Maternal HIV disclosure to young HIV-uninfected children: an evaluation of a family-centred intervention in South Africa.

Objectives: Sub-Saharan Africa has large populations of HIV-infected parents who need support to raise their HIV-uninfected children. This research evaluates the 'Amagugu Intervention' aimed at supporting mothers to disclose their own HIV diagnosis to their HIV-uninfected children.

Design: Uncontrolled pre and post-intervention evaluation.

Setting: Africa Centre for Health and Population Studies, South Africa.

Participants: Two hundred and eighty-one HIV-infected women and their HIV-uninfected children aged 6-10 years.

Intervention: This lay counsellor-led intervention included six sessions conducted with mothers at home, providing printed materials and child-friendly activities to support disclosure of their diagnosis.

Main Outcome Measure: The primary outcome was disclosure to the child (full, partial, none). The secondary outcomes included maternal mental health (General Health Questionnaire) and child mental health (Child Behaviour Checklist).

Results: One hundred and seventy-one (60%) women 'fully' disclosed and 110 (40%) women 'partially' disclosed their HIV status to their child. Women who perceived their health to be excellent were less likely to 'fully' disclose compared to those considering their health to be poorer [adjusted odds ratio 0.50 (0.26-0.98), P = 0.042]. [corrected]. Compared to those not in a current partnership, those with a current partner were almost three times more likely to 'fully' disclose [adjusted odds ratio 2.92 (1.33-6.40), P = 0.008]. Mothers reported that most children reacted calmly to 'full' (79%) or 'partial' disclosure (83%). Compared to 'partial' disclosure, 'full' disclosure was associated with more children asking questions about maternal death (18 versus 8%).

Conclusions: This intervention is acceptable in resource-limited settings and shows promise. Further research using a controlled design is needed to test this intervention.