BMJ's Journal of Medical Genetics has seven of eight papers from a single special issue. The retraction notice described an "irreparably compromised" review process: a guest editor had handpicked the reviewers, most of them affiliated with one university, and the journal's own investigation uncovered problems across nearly every article. One issue, seven retractions, years of potentially misleading data sitting in the scientific record before anyone pulled the plug.
Isolated as it sounds, the episode fits a pattern that has been widening for years. Retraction Watch now catalogues more than 64,000 withdrawn papers in its database. Recent bibliometric work examining 46,000 retractions across ten major publishers, reported by Chemistry World, confirmed that compromised peer review ranks among the most frequent reasons manuscripts are pulled. Meanwhile, a 2026 Australian survey of 139 journal editors found that 55% rate securing qualified reviewers as a significant or very significant challenge, with some editors sending upward of thirty invitations just to land two referees for a single manuscript.
How a Bad Kidney Study Reaches the Ward
Sixty-four thousand retractions sounds like a publishing industry problem. In clinical medicine, though, a poorly reviewed study on a diagnostic marker or treatment protocol does not merely clutter a reference list. It can shift the way physicians evaluate patients, choose therapies, and define thresholds for intervention. Nephrology, for instance, depends on published biomarker research to guide early detection of kidney damage, an area where delayed diagnosis often means the difference between manageable disease and irreversible organ failure. If the studies behind those decisions slip past reviewers unchecked, the mistakes can settle into clinical routines long before anyone issues a correction.
Mariia Dolinna has worked on both ends of that process. An associate professor at the Department of Clinical Pharmacology, Pharmacy, Pharmacotherapy and Cosmetology at Zaporizhzhia State Medical and Pharmaceutical University in Ukraine, she is also a nephrologist with years of clinical experience in the nephrology department of Zaporizhzhia Regional Clinical Hospital. Since 2026, she has been reviewing manuscripts for the European Journal of Medical Research and Scientific Reports; her work with the Ukrainian Journal of Nephrology and Dialysis in the same capacity began in 2024. Few reviewers bring her particular combination: she graduated with honors and holds two diplomas — in medicine and pharmacy, along with research credentials, a patent, and hands-on clinical experience in the exact field she evaluates on paper.
"A reviewer who only understands the laboratory side might miss that a proposed diagnostic cutoff would be impractical in a ward setting," she says. "Someone who only knows the clinic might not catch a flaw in the statistical model. Evaluating a manuscript properly requires both angles."
The Ward Where Standard Tests Kept Falling Short
Most of that career took shape at a single institution. After earning both diplomas at Zaporizhzhia State Medical University, Dolinna entered clinical residency, then doctoral study. Patients she encountered in the nephrology ward, particularly those with hypertension whose kidneys were quietly deteriorating, posed a question that standard indicators could not fully answer: how do you identify organ damage before it advances beyond the point where intervention helps?
Dolinna's 2016 dissertation for the degree of Candidate of Medical Sciences went after that question directly. Focused on improving the diagnosis and treatment of glomerular apparatus and tubulointerstitial kidney tissue damage in patients with hypertensive disease and chronic glomerulonephritis, the research yielded a patented diagnostic method registered in Ukraine. Where conventional assessments depend on filtration rate and urinary albumin, her approach layers in additional biomarker measurements: neutrophil gelatinase-associated lipocalin and interleukin-18, assessed in both blood and urine, to flag tissue injury and inflammatory activity at an earlier stage. Hospital departments in nephrology, cardiology, and therapeutics across Ukraine adopted the method, and university teaching programs incorporated it into their curricula.
Minsk, Astana, Tbilisi: Putting the Research in Front of Colleagues
Between 2014 and 2019, Dolinna presented her findings to professional audiences in four countries. At a 2019 meeting in Minsk, she examined kidney injury markers and the nephroprotective effects of ramipril. Astana hosted her a year earlier, where she laid out an analysis of clinical-morphological features and treatment outcomes for focal segmental glomerulosclerosis. Back in 2014, a Tbilisi talk focused on morphological patterns in patients with mesangioproliferative glomerulonephritis complicated by arterial hypertension. Additional abstracts from Ukrainian conferences between 2016 and 2019 appeared in published proceedings.
Her published journal articles on clinical-pharmacodynamic effects of ramipril in hypertensive patients, on AL-amyloidosis case analysis, and on the influence of ACE inhibitors on indicators of tubular kidney injury in chronic glomerulonephritis, put Dolinna through the same gauntlet she now runs for others: designing methodology, defending results, responding to reviewer critiques. Having been on the receiving end of that scrutiny changes how you read someone else's manuscript.
"Going through peer review as an author teaches you where the weak points usually are," she says. "You start recognizing the same patterns of unclear methodology or unsupported conclusions when you sit on the other side of the desk."
Teaching Doctors to Read a Paper with Suspicion
Catching errors in manuscripts is half the job. Making sure the next generation of physicians can spot unreliable research on their own is the other half. Within Zaporizhzhia State Medical and Pharmaceutical University, Dolinna lectures, develops methodological materials, and builds an online course on phytotherapy in nephrology, an area where published evidence ranges from rigorous clinical trials to poorly controlled observational studies, making critical appraisal unavoidable for any clinician who touches the literature.
From 2019 to 2020, she held the position of deputy dean of postgraduate education, overseeing continuing medical training for practicing physicians rather than students alone. For that audience, the challenge was immediate: working professionals who needed to separate reliable published findings from unreliable ones in real time, with real patients.
"Teaching and reviewing are closer than people think," she says. "Both force you to ask the same question: does this actually hold up under scrutiny?"
What Algorithms Still Cannot Catch
BMJ's retracted special issue will fade from the headlines. But the structural conditions that produced it, a chronic shortage of qualified reviewers, editorial workflows that can be circumvented, and a publishing ecosystem where the volume of submissions continues to outpace the supply of experts willing to evaluate them, are not going anywhere.
No software update will fix this. Artificial intelligence tools can flag statistical anomalies and detect plagiarism, but they cannot tell whether a proposed diagnostic threshold makes sense in a busy hospital, or whether a biomarker study accounts for the biological variability that clinicians encounter every day. That judgment still belongs to human reviewers and specifically to reviewers who carry enough clinical and scientific depth to exercise it.
Measured by volume, her reviewing workload represents a small fraction of the global system, one researcher's effort weighed against tens of thousands of retractions. Peer review, however, was never about volume. All that counts is whether the person reading the manuscript has enough experience to know what does not add up and enough stubbornness to say so.
© 2026 ScienceTimes.com All rights reserved. Do not reproduce without permission. The window to the world of Science Times.
