1. Bibliographic Information

1.1. Title

Prevalence, severity and risk factors for mental disorders among sexual and gender minority young people: a systematic review of systematic reviews and meta-analyses.

1.2. Authors

The authors of the paper are Jonathan O'Shea, Rebecca Jenkins, Dasha Nicholls, James Downs, and Lee D. Hudson. The affiliations provided in the paper are indicated by superscripts, suggesting they are associated with academic or medical institutions, likely in the fields of psychology, psychiatry, and public health. For instance, the lead author, Jonathan O'Shea, is a key researcher in this study.

1.3. Journal/Conference

The paper was published in Child and Adolescent Psychiatry and Mental Health. This is a peer-reviewed, open-access journal that focuses on the diagnosis, treatment, and prevention of mental health disorders in children and adolescents. It is a reputable venue in its field, known for publishing high-quality research relevant to clinical practice and public health policy for young people's mental health.

1.4. Publication Year

2024. The paper was received on April 9, 2024, accepted on August 1, 2024, and published online on August 14, 2024.

1.5. Abstract

The paper presents a systematic review of systematic reviews and meta-analyses to provide a comprehensive overview of mental health among sexual minority (SM) and gender minority (GM) young people (aged 25 and under). The authors searched four major databases and included 42 reviews, all of which were assessed as low quality. Their own meta-analysis found the prevalence of depression to be 26% among SM youth and 46% among GM youth. They also found that SM youth experience greater depression severity than heterosexual youth (Hedges’ g = 0.38). GM youth also reported more severe symptoms than their cisgender peers. The review identifies numerous proximal (internal) and distal (external) risk factors associated with these mental health disparities. The authors conclude that past reviews consistently show a heightened risk of mental disorders for SM and GM youth and urge mental health services to be aware of these disparities and adapt their care accordingly.

1.6. Original Source Link

The original source link provided is: /files/papers/69056be00b2d130ab3e04804/paper.pdf. Based on the publication details and the presence of a DOI in the supplementary information link (https://doi.org/10.1007/s00787-024-02552-1), this paper has been officially published in a peer-reviewed journal.

2. Executive Summary

2.1. Background & Motivation

The core problem the paper addresses is the well-documented but fragmented and often inconsistent body of evidence regarding mental health disparities in sexual and gender minority (SGM) young people. While it's generally known that this population faces a higher burden of mental disorders, existing research, particularly systematic reviews (SRs), suffers from several key challenges:

• Methodological Inconsistencies: SRs often combine studies with vastly different age groups, fail to use rigorous definitions for mental disorders, and are not always clear in their methodologies.

• Lack of Separation: Many reviews group sexual minority (SM) and gender minority (GM) individuals together, obscuring potentially important differences in their experiences and health outcomes.

• Narrow Scope: Existing reviews tend to focus on a single mental disorder (like depression) or a single risk factor (like bullying), failing to provide a comprehensive picture that reflects the complex, co-occurring nature of mental health issues.

  This proliferation of inconsistent and methodologically flawed reviews makes it difficult for clinicians, policymakers, and researchers to get a clear, reliable overview of the field. The paper's innovative entry point is to conduct a systematic review of systematic reviews (also known as an umbrella review). This high-level approach allows the authors to synthesize the entire body of review-level evidence, identify consistent findings, critique the quality of existing reviews, and provide a definitive summary of the current state of knowledge.

2.2. Main Contributions / Findings

The paper's primary contributions are:

1. Comprehensive Synthesis of Existing Reviews: It rigorously analyzes and synthesizes the findings from 42 separate systematic reviews, providing the most comprehensive overview of the topic to date.
2. New, Rigorous Meta-Analyses: By extracting and re-analyzing data from primary studies cited in the reviews, the authors conduct new meta-analyses with stricter criteria (age ≤ 25, separating SM and GM groups). The key quantitative findings are:
   • The pooled prevalence of depression is 26% (95% CI 21–32%) among SM young people.
   • The pooled prevalence of depression is 46% (95% CI 36–56%) among GM young people, highlighting their heightened vulnerability.
   • SM young people experience significantly greater depression severity than their heterosexual peers, with a small but statistically significant effect size (Hedges’ g = 0.38).
3. Structured Analysis of Risk Factors: The paper systematically identifies and categorizes a wide range of risk factors for mental disorders using the Minority Stress Model as a framework, separating them into proximal (e.g., internalized homophobia) and distal (e.g., family rejection, discrimination) factors.
4. Critique of the Research Field: A crucial finding is that all 42 included systematic reviews were of low or critically low quality according to the AMSTAR-2 assessment tool. This highlights a systemic problem in the quality of research synthesis in this field and points to a clear need for more rigorous review methodologies in the future.

3. Prerequisite Knowledge & Related Work

3.1. Foundational Concepts

3.1.1. Systematic Review (SR) and Meta-Analysis

• A Systematic Review (SR) is a type of literature review that collects and critically analyzes multiple research studies or papers. Unlike a traditional literature review, an SR uses a structured, explicit, and reproducible methodology to identify, select, and appraise all relevant research on a specific question. The goal is to provide an exhaustive summary of the current evidence.
• A Meta-Analysis is a statistical procedure used within a systematic review. It combines the quantitative results from multiple independent studies to calculate a pooled or overall effect. This increases statistical power and provides a more precise estimate of the effect's magnitude (e.g., how much more prevalent depression is in one group versus another).

3.1.2. Systematic Review of Systematic Reviews (Umbrella Review)

This is the methodology used in the current paper. It is a review that synthesizes the evidence from multiple existing systematic reviews on a related topic. It is particularly useful when many reviews have already been published, and their findings may be inconsistent. An umbrella review provides a high-level overview of the evidence and can also assess the quality of the reviews themselves.

3.1.3. Sexual and Gender Minorities (SGM)

The paper defines these groups as follows:

• Sexual Minority (SM): Individuals who do not exclusively identify as heterosexual (e.g., lesbian, gay, bisexual, queer).
• Gender Minority (GM): Individuals who do not exclusively identify with the sex they were assigned at birth (e.g., transgender, non-binary, genderqueer).
• SGM: An umbrella term for both sexual and gender minorities.

3.1.4. The Minority Stress Model

This is the central theoretical framework used in the paper to understand mental health disparities. Proposed by Ilan Meyer (2003), the model posits that SGM individuals experience unique stressors related to their stigmatized social status, which contributes to adverse health outcomes. These stressors are categorized as:

• Distal Stressors: External, objective stressful events and conditions. These are environmental challenges, such as prejudice, discrimination, harassment, and violence.
• Proximal Stressors: Internal, subjective processes that occur as a reaction to stigma. These include internalized homophobia/transphobia (directing negative societal attitudes inward), expectations of rejection, and the stress of concealing one's identity.

3.1.5. Effect Size (Hedges' g)

An effect size is a quantitative measure of the magnitude of a phenomenon. In this paper, Hedges' g is used to measure the difference in depression severity between two groups (e.g., SM vs. heterosexual youth).

• It is a standardized mean difference, meaning it expresses the difference between two group means in terms of their pooled standard deviation. For example, a Hedges' g of 0.38 means the average depression score of the SM group is 0.38 standard deviations higher than the average score of the heterosexual group.
• It is similar to another common measure, Cohen's d, but includes a correction factor for small sample sizes, making it more accurate in such cases. The paper interprets the size of the effect using Cohen's conventions: < 0.20 (minimal), > 0.20 (small), > 0.50 (medium), and > 0.80 (large).

3.1.6. Prevalence and Confidence Interval (CI)

• Prevalence: The proportion of a population that has a specific characteristic or disorder at a given time. It is usually expressed as a percentage.
• 95% Confidence Interval (CI): A range of values that is likely to contain the true population value. When the paper reports a depression prevalence of 26% with a 95% CI of [21%–32%], it means that while the study's best estimate is 26%, we can be 95% confident that the true prevalence in the entire population of SM youth lies somewhere between 21% and 32%. A narrower CI indicates a more precise estimate.

3.2. Previous Works

The authors position their work as a necessary response to the shortcomings of prior systematic reviews. They note that while numerous SRs exist, they are problematic. The paper critiques them on several grounds:

• Inconsistent Study Inclusion: Different reviews include different primary studies, leading to conflicting conclusions.

• Methodological Flaws:

  • Many reviews combine vastly different age groups (e.g., adolescents with adults over 25), which is inappropriate as risk factors and mental health presentations can differ significantly across the lifespan [11].
  • Some reviews have not been rigorous in defining and measuring mental disorders, sometimes including measures of general distress rather than clinically significant symptoms or diagnoses [12].

• Narrow Focus: Most SRs assess only one specific risk factor (e.g., cyberbullying) or one mental disorder (e.g., depression), failing to capture the complex interplay of factors and co-occurring conditions that SGM youth face in reality [13, 14].

  By conducting an umbrella review, the current paper aims to overcome these limitations by synthesizing the entire landscape of SRs and applying its own stricter, more consistent criteria for analysis.

3.3. Technological Evolution

The methodology in this field has evolved to handle the increasing volume of published research. The progression is as follows:

1. Primary Studies: Individual research projects that collect original data (e.g., a survey of 500 adolescents).
2. Systematic Reviews (SRs): Researchers begin synthesizing the findings from multiple primary studies to get a clearer picture.
3. Systematic Reviews of Systematic Reviews (Umbrella Reviews): As SRs themselves multiply, the field requires a higher-level synthesis to manage the "information overload" and resolve inconsistencies between reviews. This paper represents this advanced stage of evidence synthesis, aiming to provide a definitive summary of what the collective body of reviews tells us.

3.4. Differentiation Analysis

Compared to previous systematic reviews, this paper's core innovations are:

• Methodological Approach: It is an umbrella review, which provides a broader and more critical perspective than a standard SR by analyzing the reviews themselves.
• Strict Inclusion Criteria:
  • It strictly focuses on young people aged 25 and under, avoiding the issue of inappropriately mixing age groups.
  • It mandates the separation of SM and GM groups, allowing for a more nuanced analysis of their unique mental health profiles.
  • It requires mental disorders to be defined by contemporaneous validated tools or clinical diagnoses, ensuring higher quality and more clinically relevant data.
• Comprehensive Scope: Instead of focusing on a single disorder or risk factor, it aims to synthesize evidence across prevalence, severity, and a wide range of risk factors for multiple mental disorders, better reflecting the holistic experience of SGM youth.
• Quality Assessment of Reviews: A key part of its contribution is the formal quality assessment of the existing SRs using the AMSTAR-2 tool, which no single SR would do. This provides a critical commentary on the state of the research field itself.

4. Methodology

4.1. Principles

The core principle of this study is to employ a systematic review of systematic reviews (umbrella review) methodology, guided by the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework. This approach allows for a comprehensive, high-level synthesis of the vast and often inconsistent literature on mental health in SGM youth. The goal is to rigorously summarize the existing evidence on prevalence, severity, and risk factors while also critically appraising the quality of the reviews that have produced this evidence.

4.2. Core Methodology In-depth (Layer by Layer)

4.2.1. Search Strategy

The authors conducted a comprehensive search to identify all relevant systematic reviews and meta-analyses.

1. Databases Searched: MEDLINE, PsycInfo, Scopus, and Web of Science.
2. Search Dates: The initial search was conducted on March 23, 2022, and was updated on January 31, 2024, to include the most recent literature.
3. Search Terms: The search combined terms related to three core concepts:
   • Population: $LGBTQ+$, child/adolescent/young adult.
   • Outcome: mental disorders.
   • The authors intentionally did not use a filter for "systematic review" to avoid missing any relevant papers that may not have been indexed as such. Specific search terms are detailed in the paper's supplementary material (Table S1).

4.2.2. Eligibility Criteria

The authors applied strict inclusion and exclusion criteria to ensure the relevance and quality of the synthesized evidence.

• Inclusion Criteria:
  1. Must be a systematic review (SR) and/or meta-analysis.
  2. Must report on:
     • Prevalence: Studies defining prevalence as reaching a clinical cutoff on a validated mental health tool or receiving a clinical diagnosis.
     • Severity: Studies comparing SM/GM groups with heterosexual/cisgender groups on a validated mental health tool.
     • Risk Factors: Studies reporting on moderators, mediators, or associated variables related to mental disorders in SGM youth.
• Exclusion Criteria:
  1. Reviews that did not separate SM from GM groups. This was crucial for understanding the potentially different experiences of these populations.
  2. Reviews that combined participants over 25 years old with younger individuals. This ensures the focus remains on youth and young adults.
  3. Reviews that only assessed mental disorders among participants with pre-existing vulnerabilities (e.g., studies focused only on suicidal youth), to avoid confounding factors.

4.2.3. Screening and Quality Assessment

1. Screening: After removing duplicates, two researchers (JO and RJ) independently screened titles, abstracts, and full texts for eligibility. A third researcher (LH) resolved any disagreements.
2. Quality Assessment of Reviews: The methodological quality of each included SR was assessed using the Assessing the Methodological Quality of Systematic Reviews 2 (AMSTAR-2) tool. This is a 16-item checklist that provides a critical appraisal of a review's methodology, resulting in a rating of High, Moderate, Low, or Critically Low quality.

4.2.4. Data Extraction

Data was extracted from the 42 eligible SRs. If an SR described a relevant primary study but did not provide the specific quantitative data, the authors went back to the original primary study to extract it directly. This ensured all relevant data was captured for their new meta-analyses. The following data were extracted:

• Prevalence rates of mental disorders in SM and GM groups.
• Mean scores and standard deviations on mental disorder screening tools for SM/GM and heterosexual/cisgender groups.
• Identified risk factors for mental disorders.

4.2.5. Meta-Analyses and Qualitative Synthesis

The authors performed new meta-analyses on the extracted data where possible.

1. Meta-Analysis of Prevalence:

   • This was conducted for depression in SM and GM groups separately, as there was insufficient data for other disorders.
   • A random-effects model was used, which assumes that the true effect can vary from study to study.
   • The analysis was performed in Stata using the metaprop command and applied a Freeman-Tukey arcsine transformation. This transformation is used for proportions (like prevalence) to stabilize the variance before pooling, which is especially important when prevalence rates are very high or very low.
   • Publication bias (the tendency for studies with significant results to be published more often) was assessed using the Luis Furuya-Kanamori (LFK) index.

2. Meta-Analysis of Severity:

   • This was conducted only for depression severity, comparing SM youth to heterosexual youth, due to a lack of data for other comparisons.
   • The analysis calculated the Standardized Mean Difference (SMD), reported as Hedges' g, to account for the use of different depression scales across studies. The formula for Hedges' g is: $ g = \frac{M_1 - M_2}{SD_{pooled}^*} \times J $ Where:
     • $M_1$ and $M_2$ are the mean scores of the two groups (e.g., SM and heterosexual).
     • $SD_{pooled}^*$ is the pooled and weighted standard deviation of the two groups.
     • $J$ is a correction factor that adjusts for bias in small sample sizes, calculated as $J(df) \approx 1 - \frac{3}{4df - 1}$, where $df = n_1 + n_2 - 2$.
   • Publication bias was assessed by visually inspecting funnel plots (a scatter plot of effect size against a measure of study precision) and using Egger's test (a statistical test for funnel plot asymmetry).

3. Qualitative Synthesis: For all other findings where a meta-analysis was not feasible (e.g., prevalence of other disorders, risk factors), the results were synthesized and described narratively.

4. Quality Assessment of Primary Studies: Individual studies extracted from the SRs for the new meta-analyses were assessed for bias using the Newcastle-Ottawa Scale (NOS), a tool designed to evaluate the quality of non-randomized studies.

5. Experimental Setup

5.1. Datasets

In this umbrella review, the "dataset" is not a single primary dataset but rather the collection of primary studies included within the 42 systematic reviews that met the eligibility criteria.

• Source: These primary studies were published between 1996 and 2022.
• Scale and Characteristics: The 42 reviews synthesized a large but unspecified total number of primary studies. The authors note important characteristics of this collective evidence base:

  • Geographical Distribution: The research is heavily concentrated in Western, high-income regions. 91% of reviews included studies from North America (US/Canada), 33% from Europe (mostly Northern/Western), and 26% from Australia/New Zealand. Only 12% included studies from Asia.

  • Study Design: Most of the primary studies were cross-sectional in nature, capturing data at a single point in time. Many also used convenience samples (e.g., recruiting from SGM community organizations), which are not representative of the entire SGM youth population.

  • Data Type: All eligible primary studies were quantitative.

    The choice of these "datasets" (the included SRs) was determined by the strict search and eligibility criteria designed to find the most relevant and high-quality review-level evidence available.

5.2. Evaluation Metrics

The study used several metrics for its own analysis and for assessing the quality of the included literature.

5.2.1. Prevalence

• Conceptual Definition: This metric quantifies the proportion of a population that has a particular disorder or condition at a specific point in time. It is a fundamental measure of disease burden.
• Mathematical Formula: $ \text{Prevalence} = \frac{\text{Number of individuals with the disorder}}{\text{Total number of individuals in the population}} $
• Symbol Explanation: In this paper, this was calculated for specific mental disorders (e.g., depression) within SM and GM youth populations.

5.2.2. Hedges' g (Standardized Mean Difference)

• Conceptual Definition: This metric quantifies the difference between the mean scores of two groups (e.g., SM vs. heterosexual youth on a depression scale) in standardized units. It allows for the comparison of results across studies that used different measurement scales. A positive value indicates the first group has a higher mean score.
• Mathematical Formula: $ g = \frac{M_1 - M_2}{SD_{pooled}^*} \times \left(1 - \frac{3}{4(n_1+n_2-2) - 1}\right) $
• Symbol Explanation:
  • $M_1, M_2$: The means of group 1 and group 2.
  • $SD_{pooled}^*$: The pooled standard deviation, calculated as $\sqrt{\frac{(n_1-1)s_1^2 + (n_2-1)s_2^2}{n_1+n_2-2}}$.
  • $n_1, n_2$: The sample sizes of group 1 and group 2.
  • $s_1, s_2$: The standard deviations of group 1 and group 2.
  • The term in the parentheses is the correction factor for small sample sizes.

5.2.3. I² statistic

• Conceptual Definition: The I² statistic describes the percentage of total variation across studies in a meta-analysis that is due to genuine differences between studies (heterogeneity) rather than random chance.
• Mathematical Formula: $ I^2 = \frac{Q - df}{Q} \times 100% $
• Symbol Explanation:
  • $Q$: Cochran's Q statistic, which measures the total variation.
  • df: The degrees of freedom, calculated as (number of studies - 1).
  • Interpretation: A value of 0% indicates no observed heterogeneity, while higher values indicate increasing heterogeneity. Typically, >75% is considered high heterogeneity, suggesting the studies are quite different from one another.

5.2.4. AMSTAR-2 and Newcastle-Ottawa Scale (NOS)

These are not performance metrics but quality assessment tools:

• AMSTAR-2: Used to assess the methodological quality of the systematic reviews themselves. It helps determine how trustworthy a review's conclusions are.
• Newcastle-Ottawa Scale (NOS): Used to assess the quality and risk of bias in the primary non-randomized studies (like cross-sectional and cohort studies) that were included in the authors' new meta-analyses.

5.3. Baselines

In this study, the primary "baselines" or comparison groups are:

• Heterosexual Youth: Used as the comparison group for sexual minority (SM) youth when analyzing the severity of mental disorders.
• Cisgender Youth: Used as the comparison group for gender minority (GM) youth.
• General Youth Population: The prevalence rates calculated in the meta-analyses are compared narratively to established prevalence rates for depression in the general adolescent population from other major studies, providing context for the observed disparities.

6. Results & Analysis

6.1. Core Results Analysis

6.1.1. Summary of Included Systematic Reviews

The search yielded 42 systematic reviews (SRs) that met the inclusion criteria. The overall quality and focus of these reviews were a key finding:

• Quality: Using the AMSTAR-2 scale, all 42 SRs were rated as either low or critically low quality. The main reason for this poor rating was a consistent failure to investigate the causes of heterogeneity (i.e., the reasons for variation in results between studies).

• Geographical Focus: The evidence base is heavily skewed towards Western countries, with 91% of reviews including studies from North America.

• Population Focus: 57% of reviews focused on SM youth only, 31% on GM youth only, and just 12% included both groups separately.

• Disorder Focus: Depression was the most frequently studied disorder, appearing in 83% of the reviews. Other disorders like generalized anxiety disorder (36%), PTSD (14%), and eating disorders (10%) were studied far less.

  The following is the full list of included SRs from Table 1 of the original paper:

  SR/meta-analysis authors	Countries of eligible included studies within systematic review of systematic reviews	SR/meta- analysis review qualitya	Number of studies in SR/meta-analysis	Group in eligible studies within SR/meta-analysisb	Disorders assessed in eligible studies within SR/meta-analysis
  Abreu & Kenny (2018)	1 (USA)	Critically Low	27	SM	Depression
  Ancheta, Bruzzeze & Hughes (2021)	2 (New Zealand, USA)	Critically Low	6	SM	Depression
  Argyriou, Goldsmith and Rimes (2021)	3 (Netherlands, Sweden, USA)	Low	40	SM	Depression
  Baker et al. (2021)	3 (Netherlands, Spain, USA)	Low	20	GM	Depression
  Berger, Taba, Marino, Lim & Skinner (2022)	1 (USA)	Low	26	SM	Generalised Anxiety, Depression
  Bouris et al. (2010)	3 (Canada, New Zea- land, USA)	Critically Low	31	SM	Generalised Anxiety, Depression, Post-Trau- matic Stress Disorder
  Brown, Porta, Eisen- berg, McMorris, & Sieving (2020)	2 (Thailand, USA)	Critically Low	16	GM	Generalised Anxiety, Depression, Post-Trau-matic Stress Disorder
  Caldarera, Vitiello, Turcich, Bechis & Baietto (2022)	3 (Canada, Netherlands, USA)	Critically Low	9	Both	Separation Anxiety Disorder, Social Anxi- ety Disorder
  Campbell et al. (2024) 1 (USA)		Critically Low	26	GM	Eating Disorders
  DeSon & Andover (2023)	2 (Canada, USA)	Critically Low	45	SM	Depression, Alcohol Use Disorder
  Dürrbaum & Satler (2020)	3 (Canada, New Zea- land, USA)	Critically Low	7	SM	Depression, Post-Trau- matic Stress Disorder
  Freitas et al. (2017)	1 (USA)	Critically Low	13	SM	Generalised Anxiety, Depression
  Frew, Watsford & Walker (2021)	3 (Canada, Netherlands, USA)	Low	15	GM	Generalised Anxiety, Depression
  Gilbey, Mahfouda, Ohan, Lin & Perry (2020)	3 (Netherlands, UK, USA)	Critically Low	10	SM	Depression
  Hall (2018)	3 (Canada, New Zea- land, USA)	Critically Low	35	SM	Depression
  Johns et al. (2018)	1 (USA)	Critically Low	21	GM	Depression, Post-Trau- matic Stress Disorder
  Lekwauwa, Funaro and Doolittle (2022)	1 (USA)	Critically Low	18	Both	Alcohol Use Disorder, Depression
  Lucassen, Stasiak, Samra, Frampton & Merry (2017)	5 (Canada, China, New Zealand, UK, USA)	Critically Low	23	SM	Depression
  Ludvigsson et al. (2023)	2 (Netherlands, USA)	Low	24	GM	Generalised Anxiety, Depression
  Mahon (2021)	1 (USA)	Critically Low	46	SM	Social Anxiety Disorder
  Marconi et al. (2023)	3 (Australia, China, USA)	Critically Low	21	GM	Generalised Anxiety, Depression
  Marshal et al. (2008)	1 (USA)	Critically Low	18	SM	Substance Use Disorders
  Marshal et al. (2011)	2 (Canada, USA)	Low	24	SM	Depression
  McCann & Brown (2019)	2 (Canada, USA)	Critically Low	14	SM	Depression
  McCann, Donohue and Timmins (2020)	1 (USA)	Critically Low	9	SM	Depression
  McDonald (2018)	2 (Canada, USA)	Critically Low	10	Both	Conduct Disorder, Depression
  Meneguzzo et al. (2018)	1 (Norway)	Critically Low	45	SM	Eating Disorders
  Mezzalira et al. (2022)	4 (Canada, China, New Zealand, USA)	Critically Low	33	GM	Generalised Anxiety, Conduct Disorder, Depression, Separa- tion Anxiety Disor- der, Somatic Symptom Disorder
  Millet, Longworth & Arcelus (2017)	3 (Canada, Netherlands, USA)	Critically Low	25	GM	Generalised Anxiety
  Newcomb and Mus- tanski (2010)	1 (USA)	Critically Low	31	SM	Generalised Anxiety, Depression
  Pinna et al. (2022)	2 (Canada, UK)	Critically Low	165	GM	Generalised Anxiety, Depression
  Plöderl & Tremblay (2015)	4 (New Zealand, the Philippines, UK, USA)	Critically Low	199	SM	Alcohol Use Disorder, Generalised Anxiety, Borderline Personal- ity Disorder, Conduct Disorder, Depression, Oppositional Defiant Disorder
  Pompili et al. (2014)	1 (USA)	Critically Low	19	Both	Conduct Disorder, Depression, Eating Disorders, Post-Trau-matic Stress Disorder
  Shokoohi et al. (2022)	2 (Australia, USA)	Critically Low	105	SM	Alcohol Use Disorder
  Singh, Dandona, Sharma, & Zaidi (2023)	1 (USA)	Critically Low	42	SM	Generalised Anxiety, Depression, Substance Use Disorder
  Tankersley, Grafsky, Dike and Jones (2021)	2 (Canada, USA, inc. Puerto Rico)	Low	44	GM	Generalised Anxiety, Depression, Post-Trau- matic Stress Disorder, Separation Anxiety Disorder
  Tsarna et al. (2022)	1 (Greece)	Critically Low	20	GM	Generalised Anxiety, Depression
  Valentine and Shipherd (2018)	1 (USA)	Critically Low	77	GM	Depression
  Vrangalova and Savin-Williams (2014)	1 (Norway)	Critically Low	60	SM	Depression, Eating Disorders
  Williams et al. (2023)	1 (USA)	Low	22	Both	Depression
  Wittgens et al. (2022)	2 (Iceland, New Zealand)	Critically Low	26	SM	Depression
  Xu et al. (2023)	2 (Australia, USA)	Critically Low	23	SM	Depression
  a assessed with the AMSTAR 2 scale, b SM = sexual minorities, GM = gender minorities

  6.1.2. Mental Health in Sexual Minority (SM) Young People

  Prevalence of Mental Disorders

  • Depression: The meta-analysis of six studies found a pooled prevalence of 26% (95% CI 21–32%) for depression among SM youth. This is significantly higher than the 16-20% rates reported in general youth populations. The LFK index indicated a moderate risk of publication bias.
  • Other Disorders: There was insufficient data for meta-analyses of other disorders. Qualitative synthesis showed:
    • Eating Disorders: Prevalence varied widely depending on the measurement tool, from 1.3% (anorexia) using diagnostic interviews to 21.6% (bulimia) using self-report.
    • Substance Use Disorders: Prevalence was very high in homeless SM youth (e.g., 52.4% for alcohol use disorder) but lower in community samples (7-9%).

  Comparison of Severity: SM vs. Heterosexual Youth

  • Depression: The meta-analysis of 17 studies was a key finding. It showed that SM young people reported significantly more severe depressive symptoms compared to their heterosexual peers.

    • The overall effect size was Hedges’ g = 0.38 (95% CI 0.25 to 0.50), which is a small to medium effect.

    • This finding was consistent when analyzed by gender:

      • SM males vs. heterosexual males: Hedges’ g = 0.27 (small effect).
      • SM females vs. heterosexual females: Hedges’ g = 0.34 (small effect).

    • Funnel plots and Egger's test indicated no significant publication bias for this analysis.

      The forest plots below (Figures 2, 3, and 4 from the paper) visualize these findings, showing that the individual study effect sizes and the overall pooled effect are consistently above zero, indicating greater severity in SM youth.

      该图像是一幅森林图，展示了多项研究中性少数群体（SM）与异性恋者抑郁症状严重度的效应值（Hedges’s g）。整体结果显示SM抑郁症状较异性恋者严重，效应值为0.38（95% CI 0.25至0.50）。

      该图像是一个森林图，展示了三项研究中性少数群体（SM）与异性恋群体在抑郁症评分上的差异，采用Hedges’s g作为效应量，结果总体显示SM群体抑郁症状略高于异性恋者。

      该图像是一张森林图，展示了五项研究中性少数群体（SM）与异性恋者在抑郁症严重程度上的差异，图中用Hedges' g效应值及95%置信区间表示，整体效应为0.34，表明SM抑郁症状较异性恋者更严重。

  • Anxiety and Other Disorders: Qualitative synthesis found that SM youth also had significantly more severe symptoms of generalized anxiety, social anxiety, conduct disorder, and borderline personality disorder compared to heterosexual youth.

  Risk Factors for Mental Disorders in SM Youth

  The paper identified numerous risk factors, categorized into proximal and distal factors as summarized in the table below. Key findings include the strong negative impact of internalized homophobia, family rejection, and victimization/bullying. Conversely, family support, peer support, and a positive school climate were identified as important protective factors.

  The following is the summary of risk factors from Table 2 of the original paper:

  Proximal factors	Distal factors
  Maladaptive coping mechanisms/emotion regulation	Lack of closeness with and hostility from family members
  Internalised homophobia/discomfort with sexual identity	Lack of support from friends
  Lack of openness about sexual orientation/stress associated with 'coming out'	Bullying victimisation/discrimination/harassment/ microaggressions
  Vulnerable groups (asexual and bisexual young people, adolescents, females)	Lack of school support
  Early frequent sexual activity	Hostility from religious groups
  Internal religious conflicts	Stressful life events
  Unmet medical needs	Traumas experienced in childhood and adolescence

  6.1.3. Mental Health in Gender Minority (GM) Young People

  Prevalence of Mental Disorders

  • Depression: The meta-analysis of 14 studies found a pooled prevalence of depression among GM youth of 46% (95% CI 36–56%). This rate is substantially higher than that found in SM youth (26%) and the general youth population, highlighting the extreme vulnerability of this group. The analysis had high heterogeneity (I² = 96%), indicating significant variability between studies.
  • Other Disorders: Qualitative synthesis found high prevalence rates for other disorders, with anxiety disorder prevalence ranging from 13% to 85% across different studies.

  Comparison of Severity: GM vs. Cisgender Youth

  The paper found five studies comparing symptom severity. While they consistently indicated that GM young people had more severe symptoms of anxiety, depression, conduct disorder, and somatic symptom disorder compared to cisgender youth, the differences did not always reach statistical significance, or significance was not reported. More research is needed here.

  Risk Factors for Mental Disorders in GM Youth

  Similar to SM youth, risk factors for GM youth were categorized into proximal and distal factors. Key themes include the detrimental effects of internalized transphobia, low appearance congruence (feeling one's body does not match their gender identity), family rejection, and victimization. Protective factors included resilience, family support, peer/school support, and access to gender-affirming care (though evidence on the latter was mixed).

  The following is the summary of risk factors from Table 3 of the original paper:

  Proximal factors	Distal factors
  Negative coping mechanisms	Lack of acceptance/rejection from family
  Internalised transphobia	Lack of support from peers/schools
  Low appearance congruence	Victimisation/discrimination
  Detectable HIV viral load	Living in suburban area
  Vulnerable groups (non-binary, older adolescents, assigned male at birth)
  Drug use
  Trading sex

  7. Conclusion & Reflections

  7.1. Conclusion Summary

  This systematic review of systematic reviews provides a robust and comprehensive summary of the mental health landscape for SGM youth. The main conclusions are:

  1. Significant Disparities Exist: SGM young people, and particularly GM young people, experience a disproportionately high prevalence and severity of mental disorders, most notably depression.
  2. GM Youth are a High-Risk Group: With a depression prevalence of 46%, GM youth face a particularly severe mental health burden that requires prioritized attention from clinicians and researchers.
  3. Risk and Protective Factors are Well-Documented: The Minority Stress Model is a valid framework for understanding these disparities. Distal stressors like family rejection and victimization, and proximal stressors like internalized stigma, are consistently linked to poor mental health. Conversely, social support from family, peers, and schools is a critical protective factor.
  4. The State of Research is Poor: A major finding is the low methodological quality of the existing systematic review literature, which undermines the reliability of previous evidence synthesis and calls for more rigorous research standards.
  5. Clinical Implications: The evidence strongly supports the need for mental health services to be aware of these disparities, screen SGM youth for mental health issues, and adapt their care to be sensitive and affirming to the unique needs of this population.

  7.2. Limitations & Future Work

  The authors acknowledged several limitations and proposed directions for future research.

  • Limitations of the current study:
    • The findings are built upon a foundation of low-quality systematic reviews, which may compound biases.
    • The primary studies synthesized are mostly cross-sectional, preventing conclusions about causality.
    • The reliance on convenience samples limits the generalizability of the findings.
    • The evidence base is heavily Western-centric, and may not apply to SGM youth in other cultural contexts.
  • Future Work Suggested by Authors:
    • Methodological Improvements: Future research should use larger, community-based samples with longitudinal designs to track mental health over time and establish causality.
    • Broader Inclusion: Research needs to include more diverse SGM identities (e.g., asexual, agender) and focus on non-Western populations to understand cultural differences.
    • Explore Understudied Factors: More research is needed on the impact of macro-level factors like legislation (e.g., anti-discrimination laws or restrictions on gender-affirming care).
    • Adopt an Intersectional Framework: Future studies should examine how intersecting identities (e.g., race, socioeconomic status) combine with SGM identity to affect mental health.
    • Co-production: Researchers should collaborate directly with SGM young people to co-design research and services that are most relevant and beneficial to them.

  7.3. Personal Insights & Critique

  This paper is an exemplary piece of evidence synthesis that makes a significant contribution to the field of adolescent mental health.

  • Strengths:
    • The umbrella review methodology is a major strength. In a field saturated with research, it provides a much-needed "state-of-the-art" summary that is both comprehensive and critical.
    • The finding that all 42 included reviews were of low quality is a powerful meta-analytic insight. It serves as a crucial wake-up call for the research community to improve the rigor of its systematic reviews.
    • The authors' decision to apply stricter inclusion criteria (especially separating SM and GM groups and focusing on youth) and conduct their own meta-analyses adds immense value and clarity, producing more reliable estimates than what was previously available.
  • Personal Reflections:
    • The stark difference in depression prevalence between SM youth (26%) and GM youth (46%) underscores the critical importance of not treating "SGM" as a monolithic group. The unique stressors faced by gender minority individuals clearly translate into more severe health outcomes.
    • The paper effectively bridges the gap between research and practice. The findings are not just academic; they provide a clear and urgent mandate for clinicians to change how they approach mental healthcare for SGM young people. The recommendation to consider SGM identity as a risk factor in a sensitive, non-discriminatory way is practical and important.
  • Potential Issues and Areas for Improvement:
    • While the authors did their best to manage bias by assessing quality at both the review (AMSTAR-2) and primary study (NOS) levels, the review is still fundamentally limited by the poor quality of the source material. The conclusions, while the best possible given the evidence, should be interpreted with this in mind.
    • The paper's risk factor analysis is comprehensive but remains largely a list of associations due to the cross-sectional nature of the data. Future work, as the authors note, must move towards understanding the complex, dynamic interplay and causal pathways between these factors using longitudinal and experimental designs. For example, how do protective factors like parental support buffer the impact of distal stressors like bullying over time? This is the next frontier for research in this area.