Accuracy of PHQ-9 Against Psychiatric Diagnosis for Depression Among People Living with HIV: A Mixed-Method Study in Cameroon, Kenya, and the Republic of Congo

Abstract

Background: Depression is highly prevalent among people living with HIV (PLWH) and is known to negatively impact HIV care and HIV clinical outcomes. Integration of depression management into HIV care is challenging because of the difficulties identifying depression. The aim of this study was to assess the performance of the 9-item Patient Health questionnaire (PHQ-9) as a diagnostic instrument against psychiatrist diagnosis in PLWH. Methods: Data were collected between January 2018 and June 2021 across three sites in Cameroon, Kenya, and the Republic of Congo. Participants were ≥18 years, diagnosed with HIV between 3-24 months prior to enrollment. PHQ-9 was administered by study staff followed by a psychiatrist’s evaluation within 3 days. In Cameroon, in-depth qualitative and cognitive interviews were conducted with a sub-sample of participants. Results: Overall, 530 participants with complete data were included: 300 (56.6%) in Cameroon, 132 (24·9%) in Congo, and 98 (18·5%) in Kenya. PHQ-9 was highly accurate in excluding current depression, with areas under the curve for the continuous PHQ-9 score from 0·888 (95% CI: 0·784, 0·992) in Cameroon, 0·802 (95% CI: 0·603, 1·00) in Kenya, to 0·768 (95% CI: 0·589, 0·947) in Congo and negative predictive values (NPV) of 98·9% (95% CI: 96·9%, 99·8%) in Cameroon, 95·7% (95% CI: 90·2%, 98·6%) in Congo, and 96·3% (95% CI: 89·7%, 99·2%) in Kenya for a score >9. However, its sensitivity and positive predictive value (PPV) were generally low, ranging respectively, from 50·0% (95% CI: 11·8%, 88·2%) and 17·7% (3·8%, 43·3%) in Cameroon, 50·0% (95% CI: 18·7%, 81·3%) and 29·4% (95% CI: 10·3%, 56·0%) in Congo, to 66·7% (95% CI: 29·9%, 92·5%) and 37·5% (95% CI: 15·2%, 64·6%) in Kenya for scores >9. Cognitive interviews identified challenges in understanding some questions as well as processing response categories. Interpretation: PHQ-9 performed very poorly (low sensitivity and PPV) as a tool to identify psychiatrist diagnosed depression and caution must be exercised before using it as the only guide to initiating PLWH on pharmacological treatment. However, the observed high NPV suggests it can be used to rule out depression or to identify those to be referred for further evaluation.