AskDoki: Evaluation of a mobile health intervention to improve sexual and reproductive health outcomes among Kenyan adolescents living inurban informal settlements
openFIC - John E. Fogarty International Center for Advanced Study in the Health Sciences
Abstract/Project Summary
Despite recent improvements in adolescent sexual and reproductive health (ASRH) outcomes in Kenya, rates
of unsafe sex, unplanned pregnancy, abortion, and sexually transmitted infections remain high. While schoolbased comprehensive sexuality education (CSE) improves ASRH outcomes, the Kenyan government tends to
support less thorough programs, and teachers face challenges (e.g., lack of time, training, resources) that
hinder sexuality education instruction. Adolescents who do not receive adequate CSE instruction lack the
information needed to make informed sexual and reproductive health decisions. Thus, there is pressing need
to identify innovative interventions to provide CSE outside the formal educational system. Mobile health
(mHealth) technology may provide a cost-effective solution. Amref Health Africa–Kenya developed/tested an
artificially intelligent mHealth WhatsApp chat bot, AskDoki, to deliver CSE. Preliminary data suggest improved
effectiveness and outreach might be obtained, especially among adolescents who lack phones, by pairing
AskDoki with “tech savvy” youth peer leaders or “Digital Champions” (DCs). In collaboration with Amref, we will
implement AskDoki among adolescents via DC outreach and facilitation to address our specific aims: R21 Aim
1a) Examine the effectiveness of AskDoki-delivered CSE in improving key determinants of ASRH
behaviors/ outcomes via a quasi-experimental, two-group (intervention vs. control clusters), non-equivalent,
pre/post survey design with adolescents aged 15-19. Primary outcomes for comparison will be increases in
perceived control (e.g., condom-use self-efficacy), positive attitudes (e.g., condom use), subjective norms (e.g.,
peer support of condom use), knowledge (e.g., pregnancy, STIs), and intentions (e.g., to use condoms). R21
Aim 1b) Conduct a process evaluation of the DC strategy to promote/facilitate AskDoki access. We will
collect data from multiple sources to assess adolescent reach, acceptability, and uptake of the DC strategy.
We will also conduct two sets of focus groups: with Aim 1a adolescents to explore DC perceptions and
experiences and to identify improvement areas; and with DCs to explore barriers and facilitators. R21 Aim 2)
Conduct cost effectiveness analyses for the AskDoki and DC strategy. We will track start-up, operational,
and incremental per-use costs and compute cost per adolescent served and cost per adolescent who makes
progress on one and on at least three outcome measures. R33 Aim 3) Evaluate a scaled deployment of
AskDoki using RE-AIM framework. We will examine Reach (adolescents accessing AskDoki), Effectiveness
(Aim 1 outcomes, condom use, pregnancy reduction, willingness to test/treat STIs, use of SRH services);
Adoption (representativeness of AskDoki users, characteristics of curriculum completers); Implementation
(interviews with adolescents to explore barriers, facilitators, scaled deployment cost); and Maintenance
(sustained effectiveness of primary/secondary outcomes). Findings may inform scalable interventions that
provide U.S. adolescents with health information that ultimately helps to improve youth outcomes.
Up to $71K
health research