Related Experiment Video
Updated: May 3, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
Single-Item Screening for Depression in Adults (50+) with History of Traumatic Brain Injury
Summer N Rolin1, Jeremy J Davis2
1Department of Rehabilitation Medicine, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229.
Background:
Single-item depression screening has shown utility in young adult samples with limited work focused on individuals 50 and older. We examined single-item depression screening in adults 50 and older with a history of TBI one year post-injury.
Method:
This project involved secondary analysis of deidentified TRACK-TBI data. The sample (N = 508) included 61 control participants and 447 cases grouped by TBI severity. Outcome measures included the Brief Symptom Inventory-18 depression scale (BSI-D) and Patient Health Questionnaire-9 (PHQ-9). Depression was determined psychometrically by BSI-D (≥63 T) and PHQ-9 (≥10) if either score was above the cutoff. The Rivermead Postconcussion Symptoms Questionnaire depression item (RPQ-D) was used in logistic regression (LR) and receiver operating characteristic (ROC) analyses to predict depression.
Results:
Depression was observed in 16.1% of cases and 6.6% of control participants (p = .05) with non-significant variability across TBI severity (12.9-18.3%; p = .17). The LR model with age, education, TBI severity, and RPQ-D as predictors was significant with age and RPQ-D as significant predictors. The model classified 87.9% of participants correctly. The area under the ROC curve was 0.86. The optimal cutoff was RPQ-D ≥ 2.
Conclusion:
Single-item screening for depression is a suitable approach in adults 50 and older with a history of TBI.
