Adult leg length is particularly sensitive to environmental factors and diet in early childhood because that is the period of most rapid leg growth. Trunk growth is faster than leg growth after infancy and before puberty, and may be associated with the effects of serious illness and parental separation because of the child's growing sensitivity to stressful circumstances, as well as the result of the biological effects of illness.

Study objective-The aim was to describe rates of loss and assessment of representativeness during 43 In this paper representativeness of the original sample and of the contacted study population during childhood and adult life is assessed by comparison with census data from the population of the same or similar age. Rates of losses through death and emigration are compared with appropriate national data, and the extent of selective loss in particular subsections of the population described. Results THE CURRENT EXTENT OF CONTACT

No abstract

There has been recent interest in the possibility that breast cancer can have a prenatal origin (Trichopoulos, 1990). Various studies have reported on the relationship between birthweight, taken as a marker of prenatal environment, and breast cancer, but with differing results. One study (Ekbom et al, 1992; reported no association between birthweight and breast cancer, two (Le Marchand et al, 1988;Sanderson et al, 1996) an inverse relationship, and one (Michels et al, 1996) a positive association. The latter was a case-control study nested within the US Nurses' Cohort, which found that the odds of breast cancer for women who weighed 4 kg or more at birth was twice that of women who weighed less than 2.5 kg (Michels et al, 1996). The relationship was strongest for women aged 50 or younger.Various factors might explain the inconsistency of these results. In some studies the information on birthweight was based on recall by the women themselves or their mothers (Michels et al, 1996;Sanderson et al, 1996), in most no account was taken of adult-life risk factors for breast cancer (Ekbom et al, 1992;Le Marchand et al, 1988), and no account was taken in any of childhood and pubertal factors.The present study examines the relationship between birthweight and breast cancer in a UK national cohort of 2221 women who have been followed since their birth in 1946, and for whom data on birthweight, markers of childhood growth and adult-life risk factors for breast cancer have been recorded. The Medical Research Council National Survey of Health and Development (NSHD) is a socially stratified birth cohort of 2548 women and 2814 men born in the UK during the week 3-9 March 1946 (Wadsworth, 1991;Wadsworth and Kuh, 1997). The cohort comprised single, legitimate births to wives of all non-manual and agricultural workers and to 1 in 4 wives of manual workers. There have been 19 follow-up contacts with the cohort members between their birth and age 43 years, most by home interviews. The sample interviewed at age 43 years were, in most respects, representative of the native population of that age (Wadsworth et al, 1992). Since 1993, when the cohort members were 48 years of age, a postal health questionnaire has been sent annually to all women in the study with whom there was still direct contact. At these separate contacts, breast cancer diagnosis was self-reported and recorded. In 1971, when the National Health Service Central Register (NHSCR) started to record cancers occurring in the population of the UK, all cohort members (including those with whom there was no longer direct contact) were 'flagged' at the NHSCR. This provided notification of registrations of cancer, death and emigration for the cohort.Information on birthweight was extracted from the birth records of the cohort members and categorized into four groups: < 3.000 kg, 3.000-3.499 kg, 3.500-3.999 kg, ≥ 4.000 kg, in accord with previous studies (Ekbom et al, 1997;Michels et al, 1996). Data on maternal age and birth order were collected at the original h...