Annual journal · Volume I
The first annual collection of research, reviews, commentaries and scholarly work from the LASUMSA Research & Journal Club.
LASUMSA
Research + Journal Club
Annual Journal
Evidence, Method and Doubt
Vol. I · 2026
01
2 articles
01.01
F. Adeyemi, O. Eze · pp. 1-6
Studies that fail to demonstrate an effect are systematically less cited, less discussed in teaching, and less likely to be read in full. We argue that negative results are the cheapest available defence against medical error, and review three cases where a null finding changed practice more than a positive one.
publication biasevidence synthesiscritical appraisal
01.02
O. Eze, A. Balogun · pp. 19-26
Randomisation is treated as a single, decisive property of a study. It is not. This commentary separates the guarantees that randomisation of allocation provides from those it does not, and explains why readers routinely extend its protection to claims the design never supported.
randomisationstudy designinternal validity
02
3 articles
02.01
N. Bello, T. Okafor · pp. 27-38
A review of the requirements a screening test must meet before adoption in a low-resource setting, covering spectrum bias, pathway effects, capacity displacement, and the difference between diagnostic accuracy and clinical benefit. We propose a pre-adoption checklist.
screeningearly detectionimplementationhealth systems
02.02
O. Eze, T. Okafor, N. Bello · pp. 47-58
Benchmark-driven medical image segmentation optimises for a metric on a fixed dataset. We review the gap between that objective and clinical usefulness, covering boundary-sensitive metrics, out-of-distribution evaluation, and lesion types systematically excluded from public benchmarks.
image segmentationbenchmarkingevaluation methodologyneuro-oncology
02.03
A. Balogun, F. Adeyemi · pp. 59-68
Most discussions of research capacity treat supervision as a personal relationship. We argue it should be treated as institutional infrastructure with predictable failure modes, and offer a minimal specification for research mentorship in a medical school with limited faculty time.
mentorshipresearch capacitymedical education
03
1 article
03.01
A. Balogun, T. Okafor, N. Bello, F. Adeyemi · pp. 7-18
We audited abstracts published in five years of four Nigerian medical journals for adherence to CONSORT and STROBE reporting items. Of 412 abstracts reviewed, 61% met fewer than half of the applicable reporting items. Items concerning sample size justification, missing data, and conflicts of interest were reported in under a fifth of studies. Reporting quality did not improve over the study period.
research reportingCONSORTSTROBENigerian medical journals
04
2 articles
04.01
F. Adeyemi · pp. 39-46
A practical guide for medical students reading their first research papers, organised around the questions that most efficiently separate a trustworthy study from a misleading one. Intended to be read before the journal club session, not instead of it.
critical appraisalmedical educationreading skills
04.02
N. Bello, O. Eze, A. Balogun, T. Okafor, F. Adeyemi · pp. 69-80
A cross-sectional survey of 214 final-year students examining participation in research, perceived barriers to participation, and interest in postgraduate research training. Participation was associated with prior research exposure rather than with academic performance.
student researchsurveymedical educationresearch barriers
Editorial
Editorial Board, LASUMSA Research & Journal Club
Medical students are taught to be confident. We are taught the mechanisms of disease as though they were settled, the indications for treatment as though the evidence beneath them were permanent, and the guidelines as though they would not be revised within a decade of our graduation.
Almost none of that is true, and almost none of it is our fault. What we are rarely taught is how to hold a claim at the right strength — to say what a study establishes, what it merely suggests, and what it cannot address at all.
This volume is a record of a year spent trying to do exactly that. Its contents include one original audit, one methodological commentary, one student survey, and a substantial number of pieces about how we are taught to read evidence and how easily we misread it.
We have not tried to make these pages definitive. We have tried to make them honest about their own limitations, which is the standard we ask of the papers we appraise.
Volume I was assembled by the editorial board from work completed during the 2026 calendar year. Articles are published in the order in which they were accepted, and each carries the name of every member who contributed.
Where an article's conclusions are narrower than the questions its authors began with, we have preferred to say so.
How this issue’s appraisals were developed, month by month, in the open.
Read the journal club archive