Skip to main content

Table 1 Potential confounders included in models of total effects of neighbourhood environment attributes on MetS status and membership to a metabolic profile

From: The neighbourhood environment and profiles of the metabolic syndrome

Environmental attribute

Potential confounders

Population density

Person-level confounders: Sex, Age, Educational attainment, Living arrangements, Employment status, Neighbourhood self-selection variables, Smoking history

Environmental confounders: None

Commercial land use (%)

Person-level confounders: Same as above

Environmental confounders: Population density

Parkland (%)

Person-level confounders: Same as above

Environmental confounders: Population density, Commercial land use

Blue space (%)

Person-level confounders: Same as above

Environmental confounders: None

Land use mix (other)

Person-level confounders: Same as above

Environmental confounders: Population density

Street intersection density

Person-level confounders: Same as above

Environmental confounders: Population density

Air pollution (NO2 and PM2.5)

Person-level confounders: Same as above + Household income

Environmental confounders: Population density, Street intersection density, Commercial land use, Land use mix (other), Parkland, Area SES

Area SES (IRSAD)

Person-level confounders: Sex, Age, Educational attainment, Living arrangements, Employment status, Household income, Neighbourhood self-selection variables

Environmental confounders: Parkland, Blue space

  1. Abbreviations: MetS the metabolic syndrome, SES socio-economic status, IRSAD Index of Relative Socioeconomic Advantage and Disadvantage, NO2 nitrogen dioxide, PM2.5 particulate matter < 2.5 μm
  2. Land use mix (other) represents land use excluding commercial land use, parkland and blue space Minimal sufficient adjustment sets based on the Directed Acyclic Graph (DAG)