Data Collected: ALSPAC have collected self-reported data about the home (e.g. damp, new furnishings and carpets, type of cooking fuel used), modes of transport; neighbourhood perceptions, and parental occupational history. ALSPAC fieldworkers have collected air pollution data from sub-groups of participants using sensors and assayed participant biosamples (e.g. venous blood caboxyhaemoglobin and methaemoglobin). Biological measures of lead, calcium and mercury have been measured from antenatal blood sample, cord blood and in child blood (at age 30m). Child activity has been measured at various time points using sensors. Participant residential and school attended address records have been geocoded to property and postcode level. The ALGAE software has been developed to assign spatial data to participants and across periods of participant’s life (e.g. pregnancy trimesters) (Figure 1). Meaning that study collected data have been (or have the potential to be) augmented by any specially-indexed routine records (e.g. PM2.5, PM10, NO2 air pollution; green space; neighbourhood socio-demographic factors; distance to services or facilities). Examples of linkage to date include: calculating downwind proximity to powerlines; calculating the density of fast-food outlets in participants’ neighbourhoods; modelling PM2.5 and PM10 exposures; and, mapping participants to radon air pollution estimates.
Data Resource Use: The physical and social environment records have been used to investigate associations between: 1) domestic, workplace and outdoor exposures and health phenotypic and ‘omic outcomes, using both directly recorded exposure information and information derived from assayed biosamples; 2) residential location and migration on health and social outcomes; 3) movement between places on health and behavioural outcomes. These records have also been used to inform wider research: informing purposeful sampling strategies and to test for confounding relating to social position. Furthermore, ALSPAC has been a test-bed for developing new data collection and analysis methodologies.
Reasons to be cautious: ALSPAC is a regionally sampled cohort, the resulting participant clustering provides advantages in terms of testing locality-based effects and enabling multi-level modelling but is not well suited to studying issues relating to national variation. The quality and completeness of the ALSPAC address database (and thus geocoded resource) is impacted by study attrition: the rates of which are differentiated by health and social status.
Collaboration and data access: Individual level-data are available upon request from the ALSPAC Executive (www.bristol.ac.uk/alspac). To protect participant confidentiality, the use of some environmental data are subject to disclosure controls or are only handled by the ALSPAC data managers (who facilitate the investigation).
- environmental epidemiology
- environmental science
- exposure science
- Data Linkage
- longitudinal cohort study
- research platform