Medical journals are little more than industry mouthpieces. Editors-in-chief of the world’s most prestigious journals have repeatedly admitted how thoroughly corrupt and dishonest the medical publishing industry has become.
You’re better off flipping a coin than blindly trusting what’s published in medical journals—and frankly, that’s unfair to the coin. The coin isn’t taking bribes from pharmaceutical companies.
Most of the published research is junk
If research can’t be reproduced, it isn’t science—it’s storytelling. Academia suffers from a widespread reproducibility crisis, with follow-up studies failing to confirm original findings. But in medicine, bad science doesn’t just mislead—it can kill.
Millions of studies flood the literature each year—including more than 10,000 retractions in 2023 alone. And that’s just the tip of the iceberg. Most bad research isn’t retracted. It simply lingers—polluting the knowledge base and misleading generations of doctors, policymakers, and patients.
The replication crisis
Attempts to reproduce published findings reveal just how widespread the problem is.
A 2005 JAMA study examined 45 highly cited clinical studies. Only 20 (44%) were later replicated; 7 (16%) were contradicted, 7 (16%) were found to have larger effects than subsequent studies, and only 11 (24%) remained largely unchallenged.
A 2011 Nature Reviews Drug Discovery study evaluated 67 preclinical drug studies. Only 14 (21%) were reproducible. Yet findings like these can influence which drug candidates companies pursue—potentially committing hundreds of millions to their development.
A 2012 Nature study attempted to replicate 53 preclinical cancer studies. It succeeded with just 6—11%.
A 2015 study assessed 17 neuroimaging findings. Only 1 replicated—6%.
A 2025 Nature report described a large-scale biomedical reproducibility project involving 56 laboratories and 97 replication attempts across 47 experiments. Only 21% of the experiments were reproducible, while the original studies reported effect sizes 60% larger on average than those found in the replication attempts.
The problem isn’t limited to obscure journals. This is the mainstream medical literature—the very foundation of so-called “evidence-based medicine.” And much of it is irreproducible.
The replication crisis is so widespread that 90% of researchers acknowledge it.
Conflicts of interest at every level
The conflicts run through every layer of medical research and publishing—from the journals themselves to the editors, reviewers, and researchers who shape the evidence.
Medical journals
The revenue of medical journals depends heavily on pharmaceutical companies, mainly through advertising and reprint purchases—financial relationships that create obvious conflicts of interest.
In typical advertising, the goal is to inform potential customers about a product. But in medical journals, pharmaceutical ads provide financial incentives that influence editorial decisions. Similarly, reprints (purchases of specific articles) create financial incentives around which studies get published. Without lucrative reprint orders from pharmaceutical companies, many medical journals would struggle to stay afloat.
When drug companies want their studies published, they often offer substantial reprint orders if the paper is accepted. A favorable study in a high-impact journal is much more valuable than an ad—it becomes a tool doctors use to justify prescribing a product. Doctors are the pharmaceutical industry’s primary sales force, and favorable research gives them the ammunition to sell drugs.
The financial pressure is not merely theoretical. In May 2003, BMJ’s cover depicted doctors as pigs feasting at a banquet while being served by lizard drug representatives. Pharmaceutical companies reportedly threatened to withdraw £75,000 in advertising. Similarly, the Annals of Internal Medicine lost an estimated $1–1.5 million in advertising revenue after publishing a study critical of industry advertising.
![BMJ, Vol 326, Issue 7400 [Magazine Cover] (31 May 2003)](https://substackcdn.com/image/fetch/$s_!d74L!,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2Ffe2e278e-bb0d-4672-aa34-6c4cc3cecc52_3840x2160.png "BMJ, Vol 326, Issue 7400 [Magazine Cover] (31 May 2003)")
As for reprints, even The Lancet—one of Europe’s most prestigious medical journals—has acknowledged that 41% of its revenue comes from selling them.
The commercial stakes of favorable research are enormous. A 2005 report estimated that publishing a favorable paper could be worth up to £200 million to a pharmaceutical company, with a share of that value flowing to doctors who promote the company’s products.
The corruption spans the entire medical publishing industry—from prestigious journals to niche specialty publications. Transplantation and Dialysis rejected an editorial critical of a drug after three peer reviews—not on scientific grounds, but because its marketing department overruled the editors. Smaller specialty journals can be even more entangled with industry, publishing industry-sponsored content filled with brand names and promotional material. In one infamous case, Merck created an entire journal itself, packed with advertisements and favorable articles and designed to resemble independent research.
Journal editors
A 2017 BMJ study found that over half of editors at the world’s most influential medical journals were receiving payments directly from pharmaceutical and medical device companies—sometimes tens or even hundreds of thousands of dollars.
In 2014 alone, the average editor received $27,564 in personal payments, plus additional research payments. At the Journal of the American College of Cardiology, 19 editors received an average of $475,072 each in personal payments, along with another $119,407 in research payments.
Clinical trials
Clinical trials are supposed to provide independent evidence about whether drugs and other medical treatments are safe and effective. But pharmaceutical companies often control the research process—from the data generated to what investigators can access and what ultimately gets published.
A 2006 JAMA study reviewed 88 trial protocols and found that in half of them, sponsors retained the right to approve or block publication. In most remaining cases, sponsors imposed legal or logistical restrictions, including data ownership provisions and required pre-submission approval.
A 2005 New England Journal of Medicine survey found that 80% of medical schools would accept contracts giving sponsors ownership of the data, while 50% would allow sponsors to ghostwrite studies. Even after contracts were signed, 82% of institutions reported conflicts with sponsors. In one case, a company withheld final payment because it disliked the study results. Contracts were typically kept secret, making the true extent of industry influence difficult to assess. Yet 69% of administrators said funding pressures often forced them to accept unethical terms.
Industry funding now dominates clinical research. In the United States, industry’s share rose from 32% in 1980 to 62% by 2000. Meanwhile, the share going to academic medical centers fell from 63% in 1994 to 26% in 2004, as private contract research organizations (CROs) took over an increasing share of the work. Many CROs are also involved in marketing, further blurring the line between research and sales.
Universities therefore have strong financial incentives to court industry funding, offering pharmaceutical companies access to patients and clinical staff. Doctors are reduced to recruiters, offering up their patients in exchange for perks and publications. In some cases, drug sponsors pay up to $42,000 per patient enrolled.
Peer reviewers
Between 2020 and 2022, over half of peer reviewers for JAMA, the New England Journal of Medicine, BMJ, and The Lancet received payments from pharmaceutical and medical device companies, totaling $1.06 billion.
58.9% received industry payments
54% took general payments (like gifts, speaking fees, etc.)
31.8% accepted research funding
The median research payment was $153,173.
These aren’t fringe journals. They are the so-called gold standard of medical publishing. And the people deciding what gets published are receiving money from the very industry they are supposed to keep in check.
Peer reviewers also aren’t required to disclose their conflicts in the same way as editors and authors. Journals generally keep reviewers’ identities and conflicts hidden from the public.
Biases
Bias enters medical research at virtually every stage—from how studies are designed to which results get published and how researchers are rewarded.
Bias toward positive results
One of the most pervasive biases in medical research is the preference for positive findings. Studies that support a treatment or intervention are more likely to be published, cited, and taken seriously—regardless of their quality. This bias operates across the research pipeline, from study design to publication decisions.
A 2010 JAMA Internal Medicine study found that papers with positive results were accepted 97.3% of the time, compared with just 80% for papers with negative findings.
The pharmaceutical industry exploits this bias to its full advantage. Drug companies are far more likely to publish studies with favorable outcomes and bury the rest. In a 2008 New England Journal of Medicine analysis of antidepressant trials submitted to the FDA, 94% of positive studies were published—but only 8% of negative ones. Based solely on the published literature, antidepressants would appear overwhelmingly effective. In reality, only 51% of all trials showed any benefit.
This is the file-drawer problem in action: negative or inconclusive results are quietly shelved while positive ones flood the journals. The result is a systematic distortion of the medical evidence base.
Publish-or-perish culture
Academia rewards publication volume more than research quality. Researchers are pressured to publish constantly or risk losing funding, promotions, or their careers.
Instead of pursuing careful, meaningful research, academics are incentivized to produce papers that are more likely to get published.
The result is a flood of low-quality studies—designed to satisfy publication criteria, pass peer review, and pad résumés. The more papers researchers publish, the more they are rewarded, regardless of whether those papers meaningfully advance medical knowledge.
Bias in study design
Bias often begins before the research even starts—during the design phase. Researchers make choices about what to measure, how to measure it, and how to frame questions that steer results toward their hypotheses. Even without deliberate misconduct, these decisions can introduce bias from the outset.
Combined with the pressure to publish and a lack of transparency, these structural incentives create fertile ground for biased and unreliable research.
Fraud and other misconduct
Scientific misconduct takes many forms, from ghostwriting and statistical manipulation to outright fraud.
Fraud
The problem isn’t just bias or conflicts of interest—it includes widespread, deliberate fraud. And the true extent of misconduct is likely far worse than the following numbers suggest, because researchers are unlikely to admit to fraud, even anonymously.
A 2005 Nature survey of 3,247 NIH-funded U.S. scientists found that 33% admitted to questionable research practices in the previous three years, including:
16% who altered study design, methodology, or results because of funding pressure
15% who dropped inconvenient data points
14% who used inappropriate or inadequate research designs
A broader 2009 review of 21 surveys on research misconduct found:
Up to 5% of scientists admitted to falsifying or fabricating data
Up to 33% had personal knowledge of colleagues who falsified data, while up to 72% knew of other questionable practices
Up to 34% admitted to other forms of misconduct
One of the surveys found that 81% of biomedical research trainees were willing to omit or fabricate data to secure a grant or publication.
Ghostwriting
Ghostwriting is widespread in medical research: pharmaceutical companies hire writers to produce research papers that are then published under the names of prominent academics or researchers.
A 2007 study of industry-funded trials found that 75% had ghost authors. That figure rose to 91% when researchers who should have been credited as authors were instead listed only in the acknowledgments.
P-hacking
P-hacking is a way of manipulating statistical analyses to make research findings look stronger than they really are. Instead of deciding in advance what they will test, researchers can try different outcomes, different ways of analyzing the data, or different groups of patients until something produces a statistically significant result.
A researcher might test whether a drug improves ten different outcomes and, even if the drug does nothing, one of those ten outcomes may appear positive purely by chance. The researcher can then highlight that one positive result while ignoring the nine that were negative.
The result is a false impression of effectiveness: random chance is presented as evidence that a treatment works. P-hacking turns weak or meaningless findings into apparently convincing medical evidence.
Censorship
Censorship in medical publishing is not limited to rejected papers. Researchers who challenge established medical narratives can face retractions, institutional sanctions, and serious professional consequences. The cases of Dr. Andrew Wakefield and Dr. Paul Thomas are just a couple of examples that illustrate how contentious research can trigger consequences extending far beyond the publication process.
In 1998, The Lancet published a paper coauthored by Dr. Andrew Wakefield reporting a possible link between the MMR vaccine and gastrointestinal symptoms in children with developmental disorders. The study was cautious in its conclusions and called for further investigation, but it was later retracted, and Wakefield was struck off the UK medical register.
A similar controversy surrounded Dr. Paul Thomas, who in 2020 published a study comparing health outcomes among vaccinated and unvaccinated children in his pediatric practice. The study reported fewer chronic conditions among the unvaccinated children, was later retracted, and Thomas’s medical license was suspended.
This kind of censorship isn’t limited to high-profile cases. Research that questions pharmaceutical products, vaccine policy, or dominant public health narratives can face significant barriers to publication, while researchers who pursue such questions face severe professional consequences.
Peer review
Peer review is supposed to serve as the quality-control system for academic publishing. In practice, it is riddled with flaws and biases that allow poor research to pass while making it harder for findings that challenge established narratives to get published.
Reviewers may overlook serious methodological problems, misunderstand complex methods, or favor research that aligns with prevailing views. The result is a gatekeeping process that legitimizes flawed research while filtering out inconvenient findings.
Reviewers don’t agree
Reviewers frequently disagree with one another, sometimes substantially, when evaluating the same research. A 2010 meta-analysis of inter-rater agreement found that peer reviewers generally showed low agreement when assessing the quality of research.
Peer review fails to catch obvious nonsense
Even reputable journals have published or accepted research that contained obvious nonsense.
In 2009, the BMJ published a paper on a fictional condition called “cello scrotum.” The authors had submitted it as a joke, expecting reviewers to catch it. They didn’t.
In 2013, a journalist submitted a deliberately bogus paper riddled with basic chemistry errors to more than 300 journals. More than half accepted it for publication.
Peer review fails to catch basic errors
Peer reviewers can miss even obvious errors in submitted papers. In a 1998 experiment led by Fiona Godlee, then editor of the BMJ, eight deliberate errors were inserted into a paper and sent to 420 reviewers. Of the reviewers who responded, the average identified just two errors, and 16% failed to identify a single error.
In a second experiment conducted by Godlee in 2008, nine major and five minor errors were inserted into a paper and sent to 607 reviewers. Again, the reviewers detected only a small proportion of the flaws, identifying an average of just 3 out of the 9 major errors.
Academics misunderstand statistics
Even researchers with statistical expertise can struggle to correctly interpret research findings. This matters for peer review, where reviewers must assess whether the statistical methods and conclusions of a paper are sound.
In a 2015 study, approximately 80% of epidemiologists incorrectly interpreted a statistical result concerning a drug intervention when the p-value exceeded 0.05, incorrectly concluding that there was no meaningful difference between the groups.
Biases of peer reviewers
Peer reviewers tend to favor research that aligns with their own perspectives or prior beliefs, reinforcing established ideas while making novel or controversial findings harder to publish.
A 1977 study found that reviewers rated papers reporting positive results in line with their own beliefs as higher quality, while rating papers that contradicted their views more negatively. A 1993 study similarly found that scientists were more likely to accept papers that agreed with their prior beliefs.
Impact factor
Journal prestige is built on a highly manipulable metric: impact factor, which measures how often a journal’s papers are cited. Pharmaceutical companies can exploit citation networks by promoting industry-sponsored or ghostwritten publications that cite favorable trial results, increasing the visibility and number of citations those findings receive.
The New England Journal of Medicine provides a striking example: trials of Pfizer’s voriconazole were published with misleading conclusions, yet these flawed studies were cited hundreds of times in papers that propagated the findings, likely with the involvement of sponsor-directed ghostwriting.
What real peer review looks like
The traditional peer-review system is a closed process: anonymous reviewers evaluate papers behind the scenes, while editors decide what ultimately gets published. Readers often have little visibility into who reviewed a study, what criticisms were raised, or how those criticisms were resolved.
Real peer review happens out in the open—on platforms like Substack, X, independent blogs, and open-access journals—where anyone can read, critique, and engage with the material directly. No anonymous panels or backroom deals.
More researchers are publishing in open-access and preprint journals, where findings are freely available and subject to public scrutiny. There’s also growing pressure on institutions to publish all results, including negative ones—not just the ones that serve commercial interests.
Conclusion
Journal prestige should never be treated as a substitute for scientific integrity. A prestigious publication does not make research immune to conflicts of interest, methodological flaws, bias, or misconduct.
Arnold Relman, former editor of The New England Journal of Medicine, warned decades ago:
“The medical profession is being bought by the pharmaceutical industry—not only in terms of the practice of medicine, but also in terms of teaching and research. The academic institutions of this country are allowing themselves to be the paid agents of the pharmaceutical industry. I think it’s disgraceful.”
Pharmaceutical influence is not confined to individual researchers or isolated conflicts—it reaches across the institutions responsible for producing, evaluating, and disseminating medical knowledge.
Research should be judged on its own merits—openly, transparently, and case by case. The more scientific scrutiny moves beyond closed editorial systems and into open, decentralized spaces, the harder it becomes for any single institution to control what gets heard.
Scientific authority should come from the strength of the evidence—not the prestige of the journal that publishes it.
This article is part of the Epistemic Capture of Medicine: The Definitive Crash Course series:
The Substack articles are deliberately provocative. The PDF edition takes the same material and presents it in a more formal, professional format—with the inflammatory phrasing removed and the research, sources, and evidence brought together in one cohesive volume.
Exclusive to paid subscribers, with access to all future revised and expanded editions.
Note: The PDF button may not work when opened in the Substack app or an email app. If so, please open this post in a web browser to access it.
I wrote an article back in December on this very subject. People are totally unaware of how corrupt and bereft of scientific integrity the medical industry really is. You cannot trust a word they say.
See my article The Medical Profession is a Cesspool of Fraud, greed, Corruption and Bad Science.
https://stephenmcmurray.substack.com/p/the-medical-profession-is-a-cesspool
As a board-certified subspecialist in practice for over 30 years, I have witnessed the corrosive effects of pharma’s influence over clinical medicine. It is as bad as Dr. Wojak says.