O1007: Discharge Barriers and Postoperative Length of Stay among Infants with Severe Congenital Heart Defects

Background: Hospital length of stay (LOS) is associated with higher costs, increased risk of complications, and negative health outcomes for patients. Yet, the discharge barriers that influence LOS following pediatric cardiac surgery remain poorly understood. The objective of this study was to compare the contributions of discharge barriers versus clinical factors to postoperative LOS among infants who received congenital heart surgery.

Methods: This is a retrospective cohort study of infants who underwent initial cardiac surgery at < 12 months, between 2015-2021, at a high-volume congenital heart surgical center. The primary outcome was postoperative LOS, in days. The primary exposure was discharge barrier category, a six-level categorical variable measuring provider-documented discharge barriers. The six categories were determined a priori using a mixed-methods approach (Table 1): first, discharge barriers were extracted as excerpts from electronic medical records, then iteratively categorized into six categories based on expert clinical consensus, and finally tested using regression. Categories were not mutually exclusive; some patients could experience multiple discharge barrier categories. Additional covariates included surgical complexity (STAT category), postoperative complications (e.g., poor weight gain), equipment needs at discharge (e.g., nasogastric tube), comorbidities (e.g., poor weight gain), sociodemographics (e.g., race), and a COVID-19 indicator. Log-linear models estimated associations with postoperative LOS. A Shorrocks-Shapley decomposition estimated the degree to which each covariate contributed to explained variation in postoperative LOS.

Results: 753 infants met inclusion criteria; 183 (24%) reported a discharge barrier. The most common barrier category was medical equipment delivery or prior authorization (n=52); the least common was social determinants (n=14, Table 1). Median postoperative LOS was 26 (IQR 14-52) days for those with discharge barriers, and 8 (IQR 5-15) days for those without. In the adjusted model, compared to patients who experienced no discharge barrier, those with a barrier due to home nursing or bed availability had a 255% (95%CI 134-438%) higher postoperative LOS, as did those whose discharge was delayed due to a supervised parental stay (89%, 95%CI 48-142%), those with unexpected clinical instability (82%, 95%CI 38-141%), and those awaiting medical equipment delivery and/or insurance prior authorization (32%, 95%CI 3-68%, Table 2). Discharge barriers and STAT category contributed a similar amount of explained variation in postoperative LOS (21.8% and 22.0%, respectively).

Conclusion: In this cohort of infants who received cardiac surgery for a severe congenital heart defect, discharge barriers contributed as much toward variation in postoperative length of stay as surgical complexity. However, unlike surgical complexity, many discharge barriers that patients experienced were modifiable. Future research and clinical efforts to mitigate discharge barriers may yield improvements to length of stay.