Abstracts Category : Other

Feasibility and Effectiveness of Integrating HIV Prevention and Testing into FamilyPlanning Services in North West Province, South Africa: A cluster randomised trial

by S Mullick

Background:South Africa has one of the highest HIV prevalence rates in the world and highcontraceptive use among women aged 1549 (65.3%). Family planning (FP) servicesremain a missed opportunity to integrate services for HIV. Recent reviews highlighted thelack of rigorously conducted studies of the effectiveness of integrated services. A clusterrandomised trial (CRT) was conducted to evaluate the effectiveness of a model ofintegrating HIV into FP services compared with standard practice. The study sought tomeasure the effect of integrated services (Balanced Counselling Strategy Plus) on HIVtesting in the previous year; use of dual protection and quality of HIV and FP care.Methods: A CRT was conducted in 12 clinics in North West province, South Africa. Structured clientproviderobservations (CPOs) and client exit interviews (EIs) were conducted preinterventionand one year later with FP clients aged >=18 years. Primary outcomes werecondom use at last sex and testing for HIV in last year. The quality of care scores wereconstructed to assess HIV and FP quality of care. Analysis of effectiveness used statisticalmethods for CRTs.Findings: A total of 1,111 CPOs and 1,111 EIs were completed at baseline and 1,223 CPOs and1,264 EIs at follow-up. At follow-up 33.2% of women in the intervention arm had tested forHIV in the last year compared with 21.4% in the control arm; RR=1.56 (95%CI: 1.13 2.15;p=0.01). Condom use at last sex was 43.7% in the intervention arm and 39.4% in thecontrol arm; RR=1.10 (95%CI: 0.851.43; p=0.14).InterpretationThere was strong evidence of higher frequency of HIV testing among FP clients atintervention clinics. However, condom use at last sex was similar across intervention andcontrol arms. All QOC scores were higher in intervention clinics, but there was substantialvariation across clinics and these differences were not significant.