NHANES (CDC), derived Conicity Index percentiles
About this reference
No published study provides sex- and age-stratified Conicity Index percentile tables for US adults. NHANES measures waist circumference (BMXWAIST), standing height (BMXHT), and weight (BMXWT) on the same individuals in the body measures examination. FitnessNorms derived Conicity Index percentiles by downloading public XPT microdata for three NHANES blocks, computing the Valdez 1991 index per participant, and taking weighted empirical quantiles (P5, P25, P50, P75, P95) for each sex and 5-year age bracket from 20-24 through 80+. Three cycles were pooled to improve tail stability: 2015-2016 (BMX_I / DEMO_I, WTMEC2YR), 2017-March 2020 (P_BMX / P_DEMO, WTMECPRP, the CDC pre-pandemic weight), and 2021-2023 (BMX_L / DEMO_L, WTMEC2YR). Pooling divided each cycle's MEC weight by 3. After the waist, height, and weight filters, the post-filter weight totals are close across cycles (2015-2016: 221.8M, 2017-2020: 229.7M, 2021-2023: 232.8M; relative spread 4.8%), so the divide-by-3 rule gives approximately equal cycle contribution. The divide-by-N rule is a FitnessNorms design choice, not a normalisation that would force equal contribution if cycle weight totals diverged, and is not official CDC guidance for this combination of releases. 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. 8 of 130 sex/age/quantile cells crossed the 0.02 red-flag threshold (largest difference 0.033); each is listed with a reason in the derivation script's ACCEPTED_DEVIATIONS allow-list, and the script exits non-zero (refusing to overwrite the published table) on any unlisted breach. Pooled is retained as primary because it gives tighter tails than halving sample size to the single 2021-2023 cycle. Weighted quantiles use linear interpolation on the empirical cumulative distribution (analogue of Type 7). Survey-design standard errors were not computed (this is an internal derivation, not a peer-reviewed publication).
Known limitations
- US population only. Percentiles reflect the US adult population, which had an obesity rate of 42.4% by BMI (Hales et al. 2020). We did not find a directly comparable non-US Conicity Index percentile table, so we cannot quantify how these norms compare internationally; given the high US obesity prevalence, values here may run higher than in leaner populations, but we present that as a hypothesis rather than a sourced fact.
- This is an internal derivation by FitnessNorms from CDC public microdata, not a peer-reviewed publication. Weighted empirical quantiles are used; survey-design-correct standard errors are not reported.
- Equal-cycle pooling (dividing each cycle's weight by 3) is a FitnessNorms design choice, matching the WHtR/WHR/BRI references. CDC does not prescribe a pooling method for this combination of releases, particularly across the pandemic break. The divide-by-3 rule approximately equalizes cycle contribution only because the measured post-filter cycle weight totals (221.8M, 229.7M, 232.8M; relative spread 4.8%) are close; it would not force equality if they diverged.
- 8 of 130 sex/age/quantile cells differ by more than the 0.02 internal threshold between the pooled and 2021-2023-alone tables (max 0.033); these cells are enumerated in the derivation script's ACCEPTED_DEVIATIONS allow-list. The discontinuity check exits non-zero on any breach not in that list, so future regenerations cannot silently ship shifted values. The accepted cells are: male 45-49 P75 (pooled 1.362 vs 1.395), male 45-49 P95 (1.457 vs 1.481), male 50-54 P5 (1.206 vs 1.227), female 20-24 P75 (1.290 vs 1.269), female 35-39 P25 (1.223 vs 1.202), female 70-74 P5 (1.234 vs 1.213), female 75-79 P95 (1.499 vs 1.466), and female 80+ P5 (1.227 vs 1.207). They are scattered across both sexes and four quantiles with mixed sign, consistent with sampling noise rather than a systematic pandemic-cycle shift.
- No pregnancy exclusion is applied; pregnant participants are included as measured (consistent with the WHtR, WHR, and BRI derivations on this site).
- RIDAGEYR is top-coded at 80 in NHANES public data; the 80+ bracket includes all participants aged 80 and over.
- Cross-sectional design: values reflect population averages at the time of survey, not individual trajectories.