Evaluating the impact of community health worker integration into prevention of mother-to-child transmission of HIV services in Tanzania

Publication Details

Nance, N, McCoy, S, Ngilangwa, D, Masanja, J, Njau, P and Noronha, R, 2017. Evaluating the impact of community health worker integration into prevention of mother-to-child transmission of HIV services in Tanzania, 3ie Impact Evaluation Report 61. New Delhi: International Initiative for Impact Evaluation (3ie)

Link to Source
Nerissa Nance, Sandra McCoy, David Ngilangwa, Joseph Masanja, Prosper Njau and Rita Noronha
Sub-Saharan Africa (includes East and West Africa)
Health Nutrition and Population
HIV and AIDS, Preventive Health and Health Behavior, Sexual Behavior
Gender analysis
Equity Focus
None specified
Evaluation design
Randomised Control Trials (RCT)
3ie Series Report
3ie Funding Window
Integration of HIV services


The study uses a cluster randomised controlled design to measure the impact of using community health workers (CHWs) in integrating maternal, newborn, and child health (MNCH) services and prevention of mother-to-child transmission (PMTCT) of HIV services in Tanzania


Nearly one out of every five new HIV infections is due to mother-to-child transmission in Tanzania. Evidence suggests that nearly one quarter of HIV-positive pregnant women still do not receive antiretroviral therapy (ART) for PMTCT. In response, Tanzania adopted Option B+ in 2013. Nevertheless, challenges still exist in increasing treatment and care retention rates. To address these challenges, UC Berkley implemented a programme integrating MNCH services and PMTCT services. The evaluation provides insights into how best to prevent attrition from the PMTCT treatement adherence and retention in care cascade, which is a critically important component, as Tanzania begins scaling up Option B+.

Research questions

  • What is the impact of integrating community-based MNCH and PMTCT services using CHWs:
    • The timing of ART initiation among HIV-infected pregnant and postpartum women;
    • Retention of HIV-infected women in care 90 days postpartum; and,
    • Adherence to ART at 90 days postpartum.
  • Is the impact of integrating community-based MNCH and PMTCT services on retention in care and adherence greater among women who had more contact with CHWs?


Intervention design

The intervention involved using CHWs in integrating MNCH and PMTCT services. CHWs were formally linked with reproductive and child health (RCH) facilities. They received mentorship, supportive supervision and coordination of tasks. These CHWs were then tasked with distributing user-friendly action birth cards (ABCs) to pregnant and post-partum women, regardless of HIV status. The cards are an interactive planning tool that link community-level MNCH services with PMTCT services and HIV treatment delivered at the facility. CHWs also provided monthly ART adherence counselling, and helped track down women who were lost to follow-up.

Theory of change

The theory of change proposes several effects. First, the authors assume that patients face a number of barriers in accessing both MNCH and PMTCT services. The theory proposes that personal contact from CHWs, providing reminders and encouragement, will address these barriers. They assume that CHWs are willing and capable of providing accurate and timely information about adherence in a way that minimises or prevents stigma. This is partially based on the assumption that the clinic staff (to which CHWs are linked) are willing and able to work with CHWs to provide mentorship and support, and this will result in improved service and counselling provision by CHWs. The authors also posit mothers and providers will use the new ABCs to track care plans and to facilitate discussions about barriers and facilitators to care. They then suggest that that the integration and CHWs will result in increased ANC and postnatal visit attendance, exposed infant early diagnosis, and delivery in a health facility. It will catalyse these women to initiate ART earlier, improve adherence and stay in care.

Evaluation design

The study used a cluster-randomised design on 32 facilities in the Shinyanga region; 15 facilities were allocated to the treatment group and 17 to the control. The study sample included all villages with an active CHW cadre within the service area of a RCH facility that provides Option B+. Baseline data were collected for 1,152 mother-infant pairs in May 2015, and end line data were collected for 678 pairs in April 2016. The study used a mixed-method approach and difference-in-difference analyses.

Main findings


In comparison sites, there were significant improvements over time in the timing of ART initiation, retention in care, and medication possession ratio (MPR). Among women with evidence of prior HIV care, the intervention was associated with a non-significant 5.0 percentage point improvement in retention in care at 90 days postpartum and non-significant 9.5 percentage point increase in ≥95% MPR. Sub-group analyses indicate that a higher intensity implementation produced significant improvements in ≥95% MPR and ART initiation during pregnancy. Qualitative data suggest that CHWs uniquely supported HIV-infected women by encouraging and organising clinic visits, which may have contributed to the women having fewer days of medication non-possession. 

Cost efficiency analysis

The costs of the intervention, including the CHW allowance, the ABCs, cost of the staff time to meet with CHWs, and costs of trainings, added USD51,145 over an 11-month intervention period.  The investment translates to an overall cost of approximately USD4,525 per every percentage point increase in 95 per cent medication adherence in their analysis of 820 women.

Implications for implementers

The ABCs may have been instrumental in nudging women to attend clinic visits, including HIV care. CHWs also played a vital role in support and encouraging women to attend clinic regularly, and to adhere to their medication.

Implications for policy and practice

While the healthcare workforce crisis still needs to be addressed, CHWs may help alleviate some of the burdens of this crisis, as they are a part-time volunteer force that does not require extensive training.

Implications for further research

Further research could follow a larger cohort of HIV-infected pregnant and postpartum women, and scale-up efforts may consider providing greater infrastructure and monitoring around the intervention.

Scroll to Top