Comparative Discrimination of BMI and Waist-Based Adiposity Indices for Doctor-Diagnosed Arthritis: A Cross-Sectional Analysis of NHANES 2001-2018 With External Validation in HRS 2022
Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders · 2026-10 · 19
DOI: 10.1177/11795441261507667 ↗Kurz gefasst
Querschnittstudie in der Zeitschrift Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders, online erschienen am 20.11.2026. Verfasst von Yong Hu und 3 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben.
Abstract (Originalsprache)
Background
Doctor-diagnosed arthritis in population surveys is a pragmatic but heterogeneous outcome with clear relevance to musculoskeletal clinical epidemiology. Obesity is a modifiable correlate, yet it remains unclear whether central adiposity indices provide materially greater discriminatory value than body mass index (BMI) once standard covariates are considered.
Methods
We analyzed adults aged ⩾20 years from NHANES 2001-2018 with measured anthropometrics and self-reported doctor-diagnosed arthritis. Survey-weighted logistic regression, area under the receiver operating characteristic curve (AUC), incremental ΔAUC, and Brier scores were used to compare BMI, waist circumference (WC), waist-to-height ratio (WHtR), a body shape index (ABSI), and body roundness index (BRI) beyond a covariate-only base model. Guideline-based cut-offs were also evaluated. External validation was conducted in the 2022 Health and Retirement Study (HRS) physical-measures subsample.
Results
The NHANES complete-case sample included 30 652 participants, and the weighted prevalence of doctor-diagnosed arthritis was 12.4%. Fully adjusted per-standard-deviation associations were similar for BMI, WC, WHtR, and BRI, whereas ABSI was near null. Discrimination was high but nearly identical across BMI and waist-based indices in NHANES: the base-model AUC was 0.835, increasing to 0.841 with BMI and to 0.840 with WC, WHtR, or BRI; ABSI provided essentially no gain. In HRS (n = 4291), the base-model AUC was 0.694, and adding BMI, WC, WHtR, or BRI increased the AUC modestly to 0.707-0.708, whereas ABSI increased it by only 0.001. Brier scores showed concordant modest improvements in both cohorts.
Conclusion
WC, WHtR, and BRI were associated with prevalent doctor-diagnosed arthritis but did not materially outperform BMI for discrimination once standard covariates were included. External validation reproduced the relative ordering of indices and the minimal magnitude of incremental gains, indicating limited practical benefit from substituting BMI with alternative waist-based indices for the pragmatic discrimination of prevalent doctor-diagnosed arthritis in population-based data.
Zitieren
APA
Hu, Y., Wang, M., Zhai, Y., & Pan, C. (2026). Comparative Discrimination of BMI and Waist-Based Adiposity Indices for Doctor-Diagnosed Arthritis: A Cross-Sectional Analysis of NHANES 2001-2018 With External Validation in HRS 2022. Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders, 19. https://doi.org/10.1177/11795441261507667
Vancouver
Hu Y, Wang M, Zhai Y, Pan C. Comparative Discrimination of BMI and Waist-Based Adiposity Indices for Doctor-Diagnosed Arthritis: A Cross-Sectional Analysis of NHANES 2001-2018 With External Validation in HRS 2022. Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders. 2026;19. doi:10.1177/11795441261507667
BibTeX
@article{Hu2026,
author = {Hu, Yong and Wang, Min-Cong and Zhai, Yan-Zhen and Pan, Cheng-Long},
title = {Comparative Discrimination of BMI and Waist-Based Adiposity Indices for Doctor-Diagnosed Arthritis: A Cross-Sectional Analysis of NHANES 2001-2018 With External Validation in HRS 2022},
journal = {Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders},
year = {2026},
volume = {19},
doi = {10.1177/11795441261507667}
}
EUVIMED