Validation of single and combined screening tools for the detection of poststroke mood disorders in acute stroke patients in Singapore.

OBJECTIVE: Screening tools such as the Patient Healthcare Questionnaire (PHQ) and Hospital Anxiety and Depression Scale (HADS) are useful for detecting poststroke depression (PSD). However, validation of these tools has yet to be conducted in the Singapore stroke population. METHOD: A total of 138 adults were administered the HADS and both two- and nine-item versions of the PHQ. Receiver operating characteristic (ROC) analyses were utilized to validate these scales against the gold standard diagnosis of PSD through the Structured Clinical Interview for DSM-IV (SCID-IV). RESULTS: All the tools examined in this study had good convergent validity (rs = 0.55-0.89) and discriminative power (Area Under Curve: 0.849-0.887). The optimal cut-off scores were ≥ 7 for the HADS depression and anxiety subscales, ≥ 10 for the total score, ≥ 2 for the PHQ-2 and ≥ 8 for the PHQ-9. Additional analyses suggest that the use of both the PHQ and the HADS in specific combinations further improved diagnostic performance. CONCLUSION: Both the PHQ and the HADS are acceptable tools for screening for poststroke depression in Singaporean inpatient rehabilitation care settings. Furthermore, our findings lead us to recommend using PHQ-9 with HADS-A for superior screening performance at sensitivity = 83.33%, specificity = 88.33%, positive predictive value = 51.72%, and negative predictive value = 97.25%.