EPIDEMIOLOGY OF EARLY BREAST NEOPLASIA As in many other epithelial cancers, histopathology studies suggest that the natural history of carcinoma of the breast starts with proliferative epithelial dysplasia, some of which progresses to carcinoma-in-situ, some of which in turn progresses to invasive carcinoma. If this progression truly reflects stages in the same disease process one would expect to see similar distributions of risk factors in the early stages and the late, as has been shown for example in relation to dysplasia, carcinoma-in-situ, and invasive cancer of the cervix.2 A case-control study3 has now provided evidence on the relative risks of several aetiological factors in four stages of breast disease-namely, benign disease, in-situ carcinoma, invasive carcinoma up to 1 cm in diameter, and invasive carcinoma greater than 1 cm in diameter. Over 1500 patients with cancer, and comparable patients with benign biopsy results and controls, were drawn from women who had attended for screening in the U.S. Breast Cancer Detection Demonstration Program (BCDDP), the benign and control groups being matched with the cancer group for age, race, and number of screens. The relative risks for both small and larger invasive cancers, compared with controls, were very similar for the recognised reproductive and other risk factors associated with breast cancer.4 In-situ carcinomas also carried similar relative risks, except that no significant increase in risk was seen in obese women, nor in those who had a sister with breast cancer and there was no reduction in risk for those who had had oophorectomy. These differences could not be attributed to the younger age of insitu patients since age-specific analyses did not alter the findings (although numbers were small). This led the authors to speculate that while risk factors operating early in life are common to in-situ and invasive cancer, the later hormonal influences of obesity and oophorectomy enhance or postpone progression of in-situ disease to invasion. A history of previous breast biopsies was the only factor significantly associated with benign disease and the only factor common to benign disease and all malignant categories. The benign disease category, however, was nonspecific since the sample was drawn from all patients in the programme who had non-malignant biopsy findings. Had the analysis been confined to those with proliferative epithelial lesions, associations similar to those in the malignant categories might have been found. This study lends weight to the hypothesis that in-situ and invasive breast carcinoma are different stages in the same disease process, and by implication that it is a good policy to diagnose and treat the in-situ phase. One of the controversial aspects of screening, however, is the possible overdiagnosis and overtreatment of preinvasive neoplastic lesions which, if left undiagnosed, might not have progressed to invasive cancer within the woman’s lifetime.5 Necropsy series have revealed that a fifth or more of elderly women dying from other causes have proliferative mammary dysplasia and 6% have in-situ carcinoma.6 Current screening programmes,
2. 3. 4. 5. 6.
cancer. Adv Cancer Res 1980; 31: 287-314. Barron BA, Richart RM. An epidemiologic study of cervical neoplastic disease. Cancer 1971; 27: 978-86. Brinton LA, Hoover R, Fraumeni JF Epidemiology of minimal breast cancer. JAMA 1983; 249: 483-87. Kalache A, Vessey M. Risk factors for breast cancer. Clin Oncol 1982, 1: no 3: 661-78 Chamberlain J. Screening and natural history of breast cancer. Clin Oncol 1982; 1, no. 3: 679-701. Kramer WM, Rush BF. Mammary duct proliferation in the elderly. Cancer 1973; 31: 130-37.
Wellings SR. Development of human breast
finding that 20-30% of cancers in-situ phase.’ The only way to determine whether or not these represent overdiagnosisisto compare the cumulative incidence of in-situ and invasive carcinomas over a decade or more in both a group of women offered screening and a control group. This information should eventually emerge from current trials of screening policy8 but, if overdiagnosis is shown to be a real factor, the dilemma about overtreatment will still remain.
including the BCDDP, diagnosed are in the
YOUNG MOTHERS AND THEIR CHILDREN As a political concern, transmitted deprivation seems to have its own cycle. Ten years ago Sir Keith Joseph, Secretary of State at the Department of Health and Social Security, argued in a speech to the Pre-School Playgroups Association that welfare policies in Britain had failed to break the link of deprivation between one generation and another. Now, as a member of Mrs Thatcher’s Family Policy Group, Sir Keith has returned to the issue: "By hypothesis most of those who provided bad parenting tend to be the least self-disciplined and with short time horizons. Those girls who are most at risk will tend neither to restrain themselves nor to insist on or use contraceptives nor to have sufficient grip even to consider abortion in sufficient time. Can their attitudes be changed?"’ The ten years since 1972, however, have seen a com. mendable exploration of the issue of transmitted deprivation by the Social Science Research Council, and incidentally show the value of the Rothschild customer/contractor principle established in the same year. The arguments have been more clearly defined, although, with so many factors involved, there remain areas of doubt. In a comprehensive summary of the research, Brown and Madge comment,2 "Much deprivation is deeply rooted in the structure of our society and affected by the network of unequal opportunities and life chances that the structure maintains. The individual’s experience of deprivation, and his response to it, are determined by the interaction of his natural endowment and family resources with that structure and pattern of life chances". Two national longitudinal studies3,4 of babies born during one week in 1946 and 1958, respectively, have contributed to our understanding of transmitted deprivation. A third cohort, of babies born in 1970, was examined by the Child Health and Education Study in 1975, and seven years later the first results have been published.5Taylor, Wadsworth, and Butler compared the morbidity of 1031 singleton children born to teenage mothers with those of 10 950 singleton children born to older mothers. Hospital admissions for accidents, gastroenteritis, and all causes were more frequent in the teenage mothers, although no 7. Beahrs
OH, Shapiro S, Smart C. Report of the working group to review the National Cancer Institute - American Cancer Society breast cancer detection demonstration projects. J Natl Cancer Inst 1979: 62: 640-709. 8. U.K. Trial of Early Detection of Breast Cancer Group. Trial of early detection of breast cancer: description of method. Br J Cancer 1981; 44: 618-27 1. Quoted by M Dean. Guardian Feb. 19, 1983, p. 1. 2. Brown M, Madge N. Despite the welfare state: report on the SSRC/DHSS programme of research into transmitted deprivation. London: Heinemann Educational, 1982 3. Atkins E, Cherry NM, Douglas JWB, Kiernan KE, Wadsworth MEJ. The 1946 British birth-cohort survey: an account of the origins, progress and results of the National Survey of Health and Development. In: Mednick SA, Baert AE, eds Prospective longitudinal research. Oxford: Oxford University Press, 1981 4. Fogelman K, Wedge P The National Child Development Study (1958 birth cohor) In. Mednick SA, Baert AE, eds. Prospective longitudinal research. Oxford- Oxford University Press, 1981. 5. Taylor B, Wadsworth J, Butler NR. Teenage mothering, admission to hospital, and accidents during the first 5 years. Arch Dis Childh 1983; 58: 6-11.
differences were found for respiratory illnesses; all accidents, including those not resulting in admission to hospital, were also reported more frequently by the teenage mothers. Taylor and co-workers conclude that "low maternal age appears to be a health hazard for children" and suggest trial programmes of prenatal and postnatal social, medical, and educational support for the children of young mothers. The report, however, leaves some questions -open. The analysis compares teenage mothers with all older mothers, whereas the effect of parenting might have been shown better by comparing first children only. No reliability study of the survey methods-home interviews with over 500 items by health visitors-has yet been published, and the validity of the mothers’ recall in this setting must be questioned. It is also possible that casualty officers more readily admitted the children of young mothers because, as a group, they believed them to be at higher risk. It is important to bring out these reservations about cross-sectional association studies, because an apparently simple finding may be taken as a ruleof-thumb in clinical practice, leading to the "ecological fallacy" that what applies broadly to the group may not apply to individuals within the group. As a parallel example, 100% hospital delivery for women was advocated6 on the grounds that perinatal mortality rates are higher for home deliveries. But in fact many home deliveries now are unplanned, as shown by the high proportion (40% in 1979) that are illegitimate births. For mothers aged 25-29 years the born at home is perinatal mortality rate of legitimate babies lower than that of those born in hospital.7 About one in four women in Britain under 21 become pregnant,8 and most do not have abortions. Writing of America, Bolton suggests9 that many people believe that pregnancy is used deliberately by young women to obtain welfare benefits such as housing and income. An alternative sociological view is that having a child is a way of increasing self-esteem and of expressing control of the environment for individuals otherwise without power in society. But Bolton reminds us that pregnancy is usually unpremeditated, and how sense is made of it afterwards will very much reflect the future relationship between mother and child. Furstenberg’s detailed controlled study of young working-class mothers in New York cityl0 shows this. Instead of looking at causes, he describes the consequences and adaptations to pregnancy that are needed, such as postponement of marriage, rescheduling of education, and control of fertility. And he points out that crisis-oriented services often cease to offer services later on, when many of the gravest problems for the adolescent mother arise. In Britain this is the work of health visitors and social workers. The Child Health and Education Study, despite its complexity, is unable to show how far the current preventive services were already effective, but Taylor and colleagues propose more specialised support for teenage mothers. Yet Wadsworth,11 using longitudinal data from the 1958 birth cohort, showed that the ordinary information available to health visitors was a poor predictor: surveillance of the top 6 Department of Health and Social Security.Domiciliary midwifery and maternity bed needs. report of the sub-committee (chairman: Sir J. Peel). London: H.M. Stationery Office, 1970. 7 Campbell R,Davies IM, Macfarlane A. Perinatal mortality and place of delivery. Population Trends1982;28:9-12. Teper S Recent trends in teenage pregnancy in England and Wales.J Biosoc Sci 1975; 7: 141-52. 9. Bolton FG.The pregnant adolescent.London: Sage, 1980. 10. Furstenberg FF.Unplanned parenthood: the social consequences of teenage childbearing.London: Collier MacMillan, 1977 11. Wadsworth MEJ,Morris S.Assessing chances of hospital admission in pre-school children Arch Dis Childh1978; 53: 159-63
20% at risk identified only 27% of children who were admitted to hospital, and to achieve 60% pick-up half of the children would have needed special surveillance. Responding to deprivation by individual intervention is open to the charge of victim-blaming. 1The Black report took a more materialist view, emphasising the need to abolish child poverty and improve housing for the disadvantaged. Whether much more can be learnt from large multifactor cross-sectional studies is doubtful, and indeed it was decided not to have a national births survey in 1982. More detailed studies of parenting are required-the stereotype of "teenage mother" is not enough.
OPIATES OR OPIOIDS? As science advances, the terminology used to describe new discoveries must keep pace. One example is the distinction originally made between antibiotics-natural compounds manufactured by living microorganisms-and antibacterial chemotherapeutic agents synthesised by chemical means. Now that some antibiotics such as chloramphenicol are commercially synthesised de novo, and others are chemically manipulated to enhance their efficacy, the distinction has become blurred; the term antibiotic is now commonly used for all agents acting against microorganisms. We are now faced with another dilemma. The terms opiates and opioids are confusingly coming to be used interchangeably. The word opiates originally referred to alkaloids derived from the juice of the opium poppy, while opioids (literally "opium-like") was intended to refer to compounds other than opium alkaloids having pharmacological effects similar to those of opium. However, technology has reached the stage where any of these compounds can be synthesised, and this terminological distinction has outlived its taxonomic usefulness. The discovery of endogenous morphine-like compounds brought the worrying possibility that yet another title might be coined. In fact, it was: endorphins, which are, however, also known as endogenous opiates (or opioids) although the less ambiguous term opioid peptides is gaining ground. Receptor-binding studies demonstrate that all these effects are mediated by a single receptor with four or more main subtypes. It does seem illogical that both opiates and opioids should bind to the one receptor. It is high time we adopted a single terminology, and at least one textbook has already done so: Goodman and Gilman1 refers entirely to opioids; opiates receive no mention even in the index, and all opioids bind to the opioid receptor. The Advisory Council on the Misuse ofDrugs2 has opted for the terms opioid addiction and opioid addicts. But the latest issue of the British Medical Bulletin, which is devoted to opioid peptides and their receptors,3 illustrates that even the foremost researchers are inconsistent; the opiates and opiate receptors of one article become the opioids and opioid receptors of the next. The painless way to end the abuse and ambiguity is to opt for a uniform terminology, and opioids is the front-runner. We declare that opioids have won the day.
Ryan W. Blaming the victim. New York Vintage Press, 1976. Working Group. Inequalities in health: report (chairman: Sir Department of Health and Social Security, 1980
1. Goodman Gilman A, Goodman A, Gilman LS, eds. The pharmacological basis of therapeutics, 6th ed. New York: Macmillan, 1980. 2. Treatment and rehabilitation. Report of the Advisory Council on the Misuse of Drugs London: HM Stationery Office, 1982. 3. Hughes J, ed. Opioid peptides. Br Med Bull 1983; 39: 1-100.