…or are they?BMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6976.399a (Published 11 February 1995) Cite this as: BMJ 1995;310:399
EDITOR,—Although the statistics of health inequalities in Britain discussed in George Davey Smith and Jerry Morris's editorial1 seem impressive, the absolute scale and nature of the problem, and what indeed is to be done to resolve it, remain elusive. Thus, working as a general practitioner in inner city London with a wide social mix of patients I find it difficult to relate, for example, the twofold differential in infant mortality between social classes I and V to clinical practice. Death in infancy is nowadays very rare and when it does occur, whether from the sudden infant death syndrome or childhood malignancy, it occurs just as often in professional households as working class households.
The discrepancy between the raw statistics of health inequalities and everyday experience is clarified by a closer examination of the causes of infant mortality (table), which confirms the twofold or 200% excess deaths between social class V and I. The overwhelming majority of infants, however, do survive the first year, and when the same data are presented as survival rates a slightly different picture emerges. Here the relevant figures for social class I and V are 994.4 and 988.8 per 1000 live births respectively, a survival advantage of only 0.6% for children born into professional families. The differential in infant mortality is accounted for predominantly by three major categories of disease—congenital anomalies, prematurity (including the respiratory distress syndrome), and the sudden infant death syndrome.
There are many causes of congenital anomalies including chromosomal and genetic abnormalities and intrauterine infections, but with the possible exception of folic acid supplementation for neural tube defects they are not readily amenable to prevention. The causes of prematurity are not known and again are not readily amenable to preventive intervention. By contrast, the sudden infant death syndrome is now recognised to be strongly influenced by sleeping position and has declined by 70% since the public health campaign advising that babies should sleep on their backs.
None of these three categories of disease is obviously related to poverty. So the vast edifice of argument about health inequalities mounted on the back of the twofold differential in infant mortality between the social classes is reduced to focusing on one truly preventable condition—the sudden infant death syndrome. Further, the actual excess in the number of deaths from this condition in social class V compared with social class I is only seven.
Despite the vast literature on health inequalities generated over the past decade this type of detailed examination of the statistics has never been undertaken. Why might this be? The findings contradict the argument that the health differential between the social classes is readily attributable to “poverty” and also raises doubts about the proposed solutions—a major political programme of income redistribution. The Black report, for example, called for “greater restrictions on the amount of wealth which can be inherited, establishment of minimal—and maximal—earnings, the recognition of right to full employment and a major upgrading of child and family allowances.”3 There may well be good political arguments for such policies but it is hard to see how they could have much influence on the causes of the social class differential in infant mortality.