The effect of a text-messaging intervention on retention in care for women living with HIV and their infants in Kenya

Background: Retention in prevention of mother-to-child transmission (PMTCT) of HIV care is crucial to reduce vertical HIV transmission, and to improve the health and survival of women living with HIV and their infants. HIV-related stigma, social concerns, and non-disclosure of HIV status are barriers to participate in PMTCT care, which potentially could be addressed by supporting women with an interactive text-messaging intervention. The aim of this thesis was to increase knowledge of the effect of an interactive text-messaging intervention on retention in PMTCT care, and to shed light on pregnant women’s HIV status disclosure and their social and emotional barriers to participate in PMTCT care in Kenya. Methods: We recruited pregnant women living with HIV to a randomised controlled trial at six antenatal care clinics in western Kenya between June 2015 to July 2016. The participants in the intervention group received weekly text messages until 24 months postpartum, to which they were requested to respond within 48 hours if they were okay or if they had a problem. Interview data collected at study enrolment at four of the clinics were used to investigate participants’ social concerns to participate in PMTCT care and their association to HIV status disclosure (paper I). We compared the uptake of early infant HIV testing (paper II), and 18-month retention in PMTCT care (paper III) between the intervention and control group at all six clinics. We also investigated women’s adherence to respond if they were okay or had a problem in a cohort study of the intervention group participants (paper IV). Results: In paper I, we observed that 80% of the women who were married or living with a partner had disclosed their HIV status to their partner. Women 35-44 years old had lower odds of HIV status disclosure to a partner (odds ratio [OR]: 0.15; 95% confidence interval [CI]: 0.05-0.44) compared to women 18-24 years old. Women reporting concerns about isolation or lack of support from family or friends (OR: 0.33; 95% CI: 0.12-0.85), separation from a partner (OR: 0.17; 95% CI: 0.05-0.57), or conflict with a partner (OR: 0.18; 95% CI: 0.05-0.67) had lower odds of HIV disclosure to a partner. In paper II we observed no effect of interactive text messaging on uptake of early infant HIV testing in the intervention group compared to the control group (rate ratio [RR]: 0.99; 95% CI: 0.90-1.10; p=0.89). In paper III, 18-month retention in care was similar in the intervention (70%) and the control group (69%) (RR: 1.02; 95% CI: 0.92-1.14; p=0.70). In paper IV we observed that women in the intervention group responded to 49% of the text messages sent to them (48% were okay responses and 1% were problem responses). Women 18-24 years old (OR: 2.20; 95% CI: 1.03-4.72) were more likely to respond to <50% of the text messages compared to women 35-44 years old. Women with higher education (OR: 0.28; 95% CI: 0.13-0.64) were less likely to respond to