Abstract
AIMS: Common adiposity indices that are used to assess obesity include body mass index (BMI), waist circumference (WC) and waist-to-height ratio (WHtR). We aim to investigate change rates of the three indices in British middle-aged adults and compare effects of different adiposity indices' changes on incident type 2 diabetes. MATERIALS AND METHODS: Three repeated measures of BMI, WC and WHtR were collected over a decade (1991-2004) in 5666 participants without diabetes from the Whitehall II cohort study of British civil servants. Linear mixed-effects models were used to estimate standardised individual change rates of the three indices, and participants were then followed up for incident diabetes until 2023. We examined the prospective associations between change rates of the three indices and diabetes by Cox regression analysis. RESULTS: The mean change rates for BMI, WC and WHtR were 0.04, 0.06 and 0.07 standard deviation (SD)/year, respectively. There were 633 incident diabetes cases after a median follow-up of 17.6 years. For every 1-SD increase per decade, WC change (HR: 3.01, 95% CI: 2.15-4.22) was associated with a higher diabetes risk than changes in WHtR (HR: 2.67, 95% CI: 2.00-3.55), and BMI (HR: 1.98, 95% CI: 1.53-2.56). No significant interactions were found between the change rates and either sex or age. CONCLUSIONS: The changes in WC and WHtR over a decade were associated with a higher risk of developing diabetes than the change in BMI in this British adult population. Routine monitoring of WC in general practice may provide more benefit than BMI management alone in preventing diabetes.