SHBC1343

H.S.M.NG¹, L.G.CHAN¹, N.D.LIM¹, L.MOHAPATRA¹, O.H.CHAN¹

Tan Tock Seng Hospital¹

The present study aims to validate and compare the PHQ-9, with the HADS to determine the utility of these tools for screening for mood disturbance in acute post-stroke patients in a local Singaporean setting.

Stable patients within 30 days of a stroke without delirium, aphasia, history of substance abuse, current mental illnesses at Tan Tock Seng Hospital were eligible for the study. Patients were administered both the HADS and PHQ-9 along with the Structured Clinical Interview for DSM-IV (SCID-IV) for Mood Disorders. DSM-IV

136 participants were recruited for the study. Both the HADS and PHQ-9 had acceptable internal consistency (α=0.87 and α=0.75 respectively). The diagnostic accuracy for both scales were also excellent with the areas under the ROC curve at .85 for HADS and .88 for PHQ-9. The best PHQ-9 cutoff of ≥ 9 had better screening performance (sensitivity=70.6%, specificity = 90.8%) compared to the HADS-Anxiety subscale (best cutoff ≥7, sensitivity = 58.8%, specificity = 95.0%), HADS-Depression subscale (best cutoff ≥7, sensitivity = 64.7%, specificity = 89.9%) or HADS-Total score (best cutoff ≥10,sensitivity = 70.6%, specificity = 85.7%).

The ease of administering both the PHQ-9 and HADS and their favourable psychometric properties of make it useful for screening for post-stroke mood disturbances. However, the PHQ-9 was slightly better in detecting mood disturbance in acute post-stroke patients in Singapore.