Dr Montreal Children's Hospital - McGill University Health Center Montreal Children's Hospital - McGill University Health Center Montreal, Quebec, Canada
Background: In a healthcare system with limited resources, hospital organizational factors such as unit occupancy and nurse-to-patient ratios may contribute to patient outcomes.
Objective: We aimed to assess the association of NICU occupancy and nurse staffing with outcomes of very preterm infants born <33 weeks gestational age (GA).
Design/Methods: This was a multicenter retrospective cohort study of infants born 23-32 weeks GA without major congenital anomaly, admitted within 2 days after birth to one of four Level 3 NICUs in Quebec, Canada (2015-2018). For each 8h shift, data on unit occupancy were obtained from a central provincial database (SiteNeo) and linked to the hospital nursing hours’ database (Logibec). Unit occupancy rates and nursing provision ratios (nursing hours/recommended nursing hours based on patient dependency categories) were pooled for the first shift, 24h and 7 days of admission of each infant. Patient data were obtained from the Canadian Neonatal Network database. Primary outcome was mortality and/or morbidity (severe neurological injury, bronchopulmonary dysplasia, necrotizing enterocolitis and late-onset sepsis, severe retinopathy of prematurity,). Adjusted odds ratios (AOR) for association of exposure with outcomes were estimated using generalized linear mixed models with a random effect for center while adjusting for confounders.
Results: Among 1870 infants included in analyses, 796 (43%) had mortality/morbidity (Table 1). Median occupancy was 89% (IQR 82-94) and median nursing provision was 1.13 (IQR 0.97-1.37). Overall higher NICU occupancy on shift of admission, first 24h and 7 days were associated with higher odds of mortality/morbidity (Figure 1) but nursing provision was not (Figure 3). Subgroup analysis by GA (<29 and 29-32 weeks) showed similar results (not showed). Generalized linear mixed model analyses showed that a 5% reduction in occupancy in the first 24h of admission was associated with a 6% reduction in mortality/morbidity (Figure 4). Conclusion(s): NICU occupancy is associated with mortality/morbidity among very preterm infants and may reflect lack of adequate resources in periods of high activity. Interventions aimed at reducing occupancy and maintaining adequate resources need to be considered as strategies to improve patient outcomes.
Figure 1. Adjusted odds ratio for association of unit occupancy on the shift of admission, first day of admission, first seven days with mortality or major morbidity by site
Adjusted for covariates (gestational age, small for gestational age, sex, outborn, SNAPII > 20 and, mode of delivery), nursing provision ratio and site (random effect).
Figure 2. Adjusted odds ratio for association of nursing provision on the shift of admission, first day of admission, first seven days with mortality or major morbidity by site
Adjusted for covariates (gestational age, small for gestational age, sex, outborn, SNAPII > 20 and, mode of delivery), unit occupancy and site (random effect).
Figure 3. Fitted curve for association of occupancy rate in the first 24 h of admission with mortality/morbidity (obtained by generalized linear mixed model with site as random effect)
Authors/Institutions: Marc Beltempo, Montreal Children's Hospital, Montreal, Quebec, Canada; Robert Platt, McGill University, Montreal, Quebec, Canada; Anne-Sophie Julien, Universite Laval, Quebec, Quebec, Canada; Régis Blais, Universite de Montreal Ecole de Sante Publique, Montreal, Quebec, Canada; Valérie Bertelle, Centre integre universitaire de sante et de services sociaux de l'Estrie Centre hospitalier universitaire de Sherbrooke du Quebec, Sherbrooke, Quebec, Canada; Anie Lapointe, Centre Hospitalier Universitaire Sainte-Justine, Montreal, Quebec, Canada; Guy Lacroix, Universite Laval, Quebec, Quebec, Canada; Sophie Gravel, Centre Hospitalier Universitaire Sainte-Justine, Montreal, Quebec, Canada; Michèle Cabot, CHU de Quebec-Universite Laval, Quebec, Quebec, Canada; Bruno Piedboeuf, Universite Laval Faculte de medecine, Quebec City, Quebec, Canada