Criterion-related validity and reliability of the Urdu version of the patient health questionnaire in a sample of community-based pregnant women in Pakistan

Background. Depression is one of the most prevalent, yet unrecognized but treatable mental disorders in low and middle income countries (LMICs). In such locations, screening tools that are easy-to-administer, valid, and reliable are needed to assist in detecting symptoms of depression. The Patient Health Questionnaire (PHQ-9) is one of the most widely used depression screeners. However, its applicability to community-based settings of Pakistan is limited by the lack of studies examining its validity and reliability in such settings. The current study aimed to demonstrate the criterion related validity and internal reliability of the Urdu version of the PHQ-9 in a sample of community-based pregnant women in Pakistan compared to a diagnostic clinical interview, the Structured Clinical Interview for DSM disorders (SCID), using data from a depression treatment cluster randomized trial in rural Pakistan. Methods. Pregnant women in a rural, low income sub district in Pakistan were approached between October 2014 and February 2016 and, after providing informed consent, screened for depression using the Urdu version of the PHQ-9, with a cutoff of >= 10 used to indicate significant depressive symptoms. Following the PHQ-9, the diagnostic module for current major depressive episode of the SCID was administered. We examined the psychometric properties of PHQ-9 compared to SCID as a gold standard, using sensitivity, specificity, and negative and positive predictive value to measure the criterion-related validity of the PHQ-9 as an indicator of symptoms of depression. We computed area under the receiver operating characteristic curve to determine diagnostic accuracy, and used Cronbach's alpha to assess internal reliability. Results. A total of 1,731 women in their third trimester of pregnancy were assessed for major depressive disorder. Of these women, 572 (33%) met the cutoff for significant depressive symptoms on PHQ-9, and 454 (26%) were assessed positive for depression using the SCID. The sensitivity and specificity of PHQ-9 at a cutoff of >= 10 was 94.7% and 88.9%, respectively. The positive and negative predictive values were 75.2% and 97.9%, respectively; and the area under the curve was 0.959. Internal reliability, as measured by Cronbach's alpha, was 0.844. Discussion. Valid and reliable screening tools to assist in detecting symptoms of depressive disorder are needed in low income settings where depressive disorders are highly prevalent. The Urdu version of the PHQ-9 has not been previously validated against a well-known assessment of depression in a community setting among pregnant women in Pakistan. This study demonstrates that the Urdu version of the PHQ-9 has acceptable criterion-related validity and reliability for screening for depressive symptoms in Pakistan among community-based pregnant women; and when the recommended cut-off score of >= 10 is used it can also serve as an accurate screening tool for major depressive disorder.