A Community-Based, Mobile Electronic Medical Record System App for High-Quality, Integrated Antiretroviral Therapy in Lilongwe, Malawi: Design Process and Pilot Implementation
Bibliographic record
Abstract
BACKGROUND: Differentiated service delivery (DSD) increases antiretroviral therapy (ART) access in sub-Saharan Africa by moving clients out of congested ART clinics to communities for care. However, DSD settings challenge provider adherence to complex, chronic care treatment guidelines and have burdensome systems for client monitoring and evaluation (M&E), reducing data for decision-making. Electronic medical record systems (EMRS) improve client outcomes and reduce M&E workload. Traditional EMRS cannot operate in most DSD settings with unreliable power and poor connectivity. OBJECTIVE: This study aims to detail the human-centered design (HCD) process of developing a mobile EMRS for community-based DSD services in Lilongwe, Malawi. METHODS: Lighthouse Trust (LT) operates 2 Ministry of Health (MoH) clinics in Lilongwe, Malawi, with a combined total of >35,000 ART clients. LT's real-time, point-of-care EMRS collects complex client M&E data and provides decision-making support, ensuring adherence to integrated HIV and tuberculosis guidelines that optimize client and program outcomes. LT's EMRS scaled to all large MoH ART clinics. LT also implements a nurse-led community-based ART program (NCAP), a DSD model to provide ART and rapid assessment for 2400 stable LT clients in the community. LT, alongside collaborators, from the University of Washington's International Training and Education Center for Health and technology partner, Medic, used the open-source Community Health Toolkit (CHT) and HCD to develop an open-source, offline-first, mobile EMRS-like app, "community-based ART retention and suppression" (CARES). CARES aims to bring EMRS-like provider benefits to NCAP's DSD clients. RESULTS: CARES design took approximately 12 months and used an iterative process of highly participatory feedback sessions with provider, data manager, and M&E team inputs to ensure CARES optimization for the NCAP and LT settings. The CARES mobile EMRS prototype supports NCAP providers with embedded prompts and alerts to ensure adherence to integrated MoH ART guidelines, aiming to improve the quality of client care. CARES facilitates improved data quality and flow for NCAP M&E, aiming to reduce data gaps between community and clinic settings. The CARES pilot demonstrates the potential of a mobile, point-of-care EMRS-like app that could benefit NCAP clients, providers, and program teams with integrated client care and complete M&E data for decision-making. CARES challenges include app speed, search features to align longitudinal records, and CARES to EMRS integration that supports timely care alerts. CONCLUSIONS: Leveraging the CHT and HCD processes facilitated the design of a locally specified and optimized mobile app with the promise to bring EMRS-like benefits to DSD settings. Moving from the CARES prototype to routine NCAP implementation should result in improved client care and strengthened M&E while reducing workload. Our transparent and descriptive process shares the progress and pitfalls of the CARES design and development, helping others in this digital innovation area to learn from our experiences at this stage.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".