# Validation and limitations of the distress thermometer in identifying depression among metastatic breast cancer patients in Nigeria: Methodological challenges in depression-specific screening validation

## TL;DR

The Distress Thermometer is not effective at identifying depression in Nigerian metastatic breast cancer patients, likely due to cultural and methodological issues.

## Contribution

This study is the first to evaluate the DT's diagnostic validity for depression in advanced cancer populations in a sub-Saharan African context.

## Key findings

- The DT had poor diagnostic performance with an AUC of 0.414, significantly below acceptable accuracy thresholds.
- At the commonly used DT threshold of ≥4, sensitivity was 34.7% and specificity was 50.8%.
- The study highlights the need for culturally appropriate and construct-matched depression screening tools in advanced cancer populations.

## Abstract

Psychological distress is highly prevalent among patients with metastatic breast cancer (MBC), yet validated screening tools remain scarce in low- and middle-income countries (LMICs). The Distress Thermometer (DT), a globally endorsed screening instrument, lacks robust validation in advanced-stage cancer populations within LMIC contexts, where cultural, linguistic, and healthcare system factors may significantly influence its diagnostic performance.

To assess the diagnostic validity of the Distress Thermometer in identifying clinically significant depressive symptoms among Nigerian women with metastatic breast cancer, using the Beck Depression Inventory-II (BDI-II) as the reference standard.

Cross-sectional diagnostic validation study employing receiver operating characteristic (ROC) analysis to evaluate DT performance, with optimal cutoff identified using Youden’s Index.

A total of 309 histologically confirmed de-identified metastatic breast cancer patients were consecutively recruited from the NSIA-LUTH Cancer Centre, Lagos, Nigeria, between September 2020 and February 2022. All participants completed the DT, BDI-II, and EORTC QLQ-C30/BR23 modules.

The mean DT score was 3.4 (SD ± 1.6), with 47.0% of participants meeting the NCCN threshold for clinically significant distress (DT ≥ 4). In contrast, only 15.9% had BDI-II scores ≥20, indicating moderate-to-severe depressive symptoms. The DT demonstrated diagnostic performance significantly worse than random classification, with an AUC of 0.414 (95% CI: 0.326–0.503), significantly below the threshold for acceptable diagnostic accuracy (p < 0.001). The Youden-optimal cutoff was DT ≥ 7.5, yielding a sensitivity of 2.0% and specificity of 98.5%. At the commonly used DT ≥ 4 threshold, sensitivity was 34.7% and specificity was 50.8%, indicating poor overall diagnostic utility.

In this sub-Saharan African cancer cohort, the Distress Thermometer performed poorly in detecting clinically significant depression in this Nigerian MBC cohort when benchmarked against the BDI-II. This may reflect construct mismatch, somatic symptom confounding, cultural factors, or disease-specific characteristics of advanced cancer populations. The DT should not be used as a depression-specific screening tool in advanced cancer populations in LMICs, though its utility for identifying general distress remains unclear given the methodological limitations of this study. The poor concordance observed likely reflects construct mismatch, somatic symptom confounding in the reference standard, and the fundamental challenges of depression assessment in advanced cancer populations using self-report instruments. These findings underscore the critical need for appropriate reference standards (structured clinical interviews) and highlight methodological considerations in validating psychosocial screening tools across different constructs and cultural contexts.