Obesity prevalence varies markedly by definition: multiple cross-sectional and prospective studies

Multiple operational definitions of obesity have been proposed. We compared obesity prevalence and incidence according to World Health Organization (WHO), National Institute of Health (NIH), and European Association for the Study of Obesity (EASO) using the baseline survey and subsequent follow-ups of a population-based cohort (CoLaus|PsyCoLaus), in Lausanne, Switzerland. Sample included adults aged 35 to 75 years (n = 6733 at baseline). Obesity was defined by WHO (body mass index [BMI]), NIH (BMI and waist circumference), and EASO (BMI/waist-to-height ratio plus obesity-related complications such as hypertension, type 2 diabetes, cardiovascular disease, or chronic kidney disease). The joint prevalence of overweight and obesity did not change, but there was a twofold increase in obesity prevalence between WHO and NIH/EASO (average obesity prevalence combining all follow-ups: 17.4%, 33.9% and 37.5%, respectively), at the expense of the overweight category (average prevalence of 38.7%, 22.2% and 18.6%, respectively), with a particularly marked shift at older ages: obesity prevalence in age group +75 years old for follow-up 2 was 19.2%, 48.9% and 55.9% for WHO, NIH and EASO, respectively. Regarding the incidence analysis (n = 2334, 34.7% of the baseline cohort), after a median follow-up of 14.5 years, the incidence of obesity among participants without obesity at baseline was 5.8% (WHO), 16.8% (NIH), and 20.6% (EASO). For EASO definition, final prevalence of obesity (20.6%) exceeded overweight (18.3%). We conclude that broader definitions, notably EASO, identify more individuals at potential risk but substantially expand the obesity category, with implications for case classification, clinical management, and resource allocation.