Body Roundness Index: NHANES Microdata Derivation
NHANES measures waist circumference (BMXWAIST) and standing height (BMXHT), the two inputs to the Body Roundness Index introduced by Thomas et al. (2013). FitnessNorms calculated the published percentiles directly from CDC public microdata.
- Formula:
BRI = 364.2 − 365.5 × √(1 − (WC/(2π))² / (0.5×Ht)²), with waist circumference and height in metres. Lower values correspond to a narrower waist relative to height; higher values correspond to a rounder profile. - Published data: NHANES 2021-2023 (
BMX_L / DEMO_L). The analysis includes adults aged 20 and over with complete waist and height measurements and excludes participants recorded as pregnant (RIDEXPRG = 1), leaving n = 5,718. - Weights: MEC examination weights (
WTMEC2YR) were applied without a pooling adjustment because the published table uses one release. - Sensitivity analysis: we also calculated a pool spanning 2015-2016, 2017-March 2020, and 2021-2023. Contributions were scaled by survey duration: 2/7.2, 3.2/7.2, and 2/7.2. CDC documents the 3.2-year duration and duration-based adjustment for analyses involving the pre-pandemic file; extending that principle across the post-pandemic release is a FitnessNorms sensitivity-analysis choice, not CDC guidance for this exact combination.
- Publication decision: 71 of 130 sex, age, and percentile cells differed by more than 0.15 BRI units between the recent release and the pool. The median absolute difference was 0.160. The largest difference was male ages 45-49 P95: 11.52 in 2021-2023 versus 9.62 in the pool. The comparison cannot distinguish temporal change from sampling variation or nonlinear amplification, so the recent release is published and the pool is retained only as a sensitivity artifact.
- Quantiles: weighted empirical P5, P25, P50, P75, and P95 values were calculated separately for each sex and five-year age bracket from 20-24 through 80+. Linear interpolation was applied to the weighted cumulative distribution. Survey-design standard errors were not calculated.
- Formula guard: rows with a negative square-root term would be excluded. No eligible row failed that check.
Interpretation and limits
These are population percentiles, not diagnostic thresholds. Published BRI cut-offs vary by population and outcome, and no single clinical boundary is established. The public NHANES age variable is top-coded at 80, so the oldest bracket combines everyone recorded as age 80 or older.
This is an internal FitnessNorms derivation rather than a peer-reviewed percentile study. See the reference page for a compact provenance and limitations record.