Collecting biomarker data in longitudinal surveys
Jul 26, 15:20
This paper will provide an overview and summary of methodological issues in collecting biomarker data in longitudinal surveys. To motivate the paper, we will begin with a discussion of why it is particularly valuable to collect biomarker data in longitudinal surveys. We will identify the unique scientific value of the longitudinal context, characterising research types that benefit from a) prior longitudinal measurement of social/health outcomes, b) subsequent longitudinal measurement of social/health outcomes, c) repeat collection of biomarkers. The main part of the paper will then focus on methodological issues in biomarker collection where the longitudinal survey context makes a difference. Examples may include sample selection for biomarker collection based on data from previous waves; taking advantage of multiple visits to each sample person/household; obtaining consent/participation (where both the methods and the outcomes could differ from cross-sectional surveys, due to prior information, panel loyalty, etc); possible panel conditioning effects if the results of medical tests are fed back to participants; etc. We will include some empirical evidence from UK longitudinal surveys in addition to summarising findings from the literature. The paper will conclude with a brief discussion of likely future methodological advances in biomarker collection in longitudinal surveys and of relevant knowledge gaps.