Thank you very much, Gary, for your thoughts and your opinion, which deserve a detailed response.

I deliberately chose "medicine" over "biomedical science" in the title for a simple reason. If biomedical research, being the bedrock of the practice of medicine, is compromised, then so is the entire edifice of medicine.

Yes, Heathers' paper has been criticized for lacking a rigorous method to derive the 1-in-7 number, which he himself readily acknowledges. That's because there exists no gold-standard method for the estimation of fakery, which is why he took the meta-analytical approach that Fanelli found fault with. Interestingly, Fanelli herself used meta-analysis in her 2009 paper for her estimation of the percentage of researchers fabricating or falsifying study data. She reported weighted means of 2% and 34% of scientists admitting to having used falsification and other questionable data manipulation respectively. Mind you, these figures are for misconduct admitted by polled researchers only. The numbers for observed misconduct by peers are substantially higher. For obvious reasons, it is difficult to translate these numbers into a percentage of falsified papers (are the cheaters more or less prolific authors than the honest ones?). But in his 2024 paper (Quantitative research assessment: using metrics against gamed metrics), John Ioannidis cites the editor of Anesthesia as reporting that 30-40% of submitted papers contain obvious data manipulations and fakery.

I respectfully disagree with your suggestion that my article's headline is harmful ("…in a world where trust is at an all-time low- supporting people’s denial of expertise in anyone other than themselves"). On the contrary. When trust is undeserved, it needs to be withdrawn. There will always be anti-vaccine imbeciles and those who willingly expose themselves to avoidable health risks. My work is not about abetting this type of behavior, it's about enabling lay readers to find their individual path toward their health objectives, which in my realm of science and research is healthy and functional longevity. Since we cannot translate the group-based findings of (credible) RCTs to a given individual, N-of-1 is a legitimate and FDA sanctioned approach that deserves a much more prominent presence in the practice of medicine than it is currently afforded. The "large grain of salt" that you suggest is exactly the same grain that applies to group-based statistics. We can do a lot of mischief with the statistical method, but when applied correctly it is the one and only way to draw reliable conclusions from the data. The problem with the N-of-1 method is not that it is less valuable than group-based statistics, the problem is that it remains largely ignored in the practice of medicine and completely absent from the medical curriculum.