Efficacy of Mobile phone use on adherence to Nevirapine prophylaxis and retention in care among the HIV-exposed infants in prevention of mother to child transmission of HIV: a randomized controlled trial

HIV is a major contributor to infant mortality and in the absence of intervention, the risk of infant HIV acquisition from a HIV-positive mother is close to 40%. In 2018, an estimated 160,000 new HIV infections occurred among children under the age of 15, with 90% of these children getting the infection through vertical transmission [1, 2]. Cumulatively, globally nearly 3 million children are living with HIV and almost 250,000 HIV related deaths are reported yearly. Prevention of mother-to-child transmission is an important strategy in reducing infant mortality from HIV [1]. Kenya is one of the 22 priority countries for eliminating mother to child transmission of HIV [1, 3]. Despite overall progress, infant antiretroviral uptake significantly lags that seen in adult population and this may be attributable to poor retention and follow-up mechanisms within and outside health care systems [1].

There is growing evidence that mobile phones can improve antiretroviral therapy adherence among HIV-positive patients and improve health outcomes [4‐7]. Information technology offers an opportunity to achieving an HIV-free generation through optimized communication and encouragement for retention of the mother-baby pairs. Expansion of mobile phone coverage in Africa, and in Kenya, where current coverage is 80%, presents an opportunity to scale up the HIV/AIDs response [8‐10].

This study evaluated the effect of mobile phone use on adherence to Nevirapine prophylaxis, retention in care, early infant diagnosis (EID) and breastfeeding, using a randomized clinical trial design in Kisumu, Kenya. We recruited 150 mother infant pairs. We hypothesized that the phone call intervention would improve adherence and retention in PMTCT compared to the standard of care of giving an appointment at time of visit without follow-up reminders. Women in the intervention arm received phone calls fortnightly, while those in the control arm did not receive calls.

There are three key findings from this study. The first is that mobile phone reminders improved adherence to infant NVP prophylaxis. At six weeks, 90.7% of the infants in the intervention arm were reported to have received the NVP prophylaxis, compared to 72% in the control arm of the study, a 19% difference. Assuming HIV transmission rate of 20% at six weeks for babies who did not access NVP prophylaxis, assuming a population of 100 HIV-exposed infants, the phone intervention would have averted four infant infections. The rate of adherence to infant NVP prophylaxis was higher than national data which demonstrated that only 63% of exposed infants receive ARV prophylaxis at six weeks [11]. The results of this study have important implications as Kenya strives to meet its stated goals of elimination of MTCT [12].

The second important finding was that the fortnightly reminders improved retention at both times of observation. At six weeks 78.7% of the mother-infant pairs in the intervention arm were still in care compared to 58.7% in the control arm, a 20% difference in retention in care. Put in another way, four out of five HIV-exposed infants in the intervention arm were still in care compared to just less than three of five in the control arm of the study. Mobile phone calls have been shown to increase ARV adherence and retention in care among adult populations, however, there is limited data regarding this strategy in the context of prevention of mother to child transmission of HIV or HIV-exposed infant care. Our study is an important contribution towards improved quality of care for HIV-exposed infants. The findings of this study are comparable to a Kenyan study in which short message service (SMS) reminders improved adherence to ART among patients attending a rural clinic (90% after 48 weeks of follow-up among those in the intervention arm compared to 40% among the controls) [6].

A third observation was that there was a strong trend towards higher rates of early infant diagnosis in the intervention arm (70.7% versus 58.7% at six weeks). The rate of EID in this study is higher than the national rate of 35% [11] and this could be attributed to the controlled study setting. On exclusive breastfeeding, although the results were not statistically significant at 10 weeks, we showed a difference between the two groups (97.3% compared to 90.7% = 0.1). This was a relatively small study and with a larger study we might be able to show a difference. This could be attributed to the level of counseling in the PMTCT program.

Our study involved direct phone calls in which the mother had the opportunity to ask questions including those related to their child’s health. A previous study at a Kenyan tertiary facility, demonstrated that weekly phone calls during the six postpartum weeks to primiparous mothers improved infant survival at six and 10 weeks (Kihara). Retention in care for HIV exposed infants remains an important challenge with estimates in Kenya still largely unknown. Although our study did not follow infants through the first year of life, the findings suggest that even as early as 10 weeks, retention was impacted by regular phone reminders. These results suggest that there is potential to use this tool for other health related aspects.

Randomization was successful, and the two groups were comparable on most characteristics, however women in the control arm were more likely than those in the intervention arm to be employed (2 fold higher) and have received tertiary education (2 fold higher). Both of these factors are well recognized drivers of good health seeking behavior [13, 14]. Consequently the relative imbalance on the two characteristics could have attenuated our findings as a mother’s education can exert a positive influence on her children’s health and survival and a more educated mother is more likely to adhere to instructions than an uneducated one.

There were a number of limitations to our study. There might have been an overestimate to report ART adherence since the study relied on self-reports. Facility level adherence data such as pharmacy refill to verify adherence to NVP prophylaxis in infants and infant HIV status were not ascertained. Nevertheless, there was strong corroboration between the rate of adherence and retention in care suggesting that in this study self-reports were reasonably accurate. Another limitation was the customary use of mobile phones by more than one individual in some homes which poses a confidentiality risk and which may impact on continuity of care. It is illustrative that although retention in the intervention arm of the study was significantly higher, 29% of the women were lost to follow-up by the tenth week of the postnatal visit. This may have resulted in differential loss to follow-up given that mobile phone ownership is related to socioeconomic status. Further research on mobile phone use will need to continue to address issues of confidentiality.

Overall, these results suggest that phone calls can be an important tool to improve adherence to infant NVP prophylaxis and retention in care for HIV-exposed infants. Given the high rate of mobile phone penetration (above 80%) there is potential to scale up this intervention to other parts of the country and region. The ART regimen for PMTCT changed from Nevirapine to a two-dose regimen in 2016. Despite the regimen changes after completion of the findings reported in the current study, the problem of non-adherence to ART regimens in PMTCT remains and evidence on effective approaches for improving ART adherence through patient follow up is still needed. Mobile phone coverage has increased consistently including in the period following study completion. In this regard, the evidence presented in this study remains relevant to PMTCT care given that ART has been retained as the mainstay of PMTCT care and mobile phone coverage is increasing. A key concern however is that the poorest sectors of the population either may not afford a mobile phone, or will share one handset leading to unique challenges in rolling out m-health interventions such as the one evaluated here. In addition, cost effectiveness studies will be critical in understanding the true benefits and best implementation strategies of mobile phone use for follow-up in PMTCT programs.

These results suggest that phone calls are potentially an important tool to improve adherence to infant NVP prophylaxis and retention in care for HIV-exposed infants.