A comprehensive scientific review confirms inequalities in mental health amongst LGBTQIA+ people

For me, the starting point is that sexual and gender diversity is not a pathology. I would focus on those who orchestrate, perpetrate and legitimise violence, and on the institutions that allow it to continue. Sometimes we scrutinise the distress of those on the receiving end of the attacks whilst largely overlooking those who perpetrate them. The minority stress model helps us to understand how one thing is connected to the other. On top of the difficulties anyone might face, there is an additional burden linked to discrimination. There are insults, attacks or rejection by one’s family, but the strain continues when you have to anticipate the next incident, watch what you say or work out where you can mention your partner without putting yourself at risk. You may even end up internalising the negative messages you receive. Hiding oneself can offer protection in a hostile environment, but having to make these calculations constantly takes its toll. This build-up of stress can contribute to anxiety, depression or suicidal behaviour. It is not an inevitable consequence of being LGTBIQA+.

I find the article to be a well-structured piece of work that organises the evidence and distinguishes which associations have the strongest statistical backing. Among its most robust findings, lesbian, gay and bisexual people are at approximately twice the risk of depressive disorders and 2.5 times the risk of suicide attempts during the previous year, compared with the control group. It also finds more symptoms of anxiety and non-suicidal self-harm amongst bisexual people than amongst lesbians and gay men, and a higher frequency of certain eating behaviours, such as purging, amongst gay teenage boys compared with their heterosexual peers. That said, I would distinguish the quality of this review from that of the studies it brings together. The included reviews are of low or critically low methodological quality; cross-sectional studies predominate; and a large proportion of the research originates from North America. Some associations have consistent statistical support, but this does not prove causality nor does it allow the figures to be simply extrapolated to Spain or to all the contexts encompassed by the acronym.

These findings are consistent with evidence accumulated over decades. In our review on protective factors, we identified the importance of support from family, school and friends. Our study with professionals in Spain revealed shortcomings in training and resources. And in the Spanish national CLICK report, we document both experiences of violence and vulnerability, as well as pride, joy and a sense of belonging when safe spaces exist. I am keen to highlight both aspects, because LGBTIQA+ lives are not reduced to suffering, and the support available can significantly change the experience of growing up.

Nor would I overlook the social and political context. The rhetoric of VOX, Milei, Trump or Bolsonaro—which portrays LGBTIQA+ diversity as a threat or its recognition as indoctrination—can contribute to normalising rejection. This is not merely a discussion for the airwaves. Drawing on the minority stress model, we need to analyse how such messages can legitimise discrimination and heighten the sense of threat felt by those who are singled out. The article does not assess the specific impact of such rhetoric, but I highlight it as part of the context we need to investigate and transform. For me, the implications involve intervening with those who perpetrate abuse, ensuring institutional responses, and sustaining support networks. Psychological care is necessary when distress is present, of course, but it makes little sense to ask a teenager to learn to cope better whilst leaving the environment that is harming them untouched.

The article presents a review of 22 systematic reviews, a highly robust methodology for generating and consolidating scientific evidence. The findings are clear: the LGBTIQA+ population experiences poorer mental health outcomes. Furthermore, it includes comparative data for adolescents and adults, as well as comparisons amongst the different members of the LGBTIQA+ community.

There is strong evidence of a higher prevalence of depression or depressive symptoms, anxiety, self-harm or suicide attempts (particularly amongst young people), as well as eating disorders and issues with body image or post-traumatic stress (which is particularly prevalent amongst the trans population). The study links these findings to exposure to trauma and childhood trauma, discrimination and bullying, and other life stressors such as family rejection, victimisation due to their identity, or the risk of difficulties accessing housing.

Furthermore, the LGTBIQA+ population faces stigma and discrimination within the healthcare system, which acts as a barrier to accessing healthcare and mental health prevention, leading to a risk of lack of access to treatment or a high risk of suicide.

It is worth noting that trans, intersex and non-binary people are under-represented in systematic reviews specifically focusing on these groups, which highlights the need for further research to better understand their realities. Nevertheless, the evidence generated by these studies is quite robust, including large-scale studies with a low risk of bias.

In summary, the study brings together evidence on how the social discrimination and stigma experienced by the LGTBIQA+ population lead to poorer mental health, with a higher risk of suicide and greater psychological and physical suffering. A healthcare system that is welcoming and respectful towards the LGTBIQA+ population would act as a protective factor for their mental health.

The study by Oliva et al. in World Psychiatry is an important contribution to our understanding of the differences in mental health between LGBTQIA+ populations and the general population. The authors have carried out a thorough statistical analysis of a considerable body of previous research, which allows for a more precise assessment of the strength of the observed associations, particularly in relation to depressive disorders and suicidal behaviour.

Its main strength is also its main limitation: it provides a highly rigorous quantitative description of the existing differences, but does little to explain the mechanisms underlying them. The authors highlight the role of stigma and discrimination – factors that are undoubtedly relevant – although the design of the studies analysed does not allow for the specific contribution of these factors to be determined, nor for causal relationships to be established.

From a clinical perspective, it would have been interesting to explore certain variables in greater depth, particularly sex and exposure to hormone treatments in transgender populations. Sex should have been considered as a cross-cutting factor across the various analyses, as there are well-known differences between cisgender women and men in the prevalence of mental disorders, particularly depression, which is more common in women. These differences may influence the associations observed when comparing populations with different sex compositions. It is therefore important to distinguish between effects related to sex and those associated with sexual orientation and gender identity – dimensions that are not equivalent and which may interact with one another.

Another aspect worthy of attention is the role of stigma. In the scientific literature, there is a problem regarding its definition and how it is measured, as a distinction should be made between structural stigma—related to social norms and institutional policies—interpersonal stigma, which encompasses experiences of discrimination, harassment or rejection; and internalised stigma, referring to the internalisation of negative judgements about one’s own identity. Although the authors recognise these dimensions, it is not possible to determine from their analyses the specific contribution of each to the observed differences in mental health.

The findings relating to body image also merit special attention, as they reveal differences between women and men according to their sexual orientation. These differences could reflect both biological and sociocultural factors, including ideals of beauty, social pressures regarding physical appearance, and expectations associated with gender and sexual orientation. These are likely to be interacting processes, and it would be interesting to investigate their relative contributions.

Overall, I consider this to be a valuable piece of work, both for the considerable effort put into synthesising the literature and for the questions it leaves open for future research.

Overall, I consider the article to be a significant academic contribution, as it synthesises a vast body of scientific literature, classifies the findings according to different levels of reliability of the research results (from the most convincing to the weakest evidence) and identifies various gaps in knowledge that require further attention. These include the under-representation of certain groups within the LGBTQIA+ population, the lack of data on access to mental health care and on certain serious psychiatric disorders, the difficulties in determining which factors contribute to creating or perpetuating the observed inequalities, and the need to conduct methodologically sounder research in contexts other than North America.

Furthermore, the findings support the existence of certain inequalities in mental health, particularly in relation to depression and suicidal behaviour, and the review identifies greater vulnerability in some subgroups, for example, among bisexual people in relation to certain mental health indicators.

However, the limited methodological quality of some of the included reviews, the higher proportion of cross-sectional studies, and the fact that the majority of the research originates from North America make it difficult to establish causal relationships and limit the generalisability of the findings to other contexts.

One of the main implications of the article is that LGBTQIA+ identities should not, in themselves, be considered the cause of mental health problems. On the contrary, the findings highlight the importance of considering social factors such as stigma, discrimination, rejection and a lack of support, which can affect psychological well-being. In this regard, it is essential to promote preventive measures, ensure adequate healthcare and develop public policies that reduce the impact of these factors and improve the mental health and quality of life of these individuals. However, further research is needed to assess the effectiveness of interventions aimed at reducing these inequalities.

From a critical perspective, whilst the article presents a valuable synthesis strategy and helps to identify which associations have the strongest scientific support, the strength of the evidence varies depending on the results analysed. Consequently, its conclusions must be interpreted whilst taking into account the methodological limitations of some of the included studies and the need to continue expanding research in this field.

This study highlights the importance of recognising that the mental health of LGBTQIA+ people is not explained solely by individual factors, but by the social and structural context in which they live. The authors emphasise the role of structural stigma (laws, cultural norms and institutional policies), interpersonal discrimination — such as bullying or rejection by family — and barriers within the healthcare systems themselves, at a time when established rights are being eroded in several countries. In my view, sexual orientation and gender identity are still not systematically captured or analysed in much of the mental health research, which contributes to rendering these inequalities invisible. This umbrella review, which re-analyses 22 meta-analyses comprising 398 studies, helps to organise the existing evidence on the subject to date. Its most robust findings confirm that lesbian, gay and bisexual people are approximately twice as likely to experience depressive disorders and more than twice as likely to attempt suicide as the heterosexual population, and that gay adolescents exhibit more dieting and purging behaviours than their heterosexual peers. Within this group, bisexual people report higher levels of anxiety and non-suicidal self-harm than lesbians and gay men.

However, it should be borne in mind that the methodological quality of the included meta-analyses is low or critically low according to AMSTAR-2, a standard tool for assessing the quality of systematic reviews. This limits the reliability of some of the conclusions. Around 90 per cent of the studies are cross-sectional, which prevents inferring causality or trajectories, and the evidence comes mainly from North America and focuses more on lesbian, gay and bisexual people than on trans and non-binary people. There are no quantitative data on access to or quality of care, and almost none on severe mental disorders. Furthermore, these are differences between groups, not a risk applicable to every individual. Nevertheless, the findings support the need for early detection and clinically sensitive care that is responsive to diversity, for measures to combat stigma and discrimination, and for further longitudinal studies to clarify the mechanisms involved. This could help in designing measures to reduce social and structural stigma.

The study has a clear strength in that it brings together research that was previously scattered across different mental health issues and different population groups. Specifically, it incorporates 22 reviews covering 398 studies and applies common criteria to assess the consistency of their findings. This makes it possible to identify which inequalities emerge most consistently and which are still supported by limited evidence. However, it is important to distinguish between the quality of the procedure used to compile and analyse this information, and the quality of the studies from which it is derived. A well-organised review remains subject to the limitations of the available data on which it is based.

The first important point is that the existence of these inequalities is nothing new. The review by King et al., published in 2008 and included in this study, had already found a higher prevalence of depression and suicidal behaviour among lesbian, gay and bisexual people. The contribution of the new study lies in organising this accumulated knowledge, assessing its consistency and demonstrating that the differences are not distributed uniformly within the group. The latter point is of interest because referring to LGBTQIA+ populations as if they shared the same experiences and needs can obscure significant differences.

Among the most consistent findings is an incidence of depression approximately twice as high, and of suicide attempts approximately two and a half times higher over the past year, among lesbian, gay and bisexual people compared with the comparison groups. There are also more symptoms of anxiety and self-harm without suicidal intent among bisexual people than among lesbians and gay men. These figures describe differences between the populations studied. They do not allow us to predict the situation of a specific individual, nor do they imply that these problems have a direct link to, or stem from, experiences associated with their sexual orientation.

A similar situation applies to the findings on eating behaviour. Among gay adolescent boys, there are more instances of dieting and purging than among heterosexuals. Here, it is important to clarify what has been measured, as behaviour associated with an eating disorder does not, in itself, equate to a diagnosis. For example, knowing the frequency of certain weight-control practices does not allow us to conclude how many people meet the criteria for anorexia or bulimia.

Furthermore, particular caution is required due to the limitations of the original studies. All the included reviews were rated as having low or critically low quality. The authors highlight problems with the search strategies used in the previous studies, the justification for excluding certain studies, and the consideration of potential biases. This does not prove that the disparities are false, but it does necessitate a more nuanced interpretation of the term ‘high-credibility evidence’. In this paper, that phrase refers to compliance with certain statistical criteria. The fact that several studies find similar results lends credibility, but does not rule out possible common errors in participant selection or in the way problems were assessed.

Caution is also required when interpreting the causes. Most of the studies provide a snapshot of a specific point in time (they are cross-sectional). This allows us to observe that two groups differ, but it is insufficient to reconstruct how these differences have arisen. The authors interpret the findings using models that highlight the role of stigma, discrimination and rejection. This interpretation is consistent with previous research, but this review does not determine the extent to which each factor explains the findings. Furthermore, much of the data comes from North America, and the most robust results are concentrated among lesbian, gay and bisexual populations. It would therefore be inappropriate to apply all the figures directly to Spain or to extend them equally to all people falling under the LGBTQIA+ umbrella.

From a care perspective, the study reinforces the need to recognise these inequalities and to address each person’s individual circumstances. For example, it can guide professionals to explore experiences of rejection, available support or difficulties in seeking help, alongside the issues that led to the consultation.

In the field of public policy, the findings support paying attention to both care needs and the social conditions that may undermine well-being. However, the study does not compare interventions nor does it allow us to determine which would be the most effective. Nor did it find any systematic reviews with meta-analyses that met its criteria for assessing access to mental health care. To evaluate the healthcare response, it would be necessary to examine whether people receive help when they need it, whether that care is tailored to their difficulties, and whether they encounter barriers or experiences of discrimination within the system itself. Identifying an inequality is a necessary step; verifying that the measures taken succeed in reducing it requires further evaluation.