No abstract

Britain’s oldest birth cohort study, the MRC National Survey of Health and Development (NSHD) provides data to explore life time influences on ageing. The latest data collection was undertaken between 2006 and 2011 when study members were aged 60–64 and consisted of postal and pre-assessment questionnaires to eligible study members, followed by invitation to attend one of six clinical research facilities (CRFs) across the UK for clinical assessments, and dietary diaries and activity monitors in the days following the CRF visit. The option of a home visit for clinical assessments was provided if the study member refused or was unable to attend the CRF. We examined response and attrition, here describing rates overall and for postal and clinical assessment modes of data collection, identifying socioeconomic and health-related predictors of response, and assessing the continued representativeness of the sample. In total, 2,661 (84 % of the target sample) responded. Lower educational attainment, lower childhood cognition and lifelong smoking independently predicted lower likelihood of both overall response and CRF cooperation. At 53 years, not owning one’s home and not being married predicted lower likelihood of overall response whereas manual social class and obesity predicted lower likelihood of CRF cooperation. Providing for collection of biomedical data in the home and use of assessment instruments and modes to retain study members with lower education attainment, lower cognition and poorer health behaviours should be priorities for helping reduce attrition amongst vulnerable ageing study members.

Objective: To test the validity of age at menarche self-reported in adulthood and examine whether socioeconomic position, education, experience of gynaecological events and psychological symptoms influence the accuracy of recall. Design: Prospective birth cohort study. Setting: England, Scotland and Wales. Participants: 1050 women from the Medical Research Council National Survey of Health and Development, with two measures of age at menarche, one recorded in adolescence and the other selfreported at age 48 years. Results: By calculating the limits of agreement, k statistic and Pearson's correlation coefficients (r), we found that the validity of age at menarche self-reported in middle age compared with that recorded in adolescence was moderate (k = 0.35, r = 0.66, n = 1050). Validity was improved by categorising age at menarche into three groups: early, normal and late (k = 0.43). Agreement was influenced by educational level and having had a stillbirth or miscarriage. Conclusions: The level of validity shown in this study throws some doubt on whether it is justifiable to use age at menarche self-reported in middle age. It is likely to introduce error and bias, and researchers should be aware of these limitations and use such measures with caution.