Long-term effects on depressive symptoms among Ugandan mothers - Findings from a follow-up of a cluster-randomized education trial in a rural low-resource setting.

Introduction: Depression is increasingly affecting mothers in poor countries such as Uganda. Various interventions have been implemented to tackle this problem, but their sustainability is under-researched. Here we present follow-up data on maternal depression six years after a cluster-randomized controlled maternal education trial in rural Uganda.

Methods: The intervention lasted six months and consisted of nutrition, hygiene, sanitation and child stimulation education, delivered to 511 mothers of 6 to 8 months' old children. Six years later we assessed maternal depressive symptoms using two psychometric tools; the Beck Depression Inventory II (BDI-II) and Center for Epidemiologic Studies Depression scale (CESD).

Results: For this follow-up study, data was available from 307/511 (60 %) mothers. Intention-to-treat analyses adjusting for clustering showed that the intervention mothers had non-significantly less depression symptoms (absolute score difference - 2; 95 % CI -5 to 0; p = 0.07) on BDI-II, and borderline significantly less depression symptoms (absolute score difference - 3; 95 % CI -5 to 0; p = 0.05) on CES-D compared to the controls. For categorized depression scores, the control mothers had significantly higher proportion of women classified in the worse depression categories for both BDI-II and CESD. We did not find any baseline characteristics associated with maternal depression.

Limitations: The BDI-II and CES-D tools are both self-reported and we cannot rule out the possibility of social desirability bias in reporting of depression symptoms.

Conclusion: Six years after the maternal education trial, some benefits on maternal mental health were sustained. More studies are warranted on sustainability and scale-up of such interventions.