308234
A propensity matched analysis of population movement implicating area contributions to increased cardiometabolic risk over time
Longitudinal cohorts have increasingly sought to account for population movement and its impact on health. In research on place and health, the focus has centred on the self-selection of residents into local-areas, and potential biases for geospatial epidemiological analyses. However, of equal importance are the area-level drivers of population movement and, thus corresponding links with health outcomes. Informed by geographic mobility theories, this study aimed to implicate area-level influences by accounting for individual factors using propensity matched pairs of ‘movers’ and ‘non-movers’ to assess change in cardiometabolic risk across two time points.
Method:
Data were utilised from Wave 1 (n=4041; 2000-03) and Wave 2 (n=3507; 2005-06) of the North West Adelaide Health Study. The outcome measure, the count of clinically measured cardiometabolic risk factors, and socio-economic and demographic information, were collected from urban-dwelling adult participants linked by residential address using a geographic information system. Matched propensity scores were estimated by a logistic regression model in which residential mobility was regressed on: a change in marital status, work status and household income, as well as gender, age cohort and housing tenure. Comparison between time 2 vs. time 1 change between the ‘mover’ and ‘non-mover’ matches for cardiometabolic risk scores (count of six risk measures) was evaluated by paired t-test.
Results:
Four hundred and thirteen ‘movers’ were pair-matched with ‘non-movers’ for individual-level predictors of residential movement. ‘Non-movers’ had an increase in the count of elevated cardiometabolic risk factors (mean 0.04) than ‘mover’ counterparts (mean -0.11).
Discussion:
'Non-movers' had a greater increase in risk of cardiometabolic disease. In so far as this analysis accounted for individual-level factors that contribute to re-location, area-level influences are potentially implicated for advancing understandings of population movement.
Learning Areas:
EpidemiologySocial and behavioral sciences
Learning Objectives:
Demonstrate propensity matched analysis of population movement for assessment of change in cardiometabolic risk across two time points.
Keyword(s): Geographic Information Systems (GIS), Epidemiology
Qualified on the content I am responsible for because: Natasha is a Post-Doctoral Research Fellow within the Spatial Epidemiology and Evaluation Research Group, Sansom Institute for Health Research at the University of South Australia. She is a Social Geographer with a keen interest in exploring the social and spatial inequalities of cardiovascular risk behaviours and well-being. Her work experience spans both the Health and Social Sciences, applying population approaches to investigate how the social and built environment enables and promotes cardiometabolic health and well-being.
Any relevant financial relationships? No
I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.