Supplementary Material for: The association between the obesogenic environment and BMI independent of familial factors: a cross-sectional twin study

Objective To investigate the association between the obesogenic environment and body mass index (BMI), independent of familial factors shared by twins (e.g., genetic and early life environmental factors). Methods This cross-sectional twin study included 2764 adults (77% female, 72% monozygotic [MZ], mean age 40 [SD 14]) from the Dutch Twin Registry. The Obesogenic Built Environment CharacterisTics (OBCT) index was used as neighbourhood “obesogenic” exposure, integrating food‑environment healthiness, walkability, drivability, and sports facilities within a 1,000 m Euclidean buffer around the residential address. Univariate twin models established the heritability estimates of the OBCT-index and BMI. Full-informed maximum likelihood linear regressions determined individual and pairwise difference (i.e., within-twin-pair differences in outcome and exposure) associations between the OBCT-index and BMI. Results The heritability of the OBCT-index was 9% and common environmental factors explained 27% of its variation, whereas the BMI heritability was 75%. Living in a 10% higher OBCT-index environment was associated with a mean [95% CI] 0.15 [0.07, 0.24] kg/m2 higher BMI in all twin individuals. However, this association was non-significant in the MZ twin pairwise difference analyses. Conclusion This study found a small association between a more obesogenic environment and a higher BMI, yet the association was explained by familial confounding. Future research in larger, more environmentally diverse international twin cohorts is needed to elucidate gene-environment interactions.