Association between six anthropometric indices and incident arthritis: prospective findings from the English...

Arthritis, predominantly osteoarthritis in older adults, imposes a substantial burden on quality of life and healthcare systems. Obesity is a well-established modifiable risk factor; however, traditional body mass index (BMI) has limitations in capturing visceral adiposity and central obesity. Novel anthropometric indices, including the Body Roundness Index (BRI), A Body Shape Index (ABSI), weight-adjusted waist index (WWI), waist-to-height ratio (WHtR), waist circumference (WC), and BMI, may offer additional insights. This study examined their associations with incident arthritis in a prospective cohort of elderly individuals.

We analyzed data from 4,112 participants without arthritis from the English Longitudinal Study of Ageing (ELSA) Wave 2 (2004-2005) and followed them until Wave 9 (2018-2019). Height, weight, and WC were objectively measured by trained nurses. Anthropometric indices were calculated using standard formulas and standardized to per-standard deviation (SD) increments and tertiles (Q1: lowest; Q2: medium; Q3: highest). Incident arthritis was defined as the first self-reported physician-diagnosed arthritis during the follow-up period. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs), after adjusting for age, sex, ethnicity, education, marital status, smoking status, drinking status, hypertension, diabetes, stroke, high cholesterol levels, congestive heart failure, psychological problems, the Control, Autonomy, Self-realization and Pleasure-19 (CASP-19) scale scores, and physical activity. Subgroup analyses and interaction tests were performed. Kaplan-Meier curves and restricted cubic splines (RCS) were used to assess cumulative incidence and dose-response relationships.

During a median follow-up of 14.2 years, 944 incident arthritis cases occurred (23%). After fully adjusting for all covariates, per-SD increases in anthropometric indices such as the BRI (HR: 1.17, 95% CI: 1.10-1.25), WHtR (HR: 1.18, 95% CI: 1.10-1.27), WC (HR: 1.21, 95% CI: 1.13-1.31), BMI (HR: 1.20, 95% CI: 1.12-1.28), and WWI (HR: 1.09, 95% CI: 1.01-1.18) were significantly associated with a higher incidence of arthritis. However, the ABSI showed no association (HR: 1.01, 95% CI: 0.92-1.10). RCS analyses confirmed that there are predominantly linear positive dose-response relationships for the BRI, WHtR, WC, and BMI (all p values for non-linearity > 0.05). Tertile analyses revealed clear dose-response trends for the majority of indices, with participants in the high tertile having significantly greater risk of incident arthritis compared to those in the low tertile.

Anthropometric indices, excluding the ABSI, were associated with increased incident arthritis risk in middle-aged and older adults.