Babies of birth weight less than 1000 g (ELBW) often pose a heavy burden on available facilities in neonatal intensive care units (NICU) the world over. Reports from various countries show that survival rates among this category of babies still remain relatively low. [1][2][3][4] In a previous report from Afif General Hospital, the 28-day survival rate among ELBW babies was 30%, but within this category, there was no survival among the subset of birth weight less than 750 g. 5 Anecdotal evidence seems to suggest an increase in the proportion of these babies born in this hospital in recent times, and a cursory inspection appeared to show a higher incidence compared to data from an earlier audit by Al Faraidy et al., 6 pooled from other hospitals in Saudi Arabia. The authors, therefore, set out in this report to determine the circumstances related to the delivery of this group of babies in this hospital during the sixyear period from January 1992 to December 1997, paying particular attention to possible trends and other timerelated changes.
Subjects and MethodsThe subjects consisted of mothers with newborn babies of birth weight 500 g and above, including stillbirths, during the six-year period. The focus of interest was the group of babies of birth weight less than 1000 g, who were selected as a subset. Data including birth weight, infant sex, month of delivery, outcome, maternal age, gravidity and parity were retrieved from the delivery records. For the comparative study, two groups of mothers were used-one consisting of mothers of singleton babies of birth weights ≥2500 g (normal birth weight), and the other of singleton babies of birth weights <1000 g (ELBW babies). Two mothers were selected from the former group-that of the baby born immediately before and the other immediately after each ELBW baby. Twins, macerated stillbirths and grossly abnormal fetuses were excluded from the comparative study because of the possibility of their disproportionate contribution to low birth weight.The birth weight distribution in the Afif General Hospital dataset was first compared with that published in 1993 by Al Faraidy et al., 6 using Poisson probability test. An expected number of ELBW births and a ratio of observed to expected ELBW births, with their corresponding 95% confidence intervals, were derived while noting particularly the magnitude of the departure of the observed number of ELBW births from the expected number.
Statistical AnalysisData were analyzed using the Software Statistical Package PEPI Version 2 (Computer Programs for Epidemiologic Analysis).7 The annual delivery rates of ELBW babies were tested for possible trends using CoxStuart method. Monthly rates of delivery of ELBW babies over the combined 12 months of the study period were tested for patterns of variations by Hewitt's rank sum test for seasonal peak. Column statistic was carried out on the ages of all the mothers, deriving the median, first and third quartiles for age, and the interquartile range (IQR). Subsequently, two target ages were delinea...