Social class inequality in infant mortality in England and Wales: 1921 through 1984

Although the existence of substantial social class differences in infant mortality in virtually all industrialized societies is widely acknowledged, the secular trend in the level of inequality remains a matter of debate. Fueling this debate are documented inadequacies in the data themselves and in the methods commonly employed in their analysis. This analysis examines the trend in social inequality in infant mortality in England and Wales between 1921 and 1980-84, using data on infant mortality according to the social class and occupation of the father. It employs a summary measure of inequality that incorporates the mortality rate and size of each class or occupation, converts easily from an absolute to relative measure, and can be evaluated in terms of its sensitivity to errors using accepted diagnostic techniques. Occupations that played a significant role in determining the time trend in inequality are identified, and the results are used to evaluate the trends in social class inequality in neonatal and postneonatal mortality and in five cause of death categories. In addition, mortality among unclassified and out-of-wedlock infants, groups usually excluded from analyses by social class, is examined. The results of the analysis show that absolute inequality has declined in concert with the reduction in infant mortality in England and Wales between 1921 and 1980-84. Relative social class inequality, on the other hand, was higher in 1949-53 than in 1921 for all age and cause groups. In the period after World War II, social class inequality in postneonatal mortality and in most of the causes commonly associated with it declined to a level below that for 1921, while relative inequality in neonatal mortality and in its associated causes continued to increase. However, the inclusion of unclassified and out-of-wedlock infants alters these trends, particularly for postneonatal mortality and its associated causes. These results indicate that more than one factor is responsible for the persistence of large relative differences in infant mortality across social strata