More on the McMaster Mess

How Gordon Guyatt's comments misrepresent the role of systematic reviews.

Photo by GV Chana on Unsplash

The controversy surrounding Gordon Guyatt’s public statements on the McMaster systematic reviews of pediatric gender medicine has already been the subject of extensive commentary. The facts have been reviewed in detail in a two-part piece by Jesse Singal, which includes an interview with Guyatt, an article by Jonathan Kay in Quillette, and an article by Benjamin Ryan in UnHerd and further reporting on his Substack. Nevertheless, there is still more to be said on why Guyatt’s views are wrong and harmful.

The basic facts are that Gordon Guyatt is one of the leaders in the development of the principles of evidence-based medicine and a Distinguished Professor in the Department of Health Research Methods, Evidence and Impact at McMaster University in Hamilton, Ontario. The Society for Evidence Based Gender Medicine entered into a contract with McMaster to prepare a series of systematic reviews on various aspects of pediatric gender medicine. The reviews on puberty blockers, cross sex hormones and mastectomies have been published. In each case, they concluded that the evidence that the treatments are beneficial is of low certainty. Reviews of social transition and binding and tucking have been completed but are still pending.

The reviews led to an outraged response by transgender activists at McMaster. This led Guyatt and four of his colleagues to issue a public letter in which they claimed that their work was being misused to support bans on medical transition.

Science in Public Debate

The letter shows a serious misunderstanding of the proper role of science in public debate. The opening sentence reads:

Authors of scientific articles have a responsibility to attend to how their contributions will be used and to modify their presentation in the articles, or other communications, accordingly.

It is not clear what Guyatt and colleagues mean by “modify their presentation.” Presumably, this does not mean suppressing or modifying actual conclusions. There might be circumstances where it may be necessary to provide a clarification of scientific issues if policy makers are misstating the conclusions of a study, but editorializing on broader ethical and political issues risks compromising the impartiality of research.

Susan Bewley, another eminent figure in the field of evidence-based medicine, explains this concern in an X/Twitter thread:

I find it hard to agree here. What made Guyatt (a highly respected scientist) realize this extra responsibility? Why step 'out of lane' & out of academic fora to discharge it now & opine to policymakers (a kind of global call of 'how to mark my homework')? He didn't before. Gordon guided me on the dividing line between scientists & policy makers, each responsible for their sphere. We were a united team who exposed how policy makers chose to handle higher quality evidence of serious harms to pregnant women living with HIV https://bmj.com/content/358/ I don't understand the generic, i.e. 'pre-specified' rule that all scientists should follow as to 'when to break impartiality' & condemn a funder or policy maker for misusing their work after they & university must have done due diligence? It happens often but isn't our fault.

The role of scientists in public debate is similar to that of an expert witness in a court case. This was recently discussed in the Australian case of Re Devin which was a family court dispute where the parents disagreed on medical transition of a child. Judge Strum explained that the role of an expert witness is to “give an objective and unbiased opinion that is also independent and impartial on matters that are within the expert witness’s knowledge and capability.” Although expert witnesses are usually called by the parties, their first duty is to the court. Judge Strum strongly criticised Michelle Telfer, the head of medicine at Melbourne Children’s Hospital, who gave evidence for the mother, for overstepping her role as an expert and acting as an advocate.

It is significant that Guyatt and his colleagues donated money to Egale Canada on the recommendation of community advocates. Egale Canada is one of the plaintiffs in a constitutional challenge to Alberta legislation restricting medical transition for minors. A judge recently granted an injunction against the implementation of the law. The government has appealed that decision and the case will proceed to trial. (See this article by Geoffrey Sigalet to learn more about the case.) Evidence of the results from systematic reviews will form part of the government case. The effect of the donations will be to compromise the impartiality of Guyatt and his colleagues as potential witnesses.

It is also important to note that in 2023, Egale Canada received nearly 69% of its operating budget from the Canadian federal government. The federal government is also the major funding source for university based medical research in Canada.

Use of the Systematic Reviews

Guyatt’s letter goes on to say:

Specifically, we are concerned that the assessments of the certainty of evidence using established and standard methodology are interpreted as evidence supporting denial of care to trans, nonbinary, and gender-diverse (TGD) individuals, including youth. We are concerned our findings will be used to justify denying care such as puberty blockers and hormone replacement therapy to TGD individuals. Indeed, our prior work has been used in exactly this highly problematic way.

It is not clear what is problematic. The point of a systematic review is to assist in decisions to provide or not to provide a treatment. If systematic reviews have found that there is low or very low certainty evidence of benefits from hormones and puberty blockers and high to moderate certainty evidence of risk, then refusing to provide these treatments cannot be described as denial of care.

The Role of Autonomy

The following paragraphs invoke the principle of autonomy, shared decision making and patient values and preferences, but misrepresent all of them and their application in gender medicine:

Following fundamental principles of humane medical practice, clinicians have an obligation to care for those in need, often in the context of shared decision making. It is unconscionable to forbid clinicians from delivering gender-affirming care. 

 Moreover, following the principles of evidence-based decision-making, clinicians should always have a high respect for the autonomy of patients and their advocates. The high respect for autonomy becomes particularly important when the certainty of the evidence is low or very low. In such circumstances, clinicians should work with patients to ensure that care reflects the experience, goals, and priorities of those needing care – that is, their values and preferences.

Eithan Haim has responded to Guyatt on this issue. Respect for autonomy, values and preferences is not a justification for providing treatments which are not beneficial. The principle of autonomy gives patients an absolute right to refuse a treatment proposed by a doctor but it does not give patients the right to demand a treatment which has not been shown to be safe and effective.

Shared decision-making means just that. The process is a shared one where scientific evidence; the doctor’s clinical judgment, and the patient’s values and preferences all play a role. A shared decision-making process does not justify providing treatments that are not supported by evidence of benefit.

Guyatt’s own work shows that he understands the appropriate use of patient values and preferences. Just a few days ago he posted a link on his X account to an article entitled “The GRADE Evidence to Decision (EtD) framework for health system and public health decisions” which describes a process for applying evidence from systematic reviews to decision making to clinical and public health decision making. Patient preferences play a role in the process, but they need to be considered alongside factors such as evidence of benefits and harms, resource use and feasibility.

There are circumstances where patient preferences may be expressly overridden. In 2017, Guyatt was part of a team that developed a Guideline for opioid therapy for chronic noncancer pain. The values and preferences consultation found that patients in the advisory group often placed little value on avoiding addiction and the risk of overdose. The values and preferences report noted:

Out of desperation for any relief from chronic pain, patients with chronic noncancer pain are often willing to trade rare but serious harms for small but important pain relief, and this may leave some patients vulnerable to short-term decisions that are inconsistent with societal values. The patients on our advisory committee who had actually suffered serious adverse events placed a much higher value on their avoidance, consistent with societal attitudes toward such events.

The guidelines panel made a choice to prioritize values and preferences, favouring protection against addiction and overdoses and recommended a restrictive approach to prescriptions.

Guyatt has said that he disagrees with SEGM on the application of the principle of autonomy. In fact, SEGM does not have an official position on autonomy or treatment bans. The real difference of opinion is between most clinicians and researchers, including Guyatt in other cases, who see autonomy as one of a group of principles which inlcudes non-malificence, beneficence and justice, and the transgender activist approach, which regards autonomy as absolute.

Recommendations Based on Low-Certainty Evidence

The following paragraph of the letter misrepresents another key aspect of decision making in evidence-based medicine:

It is profoundly misguided to cast health care based on low-certainty evidence as bad care or as care driven by ideology, and low-certainty evidence as bad science. Many of the interventions we offer are based on low certainty evidence, and enlightened individuals often legitimately and wisely choose such interventions.

A 2021 article by Guyatt describes situations where it may be appropriate to make strong recommendations for treatment despite low certainty evidence. A treatment that is supported only by low-quality evidence might be recommended in life-threatening situations where no superior alternatives are available. This is not applicable to childhood gender dysphoria, which is not life-threatening and may resolve without treatment. A second example in the article is that a strong recommendation may be made against a treatment which has low certainty of benefit but high certainty of harm. This describes the current evidence on puberty blockers and cross sex hormones in minors. In any event, children and adolescents who often have co-morbid mental health conditions can hardly be considered “enlightened individuals.”

Treatment Bans

The issue of banning treatments raises issues of ethics and political philosophy that I discussed in a recent article. It is not justifiable to prohibit a treatment simply because it is supported only by low-certainty evidence. The argument for banning pediatric gender treatments considers many other factors, including the lack of a clear diagnosis, uncertainty over which patients will persist in cross-sex identity, the high risk of serious harm, the vulnerable patient population and evidence of inadequate safeguarding in gender clinics.

However, the key argument for treatment bans is that there has been a failure in self-regulation. Pressure from transgender activists has caused the leadership in the medical and mental health professions to ignore and suppress scientific debate and clinical caution in promoting the gender affirming care model. By showing how easily even the most eminent researchers and clinicians can be intimidated by activist pressure, the campaign against Guyatt and his colleagues strengthens the case for legislative action.

What Next?

Guyatt and company are learning the hard way that you can never grovel enough to satisfy the transgender activist community. The Instagram Account @segm_x_mcmaster has posted a response to the statement and they are not impressed:

Trans rights activists are NOT PLACATED — this is simply more laundering of anti-trans pseudoscience, with more lies and insults. A literal half-assed attempt, this refried, convoluted anti-trans statement was signed by HALF of the authors — not even including the lead author, Anna Miroshnychenko!

The trans-activists would like to prevent publication of the remaining reviews and they may be successful here. They would also like to see the existing reviews retracted, but this will be more difficult as they have not been able to identify any errors in the reviews themselves. The longer-term objective of the campaign against McMaster is to deter other universities from publishing systematic reviews on gender medicine without the approval of the transgender activist community. The restrictions that WPATH imposed on the publication of the reviews it commissioned from Johns Hopkins University show what the activists have in mind: all manuscripts had to be approved by WPATH but had to carry a disclaimer that the authors were solely responsible for the content.

This campaign is not likely to succeed. The number of countries where transgender activists have the influence they still do in Canada is rapidly dwindling. High-quality research on gender medicine will still be done but the reputation of McMaster as a source of such research will be seriously compromised.