First day neonatal mortality since 1935: re-examination of the Cross hypothesis.

Abstract

OBJECTIVES--To describe the change in first day infant mortality during 1935-87. To examine the hypothesis that excess first day mortality in the 1950s and 1960s was attributable to restricting oxygen for sick newborn infants. DESIGN--Time series analysis of first day infant mortality and stillbirth rates. SETTING--England and Wales and the United States of America. SUBJECTS--All first day infant deaths, all neonatal deaths, and all stillbirths. MAIN OUTCOME MEASURES--Rate of fall in mortality, dates of deviation of mortality from established fall, and correlation with stillbirths. RESULTS--In England and Wales first day infant mortality fell by 3.1% a year, except between 1951 (95% confidence interval 1951 to 1954) and 1980 (confidence interval less than 1 year). During these years there were 37,000 excess deaths. In the United States an annual fall of 2.7% was interrupted in 1955 (1951 to 1954) and resumed in 1980 (1978 to 1980), resulting in 195,000 excess deaths. A similar pattern was observed in stillbirth rates. CONCLUSIONS--Restriction of oxygen in sick newborn infants cannot be the sole cause of the interruption in fall of first day neonatal mortality as stillbirth rates were also affected. The timing of onset and the course of the deviation is not consistent with the oxygen restriction hypothesis. Further investigation is needed to identify a factor affecting both fetal and newborn survival between 1950 and 1980.