In 1968, a prospective study was started in collaboration with the Family Planning Association to try to provide a balanced view of the beneficial and harmful effects of different methods of contraception. This investigation is now in progress at seventeen clinics and over 17,000 women are under observation. At the time of recruitment, all these women were married white British subjects, aged 25–39 years, who voluntarily agreed to participate. Fifty-six per cent were using oral contraceptives, 25% were using a diaphragm and 19% were using an intrauterine device (IUD). During follow-up each woman is questioned at return visits to the clinic and a record of pregnancies and their outcome, hospital referrals (inpatient or outpatient), changes in contraceptive methods and the results of cervical smears, is accumulated. Women who default are sent a postal questionnaire and, if this is not returned, are telephoned or visited in their homes to collect the necessary information.

So far, data obtained during 56,000 woman-years of observation are available for analysis. Follow-up has been maintained with an annual lapse rate of about 0.3%due to withdrawal of co-operation or loss of contact; adherence to the method of contraception in use at recruitment has been reasonably good, and the reporting of pregnancies and hospital admissions appears to have been both reliable and unbiased.

The present data include only 24 deaths, so the mortality associated with different contraceptive methods cannot yet be estimated. With regard to morbidity, however, our preliminary results closely resemble those obtained in the prospective study carried out by the Royal College of General Practitioners (1974) and in the principal retrospective studies carried out in Britain and the United States. Women who used oral contraceptives at the start of the study experienced a deficiency of hospital referrals for cancer, benign lesions of the breast, menstrual disorders other than amenorrhoea, duodenal ulcer, and retention cysts of the ovary; they showed an excess of referrals for cerebrovascular disease, cervical erosion, skin disorders, self-poisoning, migraine, venous thrombosis and embolism, hayfever, gallbladder disease, amenorrhoea, and sterility. Women who used a diaphragm showed a deficiency of hospital referrals for carcinoma-in-situ and dysplastic lesions of the cervix uteri and accidental injury; and an excess of referrals for haemorrhoids and cystitis. Women who used an IUD experienced an excess of hospital referrals for anaemia, varicose veins and salpingitis. About half of these differences (12 out of 23) were predicted from other studies while some suggestive evidence already existed for a further five. Of the remaining six, some probably reflect the influence of selective factors or chance.

Multiple pregnancies, stillbirths, malformations, the sex ratio and birthweight showed no consistent relationship to method of contraception. The outcome of unplanned pregnancies occurring in women using an IUD, however, was remarkably unfavourable both in terms of ectopic gestation and miscarriage.

Clear evidence was found of impairment of fertility after discontinuation of oral contraceptives. Whether this is likely to lead to permanent sterility in some women is uncertain.

The study has provided data on the efficacy of a wide variety of contraceptive methods. In general the failure rates are in keeping with those obtained in other large-scale studies, save that those for the diaphragm and the sheath are much lower than those usually quoted.

The available evidence does not yet allow a final balance to be struck between the benefits and risks associated with the new methods of contraception that have become widely used during the last two decades. It seems clear, however, that there are no material risks associated with the use of the diaphragm apart from the risk of pregnancy and that there may be some unintended benefits.