UCSF

Background: In many parts of Sub-Saharan Africa, adolescent sexual and reproductive health is often of low political priority and there are often restrictive laws and policies that are in contradiction with international agreements and commitments.

Objectives: The main objectives of this dissertation research were 1) to apply the Public Arenas Model to appraise the environments, definitions, competition dynamics, principles of selection, and current actors involved in problem-solving and prioritizing adolescent pregnancy as a policy issue; 2) to qualitatively explore the generation, process, constraints, dilemmas, and institutionalization of political priority for adolescent sexual and reproductive health in Kenya, and 3) to compare pregnancy incidence among women using depo medroxy progesterone acetate levonorgesteral implants and intra uterine copper devices within a multicenter, open-label, randomized clinical trial.

Methods: The research used mixed research methodology. For Aim 1, We applied the Public Arenas Model to critically appraise the environments, definitions, competition dynamics, principles of selection, and current actors involved in problem-solving and prioritizing adolescent pregnancy as a policy issue. To achieve Aim 2, a postmodern, interpretive focused ethnographic approach was used. This included a critical review of the empirical and theoretical literature, existing national documents, and participant interviews. We used the Shiffman and Smith policy framework consisting of four categories—actor power, ideas, political contexts, and issue characteristics—to analyze factors that have shaped political prioritization of adolescent sexual and reproductive health. We undertook semi-structured interviews with members of adolescent sexual and reproductive health networks at the national level and conducted thematic analysis of the interviews. For Aim 3, we analyzed data from the ECHO Trial, which assessed HIV incidence among 7829 women from 12 sites in Eswatini, Kenya, South Africa, and Zambia who were seeking effective contraception and consented to be randomized to DMPA-IM, a copper IUD, or an LNG implant. Cox proportional hazards regression adjusted for condom use was used to compare pregnancy incidence during both perfect use (defined as from initiation of method until first discontinuation for any reason). Additional analyses explored more typical use (i.e., until decline or change to a different contraceptive method).

Summary of Findings:

In Aim 1, we found existing definitions center around adolescent pregnancy as a “disease” that needs prevention and treatment, socially deviant behavior that requires individual agency, and a national social concern that drains public resources and therefore needs to be regulated. These conflicting definitions contributed to the rarity of the topic achieving traction within the political agenda and may result in conflicting solutions, such as lowering the legal age of consenting to sex, while limiting access to contraceptive information and services to minors.

In Aim 2, we found that the adolescent sexual and reproductive health community was diverse and united in adoption of international norms and policies, but lacked policy champions to provide strong leadership, and policy windows were often missed. Community members lacked consensus on a cohesive public positioning of the problem. Moreover, the perception of adolescents as lacking political power made politicians reluctant to act on the existing data on the severity of adolescent SRH. There was also a lack of consensus on the nature of interventions to be implemented. The sectorial funding by donors and government treasury brought about tension within the different government ministries resulting in siloed approaches, lack of coordination, and overall inefficiency.

In Aim 3, 7710 women contributed to this analysis. Seventy pregnancies occurred during perfect use and 85 during more typical use. Perfect use pregnancy incidence rates were 0.61 per 100 woman-years for DMPA-IM (95% CI 0.36-0.96), 1.06 for copper IUD (95% CI 0.72-1.50), and 0.63 for LNG implants (95% CI 0.39-0.96); differences between methods were not statistically significant (p>0.05). Typical use incidence rates were generally similar, although slightly higher for copper IUD (1.11 per 100 woman-years, 95% CI 0.77-1.54).