Waist-to-Hip Ratio — NHANES Microdata Derivation
No published paper provides sex- and age-stratified WHR percentile tables for US adults. The CDC NHANES body measures examination collects waist circumference (BMXWAIST) and hip circumference (BMXHIP), but hip circumference was not measured in the 2011–2016 cycles. Hip measurement was reinstated starting with the 2017–March 2020 release, so the WHR pool starts in 2017 (in contrast to waist-to-height ratio, which uses three pooled cycles back to 2015). Percentiles were derived as follows:
- Data: Two NHANES releases were used: 2017–March 2020 (
P_BMX / P_DEMO) and 2021–2023 (BMX_L / DEMO_L). Adults aged 20 and over with complete waist and hip measurements were included after filtering biologically impossible values (n ≈ 13,727). - Weights: MEC examination weights were applied:
WTMECPRPfor the 2017–March 2020 release (the CDC-prescribed pre-pandemic weight) andWTMEC2YRfor 2021–2023. Each cycle's weight was divided by 2 before pooling. Post-filter cycle weight totals turn out to be within 0.5% of each other (229.5M vs 230.5M), so the two cycles contribute approximately equally to the pooled estimate; the divide-by-N rule is not a normalisation that would force equality if cycle totals diverged. No CDC-prescribed pooling method exists for this combination of releases. - Pandemic discontinuity: The 2021–2023-alone table was compared against the pooled table at every published quantile (P5, P25, P50, P75, P95), not only at the median. One cell narrowly crossed the 0.02 red-flag threshold: female 20–24 P75 differs by 0.022 (0.866 single-cycle vs 0.888 pooled). All other 129 (sex × bracket × quantile) cells stayed below the threshold (max 0.018). Pooled is retained as primary because halving every cell's sample size to fix one tail in one cell would degrade the other 25 brackets; the deviation is documented on the reference page. The 2021–2023-only table is kept as an internal sensitivity check.
- Quantiles: Weighted empirical quantiles (P5, P25, P50, P75, P95) were computed for each sex × 5-year age bracket using linear interpolation on the weighted cumulative distribution. Survey-design standard errors were not computed.
- Computation: WHR =
BMXWAIST÷BMXHIP. Both measurements are in centimetres, so the ratio is dimensionless.
This is an internally derived dataset, not a peer-reviewed publication. The derivation method and its limitations are summarised on the reference page. See also waist-to-height ratio methodology for the sister derivation.