The Zombie Study Behind the ‘1% Regret’ Myth for ‘Gender-Affirming’ Surgery

A critical look at the paper behind one of gender medicine’s most repeated false statistics.

Reality’s Last Stand is a reader-supported publication. Please consider becoming a paying subscriber or making a one-time or recurring donation to show your support.

About the Author

Dr. Colin Wright is the CEO/Editor-in-Chief of Reality’s Last Stand, an evolutionary biology PhD, and Manhattan Institute Fellow. His writing has appeared in The Wall Street Journal, The Times, the New York Post, Newsweek, City Journal, Quillette, Queer Majority, and other major news outlets and peer-reviewed journals.

---

One of the biggest controversies in the debate over pediatric “gender-affirming care” is the experimental nature of its most invasive interventions: puberty blockers, cross-sex hormones, and surgeries. Recent systematic reviews in Europe—most notably the UK’s Cass Review and a comprehensive new report from the U.S. Department of Health and Human Services (HHS)—have concluded that the quality of evidence supporting these treatments is very low. Meanwhile, the risks are well documented: sterility, sexual dysfunction, impaired brain development, and irreversible physical changes.

You would think that something this serious—especially for kids—would be backed by strong, long-term clinical data. But that’s not what’s happening. When critics raise concerns about the lack of good evidence, supporters often change the subject. Instead of defending the science, they focus on an emotional claim: regret is incredibly rare. If nearly everyone is happy afterward, they argue, the treatments must be working.

You would think that something this serious—especially for kids—would be backed by strong, long-term clinical data. But that’s not what’s happening. When critics raise concerns about the lack of good evidence, supporters often change the subject. Instead of addressing the lack of positive evidence of benefit, they pivot to an emotionally resonant claim: that regret is exceedingly rare. If nearly everyone is happy afterward, they argue, the treatments must be working.

This rhetorical sleight of hand—substituting low regret rates for proof of benefit—has proven remarkably effective. It is frequently claimed that fewer than 1 percent of patients who undergo “gender-affirming” surgery come to regret it. This talking point is everywhere, deployed in courtrooms, legislation, news segments, and policy debates as proof that these surgeries are safe and effective.

What most people don’t know is that this widely repeated claim comes from a single paper published in 2021: “Regret after Gender-affirmation Surgery: A Systematic Review and Meta-analysis of Prevalence” by Bustos et al. The study appeared in Plastic and Reconstructive Surgery Global Open, a pay-to-publish version of the flagship journal of the American Society of Plastic Surgeons. And although systematic reviews are generally considered the highest tier of evidence, this particular review has been shown to contain so many major flaws and data errors that it should not be relied on for any public health decision.

And yet it remains one of the most cited pieces of evidence used to justify gender-affirming surgery—including for minors. It is perhaps the most influential zombie paper in gender medicine, being dragged around like Bernie Lomax in the movie Weekend at Bernie’s by activists, journalists, and supposed “experts” who need its corpse propped up to sell the illusion that it’s alive.

The central claim of the paper is simple: that among 7,928 transgender patients pooled across 27 studies, only 77—less than 1 percent—reported regret, and just 34 expressed what the authors categorized as “major” regret. They conclude that “there is an extremely low prevalence of regret in transgender patients after GAS.” This is the origin of the “1 percent regret” statistic for gender surgeries that has been repeated ad nauseam ever since.

But serious problems emerge the moment you look more closely. Shortly after publication, a Letter to the Editor by Pablo Expósito-Campos and Roberto D’Angelo appeared in the same journal, cataloging a list of serious concerns. The authors pointed out that numerous relevant studies fitting their search criteria had been omitted from the review. At the same time, the paper included studies that should have been omitted. More troubling still were the blatant data extraction errors, with sample sizes misreported, outcomes misrepresented, and methods misunderstood.

---

🎧 Prefer to listen?

Watch or listen to a full breakdown of this paper by evolutionary biologist Dr. Colin Wright and journalist Brad Polumbo on the latest episode of the Citation Needed podcast.

---

One of the most striking examples comes from the inclusion of a 1998 study by Kuiper et al. Bustos et al. claimed that the study followed 1,100 patients, of whom 10 regretted surgery. In reality, the study identified 10 regretful patients through a much narrower and more selective process: participants were recruited via newspaper and magazine advertisements, outreach to self-help groups for transsexuals, and invitations extended to patients of the Amsterdam gender team. Individuals then had to voluntarily agree to participate in an interview. The 1,100 figure cited by Bustos et al. refers not to the number of people surveyed, but to a rough estimate of the total number of gender surgeries that had taken place in the Netherlands up to that point. Claiming this as the sample size is an extreme misrepresentation.

Compounding the error, Bustos et al. claimed that the study recruited participants using “social media”—in 1998!

This is patently absurd. “Social media” did not exist at the time. MySpace, often credited as one of the first social media platforms, wasn’t launched until 2003. Twitter didn’t exist until 2006. For Bustos et al. to describe a 1998 recruitment method as using “social media” reflects a total lack of diligence. This is not a minor mistake. It is the kind of error that suggests the authors either misunderstood or did not carefully read the study they were citing.

Another massive error results from their handling of the 2018 study by Wiepjes et al., which contributed the largest share (over half!) of patients to their review.