This protocol describes a human-centered design study to co-design, develop, and evaluate 'Dəskəs,' a social media-based AI-powered chatbot platform to support mental wellness among adolescents and young people aged 18-24 in Kenya across urban, peri-urban, and rural contexts.
Key Findings
Background
Mental health disorders are identified as a leading cause of disability among adolescents and young people in Kenya, with access severely limited by multiple barriers.
Barriers to mental health support include shortages of trained professionals, stigma, financial barriers, and geographical inequities.
Evidence on the feasibility, usability, and acceptability of AI-powered mental health tools for young people in sub-Saharan Africa is described as 'limited.'
The study is positioned to address gaps in digital mental health intervention evidence in low- and middle-income country (LMIC) settings.
Methods
The study uses a three-phase human-centered design (HCD) methodology to co-design and evaluate the 'Dəskəs' digital mental wellness platform.
The three phases are: formative research and co-design, platform development and iterative refinement, and implementation with monitoring and evaluation.
Adolescents and young people aged 18 to 24 years will be purposively recruited from three sites: Kibra (urban), Kikuyu Town (peri-urban), and Nachu (rural).
Formative data will be collected through focus group discussions to inform platform design, content, and functionality.
Participants will use the platform for 2 months during the implementation phase.
Methods
The study employs a mixed-methods data collection approach combining quantitative behavioral health screening and qualitative usability assessment.
Quantitative data will include baseline behavioral health screening using the Global Appraisal of Individual Needs-Short Screener (GAIN-SS) and platform use analytics.
Qualitative data will be collected through interviews and feedback sessions to assess usability, user experience, acceptability, and perceived value.
Data will be analyzed using descriptive statistics and thematic qualitative analysis.
Primary objectives include assessing feasibility of implementation, usability and user experience, acceptability, and preliminary applicability across diverse contexts.
Methods
The study timeline indicates that field activities began in January 2026, with formative data collection planned for completion by end of February 2026.
Usability and user experience testing was planned to run from March 2026 to June 2026.
The study was funded by the Royal Academy of Engineering, UK in March 2025.
Ethical approval was granted by the Strathmore University Institutional Scientific Ethics Review Committee (SU-ISERC; SU-ISERC 2980/25), approved from November 4, 2026.
Findings from formative work, baseline data, and usability and user experience testing will be reported in subsequent original research publications.
Background
The platform 'Dəskəs' is designed as a social media-based intervention featuring an AI-powered chatbot to deliver culturally relevant, low-stigma mental wellness support.
The platform specifically targets adolescents and young people in an LMIC context (Kenya).
The intervention is described as leveraging AI to 'expand access to culturally relevant, low-stigma mental wellness support.'
The study addresses the gap that AI-powered mental health tools designed for young people in sub-Saharan Africa lack evidence on feasibility, usability, and acceptability.
The protocol is described as 'adolescent-centered' and 'ethically grounded.'
Conclusions
The study is expected to contribute to the evidence base on responsible design of AI-powered mental health tools and inform future scale-up of digital mental wellness interventions in Kenya and similar settings.
Findings are anticipated to be relevant to other LMIC settings beyond Kenya.
The study aims to inform 'accessible, culturally relevant digital mental wellness interventions.'
The protocol explicitly frames its contribution as informing 'responsible design and implementation of AI-powered mental health tools for adolescents and young people.'
What This Means
This research describes the planned design and testing of a digital mental health tool called 'Dəskəs' — an AI-powered chatbot delivered through social media — for young people aged 18 to 24 in Kenya. The study recognizes that mental health conditions are a major cause of disability among Kenyan youth, but that access to mental health support is severely restricted due to too few trained professionals, social stigma, costs, and geographic distances from services. The researchers aim to co-create the platform with young people themselves, recruiting participants from urban, peri-urban, and rural communities across three sites, so that the tool reflects the real needs and cultural contexts of its intended users.
The study follows a three-phase process: first, gathering information through group discussions to guide how the platform is built; second, developing and refining the platform; and third, having participants use it for two months while researchers collect data on how well it works and how people feel about using it. Both survey-style questionnaires (including a standardized behavioral health screening tool) and in-depth interviews will be used to understand the platform's usability, acceptability, and real-world value.
This research suggests that AI-powered digital tools could be a promising way to reach young people in low- and middle-income countries who currently lack access to mental health support, particularly because they may reduce stigma and overcome geographic and financial barriers. The findings are intended to help build an evidence base for responsible, culturally appropriate digital mental health tools that could eventually be expanded more broadly across Kenya and similar settings. As this is a protocol paper, no results have yet been reported.
Macharia P, Misoi I, Ngaruiya N, Mbwayo A, Jabin M. (2026). Enhancing Mental Wellness in Kenyan Adolescents and Young People: Protocol for a Platform Development and Usability Study.. JMIR research protocols. https://doi.org/10.2196/92983