Implicit and explicit fat bias among adolescents from two US populations varying by obesity prevalence.

Abstract

Research shows that explicit (reported) and implicit (unconscious) bias are distinct constructs varying by socioecological context. Implicit bias better predicts poor health outcomes related to chronic psychosocial stress. Variation in fat bias was identified in two populations of adolescents in Indiana counties with relatively high (Lawrence) and low (Monroe) obesity prevalence. Adolescents (n = 185) aged 14-18 years were recruited October 2017-2018. Explicit fat bias was measured using the Attitudes Towards Obese Persons (ATOP) scale. Implicit fat bias was measured using the obesity attitude implicit association test (IAT). Both samples scored high on the ATOP, indicating more positive reported attitudes towards persons with obesity, and scores did not differ between counties (P > .05) after adjusting for age, sex, ethnicity, and BMI-for-age percentile. Both samples demonstrated unconscious anti-fat bias, with Monroe exhibiting significantly higher IAT scores (ie, greater anti-fat bias) than Lawrence after controlling for confounders (P = .039). Differences between explicit and implicit fat-bias were observed. Both samples reported positive attitudes towards obesity but also demonstrated unconscious anti-fat bias. Unconscious anti-fat bias was significantly higher in the county with lower prevalence of obesity, suggesting epidemiological context may influence implicit fat bias. Higher obesity prevalence may reduce unconscious anti-fat bias in a population or, alternatively, lower obesity prevalence may increase such bias.