Health after childbirth

We have carried out an investigation, thc first of its kind. into the nature and extent of long-term health problems beginning for the first time soon after childbirth. In particular wc have searched for potential determinants of thcse health problems from among a wide range of social, obstetric and anaesthetic factors and the findings are to be published as a book by HMSO in November (MacArthur et a / . 1991). Long-term morbidity after childbirth has becn a neglected area of research. A few studies have followed up specific problems, such as perineal management and urinary incontinence (Sleep & Grant 1987) or dcpression (Kumar & Robson 1984; Watson er al. 1984), but there has becn no comprehensive assessment of the extent of problems expericnced by women. This present study is based upon ;i sample of 11 701 womcn giving birth to their most recent child, in one hospital in Birmingham between 1978 and 1985. Information on health problems was obtained through a postal questionnaire scnt in January 1987. and this was linked to the full maternity case-notes for cach woman. The main investigation was limited to symptoms not previously experienced by the woman so that they could be related specifically to the circumstances of a particular delivery. Recurrent symptoms. including those starting during the pregnancy. were discardcd. Complaints lasting 6 weeks or less were also excluded since our main concern was with longer-term health problems. Most of the symptoms rcported by thc women had lastcd more than a year, and around twothirds (varying slightly for specific symptoms) wcrc still present at the time of the enquiry, 13 months to 9 years later. Investigation was also limitcd to symptoms first occurring within 3 months of the birth, although in fact thc majority started within the first week. The research stemmed from an initial concern of obstetric anaesthetists as to whether epidural anaesthcsia had any long-term consequences. However, thc technical necessities of the research design, the opportunities of the situation and the lack of previous research in the area, led to a widened investigation of postpartum morbidity in general, including an examination of social and obstetric factors along with the anaesthetic ones. Thc extent and the persistence of the reported morbidity surprised us: 47% of the women reported at least one ncw health problem, starting within 3 months of the delivery and lasting more than 6 weeks. The analysis of the social and obstetric associations of the symptoms was based on multivariate discriminant analysis and on multi-way tabulations. Potential determinants were those factors remaining as independently associated after taking account of inter-related factors using both forms of multivariatc comparisons. Backache was a common health problem after delivcry, reported by 1634 women (14'Yo). Many backaches had become chronic, with 65% still unresolved at the time of the enquiry. There was a powerful association with cpidural anaesthesia: 18.9% of women reported backache after an epidural compared with 10.5% after noncpidural delivcries (MacArthur et al. 1990). This difference was consistently prcsent in many different circumstances, and the excess of 'abnormal' deliveries occurring with epidural maesthesia did not seem to account for the association; nor did varying pain tolerances. Spinal anaesthesia, a long sccond stage labour and Asian ethnic group were additional independent predictors of backache. The only circumstancc in which an cpidural did not increase the risk of backache was after elective caesarcan section. We have hypothesized that the backache-producing mechanism is postural, occurring generally during labour, but exacerbated by epidural anaesthcsia, through loss of muscle tone, inability t o move, and inhibition of discomfort-feedback. Additional events in the