A completed trial that never reports its results does not disappear quietly. Participants still took the risk. The site still did the work. The funding was still spent. What disappears is the evidence, and with it the only public record that any of it was done properly.
That makes results reporting a governance function, not an administrative one. Registration entries and posted results are among the very few indicators of research quality that anyone outside an institution can verify for themselves. A funder, a regulator, a journal editor, or a community advisory board can check them without being granted special access. For African networks building trial capacity now, that fact is worth pausing on. Reporting discipline is a credibility signal that can be demonstrated rather than claimed.
The current picture is mixed rather than dire. A 2026 analysis of 3,026 completed or terminated randomised controlled trials with at least one site in sub-Saharan Africa found journal publications for 65.5% of them, leaving 34.5% unpublished, at a median of 34.2 months from primary completion to publication (1). That is better than the global picture: the most recent Cochrane review, covering 204 studies tracking 165,135 trials, found only 53% published in full (2). African trials are not the outlier here.
But 34 months sits well outside the 12-month registry and 24-month publication timeframes set out in the World Health Organization (WHO) joint statement on public disclosure of results (3). Being ahead of the global average is not the same as being good enough.
Why results go unreported
The reasons are structural, and the evidence is now African rather than inferred. In a 2026 survey of 409 principal investigators based in sub-Saharan Africa, 74% reported never having received training in writing trial-specific manuscripts. Among those who had failed to publish at least one completed trial, the most common barriers were journal rejection (30.5%) and time constraints (29.3%). Unfavourable or non-significant findings accounted for 14.6% (4).
The distribution matters more than the total. The dominant narrative around non-reporting emphasises suppression: sponsors burying inconvenient results, investigators protecting their reputations. That happens. But in sub-Saharan Africa, the larger problem is capacity. Investigators who lack writing support, protected time, or statistical assistance to turn completed work into published evidence. Addressing suppression while ignoring capacity will not move the numbers where they most need to move.
Registry design compounds the gap. The Pan African Clinical Trials Registry (PACTR) only made its results-reporting field mandatory in 2019, following a 2018 relaunch to carry the full WHO 24-item dataset (5). A system never built to capture results will not produce them by default.
What non-reporting costs in practice
The clearest recent illustration came from the pandemic. A cohort study led from Stellenbosch University and the South African Medical Research Council tracked 357 registered randomised trials of hydroxychloroquine, corticosteroids, and vitamin D for COVID-19. By October 2022, 250 of them, 70%, had not published or otherwise released results. Many registry entries had not even been updated to record that the trial had ended (6).
That was precisely the period when African clinicians, ethics committees, and regulators were making treatment decisions in real time. Seven in ten of the relevant trials gave them nothing to work with. Trials continued to be registered for questions that already had high-certainty answers (6). The cost was not abstract: it was participants enrolled to re-answer settled questions, and clinical guidance written on a partial record.
Four commitments, four separate problems
Transparency is often discussed as a single obligation. It is at least four distinct ones, each with different barriers, and collapsing them into general ask makes all four easier to avoid.
Prospective registration. Registration before the first participant is enrolled, in a registry meeting WHO standards. PACTR, hosted by the South African Medical Research Council, is the continent’s WHO primary registry. Of 2,998 trials registered by August 2021, 55% were registered prospectively (7). The infrastructure exists. The consistency does not, yet.
Summary results reporting. Key outcomes posted to the registry within 12 months of completion, regardless of what they show (3). This is the weakest link: of 1,083 completed trials on PACTR, only 94, roughly 3%, had results available in the registry (5). Journal publication is not a substitute, and behind a paywall it does not reach the regulators and clinicians who need it.
Protocol availability. Full protocols with amendments, published so readers can assess whether the reported analysis matches the planned one. Without this, selective outcome reporting remains invisible.
Participant-level data sharing. Governed access, not open release. The H3Africa Consortium operates a working African model: controlled through a Data and Biospecimen Access Committee with continental membership, conditions favouring collaboration with African scientists, and a publication embargo protecting the researchers who generated the data (7, 8). An important distinction: that framework governs genomic data and biospecimens, not trial results. The governance principles transfer. The machinery for clinical trials does not yet exist.
Why this is strategic, not only ethical
Complete reporting is what turns locally generated evidence into something the rest of the world is obliged to take seriously. Long-acting injectable therapy for HIV had been tested primarily in European and North American populations. The CARES trial, conducted across sites in Kenya, Uganda, and South Africa, registered on PACTR and reported at 96 weeks in 2025, demonstrated non-inferiority in African adults and concluded that the regimen may be considered for African treatment programmes (9). That recommendation exists because a trial was registered, completed, and reported. Without the final step, it would still be an extrapolation from other populations: useful, perhaps, but never authoritative.
The broader incentives reinforce the same logic. Sponsors select sites on delivery history they can verify. Regulators weigh evidence they can see in full. Funders answer to their boards on documented outputs. WHO resolution WHA75.8, adopted in 2022, calls on member states to tie registration and timely reporting of both positive and negative results to research funding conditions (10). Networks with a reporting record already in place will be well positioned as that standard tightens.
This is not about performing accountability for external audiences. It is about building the institutional credibility that determines whether African-generated evidence shapes treatment guidelines, regulatory decisions, and investment flows, or whether those decisions continue to be made on data from somewhere else.
ACRN’s position
The Africa Clinical Research Network treats reporting as part of conducting a trial, not as an optional step that follows it. Trials we coordinate are registered before enrolment, and results are reported whether or not they support the hypothesis. We would rather be measured on that record than describe our intentions.
The discipline this requires
None of this demands extraordinary resources. A registry entry completed before the first participant is enrolled. A results summary filed within twelve months by someone whose job description includes filing it. Analysis and writing time budgeted at proposal stage rather than absorbed by whoever has capacity afterwards. A registry entry updated when a trial stops early.
African research is being assessed right now, by funders, by regulators, and by the communities that enrol in its studies. Reporting is the part of that assessment institutions control entirely. The credibility gap in global clinical research is real. It will not be closed by institutions content with the status quo. It will be closed by those willing to build a verifiable record, starting with the most basic commitment: when a trial is done, say what was found.
References
- Hohlfeld A, Bennie LDM, Kredo T, Clarke M. Publication bias and time to the publication of randomised trials conducted in sub-Saharan Africa: a cross-sectional study. Trials. 2026;27(1):456.
- Showell MG, Cole S, Clarke MJ, DeVito NJ, Farquhar C, Jordan V. Time to publication for results of clinical trials. Cochrane Database of Systematic Reviews. 2024(11).
- World Health Organization (WHO). Joint statement on publication disclosure of results from clinical trials. Technical document. 2017.
- Hohlfeld A, Kredo T, Clarke M. Barriers and enablers for publishing results of randomized trials: a cross-sectional survey of principal investigators in sub-Saharan Africa. Research Integrity and Peer Review. 2026;11(1):29.
- Ndwandwe DE, Runeyi S, Pienaar E, Mathebula L, Hohlfeld A, Wiysonge CS. Practices and trends in clinical trial registration in the Pan African Clinical Trials Registry (PACTR): a descriptive analysis of registration data. BMJ Open. 2022;12(1):e057474.
- Fincham L, Hohlfeld A, Clarke M, Kredo T, McCaul M. Exploring trial publication and research waste in COVID-19 randomised trials of hydroxychloroquine, corticosteroids, and vitamin D: a meta-epidemiological cohort study. BMC Medical Research Methodology. 2024;24(1):19.
- H3Africa Consortium. H3Africa Consortium Biospecimen Sharing, Access and Release Policy. 2021.
- Rebai A, Abayomi A, Andanda P, Kerr R, Herbst K, Mabuka J, et al. Responsible governance of genomics data and biospecimens in the context of broad consent: experiences of a pioneering access committee in Africa. BMJ Global Health. 2025;10(2):e016026.
- Kityo C, Mambule IK, Musaazi J, Sokhela S, Mugerwa H, Yawe I, et al. Cabotegravir and rilpivirine for treatment of HIV infection in Africa: week 96 results from the phase 3b randomized, open-label, noninferiority CARES trial. Nature Medicine. 2026;32(1):168–77.
- World Health Assemby. Resolution WHA75.8: Strengthening clinical trials to provide high-quality evidence on health interventions and to improve research quality and coordination. 2022. Report No: WHA75.8.