Independent report

Review of suicides and gender dysphoria at the Tavistock and Portman NHS Foundation Trust: independent report

Published 19 July 2024

Applies to England and Wales

By Professor Louis Appleby, University of Manchester
Department of Health and Social Care adviser on suicide prevention

Aim of this review

I have reviewed data provided by NHS England (NHSE) on suicides by young patients of the gender services at the Tavistock and Portman NHS Foundation Trust, based on an audit at the trust. The specific aim is to examine evidence for a large rise in suicides claimed by campaigners.

Summary of conclusions

1. The data do not support the claim that there has been a large rise in suicide in young gender dysphoria patients at the Tavistock.
2. The way that this issue has been discussed on social media has been insensitive, distressing and dangerous, and goes against guidance on safe reporting of suicide.
3. The claims that have been placed in the public domain do not meet basic standards for statistical evidence.
4. There is a need to move away from the perception that puberty-blocking drugs are the main marker of non-judgemental acceptance in this area of health care.
5. We need to ensure high quality data in which everyone has confidence, as the basis of improved safety for this at risk group of young people.

Suicide risk in gender dysphoria

The evidence on suicide risk in children and young people with gender dysphoria is generally poor. Most studies are methodologically weak, being based on online surveys and self-selected samples and coming from biased sources. However, there are good reasons to believe that their risk is high compared to other young people. They have often experienced prejudice and intimidation, isolation and family conflict. They may have mental health conditions such as depression and anxiety. There are high rates of autism. These are known risk factors - suicide in any group is usually the result of multiple risks acting in combination.

It therefore seems reasonable to assume that services offering non-judgemental support may contribute to lower risk. However, the evidence for “gender-affirming care” in the form of puberty-blocking drugs is unreliable. In contrast, a robust study from Finland published earlier this year (Ruuska et al, BMJ Mental Health 2024) reported that suicide risk was reduced after gender reassignment but that the improvement was explained by the treatment of co-existing mental ill-health.

Public discussion of suicide

Responsible reporting of suicide in the media is an important strand of suicide prevention, and a central feature of the national suicide prevention strategy in England. Guidance has been developed by Samaritans, originally for the news media but with wider applicability to any public discussion of suicide, and increasingly relevant to social media.

The risks include:

• alarming stories about suicide causing distress to people who are themselves at risk
• identification - when someone sees in themselves a connection with a person who has died by suicide; leading to:
• imitation and suicide clusters in people with similar characteristics

As a result, the media - and users of social media - are asked to:

• ensure that any claims about suicide are evidence-based and from a reliable source
• avoid alarming and dramatic language
• avoid the impression that suicide is the expected or likely outcome in certain situations
• avoid oversimplifying suicide by attributing it to a single cause which could be the basis of identification

Claims

The claims have been led by the legal campaign group the Good Law Project, who are challenging the decision by the previous Health Secretary to end the prescription of puberty-blocking drugs by private clinics to children and young people with gender dysphoria.

The central claim, made on X (formerly known as Twitter), is that there has been a large rise in suicide by current and recent patients of the Gender Identity Development Service (GIDS) service at the Tavistock since an earlier restriction of puberty-blocking drugs that followed a High Court decision in a case (Bell v Tavistock) in December 2020. The rise is described as a “surge” in suicides and “an explosion”, indicating a substantial and, by implication, unequivocal increase. There are multiple references to children dying in future because they are unable to access puberty-blocking drugs.

This claim is said to be based on unpublished data provided by 2 members of staff at the Tavistock, described as whistleblowers. On Twitter/X the evidence is presented in screenshots of extracts from the records of Tavistock Board meetings and other documents. These variously refer to suicides, deaths from unspecified causes and “safety incidents”. A specific claim is that there was one suicide by a patient on the GIDS waiting list in the 3 years before the High Court judgment, and 16 deaths (rather than suicides) in the 3 years after the judgment. The whistleblowers are said to have alleged a cover-up by NHSE.

These claims have been retweeted thousands of times by other campaigners and members of the public. They have been repeated by some leading journalists, though there is nothing to suggest that they have examined the evidence for themselves. They too have adopted the language of “dying children”.

NHS England appraisal of Tavistock audit 

I have examined the figures provided by NHSE on deaths in each year between 2018-19 and 2023-24. They are based on an internal audit by the Tavistock of deaths among current and former GIDS patients, divided by age (under 18 or 18 plus) and cause of death (suicide or other/suicide not confirmed).

It is important to acknowledge in describing suicide statistics that the figures are not dry data; they represent real lives lost.

The numbers are small: by this breakdown, by year, age and cause, the highest count is 2. Conventional practice in presenting small numbers, based on guidance from the Office for National Statistics, is not to present figures lower than 3, to avoid possible identification of individuals. In this review I refer to aggregate figures only.

In this period of 6 years the data show a total of 12 suicides: 6 in the under 18s, 6 in those 18 and above. In the 3 years leading up to 2020-21, there were 5 suicides, compared to 7 in the 3 years after. This is essentially no difference, taking account of expected fluctuations in small numbers, and would not reach statistical significance. In the under 18s specifically, there were 3 suicides before and 3 after 2020-21.

Alongside the figures, there is a summary of the problems faced by the young people who died. These include mental illness, traumatic experiences, family disruption and being in care or under children’s services.

These figures clearly do not support the main claim that suicides have risen steeply since the High Court judgment. They do not support the claim of one waiting list death before and 16 after the judgment. The information confirms the multiple factors that contribute to suicide risk in this group.

There are limitations to this analysis, reflecting data quality and comprehensiveness. First, the figures are for “NHS years”, April to March, so the timing of the High Court judgment does not fall neatly between years. Secondly, these figures do not include details of where in the care pathway the patients were when they died. Thirdly, they give no further information on the deaths described as being from other causes or “not confirmed as suicide”.

I have also had sight of the internal audit by the Tavistock on which the NHSE summary is based. There are small differences in the time period covered and in certain figures. There is more information about the care received and about dates of death.

The audit suggests that some of the deaths referred to as “not confirmed as suicide” are likely to have been unnatural deaths, possibly implying suicide as the cause, although there is little detail. Including these cases, however, would not affect the overall conclusions because, with small numbers, single-figure fluctuations can be expected.

The audit confirms the multiple risks that many patients had faced. It shows that the deaths occurred at different points in the care system - including waiting, inpatient care and post-discharge. Concerns are expressed about patients who are waiting or at a transition from one care setting to another. Any causal attributions to any one element of care would go beyond the data and should be avoided.

NCMD figures

NHSE have also provided figures from the National Child Mortality Database (NCMD). These are suspected suicides in all people under 18 in England. Cases are notified to NCMD with additional information about the individuals who have died, including the presence of distress relating to gender identity.

References to gender distress in the NCMD cases occur more frequently in the most recent years, the highest single year figure being in 2021-22. It is not possible at this stage to link these cases to the GIDS audit. There is no information on whether these young people were reporting gender dysphoria, seeking help from the NHS, or in contact with the Tavistock. The rise is more likely to reflect young people increasingly expressing their distress through conflicts about gender and more frequent recognition by professionals.

NCMD also report a higher figure for suicides in under 18s as a whole in 2021-22. This may be the result of random fluctuation in small numbers but it does coincide with a crucial year in the COVID-19 pandemic, when restrictions may have adversely affected the emotional health of some young people, such as autistic people. Co-occurrence of autism and gender dysphoria is well established.

In summary, I think it is unlikely that the higher NCMD figures have any link to the availability of puberty blockers but they do need further exploration.

Conclusions

1. The data do not support the claim that there has been a large rise in suicide by young patients attending the gender services at the Tavistock since the High Court ruling in 2020 or after any other recent date. The figures for the 6 years covered in this review are 12 suicides in total, 2 per year on average, of whom half were under 18. With small numbers, single-figure differences can be expected and causal explanations are unreliable.

The patients who died were in different points in the care system, including post-discharge, suggesting no consistent link to any one aspect of care. They had multiple social and clinical risk factors for suicide.

However, it is likely that there has been a rise over a longer period as young people at risk have increasingly presented with gender dysphoria and referrals to GIDS have risen.

There is a degree of uncertainty about the deaths recorded as “suicide not confirmed”. It is possible that more information on these cases would result in amended figures for individual years but the numbers remain too small to affect my conclusions.

2. The way that this issue has been discussed on social media has been insensitive, distressing and dangerous, and goes against guidance on safe reporting of suicide. One risk is that young people and their families will be terrified by predictions of suicide as inevitable without puberty blockers - some of the responses on social media show this.

Another is identification, already-distressed adolescents hearing the message that “people like you, facing similar problems, are killing themselves”, leading to imitative suicide or self-harm, to which young people are particularly susceptible.

Then there is the insensitivity of the “dead child” rhetoric. Suicide should not be a slogan or a means to winning an argument. To the families of 200 teenagers a year in England, it is devastating and all too real.

3. The claims that have been placed in the public domain do not meet basic standards for statistical evidence. To be reliable, evidence should be objective, unbiased and open to independent scrutiny. It should admit uncertainty.

Campaign groups are often selective about evidence - there is nothing wrong with this until it becomes misleading and potentially harmful. The evidence put into the public domain for an “explosion” of suicides is not unbiased nor has it been independently verified. There seems to be no suicide expertise behind the claims.

4. Suicide by any young person is a profound tragedy: it should be seen as an indictment of our society. Young people with gender dysphoria may well have experienced ostracism and abuse, and their distress is likely to be heightened if services are perceived as rejecting. It is unfortunate that puberty-blocking drugs have come to be seen as the touchstone issue, the difference between acceptance and non-acceptance. We need to move away from this perception among patients, staff and the public.

This is a group of young people who need compassion and security, skilled clinical assessment, early treatment for mental illnesses such as depression, support within their families and schools and online, and an expectation of recovery and a fulfilling future. It is vital that these are the assurances the NHS and its partner agencies are able to convey.

5. In the end this is about a group of young people at risk of suicide and our collective responsibility to their safety. This means specialist health services with the capacity to respond to rising demand and appropriate skills in general services. It means a measured public discourse, making sure we do not stoke up prejudice or cause unnecessary alarm to the young people and their families.

We need to ensure also that we have high quality data in which everyone has confidence. The number of deaths should be monitored, not only in gender services but other mainstream databases, as is now happening in NCMD and in my own unit, the National Confidential Inquiry. Future prevention will depend on it.