# An offline-first electronic health record for vulnerable populations: A mixed-methods feasibility study

**Authors:** Henry Ashista, Alanis Santiago Comas, Taylor Selby, Mohammad Yasir Essar, Jude Alawa, Samar Al-Hajj, Erica Nelson, J. Ansermino, Ziad El-Khatib, Leo Anthony Celi

PMC · DOI: 10.1371/journal.pdig.0001204 · PLOS Digital Health · 2026-02-13

## TL;DR

This study evaluates an offline-first electronic health record (EHR) in low-resource clinics, finding it feasible and beneficial for patient care despite some syncing challenges.

## Contribution

The paper introduces and evaluates a free, open-source, offline-first EHR designed for low-resource settings and vulnerable populations.

## Key findings

- After training, clinicians found the HH EHR easy to use and it reduced patient interview times by three minutes.
- The system faced challenges in settings requiring frequent syncing due to inconsistent network connectivity.
- Clinics observed improved documentation, fewer medical errors, and better patient outcomes with the EHR.

## Abstract

While there is a growing body of research showing that Electronic Health Records (EHRs) can improve health outcomes in limited resourced settings, these EHRs often require continual access to an internet connection and are challenging to customize and deploy. To address these challenges, Hikma Health (HH) has designed a free, open-source and offline-first EHR for physicians providing care in low-resource settings. This paper describes a mixed-methods study to understand the feasibility of the HH EHR in two clinics that care for displaced or rural populations in Lebanon and Nicaragua. Clinic demographics and metrics were collected through REDCap and descriptive statistics were analyzed. Using a framework analysis, in-depth interviews were conducted at both sites until thematic saturation. Interviews were coded by three authors with an inter-rater reliability kappa score of 7.6. Quantitative data showed that after about three hours of training and three weeks of use, participants were comfortable using the HH EHR and the HH EHR decreased patient interview times by three minutes. Quantitative data showed that these findings were due to the ease and simplicity of the modular workflows. However, when used in settings that required syncing multiple times during a patient’s encounter with different clinicians, the system faced challenges due to inconsistent network connectivity and the design of the sync functionality. The HH EHR was a feasible solution for offline-first settings. Both clinical sites observed that the implementation of the EHR enhanced documentation, decreased medical errors, and improved patient outcomes through tracking medications and diseases.

Research has shown that electronic health records help improve patient care and health systems management. However, electronic health records are often expensive, require continuous internet connectivity, and are not designed by clinicians. This makes implementing electronic health records in limited resource settings challenging and prevents their use in caring for patients who need them most. In this paper, we studied the implementation of a free, open-sourced, offline-first electronic health record that doesn’t require continuous internet activity across two limited-resource clinics in Lebanon and Nicaragua. We found that an off-line first electronic health record was feasible and helped improve both quantitative clinic metrics as well as clinicians’ views of their ability to care for patients. We also found that an offline-first electronic health record presents challenges when clinics require continuous syncing to the cloud throughout a patient’s encounter. More innovation is needed to help develop electronic health records in these settings. When developed in line with human-centric design processes, specifically sgeared towards use, such EHRs help improve the care provided to often displaced and migratory populations.

## Full-text entities

- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12904448/full.md

## Figures

2 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12904448/full.md

## References

22 references — full list in the complete paper: https://tomesphere.com/paper/PMC12904448/full.md

---
Source: https://tomesphere.com/paper/PMC12904448