Women's views of the place of confinement.

BACKGROUND The proportion of births taking place in the home of the mother has declined to less than 1% of the total, in spite of the lack of evidence of the benefits of hospitalization. Home confinement remains rare when supported by general practitioners; little is known of the preferences of women of childbearing age. AIMS To examine the beliefs of women of childbearing age concerning the advantages and disadvantages of giving birth in hospital or their own homes, the choices of place of birth open to them, and the information required to enable an informed choice. METHOD A self-administered postal questionnaire was sent to a sample of 340 women aged 20-40 years who were registered with the study practice--a single-handed urban general practice that supports home confinement. RESULTS Completed questionnaires were received from 68% (241) of subjects. Parity of respondents ranged from zero (24%) to five. Altogether, 76% (159) of the respondents had previously had children. A total of 97% (154) of the parous respondents had prior experience only of delivery in hospital. In all, 86% (198) of the respondents expressed a preference for hospital delivery for any future children, with 3.5% (8) preferring home and 10.5% (24) undecided. Preference for home birth was associated with multiparity and indecision with nulliparity (Kruskal-Wallis, P = 0.040). Most women believed that giving birth in hospital was safer than at home. Most women had little knowledge of the advantages and disadvantages of giving birth at home and in hospital, although there was substantial demand for further information, and a minority who were better informed than their peers. A minority of respondents spontaneously expressed hostility to all home confinements. CONCLUSIONS Women show a wide spectrum of opinion on this subject. Further information is needed for women to make a free and informed choice of place of birth; providing this in an acceptable form is likely to require knowledge of the individuals concerned.