Integrating Screening and Clinical Interviews: Advancing the Assessment of Exercise Addiction in Athletes
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background and aims
Questionnaire-based tools such as the Exercise Dependence Scale (EDS) are widely used to identify individuals “at risk” for exercise addiction (EA), but it remains unclear how accurately these classifications reflect clinically diagnosable cases. This study combined quantitative screening with structured clinical interviews to (1) determine the proportion of false-positive classifications, (2) characterize psychological and motivational differences between clinically addicted and non-addicted athletes, and (3) identify the most discriminative predictors of EA.
Methods
A total of 342 endurance athletes (126 female, 153 male) completed an online survey including the EDS, life satisfaction, and motivational factors. Athletes exceeding the EDS cut-off (>77; n = 63) were invited to structured interviews based on ICD-11 criteria for disorders due to addictive behaviors and additionally completed the SCL-90-R. Group comparisons used t-tests or Mann–Whitney U-tests, and significant variables entered a LASSO logistic regression.
Results
Of the 34 interviewed athletes, 24 (70.6%) met ICD-11 criteria for EA, while 10 (29.4%) did not, indicating a notable false-positive rate of EDS-based screening. Interrater reliability for diagnostic classification was 97%. Compared to non-addicted athletes, those with EA were older, reported lower life satisfaction, higher weight-related exercise motivation, more withdrawal symptoms, greater interpersonal sensitivity, and more comorbidities. In the LASSO model, withdrawal symptoms, past comorbidities, and interview-based anamnesis emerged as strongest predictors of EA (AUC = 0.97).
Conclusions
Combining self-report and clinical assessment revealed that questionnaire-based screening overestimates EA prevalence. Findings underscore the importance of withdrawal-related and comorbidity features for differential diagnosis and support a multi-method approach to distinguish high involvement from true addictive exercise.
