Carcinoma of the cervix is a major health problem throughout the world. After breast cancer, it is the second commonest malignancy in women, with an incidence of about half a million cases a year. In some localities-Africa, India, and certain other Asian countries-it is the commonest cancer in women; in Europe and North America it is the fourth commonest.' Data from some developed countries show a reduction in cervical cancer mortality of about 30% between 1960 and 1980, presumably because of early diagnosis through screening programmes.2 In England and Wales, however, there has been only a small decline in deaths from this cause in the past 20 years, and 2000 women a year still die of it.3 During the past decade there has been a pronounced increase in cervical intraepithelial neoplasia (CIN), particularly in young women,45 and Beral, using a computer model, has predicted a 60% increase in cervical cancer registrations and a 70% increase in mortality in women aged under 50 in 10 years' time.6Squamous carcinoma of the cervix is the culmination of CIN, which is a series of progressive epithelial changes. Although the pathogenesis of these changes is not well understood, the cytology and histology of CIN is clear,7 and invasive cancer of the cervix could be completely prevented if CIN was detected by screening tests and treated correctly. Epidemiological studies have made it possible to define some behavioural and other characteristics of women who are at relatively high risk ofcervical cancer, and recent advances in basic science have clarified, at least in part, the aetiology of the disease. The purpose of this paper is to review this evidence with particular reference to women who attend clinics for sexually transmitted diseases, and to look for ways in which the services in these clinics can be used or modified to reduce the likelihood of invasive cervical disease in this group. Epidemiological review Domenico Rigoni-Stern was chief physician of a hospital in Verona, and he analysed mortality from cancer in that city for the years 1760-1839.' He pointed out that cancer of the uterus (and he was probably referring to the cervix) was commoner in married women and widows than in unmarried women, and was unknown in those bound by monastic vows. He suggested that the liability of a uterus to cancer might depend on "the natural exercise of its functions." It is difficult to be sure what he meant by this, but he was probably referring to either menstruation or childbirth. These observations were largely forgotten until qtite recently, but in the early 1950s several studies confirmed the extreme rarity of carcinoma of the cervix in Catholic nuns.9`' In the discussion of these findings, emphasis was placed not on the celibacy of these-women but on their childlessness, because at the time chronic cervicitis and cervical lacerations during delivery were regarded as being of major aetiological importance.During the 1960s, links were noted between carcinoma of the cervix and prostitution. Studies of the inmates of women's p...