221 - Regional variations in length of stay and readmission of preterm infants in Canada

Poster Board #: 221
Publication #: 2910-HP-QA.221

Background: Previous studies have found variations in Neonatal Intensive Care Unit length of stay (LOS) between health regions for very preterm infants. However, little is known on variations in total hospitalization prior to discharge home among all preterm infants between regions and how they relate to readmission rates.


Objective: To evaluate the association between health region (Provinces) and total LOS of the birth hospitalization, 30-day and 2-year readmission rates among preterm infants in Canada.


Design/Methods: We conducted a population-based study of preterm infants born in Canada using the Canadian Institute for Health Information Database. We included infants born between 22-36 weeks gestational age (GA) admitted 2010-2015 and survived to discharge after initial neonatal hospitalization. Infants were categorized based on province of birth.
Outcomes included LOS of first neonatal hospitalization till discharge home, 30-day and 2-year readmission rates post discharge. LOS was stratified by GA group and compared between regions using generalised linear models adjusted for confounders. The 30-day and 2-year readmission rates were compared between regions using adjusted standardized ratios adjusted for same confounders. Pearson correlation coefficient (r) was used to assess correlation between the three outcomes.


Results: Among 173,789 preterm infants born alive during the study period, 170,316 (98%) survived initial hospitalization. Median LOS decreased with increased GA within each Province (Table 1). For each GA group, there were significant variations in LOS between Provinces (Table 2).
Overall, 30-day readmission rate was 6.2% (10,567) and varied significantly between provinces (Table 3). The 2-year readmission rate was 11.7% (19,971) and also varied between provinces. There was no significant correlation between provincial initial LOS and 30-day readmission rates (r=-0.44, p=0.19) or between provincial initial LOS and 2-year readmission (r=-0.22, p=0.53). There was a significant positive correlation between 30-day and 2-year readmission rates (r=0.90, p<0.001).

Conclusion(s): In this population-based cohort there was variation in LOS for preterm babies in Canada. There was no correlation between higher provincial LOS of birth hospitalisation and lower 30-day readmission rates. These variations are likely due to differences in healthcare discharge preparation, home care services and community support services which need to be explored in future studies.


Table 1. Median (IQR) length of stay per province among survivors to discharge home for each gestational age group


Table 2. Multivariable analysis for variations in length of stay among survivors to discharge home per gestational age group and province


Table 3. Provincial 30-day readmission rate among all preterm infants discharged alive with provincial standardized ratios


Table 4. Provincial 2-year readmission rate among all preterm infants discharged alive with provincial standardized ratios

---

Authors/Institutions: Marc Beltempo, Montreal Children's Hospital, Montreal, Quebec, Canada; Xiang Y. Ye, MiCare center, Toronto, Ontario, Canada; Anne Synnes, British Columbia Women's Hospital and Health Centre, Vancouver, British Columbia, Canada; Bruno Piedboeuf, Universite Laval, Quebec, Quebec, Canada; Fabianna Bacchini, Canadian Premature Babies Foundation, Etobicoke, Ontario, Canada; Prakesh Shah, Mount Sinai Hospital, Toronto, Ontario, Canada