Related Experiment Video
Updated: Sep 29, 2026
A Computer-Based Platform for Aiding Clinicians in Eating Disorder Analysis and Diagnosis
Published on: May 10, 2022
Validation of Parent-Report Questionnaires for Large-Scale Online Screening of Avoidant/Restrictive Food Intake
Diana Friskson1, Felicia Dahlbäck1, Laila L L Myrberg1
1Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.
Objective:
Studies assessing the validity of screening measures for avoidant/restrictive food intake disorder (ARFID) remain scarce. We evaluated the diagnostic performance and validity of an online, parent-reported screening approach for ARFID in a population-based sample of children.
Methods:
Participants were drawn from the ARFID Initiative Sweden (ARIES) cohort and included 65 children aged 6-14 years. Parents completed three screening questionnaires-Pica, ARFID, and Rumination Disorder Interview-ARFID Questionnaire (PARDI-AR-Q), Nine-Item ARFID Screen (NIAS), and Parent Eating Disorder Examination Questionnaire (PEDE-Q)-followed by a diagnostic interview (PARDI). Diagnostic performance indices (sensitivity, specificity, positive predictive value [PPV], and negative predictive value [NPV]) were calculated. Convergent validity was assessed via correlations between questionnaire and interview dimensions.
Results:
The combined screening algorithm demonstrated perfect sensitivity and NPV for ARFID diagnosis, indicating accurate detection of all ARFID cases and exclusion of non-cases. Specificity was high (0.83), and PPVs ranged from 0.91 to 0.95, decreasing to 0.78 under a more conservative operationalization of ARFID Criterion A4 (psychosocial impairment). Diagnostic performance varied across single ARFID criteria: PPVs were low for the medically anchored Criteria A1-A3 but high for Criterion A4 (psychosocial impairment; PPV = 0.95), which was also the most commonly met criterion. Correlations between screening measures and corresponding interview dimensions were generally moderate to strong, supporting convergent validity. Children meeting threshold ARFID criteria showed significantly greater symptom severity than subthreshold cases at screening.
Discussion:
These findings support the use of a multi-instrument, parent-reported screening approach for ARFID in large-scale pediatric research and highlight the centrality of psychosocial impairment, underscoring the need for standardized operationalization of this criterion.
