Related Experiment Video
Updated: Dec 18, 2025
Author Spotlight: Unveiling the Connection Between Sleep Disorders and Cognitive Symptoms in Depression
Published on: April 26, 2024
The validity and reliability of the PHQ-9 in screening for post-stroke depression
Piyapat Dajpratham1, Panate Pukrittayakamee2, Wanlop Atsariyasing2
1Department of Rehabilitation Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University, 9th floor Srisangwal Building, Siriraj Hospital, Wanglang Road, Bangkok Noi, Bangkok, 10700, Thailand. piyapat.daj@mahidol.ac.th.
Background:
Depression affects about 30% of stroke survivors within 5 years. Timely diagnosis and management of post-stroke depression facilitate motor recovery and improve independence. The original version of the Patient Health Questionnaire-9 (PHQ-9) is recognized as a good screening tool for post-stroke depression. However, no validation studies have been undertaken for the use of the Thai PHQ-9 in screening for depression among Thai stroke patients.
Methods:
The objectives were to determine the criterion validity and reliability of the Thai PHQ-9 in screening for post-stroke depression by comparing its results with those of a psychiatric interview as the gold standard. First-ever stroke patients aged ≥45 years with a stroke duration 2 weeks-2 years were administered the Thai PHQ-9. The gold standard was a psychiatric interview leading to a DSM-5 diagnosis of depressive disorder and adjustment disorder with a depressed mood. The summed-scored-based diagnosis of depression with the PHQ-9 was obtained. Validity and reliability analyses, and a receiver operating characteristic curve analysis, were performed.
Results:
In all, 115 stroke patients with a mean age of 64 years (SD: 10 years) were enrolled. The mean PHQ-9 score was 5.2 (SD: 4.8). Using the DSM-5 criteria, 11 patients (9.6%) were diagnosed with depressive disorder, 12 patients (10.5%) were diagnosed with adjustment disorder with a depressed mood. Both disorders were combined as a group of post-stroke depression. The Thai PHQ-9 had satisfactory internal consistency (Cronbach's alpha: 0.78). The algorithm-based diagnosis of the Thai PHQ-9 had low sensitivity (0.52) but very high specificity (0.94) and positive likelihood ratio (9.6). Used as a summed-scored-based diagnosis, an optimal cut-off score of six revealed a sensitivity of 0.87, specificity of 0.75, positive predictive value of 0.46, negative predictive value of 0.95, and positive likelihood ratio of 3.5. The area under the curve was 0.87 (95% CI: 0.78-0.96).
Conclusions:
The Thai PHQ-9 has acceptable psychometric properties for detecting a mixture of major depression and adjustment disorder in post-stroke patients, with a recommended cut-off score of ≥6 for a Thai population.
