A New York Post opinion piece claims data proves restricting gender-affirming care does not raise teen suicide rates. The researchers work for Do No Harm, an organisation with an explicit goal of ending trans medicine for minors. Their study measures population-level state suicide statistics, not what happens to individual trans young people when care is denied, and the two are not the same thing.

What the piece actually says

Jay Greene and Ian Kingsbury from Do No Harm write that they compared teen suicide rates in US states that restricted gender-affirming care from 2022 onwards with rates in states that did not. They found no statistically significant increase in teen suicides in restricting states, and conclude from this that the claim linking denied care to suicide risk is, in their words, "a manipulative bluff."

They are not subtle about their purpose. They describe the suicide argument as "a means of bludgeoning society into accepting invasive, irreversible medical interventions for children." They call the young people at the centre of this "deeply confused teens." They characterise the medical professionals, families, and advocacy organisations who support gender-affirming care as activists masquerading as experts.

That framing tells you something important before you even look at a single data point.

What their data can and cannot show

Greene and Kingsbury used CDC data on state-level teen suicide rates and looked at whether those rates changed after restrictions came into force. This is called an ecological study: it looks at aggregate population data across large geographies and draws inferences from patterns at that level.

Ecological studies have real uses. They are also well understood to have a fundamental limitation, one that has a name in epidemiology: the ecological fallacy. You cannot reliably infer what happens to individuals from what happens to populations. A state-level suicide rate is the product of thousands of causes acting on millions of people. The signal you would need to detect, whether a specific sub-population of trans young people experienced worse outcomes after being denied care, is almost certainly invisible at that resolution.

To illustrate: if you wanted to know whether denying insulin to diabetic children increased their risk of serious harm, you would not look at the overall hospitalisation rate for the state and declare the question answered when you found no population-level shift. You would look at what happened to the diabetic children. Greene and Kingsbury did not look at what happened to the trans young people. They cannot, because that data is not collected in the way they would need.

This is not a minor technical quibble. It is the entire argument. Their study design cannot answer the question they claim to have answered.

The studies they dismiss

The piece waves away the existing research with considerable confidence. Studies in the New England Journal of Medicine and JAMA Network Open are described as falling "short of scientific standards," relying on "self-reported mental health surveys," and lacking appropriate comparison groups.

Some of those criticisms have partial validity. Longitudinal studies in this field do face real methodological challenges: the populations are relatively small, attrition from follow-up is a genuine problem, and randomised controlled trials are not ethically possible. Anyone working honestly in this area acknowledges that.

But the answer to imperfect evidence is not to replace it with a cruder instrument and declare the matter settled. The body of research on gender-affirming care, including prospective cohort studies, systematic reviews, and the clinical experience of gender specialists across many countries, consistently points in the same direction: access to affirming care is associated with better mental health outcomes for trans young people. The major professional bodies in medicine, including the American Academy of Pediatrics, the Endocrine Society, and WPATH, have reviewed that evidence and reached the same conclusion.

Do No Harm disagrees. That is their prerogative. But one ecological analysis of state-level suicide statistics does not overturn that body of work.

The framing that does the most damage

The piece opens with a version of the phrase many parents of trans children will recognise: "You can choose between a live son or a dead daughter." They present this as a manipulative scare tactic deployed by activists to coerce frightened parents.

I have heard versions of that conversation. I know the context in which it happens. It is not a scare tactic. It is a clinician or a counsellor trying, sometimes clumsily, to communicate something real: that the distress trans young people experience when they are denied support, rejected by family, or left without care can be severe and, for some, life-threatening. That is not an invented threat. It is documented in the research, it is reported consistently by trans young people themselves, and it is the lived experience of families who have watched their children suffer.

The CDC data the authors themselves cite shows that 72% of trans-identifying high school students reported persistent feelings of sadness or hopelessness. They use this figure to argue that those young people had pre-existing mental health difficulties, which trans identity then latched onto. The alternative reading, supported by decades of research into minority stress, is that young people who are rejected, stigmatised, denied care, and told their identity is a confusion experience depression and hopelessness as a direct result of those conditions. Both readings cannot be equally true, and the evidence points clearly toward the second.

What Do No Harm is

The authors disclose their affiliation, and I will take that at face value as a form of transparency. Do No Harm was founded with an explicit mission to oppose gender-affirming care for minors, among other targets. Greene and Kingsbury acknowledge in the piece itself that critics may object their findings "neatly conform" to their organisation's goals. They say their response to that concern is methodological transparency.

Transparency about methods is genuinely good practice. But it does not resolve the more fundamental question: whether a study designed by an organisation committed to a particular outcome, using a methodology that cannot detect the thing it claims to have disproved, should be treated as having settled a scientific debate. It should not.

What actually keeps trans young people safe

Trans young people are not made safer by restricting their access to care. They are made safer by family acceptance, by access to affirming support, by healthcare that treats them as the experts on their own experience, and by a world that does not treat their existence as a political problem to be solved.

The distress many trans young people experience is real. It deserves a real response. That response is not a study designed to prove that their distress does not matter, published in a newspaper that has already decided what the answer should be.

If there is a news story you would like me to cover then just let Sammy know.

Dr Helen Webberley is a Gender Specialist, Medical Educator, writer, and advocate, and the founder of GenderGP.