# Distress screening in patients with high-grade glioma: diagnostic accuracy in relation to a structured clinical interview in a multicenter cluster-randomized controlled trial

## TL;DR

This study compares three screening tools for identifying mental disorders in glioma patients, finding that while they perform better than chance, none are highly accurate.

## Contribution

The study evaluates the diagnostic accuracy of three screening tools for psychiatric comorbidities in glioma patients using a multicenter trial.

## Key findings

- 31% of glioma patients had at least one mental disorder.
- The EF scale showed the best performance (AUC 0.70), followed by DT (AUC 0.69) and TSQ (AUC 0.61).
- All tools performed better than random chance but had notable false negatives.

## Abstract

Structured clinical interviews, such as the Structured Clinical Interview for DSM (SCID), are considered the gold standard for diagnosing mental disorders but are challenging in routine clinical use due to their length. Therefore, screening instruments to identify the need for further assessment are required. The National Comprehensive Cancer Network Distress Thermometer (DT) screens for psychological distress, while the Emotional Functioning (EF) scale of the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core (EORTC QLQ-C30) assesses emotional functioning. Both are frequently used in clinical routine. Additionally, three brief screening questions (TSQ), specifically developed for patients with glioma and integrated into doctor–patient consultations, may also be used for screening. This study aimed to evaluate the ability of the three tools to identify patients with psychiatric comorbidities as diagnosed by the SCID.

Using data from glioma patients treated at 13 German hospitals participating in a cluster-randomized trial, discriminative abilities were assessed using receiver operating characteristic (ROC) curves and corresponding areas under the curve (AUCs). Confidence intervals (CIs) were estimated, and hypothesis tests were conducted using bootstrapping.

Of the 691 patients interviewed, 31% presented with at least one mental disorder. The EF scale demonstrated the best discriminative ability (AUC 0.70, 95% CI: 0.66–0.74), followed by the DT (AUC 0.69, 95% CI: 0.62–0.76), and the TSQ total score (AUC 0.61, 95% CI: 0.55–0.66).

While all three tools performed better than random chance, none demonstrated convincing discriminative ability in identifying psychiatric comorbidities. In practice, screening tools can identify a substantial proportion of patients with mental disorders, however at the cost of a considerable number of false negatives.

The online version contains supplementary material available at 10.1007/s00520-025-09810-1.