Neighborhood disadvantage and body mass index: A study of residential relocation

Jerome N. Rachele, Anne M. Kavanagh, Wendy J. Brown, Aislinn M. Healy, Gavin Turrell

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)


Natural experiments, such as longitudinal observational studies that follow-up residents as they relocate, provide a strong basis to infer causation between the neighborhood environment and health. In this study, we examined whether changes in the level of neighborhood disadvantage were associated with changes in body mass index (BMI) after residential relocation. This analysis included data from 928 residents who relocated between 2007 and 2013, across 4 waves of the How Areas in Brisbane Influence Health and Activity (HABITAT) study in Brisbane, Australia. Neighborhood disadvantage was measured using a census-derived composite index. For individual-level data, participants self reported their height, weight, education, occupation, and household income. Data were analyzed using multilevel, hybrid linear models. Women residing in less disadvantaged neighborhoods had a lower BMI, but there was no association among men. Neighborhood disadvantage was not associated with within-individual changes in BMI among men or women when moving to a new neighbourhood. Despite a growing body of literature suggesting an association between neighborhood disadvantage and BMI, we found this association may not be causal among middle-aged and older adults. Observing associations between neighborhood socioeconomic disadvantage and BMI over the life course, including the impact of residential relocation at younger ages, remains a priority for future research.

Original languageEnglish
Pages (from-to)1696-1703
Number of pages8
JournalAmerican Journal of Epidemiology
Issue number8
Publication statusPublished - 1 Aug 2018
Externally publishedYes


Dive into the research topics of 'Neighborhood disadvantage and body mass index: A study of residential relocation'. Together they form a unique fingerprint.

Cite this