What’s new?‎ > ‎

2014-08-02 A commentary on the 2004 ARIF study and the Guardian article “Sex changes are not effective, say researchers”

posted 2 Aug 2014, 12:22 by Cake Kidd   [ updated 17 Aug 2014, 05:55 ]

A friend of mine told me about a recent discussion (in the comments on the article “I’m transgender and want my parents to know I’m their son, not a daughter” in The Guardian, where a commenter dug out a 2004 (!) Guardian article with the headline claiming that sex reassignment surgery is not effective. We took the time to look into this claim in more detail, and here are the findings (spoiler alert: the Guardian headline is not accurate).

The Guardian article

The article “Sex changes are not effective, say researchers” was written by David Batty and published on 2004-07-30. The main points regarding the ARIF study (see below) are:

“There is no conclusive evidence that sex change operations improve the lives of transsexuals, with many people remaining severely distressed and even suicidal after the operation, according to a medical review conducted exclusively for Guardian Weekend tomorrow.

The Guardian asked Arif to conduct the review after speaking to several people who regret changing gender or believe that the medical care they received failed to prepare them for their new lives. […]

Chris Hyde, the director of Arif, said: ‘There is a huge uncertainty over whether changing someone's sex is a good or a bad thing. While no doubt great care is taken to ensure that appropriate patients undergo gender reassignment, there's still a large number of people who have the surgery but remain traumatised - often to the point of committing suicide.’

Arif […] found that most of the medical research on gender reassignment was poorly designed, which skewed the results to suggest that sex change operations are beneficial.

Its review warns that the results of many gender reassignment studies are unsound because researchers lost track of more than half of the participants. […] Dr Hyde said the high drop out rate could reflect high levels of dissatisfaction or even suicide among post-operative transsexuals. […]”

The ARIF study

ARIF (Aggressive Research Intelligence Facility) was a specialist unit based at the University of Birmingham whose role was to improve the incorporation of research findings into health care decisions in the NHS in the West Midlands. In 2011, ARIF was succeeded by the West Midlands Commissioning Support Unit. Completed ARIF requests can be found at: http://www.birmingham.ac.uk/Documents/college-mds/haps/projects/ARIF/completed-requests.pdf.

The problem “What are the effects of gender reassignment surgery, and does the balance of positive and negative effects suggest that this procedure is clinically effective overall?” was submitted to ARIF twice, and the results were published in April 1997 and July 2004.

In 1997, “an extensive bibliography of relevant studies was compiled”. From this research, one (in numbers: 1) study published in 1990 was considered “pertinent.” Six further studies dating from 1979 to 1993 are listed under “other literature identified”.

The overall conclusion was: “The degree of uncertainty about any of the effects of gender reassignment is such that it is impossible to make a judgement about whether the procedure is clinically effective.”

The studies were summarised as: “Although the research published generally states that the effects are beneficial, it would be incorrect to say that this finding has been universal […].”

The main critical comments were:
▪ lack of a good quality, systematic review of the research literature
▪ most studies were without a control group, and not blinded
▪ high drop-out rates of study participants (some over 50 %)

An opinion on whether sex reassignment surgery is beneficial or not was not given.

When the update review was carried out in 2004, two (in numbers: 2) studies, dating from 1998 and 2002 were deemed to meet the criteria.

These two studies are thus summarised: “Both reviews while recognising net benefits to carefully selected individuals remained concerned about the quality of evidence on effectiveness […] and the biases to which available studies were open”.

In the abstract available on the university’s server, no opinion on the effectiveness of sex reassignment surgery is expressed. It is noted that “our searches confirm absence of randomised controlled trials and controlled trials to December 2001, and have identified none since then to the end of June 2004”. This is followed by a short discussion of how studies should be designed to avoid shortcomings, i.e. not meeting ARIF’s criteria.

My take on the study

The findings of ARIF are, in a way, understandable. So called double-blind trials with control groups are the gold standard of medical research. A blinded study means that neither the experimenters nor the participants know whether they received a certain treatment or not, allowing the results to be measured without bias.

One problem with medical treatment (and obviously surgery) for transsexuals is that blinding of studies is not possible. It is immediately obvious whether a participant received treatment or not, substitution by placebo will not work for obvious reasons. Clearly, all sex reassignment studies thus fail the gold standard. The next issue is including a control group into the study. This would require to properly diagnose transsexuals, making sure they meet the requirements and indications for sex reassignment surgery, and then to randomly split the participants into two groups – one receiving surgery and one not. Clearly, both groups have to be large for the result to be statistically valid. Then you could measure quality of life of the participants and compare the groups at intervals of several years. That’s the theory, anyway.

In reality, you would find the pressure transsexuals find themselves under grow so much that a large part of the untreated group commits suicide (Haas, Rodgers, Herman 2014) or seeks treatment illegally or abroad. This makes such a study highly unethical, it would never get the okay from an approvals body! You simply cannot withhold treatment from a highly stigmatised group that has a prevalence of 42 to 46 % suicide attempts, compared with 4.6 % in the general population. And suicides are only the worst outcome, there is a lot of misery and suffering before things get to the extreme.

So… what do we have… we have a medical treatment that cannot be studied using control groups or blinded trials. There are only three things that studies can do to measure the effects of SRS:
▪ measure the quality of life of participants before and after sex reassignment surgery (long-term study)
▪ measure the quality of life of participants after surgery and compare them to the average quality of life of transsexuals before surgery (demographics etc. need to be similar)
▪ simply ask transsexuals after surgery whether they feel better (yes, it is that easy!)

While these three options do not meet ARIF standards, such studies have been conducted time and time again (see http://www.cakeworld.info/transsexualism/what-helps/srs for a list of 71 relevant articles).

The outcomes have been hugely positive. One criticism ARIF made (the drop-out rate) has been addressed in newer studies, as transsexuals receive better comprehensive treatment and are less discriminated against by health care providers. As an example, Dhejne, Öberg, Arver and Landén (2014) studied all applications for sex reassignment surgery in Sweden from 1960 to 2010 and found an average regret rate of 2.2 % – with the regret rate declining significantly over time.

In view of the above, it is not clear what additional evidence the medical community and transsexuals themselves can provide. Have a look at the 71 articles mentioned, or read this article in „The Stranger“ for a good summary. Remember, everyone is entitled to their own opinions, but they are not entitled to their own facts – if you do not believe the facts presented by the research, present better research. However, from years of reading every study published, I can assure you, there is none.

I believe that ARIF, including the quoted director Chris Hyde (not a specialist on transsexualism) would have come to a different conclusion today, more than ten years later. Even in 2004, they could have made certain provisions for the nature of the researched subject.

I believe the Guardian article misrepresents the information that is available on the ARIF study. It creates a controversy where none exists. The article needed an opposing viewpoint to the transgender activism group Press for Change (PFC) and focused on one issue of ARIF’s statement only. Yes, if you do some hair-splitting, ARIF did find no gold standard studies at the time. If you focus on scientifically entirely valid criticisms of the then available studies, you can create doubt in the mind of the reader. The fact that ARIF had no official opinion, negative or positive, on the effectiveness of sex reassignment surgery was hidden behind these facts. That is twisting the facts. The title “Sex changes are not effective, say researchers” is so reduced that is a falsehood. Note that the claim the headline makes is not repeated in the article.

So, what we have here is a questionable newspaper article based on a short review of articles with little understanding of the subject at hand. And this article and the underlying study are still mentioned today, more than ten years later, to prove a point. If that is the best argument from the opponents of medical help for gender dysphoria sufferers… well, then they have no argument.