Background: some cohort studies of ageing and health supplement questionnaire-based surveys with in-home measurements of biological parameters and others have required respondents to attend assessment centres. Centre-based assessments facilitate detailed measurements and novel technologies, but may differentially influence participation. The aim of this paper is to compare the characteristics of participants who attended a centre with those who chose a home assessment and those who did not have a health assessment.
Methods: trained field workers administered a computer-assisted personal interview (CAPI) to a random sample of community-dwelling people aged 50 and over in the participants' homes. All questionnaire respondents were invited to attend an assessment centre for a comprehensive physical assessment. Participants who refused or were unable to attend a centre were offered a home assessment.
Results: of the 291 participants who completed the CAPI, 176 had a health assessment: 138 in an assessment centre and 38 in their own home. The centre, home and no visit respondents differed in demographic characteristics, behavioural factors, physical functioning and health. Lower socio-economic status, physical inactivity and current smoking were the most robust predictors of non-participation in the health assessment. Home respondents had the highest levels of physical disability and were much weaker (grip strength) and slower (walking speed) than centre respondents.
Conclusion: home and centre physical assessments are required to avoid systematically over-representing healthier and wealthier respondents.