Community health workers and contraceptive access: Findings from Burundi

The views expressed in the Let’s Talk Development blog are solely those of the author(s).

In many developing countries, restrictive social norms and limited access to modern contraception lead women to have more children than they would ideally choose. This disparity is especially pronounced in rural and low-resource communities. Burundi has one of the world’s lowest contraceptive coverage rates (UN 2022). In 2017, the most recent year with available data, only 21.5% of women aged 15 to 49 used reversible modern contraceptive methods—such as pills, condoms, injectables, implants, and intrauterine devices (IUDs)—while 6.3% relied on traditional methods. Although birth control is provided free of charge, there remains a substantial unmet need for family planning. On average, women in union have 5.5 children, exceeding their desired family size by nearly two children (DHS 2017).

Because many commonly used contraceptive methods in Sub-Saharan Africa are administered as intramuscular injections, women generally must visit health centers to receive them from trained healthcare professionals. However, relying on facility-based services creates multiple barriers that reduce contraceptive uptake. These include high transportation costs, limited availability of clinical staff, and entrenched social norms that discourage women from seeking birth control at health centers for fear of being seen doing so.

Burundi recently introduced a new generation of non-intramuscular or subcutaneous contraceptive injections, known as Sayana Press, which can be administered by less specialized health providers, including community health workers (CHWs). While women must still visit a health center for the initial evaluation and first injection by qualified clinic staff, subsequent doses can be given by CHWs. To ensure ongoing contraceptive protection, these injections must be renewed every three months.
 

Can CHWs increase the uptake of contraceptive injectables?

A recent World Bank working paper uses a cluster-randomized controlled trial and health center data to assess the impact of allowing trained CHWs to administer the new contraceptive injections directly to women during routine home visits. The training, conducted under the supervision of the Ministry of Health’s Reproductive Health National Program, was supported by the World Bank’s Investing in Early Years and Fertility project.

The study sample comprised health centers that already stocked Sayana Press but had not yet trained CHWs to apply the renewal injections. In this field experiment, half of the 138 health centers (with all but two classified as rural) were randomly assigned to either the treatment or control group. The randomization was stratified by province, population density, the total number of contraceptive products, and the proportion of alternative injectable contraceptives delivered in the previous year.
 

The intervention worked, however…

Health facility data reveals significant increases in the number of injections administered. For the average health center, we find that the number of Sayana Press injections increased by 13 per month, representing roughly a 70% increase over the mean of the control group. Of these additional injections, roughly 3.3 are initial shots. Furthermore, this increase remained statistically significant throughout the theoretical and practical training phases, as well as after CHWs were certified to administer renewal injections.

The training of CHWs to administer the subcutaneous contraceptive injections led to important substitution effects, with a shift away from long-acting contraceptive methods such as implants and IUDs. On average, health centers experienced a monthly decrease of 2.7 in the use of these devices. Given the much shorter protection period of the new subcutaneous injections compared to long-acting methods (three months versus three years), the negative impact on the use of long-acting contraceptives appears to largely offset the benefits of the intervention: exploratory estimates suggest that the intervention resulted in a net increase of only about five additional women per health center being protected throughout the analysis period. This effect on overall contraceptive coverage is modest relative to the size of the childbearing female population within the catchment areas of health centers and is statistically insignificant.
 

Policy Implications

This paper demonstrates that expanding contraceptive provision to CHWs has the potential to boost the adoption of the latest generation of contraceptive injectables in rural villages. This approach addresses persistent barriers such as long distances to health centers, limited availability of health professionals, and the pervasive adverse social norms associated with contraceptive provision. By training CHWs to administer injections locally, the scale-up of family planning services can be expedited in the country and region.

Nonetheless, it is crucial to focus on ensuring continuous protection, especially if women transition from longer-acting methods to this innovative alternative. This becomes even more critical considering evidence indicating that injectables' discontinuation rates are roughly double those of implants in Burundi. 

Policymakers should focus on sustaining contraceptive coverage for current users while expanding access for new adopters. Behavioral strategies for injection renewals, such as targeted media campaigns or mobile reminders, show promise. The approach that utilizes community health workers can be useful and relevant for other countries and contexts facing similar challenges in family planning and contraceptive access.  

Future research should explore mitigating the displacement of long-acting methods, ensuring continuous coverage, and evaluating the feasibility of self-administered contraceptive injectables, already available in some sub-Saharan African countries.

The study was a collaboration between the Burundian Ministry of Health, the World Bank health team, and the World Bank Development Impact (DIME) team. It was funded by the Strategic Impact Evaluation Fund.