12-month trajectory pathways for depressive symptoms in a Hong Kong primary care population

Abstract

INTRODUCTION: This study aimed to identify the trajectory pathways for depressive symptoms amongst primary care patients in Hong Kong by Growth Mixture Modeling (GMM) and the associated risk factors. METHOD: 721 patients were included in data analysis by GMM. Depressive symptoms were measured using the PHQ-9 at baseline, 3, 6 and 12 months. Kruskal-Wallis H or Chi-square tests were used to explore factors associated with the trajectories. RESULTS: A seven-class trajectory pathway model was found to be the best fitting model. Five trajectory pathways showed symptom reduction over 12 months, one trajectory showed no significant improvement or reduction in symptoms and one pathway showed an increase in depressive symptoms. Of these, more than half of our subjects (class 1; n=406, 54.2%) started with mild depression at baseline, as a result they recovered. Factors associated with trajectory path class included health-related quality of life scores, presence of co-morbidity and increased health services use. DISCUSSION: Depression pathways appear to be very heterogeneous in this primary care population. The findings can help to clinicians identify which patients may be at higher risk of poor prognosis over time. Clinical practice and further research can incorporate such trajectories into risk stratification in order to identify patients who are at risk of progression of depression.