# Parent–child discrepancies in screening for Internet Gaming Disorder: Evidence from a clinical sample of Japanese adolescents

## TL;DR

This study finds that parents and children often disagree on whether a child has Internet Gaming Disorder, especially when using cutoff scores for diagnosis.

## Contribution

The study reveals significant discrepancies between parent and child reports when using cutoff-based classifications for IGD in adolescents.

## Key findings

- Parent and child IGD scores were moderately correlated but parents reported higher scores.
- Categorical agreement based on cutoff scores was low (κ = 0.16), with most discrepancies showing parents flagging IGD while children did not.
- The study suggests multi-informant and dimensional approaches are needed for accurate IGD screening in youth.

## Abstract

Questionnaire‐based screening tools for Internet Gaming Disorder (IGD) are widely used in clinical and epidemiological research. However, discrepancies between child self‐reports and parent reports may complicate the interpretation of screening results, particularly when cutoff‐based classifications are applied.

Participants were 58 adolescents (aged 10–18 years) attending child and adolescent psychiatry outpatient clinics and their parents. Gaming‐related problems were assessed using parallel screening instruments: the Internet Gaming Disorder Scale for Children (IGDS‐C) and the Parental version of the Internet Gaming Disorder Scale (PIGDS). Parent–child agreement was examined using dimensional analyses (Pearson's correlation), paired comparisons (paired t‐test with Wilcoxon signed‐rank test as a sensitivity analysis), and categorical agreement indices (concordance rate, Cohen's κ, and McNemar's test) based on the conventional cutoff score.

Parent‐ and child‐reported IGDS scores were moderately correlated (r = 0.61, p < 0.001), indicating substantial dimensional concordance. However, parents reported significantly higher IGDS scores than children (mean difference = −1.09, p < 0.001), a finding confirmed by the Wilcoxon signed‐rank test. Categorical agreement based on cutoff‐based screening classifications was low (κ = 0.16), with most discordant cases reflecting parent‐positive and child‐negative classifications. McNemar's test demonstrated a significant asymmetry in these discrepancies.

Although parent and child IGDS scores demonstrate meaningful dimensional concordance, the application of fixed cutoff‐based screening classifications substantially reduces agreement, a pattern that may reflect differences in evaluative thresholds between informants. These findings highlight limitations of relying solely on self‐reported cutoff‐based measures and underscore the need for multi‐informant, dimensional approaches when interpreting IGD screening results in youth.