Skip to main content


Table 1 Loss of analytic sample due to application of inclusion criteria

From: Local descriptive body weight and dietary norms, food availability, and 10-year change in glycosylated haemoglobin in an Australian population-based biomedical cohort

  Number Reason for reduced numbers
NWAHS Wave 1 4056 -
Geocoded (Baseline) 4041 15 participants with invalid residential addresses
Residing in urban area (Baseline) 3887 154 participants outside of urban area
Did not change residential address between Wave 1 and Wave 2 3322 565 participants moved between Wave 1 and Wave 2
State Suburbs with ≥ 10 participants 3173 149 participants resided in suburbs with ≤ 9 other participants
CVD/diabetes free at baseline 2621 552 participants reported CVD/diabetes at baseline
At least one set of HbA1c and individual-level covariates data 2582 39 participants did not have at least one set of complete HbA1c and individual-level covariates data
Contextual features data 2213 369 participants did not have fast-food outlet or healthful food resource availability data at Wave 2
Compositional norms data
 Prevalence of overweight/obesity (BMI ≥ 25 kg/m2) 1890 323 participants lacked local descriptive overweight/obesity data
 Prevalence of insufficient fruit intake (<2 serves daily) 1945 268 participants lacked local descriptive fruit intake data