Virus Diseases in Pregnancy and Congenital Defects

To determine the probability that an illness of the mother during her pregnancy will give rise to a congenital malformation in the infant, essentially demands a prospective method of inquiry. In other words, the attack of illness in the pregnant woman must first observed and medically diagnosed, and the condition of the child born to her must subse quently noted. The retrospective approach, i.e. noting first the congenital deformity in the new-born child and then obtaining the history of the mother during pregnancy, inevitably omits the normal children whose mothers were affected, and is thus bound to give a highly exaggerated picture of the risks involved. The prospective inquiry is, however, extremely difficult to carry out on an adequate scale. The attack rates from infectious diseases are rela tively low in adult life and very large numbers of women must observed. It is specially important, too, that the occurrence of disease in the mother recognized and recorded at a time when she may not even know that she is pregnant, i.e. during the first month of pregnancy. One way of securing such records in Great Britain has been described in an earlier paper (Bradford Hill and Galloway, 1949). The data there used were automatically collected by the approved societies operating up to July 5, 1948, under the National Health Insurance Acts. Under these Acts, if an employed and insured married woman drew benefit for a sickness which caused her to absent from work, she had to present a sickness certificate giving the dates of her illness and the general practitioner's diagnosis of its cause. If she subsequently gave birth to a child (live or still) she would entitled to draw maternity benefit and the date of birth would therefore also recorded by the approved society. Thus the dates of the two would available in records automatically maintained. By these means information was collected on 22 cases over the space of some 2 years; they included ten cases of rubella and six of measles. The infants were specially examined not long after birth and one case of congenital heart disease following rubella in the first month of pregnancy was observed. The authors concluded their report with the hope that it might be possible to extend these observations through the records of the new health and social services (which came into being on July 5,1948). This hope has been fulfilled. With the aid of the Ministry of Pensions and National Insurance we have been able to collect similar records for 100 women reported to have suffered from rubella, measles, mumps, or chicken pox during, or shortly before, pregnancy. A further seven cases were brought to our notice privately but prospectively (i.e. before the birth of the child). It is with this total of 129 double events that the present paper is concerned.