Objective:
To investigate if cross-sectionally identified correlates of disordered eating among adolescents could also predict disordered eating prospectively over and beyond the initial level of disordered eating.
Method:
Two-wave longitudinal questionnaire study of a representative and nationwide sample of 7,751 norwegian adolescents aged 12-19 at initial testing (t1). a 12-item version of the eating attitudes test (eat) was included.
Results:
Forty percent of the girls and 25% of the boys with disordered eating at t1 also scored above the cut-off point 2 years later (t2). initial disordered eating, gender, depressed mood, body dissatisfaction, unstable self-perceptions, perceived obesity, excessive exercise, and having idols with perfect bodies all predicted change in disordered eating. however, when the initial symptom load was controlled, these variables - except gender - only contributed marginally to the prediction of disordered eating. these negative results are in line with other longitudinal studies of changes in disordered eating in unselected adolescent populations.
Discussion:
It cannot be established that psychological factors play a major etiological role in the development of disordered eating. the results are discussed with reference to a potential discontinuity between eating problems and eating disorders.