PATHWAYS - Biosocial Influences on Health
Principal Investigator: Professor Emily Grundy
PATHWAYS aimed to identify pathways that link socio-demographic circumstances and biological disadvantage to adult health, and parental family and socio-economic circumstances to infant mortality, with a particular emphasis on the mediating factors that lie on these pathways.
PATHWAYS aimed to
- promote cross fertilisation of research methodology used in the social and biomedical sciences with a view to enhancing the capacity of social scientists to investigate hypothesised causal pathways between biological and social factors while fully exploiting new biomedical information, as it becomes available in a number of important UK data sources;
- adapt, develop and facilitate the rigorous use of methodology for the identification of causal pathways, to relate their underlying assumptions to the subject-matter of interest, and to jointly address the expected limitations of observational data, namely those arising from measurement error and missing data;
- develop and provide training in all methods mentioned above, including the use of biomarkers, latent indicators to capture health related measures affected by measurement error and principled methods for handling missing data.
PATHWAYS research programme consisted of projects that explore pathways to and from particular fertility histories and their implications for later life health; pathways to and from marital status trajectories and health in mid life; pathways from parental socio-economic and demographic circumstance to birth weight of infants and infant mortality; and use of genes as instrumental variables in causal analysis.
In addition to research projects PATHWAYS ran a training programme.
Impacts (March 2013)
- PATHWAYS have derived a measure of allostatic load (a composite measure based on biomarkers conceptualised as an indicator of accumulated stress) using data in the English Longitudinal Study of Ageing. A working paper 'Allostatic load - a challenge to measure multisystem physiological dysregulation' is available online.
- PATHWAYS have developed an approach which allows the testing of a non-linear causal effect in Mendelian randomisation analyses using a binary genetic instrument and the formal comparison of the structural equation modelling and modern causal inference approaches to mediation.
- Work by PATHWAYS, comparing the ethnic breakdown of the Clinical Practice Research Databases (CPRD) to that of the most recent 2011 Census, has shown that the CPRD population is representative of the UK population in terms of ethnicity, age, gender, and geography. Therefore, population-based studies that utilize this source of routinely collected ethnicity data are generalizable to the wider UK population. The working paper Availability and use of UK based ethnicity data for health research is available online.
For further information please see PATHWAYS website. PATHWAYS was based at the London School of Hygiene and Tropical Medicine, University of London and University of Cambridge.