King's College London

Research portal

A comparative meta-analysis of the prevalence of exercise addiction in adults with and without indicated eating disorders

Research output: Contribution to journalReview article

Mike Trott, Sarah E Jackson, Joseph Firth, Louis Jacob, Igor Grabovac, Amit Mistry, Brendon Stubbs, Lee Smith

Original languageEnglish
JournalEating and weight disorders : EWD
DOIs
Publication statusE-pub ahead of print - 1 Jan 2020

King's Authors

Abstract

BACKGROUND: Exercise addiction is associated with multiple adverse outcomes and can be classified as co-occurring with an eating disorder, or a primary condition with no indication of eating disorders. We conducted a meta-analysis exploring the prevalence of exercise addiction in adults with and without indicated eating disorders.

METHODS: A systematic review of major databases and grey literature was undertaken from inception to 30/04/2019. Studies reporting prevalence of exercise addiction with and without indicated eating disorders in adults were identified. A random effect meta-analysis was undertaken, calculating odds ratios for exercise addiction with versus without indicated eating disorders.

RESULTS: Nine studies with a total sample of 2140 participants (mean age = 25.06; 70.6% female) were included. Within these, 1732 participants did not show indicated eating disorders (mean age = 26.4; 63.0% female) and 408 had indicated eating disorders (mean age = 23.46; 79.2% female). The odds ratio for exercise addiction in populations with versus without indicated eating disorders was 3.71 (95% CI 2.00-6.89; I2 = 81; p  ≤ 0.001). Exercise addiction prevalence in both populations differed according to the measurement instrument used.

DISCUSSION: Exercise addiction occurs more than three and a half times as often as a comorbidity to an eating disorder than in people without an indicated eating disorder. The creation of a measurement tool able to identify exercise addiction risk in both populations would benefit researchers and practitioners by easily classifying samples.

View graph of relations

© 2018 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454