on critical changes in the disease transmission model. Even though the overall recovery rate has reduced drastically after a sudden spike in daily confirmations, our analysis shows that there is a considerable shift in the COVID-19 confirmations to the population in the non-vulnerable age-groups. We show that the LOS of the non-vulnerable age group is considerably lower at 9 days, as opposed to 15 or 20 days in the existing literature.