In September 2025, The Geneva Learning Foundation (TGLF) launched the first Certificate peer learning programme for equity and research and practice. Health workers, a plurality of them sub-national government staff, came together for an intensive 16-day learning journey. This article shares what we learned by examining community-based peer review feedback between learners.
A thirteen-year-old girl in Nigeria, bitten by a snake, arrived at a hospital with her frantic family.
The hospital demanded payment before administering the antivenom.
The family could not pay.
The girl died.
This story, first told in a global TGLF event, did not appear in a medical journal.
It was then written by the health worker who shared it as a course project, then read, scored, and commented on by three peers in three different countries.
Community-based peer review: the radical act of reading someone else’s work
One reviewer asked whether the root cause was really about payment policy or something deeper.
Another suggested the author connect the story to national insurance legislation.
A third shared a parallel case from her own district.
The author revised her plan.
The reviewers, by their own account, learned more from giving feedback than from writing their own projects.
This is the peer review process at the center of The Geneva Learning Foundation’s model of learning.
And it may be the most underappreciated mechanism in global health education today.
The architecture of reciprocal obligation
The process of a ‘peer learning exercise’ unfolds over sixteen days.
In the first two days, professionals from dozens of low- and middle-income countries work in small groups to practice describing and analyzing situations of health inequity.
These situations are drawn from the experience of one person in each group.
This deliberate practice helps them develop an individual project, grounded in a real situation of inequity from their own work.
They analyze root causes using a structured “Five Whys” methodology, reflect on their own role in the system, develop a community-driven action plan, and connect their local story to larger patterns of structural injustice.
Then they submit it to a hard deadline, no extensions, and receive three peers’ projects to review.
This is just in the first five days.
What makes this different from every other peer review exercise in online education is a set of deliberate design choices.
The scoring system weighs the act of reviewing at sixty percent, far more than the quality of one’s own project or the quality of feedback received.
The message is architectural, not rhetorical: in this system, giving is three times more valuable than receiving.
The rubric functions not as a grading checklist but as what the course calls “your map.” Its five criteria scaffold increasingly sophisticated thinking.
A score of two on root cause analysis means “the story is clear but has no description of the larger context.”
A score of four requires “a rich description of the larger context, including the history and the power structures that affect the people in the story.”
By the time reviewers sit down to evaluate a colleague’s project, they have already practiced applying this framework collaboratively in small groups, internalized its logic, and used it to write their own submissions.
They are not checking boxes.
They are thinking through a shared analytical architecture.
Community-based peer review: the inversion
The most radical claim in this design is also the least intuitive: the reviewer learns more than the author.
The data from 230 health professionals across English and French cohorts who participated in the Foundation’s inaugural HEART course, bear this out.
When a community health worker in Cameroon reads a project from a district nurse in South Sudan, she encounters an unfamiliar context, maps it against her own experience, and constructs new knowledge in the gap between them.
The rubric gives her the vocabulary.
The deadline gives her the obligation.
The act of writing specific, constructive feedback forces a quality of attention that passive reading never achieves.
Nearly six hundred peer reviews produced detailed, criterion-by-criterion feedback.
Reviewers did not simply evaluate.
They taught.
One reviewer, assessing a project on maternal mortality in northern Nigeria, wrote: “The list captures symptoms of the problem, but the deeper why could be framed more clearly. A possible root cause might be weak health system governance and inequitable policies that fail to prioritize rural and underserved communities.”
In the French cohort, a reviewer pushed a colleague to sharpen causal reasoning: “The root cause analysis should link each cause to the next.Listing causes without causal links between them does not lead to the true root cause.”
These reviewers were not grading.
They were modeling the analytical thinking the course exists to teach.
The mirror
Community-based peer review adds a dimension that most peer review systems lack.
After receiving feedback, learners must respond to it: rate its helpfulness, explain their rating, and describe the single most important change they plan to make.
This third step is where the deepest learning surfaces.
One learner wrote: “The comments are very useful. It is an eye opener. There are areas that I was not considering but by virtue of this review, I would consider.”
A French-speaking participant reflected: “My colleague’s comments allowed me to clearly see the weaknesses of my project.The remark about the lack of clarity in the pregnant woman’s story helped me understand that I must provide more concrete details.”
These are not polite acknowledgments.
They are evidence of metacognition, the ability to think about one’s own thinking, widely recognized in learning science as a marker of genuine transformation.
Some of the most productive moments arose from disagreement.
One reviewer challenged an author’s framing directly: “This story has shown a great weakness in the health system, but I do not think it has anything to do with unfairness.”
The author’s response was immediate: “Everyone has the right to life, and I feel it is unfair for someone to die because at that critical point, he has no relative or someone to help him.”
That exchange is exactly the kind of cognitive conflict that produces durable learning.
What the machines cannot do
The Foundation permits AI tools for editing and overcoming language barriers.
But a strict Honor Code draws a bright line: AI cannot fabricate experience or generate the core analysis that should reflect genuine critical thinking.
An algorithm can summarize a framework for identifying bias.
It cannot tell you what it felt like to watch a hospital refuse treatment to a child because the family lacked connections.
It cannot review a colleague’s plan to address vaccine hesitancy among pastoralist communities and respond with the authority of having faced the same challenge in a different country.
The learning happens in the gap between two practitioners’ experiences, a gap a language model cannot occupy.
The proof
Across both cohorts, nearly ninety percent of learners rated the feedback they received as at least “helpful”. But the numbers matter less than what participants did with the feedback during community-based peer review.
One learner committed to ensuring “every patient will receive complete health information regardless of financial status or background.”
Another pledged to hold focus groups with community members before designing interventions.
A third described scheduling regular check-ins with colleagues to seek feedback earlier in her process, transferring the peer review logic from the course into her daily professional practice.
When a reviewer from Burkina Faso recognizes her country’s challenges in a Chadian colleague’s project, or when a South African health worker shares a parallel story of a woman who died because a nurse demanded documents before providing care, the learning transcends individual cases.
It reveals structural patterns that no single expert could surface alone.
This arrives at a moment when the field of global health education is grappling with how to move beyond symbolic inclusion toward genuine reciprocity and Southern-led knowledge production.
Peer review, carefully designed, does not replace expertise.
It creates a form of collective intelligence that expertise alone cannot produce.
And in a field where the gap between knowing about inequity and doing something about it remains vast, that may be the most important pedagogical innovation we have.
References
- Cope, B., Kalantzis, M., 2013. Towards a New Learning: the Scholar social knowledge workspace, in theory and practice. E-Learning and Digital Media 10, 332. https://doi.org/10.2304/elea.2013.10.4.332
- Reda Sadki (2025). Patterns of prejudice: Connecting the dots helps health workers combat bias worldwide. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/2gp3b-v4c84
- Reda Sadki (2025). From diagnosis to duty: health workers confront their own role in inequity. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/khjjv-6et24
- Reda Sadki (2025). The practitioner as catalyst: How a global learning community is turning frontline experience into action on health inequity. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/0pa3j-2×058
- Reda Sadki (2025). L’équité compte: quand les soignants du monde entier témoignent des inégalités en santé. Reda Sadki: Learning to make a difference. https://doi.org/10.59350/qg26h-ngd65