Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Currently submitted to: JMIR Medical Education

Date Submitted: Nov 13, 2026
Open Peer Review Period: Nov 16, 2026 - Jan 11, 2027
(currently open for review)

Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

Training to Build Confidence in Clinical Artificial Intelligence for Global Health: Cross-sectional evaluations of a synchronous virtual course

  • Bethany Hedt-Gauthier; 
  • Audrey Yao; 
  • Sildio Mbonyumuhire; 
  • Annie Michaelis; 
  • Joel Mubiligi

ABSTRACT

Background:

The transition from promising artificial intelligence (AI) models to improvements in routine clinical care remains limited, particularly in low- and middle-income countries (LMICs). An important barrier to clinical AI adoption in LMICs is limited workforce preparation. Existing AI training programs often emphasize introductory knowledge over critical appraisal, implementation, and governance. Evaluations of AI training programs targeting health workforces in LMICs also remain limited.

Objective:

To describe the AI for Global Health course, including course structure, participant reactions, self-reported knowledge, and attitudes toward AI.

Methods:

Partners In Health (PIH) and the University of Global Health Equity offered five two-hour synchronous online sessions to 38 participants across two cohorts. Sessions covered AI applications, machine-learning development and evaluation, ethics and equity, integration into health systems, and PIH’s mission. The Kirkpatrick–Barr framework assessed learners’ reactions (Level 1) and attitudes or perceptions (Level 2a). Participants were invited to complete seven voluntary, anonymous surveys: a pre-course survey; brief evaluations following Sessions 1 through 4; a survey preceding Session 5; and a final course evaluation. Because the surveys did not collect identifiers, responses could not be linked across time.

Results:

Thirty-four (89.5%) course participants responded to the overall course evaluation after the fifth and final session. Of these, 28 of 34 (82%) rated their satisfaction as 6 or 7, 30 of 34 (88%) rated the course as worth their time as 6 or 7, and 29 of 34 (85%) indicated a 6 or 7 for willingness to enroll again. The proportion reporting good or expert knowledge was 2 of 34 (6%) for AI bias and fairness pre-course and 23 of 34 (68%) at final evaluation; for evaluating whether AI was a good fit for a problem or setting, the corresponding proportions were 2 of 34 (6%) and 20 of 34 (59%). Strong agreement that AI could be trusted at current PIH sites was reported by 5 of 34 (15%) pre-course respondents and 9 of 22 (41%) before Session 5. Interim evaluation response rates were 30 of 38 (78.9%) following Sessions 1 and 2, 28 of 38 (73.7%) following Session 3, and 24 of 38 (63.2%) following Session 4. Median pace ratings were 5 across all four sessions. Median content ratings were 6 for Sessions 2 and 3, above the intended range of 4 to 5.

Conclusions:

Participants reported high overall satisfaction, and self-reported knowledge was higher at final evaluation across all eight domains. Surveys were anonymous, so pre-course and later differences represent group-level patterns. Future offerings will provide more time for the same core content while retaining the case examples, discussions, and applied activities that connect course concepts to participants’ work. The evaluation did not establish changes in practice or AI implementation.


 Citation

Please cite as:

Hedt-Gauthier B, Yao A, Mbonyumuhire S, Michaelis A, Mubiligi J

Training to Build Confidence in Clinical Artificial Intelligence for Global Health: Cross-sectional evaluations of a synchronous virtual course

JMIR Preprints. 13/11/2026:116898

DOI: 10.2196/preprints.116898

URL: https://preprints.jmir.org/preprint/116898

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.