(1319) Factors Associated With Direct Discharge to Home From a Rural PICU: A Retrospective Cohort Study

Introduction: Due to reduced inpatient bed availability, staffing shortages, and budget cuts there are increasing constraints on pediatric intensive care unit (PICU) beds. This leads to more children being discharged directly home from the PICU. Our study aims to compare the characteristics, predictors, and PICU readmission rates of patients discharged directly home (DDH) versus those transferred to hospital wards (THW) before discharge.

Methods: This was a retrospective observational cohort study done in a rural PICU, between January 2019 and December 2022 using the VPS database system. Children who were less than 18 years old and survived their PICU stay, and were DDH or to an inpatient ward were included. Children who were DDH were compared with THW using descriptive statistics. Logistic regression was used to identify factors associated with home discharge. Odds of readmission among DDH cases were compared to THW cases after adjusting for DDH predictors. Sensitivity analyses were conducted using lower severity THW cases based on length of stay.

Results: Of the 1,745 admissions, 1,501 children met inclusion criteria, of whom 441 (29%) were DDH from the PICU. Diagnoses associated with DDH were diabetic ketoacidosis (11.1%) followed by Hodgkin’s disease (3.9%). When children who were DDH were compared with THW, we found that children who were older with a higher weight and lower PRISM-3 scores were more likely to be DDH [Table 1]. An increase of 5 years was associated with 23% higher odds of direct discharge to home (p=0.049). When compared to the most common admission reason i.e. respiratory/ENT, poisonings, cardiovascular and neurological admissions had increased odds of DDH. After accounting for differences in age, patient origin, admission reason and other patient factors, the odds of 28-day readmission among patients DDH were not significantly higher than the odds of 28-day readmission among patients who were THW (p=0.115). Adjusted comparison between DDH and THW with stay of < 2 days showed that the odds of 28-day readmission were also not statistically significant (p=0.441).

Conclusions: Rates of PICU readmission between patients DDH and THW are similar and appear to be safe in a rural PICU. However, given small sample size, the safety of the practice requires ongoing evaluation in a larger setting.