Weight‐Control Exercise in Late Adolescence and Eating Disorder Psychopathology in Early Adulthood: A Prospective Analysis of the UK Millennium Cohort Study

OBJECTIVE: To examine whether weight-control exercise at age 17 was prospectively associated with eating-disorder psychopathology at age 23 after accounting for age-14 characteristics, and whether associations differed by repeated-report combination or sex. METHOD: Data were from 5266 UK Millennium Cohort Study participants. Weight-control exercise at age 17 was self-reported; age-23 psychopathology was assessed with the EDE-QS-5 total score (0-15). Survey-weighted linear regressions across 50 imputed data sets adjusted sequentially for sociodemographic factors, age-14 BMI-for-age z score, weight-control exercise, eating restriction, appearance dissatisfaction, and depressive symptoms. RESULTS: In the analytic sample, survey-weighted prevalence of weight-control exercise at age 17 was 62.6% (53.1% among males; 70.6% among females). The unadjusted mean difference in age-23 scores was 2.81 points (95% CI [2.53, 3.09]); it was 1.19 (95% CI [0.89, 1.48]) after full adjustment, corresponding to 0.26 outcome SD units (95% CI [0.20, 0.33]). Participants reporting weight-control exercise at age 17 only or at both ages had higher scores than those reporting at neither age, but the two groups did not differ clearly in a post hoc contrast. The association was stronger among females than males (interaction difference = 0.87, 95% CI [0.36, 1.39]). DISCUSSION: Weight-control exercise in late adolescence was associated with greater eating-disorder psychopathology in early adulthood after adjustment for earlier characteristics. The association was modest and may reflect broader patterns of weight, eating, and emotional vulnerability.