Objective: To maximize prevention of mother-to-child transmission of HIV (PMTCT) effectiveness and increase identification of HIV status in maternity units in Swaziland.
Design: With a quasi-experimental design, 3 maternity units were randomly assigned to the training intervention and 3 units were controls.
Methods: Targeted on-site training was provided to nurse-midwives in intervention sites. HIV status was recorded with testing offered to women presenting with unknown and distant negative status. Cord blood was obtained and tested for HIV antibodies and presence of nevirapine as a marker of PMTCT intervention coverage. Contingency tables and χ² tests were used to test for associations between frequencies of events.
Results: Of the 2444 enrolled women, 215 (9%) arrived in maternity with unknown status and 1398 (58%) had tested HIV negative in antenatal clinic. Significantly more HIV-negative women (45%) and women with unknown status (96%) in intervention sites were tested compared with similar women in control sites, 14% and 65%, respectively (P < 0.0001 for both). Nevirapine coverage in HIV-positive cord blood was significantly higher in intervention sites (80%) than in control sites (69%, P < 0.0001). Cumulative HIV incidence was 4% with an incidence rate of 16.8 per 100 person-years. Antiretroviral prophylaxis coverage in seroconverters was significantly higher in intervention sites 54% (13 of 24) than the control group [26% (9 of 34), P = 0.03].
Conclusions: In high HIV prevalence settings, such as Swaziland, the incidence of HIV during pregnancy is high. An on-site training intervention for maternity nurses significantly increases the identification of HIV infection and maximizes the provision of PMTCT interventions.