PinkNews August 10, 2023
Pink news ‘top surgery regret rate’ article https://www.thepinknews.com/2023/08/10/top-surgery-study-regret-rate/ and in case that doesn't work, the archive: https://web.archive.org/web/20230810161804/https://www.thepinknews.com/2023/08/10/top-surgery-study-regret-rate/
Publication date: Aug 10, 2023, Written by Amelia Hansford.
Title: Almost no one who has top surgery regrets it, study says
Subtitle: A study into the rate of regret for top surgery has confirmed that it improves life for the overwhelming majority.
I'm commenting on some of the factual statements made in this article. For the whole article, please visit the page linked above. My comments are made in italic. Note that when I included excerpts from sources that cite references, those citations will be included as numbers after a full stop, like so: text.5, or text.4-8 .
The subtitle is deceptive. The study (Bruce et al. 2023 with doi:10.1001/jamasurg.2023.3352, see https://jamanetwork.com/journals/jamasurgery/fullarticle/2808129) set out to answer the question: “What is the rate of regret and satisfaction with decision after 2 years or more following gender-affirming mastectomy?”, it did not ask or look at the quality of life of people who had undergone such a surgery for gender reasons. A statement such as made in the subtitle is not supported by the study this article discusses.
One can also discuss whether there is a difference between ‘regret and satisfaction with decision [to undergo this surgery]’ and regret and satisfaction with surgical outcome (the latter might be what some people assume is meant when people talk about surgery satisfaction/regret).
Research published on Wednesday (9 August) involved 235 patients who had undergone a gender-affirming mastectomy – top surgery – over 30 years. It tracked their levels of satisfaction across the two-year period following their surgeries.
Comment: This is a misrepresentation of the data. The study deemed 235 patients eligible for inclusion in this study but only 139 responded (59.1% response rate), meaning that the other 40.9% of eligible patients did not respond. People may not respond to a request to take part in a study like this for any number of reasons. Maybe they have moved house and the clinic was not able to find them. Maybe the person has died. Maybe the person does not want to take part (which could be for any number of reasons including having detransitioned and/or regretting their cosmetic double mastectomy).
It is not correct that the “levels of satisfaction across the two-year period following surgery” was tracked. Patients who had undergone this surgery at least 2 years prior to the start of the investigation were included meaning that the study looked at satisfaction of patients who had undergone the surgery 2 years ago or longer. A patient who underwent surgery in 2020 would have been 2 years post op (post operative, post surgery). A patient who underwent surgery in 1990 would have been 30 years post op at the time of the survey. If we look at the study data (Table 1) we can see that the median time since surgery was 3.6 years for the patients who responded (and 4.6 for the non-responders). This means that the satisfaction and regret data is for a group of patients where half of the patients had their surgery less than 3.6 years ago and the other half had their surgery more than 3.6 years ago. The median age at time of surgery was 27.1 for responders. It is possible (almost) no patients were included who were minors at the time of surgery. That is not an issue on its own but when people talk about regret being low for a cosmetic double mastectomy, they fail to understand or include the information that this may only apply to patients who underwent this surgery as adults, that this may only apply to a group that had surgery relatively recently, etc. In the grand scheme of things, 2 years since a big surgery is still quite a short period. Someone may still be very happy having undergone this surgery as they may still be in the frame of mind where this surgery makes sense to them (transgender ideology). Especially in girls and women who had this surgery when they were (very) young and/or before they reached the stage in life where they would want to start a family, it may be too early to say whether this surgery was the best for this person in the truly long-term (20+ years). When you get this surgery at 18, a 2-year follow up is way too soon as you still may have 70 years of life left during which you will change a lot, especially until you reach the age of 25-30 and become an independent adult. And then there's the next ten or so years until you reach 40, the older limit for when people start families. Some of these women may not realise what they have given up until their friends and other peers start having children.
It is interesting to note that the authors of the research paper write “Postoperative time for respondents ranged from 2.0 to 23.6 years.” This means that the patient who underwent her surgery 24 years ago did so in the year 1996 but the researchers used the time period 1990-2020. So, did no one get this surgery at their clinic between 1990 and 1996? Taking into consideration the low median time since surgery (3.6 years for responders), this suggests that the frequency at which this surgery was performed is much higher in the past handful of years than before that. Why are they not discussing this?
The authors of the research paper write “All patients met the World Professional Association for Transgender Health standards of care surgical requirements at the time of their operation prior to undergoing mastectomy.13” But these standards of care (SOC) have changed over time meaning that a patient who was approved for surgery in 1990 (the earliest year included in this study) may have ben evaluated or approved in a different way than a patient who was approved in early 2020 (the most-recent period for inclusion). This would mean that you are essentially comparing apples and oranges. A patient who underwent surgery in 1990 may have been evaluated with either the SOC of 1990 or the previous version which is from 1981; a patient who underwent this surgery in the mid 1990s would also have been evaluated using the SOC of 1990. The patient who underwent surgery early 2020 will have been evaluated according to SOC from 2012, or may have ben operated on under ‘informed consent’ meaning that no in-depth evaluation took place and the patient was merely ‘informed’ on the risks of the surgery they wanted to undergo.
Patient inclusion criteria include “were older than 18 years at the time of survey administration”. The authors are looking at people who had this surgery at least 2 years prior. But their 18 years of age minimum to be included in this study would exclude girls who underwent this surgery at/before the age of 16 (who would now be at least two years post op). It may be the case that this centre does not perform a cosmetic mastectomy for gender reasons on minors, but we know that this does happen elsewhere (see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9555285/ describing girls aged 12-17 at the time of referral for their double mastectomy for gender reasons; note in Table 1 that 60% had a history of depression, 61% had a history of anxiety, and 91% had satisfaction with 1% regret and 8% ‘not documented’…).
It is of interest to mention some of the demographic results of the study at hand showing in Table 1 that depressive disorder was seen in 68% of responders and 44% of non-responders, anxiety disorder was seen in 70% and 44% respectively, and bipolar disorder was seen in 10% and 11% respectively. This suggests that the majority of the girls/women included in this study who underwent a double mastectomy for gender reasons had at least one and possibly multiple mental health conditions. One can question whether people with such conditions are best placed to make permanent decisions about removal of organs (breasts) for cosmetic reasons.
The study from the University of Michigan (link: https://jamanetwork.com/journals/jamasurgery/fullarticle/2808129?guestAccessKey=43a62af8-3042-4678-b29d-3430c3ff98c1&utm_source=For_The_Media&utm_medium=referral&utm_campaign=ftm_links&utm_content=tfl&utm_term=080923) – published in the journal JAMA Surgery – found that the median satisfaction rate (link: https://www.thepinknews.com/2022/09/30/top-surgery-trans-young-people-study/) among those surveyed was five out of a five.
Comment: It is unclear why PinkNews links to another one of their articles/opinions because that article is about another study.
It is incomplete to make the statement that the median satisfaction rate was five out of five when you do not explain what ‘median’ means as many readers will not be familiar with this statistical concept or at least not know the name for it. The median is the value right in the middle of 50% who have a lower value and 50% who have a higher value (read more about it here: https://en.wikipedia.org/wiki/Median ). The median of something can be of value when looking at population data but it has its limitations. For instance, if we are looking at body height, we could have a median of, say, 178 cm, but this would hide the fact that the patient group consists of both males and females, two different groups with a different median height for their own subgroup. Median height of males and females lumped together may not mean very much… You would have a similar issue if you were looking at some variable, say happiness, where the median value may be ‘pretty happy’ but where a small group of extremely depressed people is hidden because they are a minority. To say that the population as a whole is ‘pretty happy’ and then conclude that ‘life is good here’ is doing a disservice to the minority who are extremely depressed. You could use the median of a whole population to compare countries with each other, or compare age groups, but to only look at a median value for a whole group and then conclude that there are no problems is lazy and neglectful of minority groups with extreme results (in this example, the minority with (extreme) depression).
PinkNews conveniently mentions the median ‘satisfaction with decision scale score’ as 5.0 (out of a scale with a max of 5.0) but it omits to include the mean (what is often called ‘average’), which was 4.8 (with standard deviation 0.6) out of a max 5.0 (see Table 3 of the research paper). This suggests that PinkNews doesn’t understand what the data mean or that they simply picked the higher number out of the results table…
Comment: This is blatant misrepresentation of the research results. The median for ‘decision regret scale score’ was indeed 0.0 on a 100-point scale. However, the mean was 4.2 (with standard deviation 12.1) (Table 3). Secondly, the paper states “No individuals in this study presented with or underwent a reversal procedure (ie, breast reconstruction) at our institution following mastectomy.” This means that there were no patients who presented with or underwent a reversal procedure at THEIR clinic. It is possible that some patients who had their mastectomy at this clinic underwent a reversal procedure elsewhere but simply did not inform this clinic about it. ‘Reversal procedure’ is not included in Table 1 of their results. It is possible that the statement the authors made on reversal procedure, flawed as it is, also only applies to the respondents and not the non-responders (who might be more likely to have undergone a reversal procedure elsewhere). Secondly, PinkNews chose to not include data on ‘revision procedure’, seen in 22% of the responders and 15% of the non-responders. This suggests that there were issues with the first surgery that they tried to fix in revision surgery/surgeries. It is important for people who are researching this surgery to know that they may end up with complications that require immediate surgery or functional or cosmetic issues that require surgery at a later stage.
The research involved only a single medical centre, so for the findings to be “generalisable”, future research will need to involve people in a number of places, the study authors said.
Comment: Interesting that PinkNews trimmed down the following from the research article to the sentence above: “Finally, participants were recruited from a single academic center with a uniform approval process to undergo surgery and may not be representative of other sites. A multicenter study with multiple practice types is needed to ensure the generalizability of these results and to examine subpopulations.” It would have been interesting had PinkNews included the information about a uniform approval process and that there may be subpopulations that are not yet identified/investigated.
They noted that a recent review of research suggested the regret rate for transfeminine and transmasculine surgeries was one per cent.
Comment: This statement is deceptive, speculative, and does not contain the necessary context. The research paper states: “Although regret is often discussed as a serious concern surrounding gender-affirming surgery, evidence that supports this claim is limited. A recent systematic review reported a regret rate of 1% following transfeminine and transmasculine procedures; however, this analysis relied on secondary coding of studies with a variety of measures and a broad range of follow-up periods from 0.8 to 9 years.4 Individual studies have also shown low rates of regret but have been limited by sample size and measurements that relied on chart review, ad hoc instruments, and use of changes in gender identity postoperatively as a proxy measure.4-9 While prospective studies have shown high short-term satisfaction with decision, follow-up time has remained limited, ranging from months to 1 year.10,11 The lack of data surrounding satisfaction with decision and decisional regret leaves clinicians unable to counsel patients on these important long-term outcome measures and legislators ill-equipped to draft evidence-based policy.” PinkNews is misrepresenting what was said and left out important context. They also did not bother to cite/link to that ‘recent review’...
Invited commentary on the survey commended the research, with the findings said to “support other studies with shorter follow-up” and demonstrate the “stability of surgical results”.
“This study contradicts claims that regret following gender-affirming surgery may not manifest for many years and highlights the disproportionate criticism encountered in gender-affirming care compared with other surgical disciplines.”
Comment: PinkNews does not cite/link to this ‘invited commentary’ but after looking online I found the following: https://jamanetwork.com/journals/jamasurgery/article-abstract/2808135. However, this isn’t publicly available so as general public we can’t judge one way or another whether this commentary is good science or not, and it is not clear whether PinkNews had access to it. Based on the research paper itself, the commentary makes no sense. The limited bit I can see from the commentary mentions the 30-year period [the time period for which they selected patients who underwent surgery] but then states the study findings “support those of other studies with shorter follow-up…”. Ok, so the current study does have a relatively short follow up? Yes, it does, see Table 1 showing a median time since surgery of just 3.6 years. And “…and demonstrate the stability of surgical results”. There is no way to say that with such a short median follow-up period. The commentary also states “This study contradicts claims that regret following gender-affirming surgery may not manifest for many years…” This is in direct contradiction with the research paper itself where the authors acknowledge “The association of time elapsed with regret specifically is unclear, with systematic review suggesting that regret may increase over time.22” The commentary continues “…and highlights the disproportionate criticism encountered in gender-affirming care compared with other surgical disciplines.” I can’t tell what they are referring to. The research paper does not seem to comment on this, so this is just the opinion of the persons who wrote the commentary. It is worth noting that the authors of the commentary, Ian T. Nolan MD, Brielle Weinstein MD, and Loren Schechter MD all work at the Division of Plastic and Reconstructive Surgery, Rush University Medical Center, Chicago, Illinois, all three are (involved with) performing gender surgeries. Their choice for line of work possibly introduces bias so I consider their commentary, whatever it is as I can’t read the full piece, as opinions and not facts. Opinions are like assholes, we all have one.
Note that the 'invited commentary' has as the title 'Low Rate of Regret After Gender-Affirming Mastectomy Highlights a Double Standard'. This is an interesting choice of words considering the research paper itself uses more precise wording: '...Regret and Satisfaction With Decision...'. The 'with decision' part is important and would have been included in the title of the commentary had the commentary authors been precise. It's also not clear what they mean with a double standard, I am guessing they discuss this in the part of the commentary that is not publicly accessible.
It is of value to include the limitations of the research study as written by the authors of the study: “This study had several limitations. Although this study includes robust long-term data surrounding regret and satisfaction with decision following gender-affirming mastectomy, the cross-sectional nature of this study and variation in follow-up time among participants pose the possibility of recall bias. The association of time elapsed with regret specifically is unclear, with systematic review suggesting that regret may increase over time.22 There were also 13 individuals with completely discordant regret and satisfaction scores, which may represent difficulty with following survey instructions, as these instruments are reverse coded from each other. These individuals were not included in the analysis given this concern. Unaccounted responder bias may also skew results. Interestingly, the only significant differences between responders and nonresponders were time since surgery and the rates of depression and anxiety diagnoses at the time of surgery in the medical record, with responders having higher diagnosis rates. While the rates of diagnoses differed between groups, the rates of medication use associated with anxiety and depression at the time of surgery did not (table 3 in Supplement 1). This discrepancy suggests that variation in documentation may contribute to this difference between responders and nonresponders. Furthermore, we were unable to determine procedures performed outside of our institution, which might have led to an underestimation of procedures following mastectomy. Clinician-level factors, such as patient counseling, were not collected and have been shown to be associated with regret and satisfaction with decision.16,17,22 Finally, participants were recruited from a single academic center with a uniform approval process to undergo surgery and may not be representative of other sites. A multicenter study with multiple practice types is needed to ensure the generalizability of these results and to examine subpopulations.”
A cross-sectional study is a study that collects data from a population at a specific point in time. In this case, this was in the period February 1 through July 31, 2022 during which they collected data for patients who underwent this surgery between January 1, 1990, and February 29, 2020. Some patients will have only been 2 years after this surgey whereas others may have had this surgery 30 years ago. It is possible that people's view on the decision to have this surgery changes as more time has passed since their surgery. There may also be many other factors that might influence decision regret such as the age at which someone had this surgery, whether the patient already had children before she underwent this surgery, whether the patient was and remained childfree for life (the choice to never have children), what the size of her breasts was before surgery (some of these women may have been so uncomfortable with having very heavy breasts they found gender issues and a mastectomy as a solution, when she may have been equally happy or possibly happier had she undergone a breast reduction instead), etc.
Recall bias means that in the case of this study, patients who had a certain kind of outcome may more often have responded to be included in this study than patients who had another kind of outcome. For instance, if you had no regret with your decision to undergo a cosmetic double mastectomy for gender reasons, you may have been happy to be part of this study, sharing your positive experience with your surgery decision. On the other hand, if you regret the decision to undergo this surgery, you may not want anything to do with these doctors that, in your view, hurt you. Another example could be that patients who regretted their decision to have this surgery are more likely to have ended their life or died of other causes than patients who did not regret this decision. The existence of such outcomes would all distort the result of this study towards a more favourable outcome for this kind of surgery.
Regret rates for gender-affirming surgeries are often overstated by right-wing, anti-LGBTQ+ groups (link: https://www.thepinknews.com/2023/05/20/nebraska-megan-hunt-gender-affirming-care/) in an attempt to justify restrictions on trans people’s care.
Comment: PinkNews simply link to another one of their opinion pieces that does not even support the statement they make here, yet that piece does include proven medical misinformation/falsehoods…
These restrictions often extend to trans youth, despite no medical organisation offering gender-affirming surgeries to under-16s, while under-18s are generally permitted to undergo them only under exceptionally rare circumstances.
Comment: This is a blatant lie. See this study on minors undergoing a cosmetic double mastectomy for gender reasons: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9555285/. There are also hundreds if not thousands of accounts of minors and adults talking about when they were minors undergoing this surgery for gender reasons. These are also discussed in The New York Times article from September 26, 2022 by Azeen Ghorayshi titled ‘More Trans Teens Are Choosing ‘Top Surgery’ ‘ (see https://www.nytimes.com/2022/09/26/health/top-surgery-transgender-teenagers.html).
Final comment: It is always interesting to take a look at how a study was funded and what kinds of ties/interests the authors may have. We can see some of the authors of the research paper received funding from The Plastic Surgery Foundation, and some of the authors perform(ed) gender-affirming surgeries. One can argue that these things can introduce bias and these people may not be the best researchers for a study on surgery decision satisfaction/regret. The same can be said for the authors of the 'invited commentary'. Ideally, research on outcomes of medical transition would be performed by people who are not at all involved with the process of medical transition, but it may be hard to find researchers with the right knowledge and experience who are not also directly involved in transing patients.