1. Home>
2. Specialties>
3. Pediatrics and Adolescent Medicine>
4. Summary and Comment

Variations in Mortality Rates at Neonatal ICUs

Significant differences in neonatal mortality at different centers could help us find ways to improve practice.

Data from the Canadian neonatal network allowed researchers to examine the deaths of 19,265 infants who were admitted to 17 tertiary level neonatal intensive care units (NICUs) between January 1996 and October 1997. Using multivariate analysis, the researchers were able to adjust for risk factors associated with baseline population characteristics and severity of illness at the time of admission; this approach enabled meaningful comparison of variations in mortality rates among the NICUs.

The overall mortality was 4%; 40% of deaths occurred within 2 days of NICU admission. Five-minute Apgar scores and severity of illness on admission (according to the Score for Neonatal Acute Physiology-II [SNAP-II]) provided the best means of evaluating the equivalence of illness severity. There was significant variation in the risk-adjusted mortality among the NICUs (range, 1.6%-5.5%). Preterm birth, birth at a hospital other than the one in which the NICU is located, and congenital anomalies were the variables most often associated with death in the NICU.

Comment: Canada's healthcare system provides an ideal opportunity for comparing different centers. SNAP-II is a useful tool for comparing outcomes at the individual centers and for evaluating specific practices that affect outcomes. This approach can help institutions determine whether changes in their practices could improve care.

— Judith G. Hall, OC, MD

Published in Journal Watch Pediatrics and Adolescent Medicine May 28, 2002

Citation(s):

Sankaran K et al. Variations in mortality rates among Canadian neonatal intensive care units. CMAJ 2002 Jan 22; 166:173-8.

• Original article (Subscription may be required)
• Medline abstract (Free)

2011 Year in Review