https://dx.doi.org/10.24016/2025.v11.459
ORIGINAL ARTICLE
Intimate partner violence in lesbian, gay, transgender, men who have sex
with men, women who have sex with women, and bisexual people: A systematic
review and meta-analysis of prevalence
Juan Trujillo-Guablocho 1,*, Cristian Mosquera Minaya 1, Gianfranco
Centeno-Terrazas 2,3
1 Universidad
Nacional Mayor de San Marcos, Lima, Peru.
2 Instituto
Peruano de Orientación Psicológica, Lima, Peru.
3 Digital Health Research Center, Lima, Peru.
*
Correspondence: juan.trujillo3@unmsm.edu.pe
Received: May 10, 2025 |
Revised: May 21, 2025
| Accepted: June 01, 2025 | Published
Online: June 02, 2025
CITE IT AS:
Trujillo-Guablocho, J., Mosquera Minaya, C., & Centeno-Terrazas,
G. (2025). Intimate partner violence in lesbian, gay,
transgender, men who have sex with men, women who have sex with women, and
bisexual people: A systematic review and meta-analysis of prevalence. Interacciones, 11, e459. https://dx.doi.org/10.24016/2025.v11.459
ABSTRACT
Background: Intimate partner violence (IPV) in LGBT populations represents a major public health problem, and although research on the topic is increasing, knowledge remains limited, as current reviews have focused on specific populations. The prevalence of IPV in some studies reaches up to 48% in lesbian populations and 33% in MSM, while among transgender individuals, 37.5% have experienced physical violence and 25% sexual violence. Various factors aggravate the impact and make data collection more difficult.
Objective: This systematic review and meta-analysis aimed to synthesize the available evidence on the prevalence of intimate partner violence among lesbian, gay, bisexual, and transgender adults, considering the different forms of IPV and providing more precise estimates to inform future interventions and policies.
Methods: Our study is a systematic review. We searched four specialized databases of scientific articles: Scopus, Web of Science, PsycINFO, and PubMed. We included studies where the population was adults 18 to 65 years of age, who are in a casual or formal same-sex or same-gender partner relationship. We included cross-sectional studies and baseline cohort study measurements. We used the JBI Systematic Reviews "Checklist for Prevalence Studies" tool to assess the risk of bias for each study. Our study was registered in PROSPERO (CRD42024529982).
Results: Twenty-six studies met inclusion criteria; 17 were included in the meta-analysis, comprising 17,144 participants from various LGBT subgroups. The pooled prevalence was 29.5% (95% CI: 20.8%–39.0%), with high heterogeneity across studies (I² = 99.2%). Prevalence rates varied widely, especially among men who have sex with men (MSM) (8.1% to 54.5%) and transgender individuals (15.2% to 57.0%), highlighting significant variability depending on the subpopulation analyzed.
Conclusions: Our study
concluded that IPV represents a significant global concern for both MSM and
transgender individuals. Notably, psychological and emotional violence emerged
as the most prevalent form of IPV in both groups. On the other hand, the need
for more inclusive research that reflects diverse cultural and social contexts
is highlighted.
Keywords: Systematic Review Meta-analysis, Sexual and Gender Minorities, Intimate
Partner Violence, Prevalence.
INTRODUCTION
Intimate
partner violence (IPV) is a global public health problem affecting individuals
of all sexual orientations and gender identities. The World Health Organization
(WHO, 2021) defines IPV as any behavior within an intimate relationship that
causes physical, psychological, or sexual harm to the victim. In recent
decades, researchers have increased their attention to IPV in lesbian, gay,
bisexual, and transgender (LGBT) populations, revealing significant prevalence
and unique patterns of victimization (Badenes-Ribera
et al., 2016; Edwards et al., 2015). However, the global understanding of this
phenomenon in the LGBT community remains limited due to the scarcity of studies
that comprehensively cover all subgroups of this population (Badenes-Ribera et al., 2019; West, 2012).
The prevalence
of IPV in LGBT populations varies considerably across studies, reflecting the
complexity of the phenomenon and the methodological challenges in its
investigation (Finneran & Stephenson, 2013; Longobardi & Badenes-Ribera, 2017). In women who identify as lesbian,
one review reported that the average lifetime prevalence rates of IPV
victimization reach 48%, while perpetration stands at 43% (Badenes-Ribera
et al., 2014). Another review for the same population found that IPV prevalence
fluctuates between 17% and 75%, with psychological and emotional violence being
the most frequent form, with a prevalence of 14.7% (Badenes-Ribera
et al., 2016). In men who have sex with men, a review found a combined IPV
prevalence of 33% for victimization and 29% for perpetration (Liu et al.,
2021). For transgender individuals, a review reported an average lifetime
prevalence of physical IPV of 37.5% and sexual IPV of 25.0%. Additionally,
transgender individuals were found to be 1.7 times more likely to experience any
type of IPV compared to cisgender individuals (Peitzmeier et al., 2020)
Research on
IPV in LGBT populations faces several methodological challenges that contribute
to the variability in prevalence estimates. One of the main obstacles is the
lack of validated measurement instruments specifically for LGBT relationships
(Stephenson & Finneran, 2013). Furthermore, most studies rely on
convenience samples, which limit the generalization of results (Edwards et al.,
2014; Badenes-Ribera et al., 2016). The geographical
distribution of studies also presents a significant bias, with 69% of research
conducted in the USA, followed by China (19.2%), and to a lesser extent, other
countries (Liu et al., 2021). This geographical concentration underscores the
need for a global synthesis that can provide more internationally
representative estimates.
The
consequences of IPV in the LGBT population are significant and may be
exacerbated by factors such as minority stress and lack of adapted support
services (Edwards et al., 2015; Rollè et al., 2018).
Stigmatization and heteronormative stereotypes can lead to underreporting and
minimization of experienced violence, further complicating the obtaining of
accurate prevalence estimates (Badenes-Ribera et al.,
2014; Badenes-Ribera et al., 2016). A recent review
identified specific risk factors for IPV in sexual minority women, including
previous trauma, psychological and emotional problems, substance use, and
minority stressors (Porsch et al., 2023). Additionally, lesbian and bisexual
women have been observed to experience a disproportionately high burden of IPV
victimization compared to their heterosexual peers, with approximately half of
them reporting long-term negative impacts and trauma (Falluji
et al., 2024). These factors underscore the crucial importance of understanding
and addressing IPV in the LGBT community and the need for accurate prevalence
estimates to inform public health policies and practices.
Although
previous systematic reviews and meta-analyses on IPV in specific subgroups of
the LGBT population exist (Buller et al., 2014; Badenes-Ribera
et al., 2014; Badenes-Ribera et al., 2016; Peitzmeier
et al., 2020; Callan et al., 2021; Liu et al., 2021;), to date, no
comprehensive synthesis has been conducted that encompasses the entire LGBT
community, considers multiple forms of IPV. Existing reviews have focused on specific
populations without making comparisons among them, which does not allow for a more
accurate understanding of the phenomenon under study, such as lesbian women (Badenes-Ribera et al., 2016), men who have sex with men
(Finneran & Stephenson, 2012; Liu et al., 2021), or transgender individuals
(Peitzmeier et al., 2020), and that addresses the methodological limitations
identified in the literature ( West, 2012; Badenes-Ribera
et al., 2019). This knowledge gap limits our ability to develop effective,
evidence-based interventions and public health policies to prevent and address
IPV in these populations. Therefore, the objective of this systematic review
and meta-analysis is to synthesize the available evidence on the prevalence of
intimate partner violence in lesbian, gay, bisexual, and transgender adults,
considering the different forms of IPV and providing more precise estimates to
guide future interventions and policies.
METHODS
Design and register
This study is based on a systematic review to analyze methodological and
conceptual approaches to research on intimate partner violence among lesbian,
gay, transgender, men who have sex with men, women who have sex with women and
bisexual people. Our review followed the international standards proposed by
PRISMA (see Supplementary Material 1). In addition, the study protocol was
registered in PROSPERO (CRD42024529982). No significant methodological
variations were recorded between the protocol registered in PROSPERO and the
final manuscript. The only modification made was the change in statistical
analysis software, replacing Stata with R.
Eligibility criteria
The objective of our study is to determine the prevalence of intimate
partner violence in lesbian, gay, transgender, and bisexual people. Our
inclusion criteria were:
Population: adults 18 to
65 years of age, who are in a casual or formal same-sex or same-gender partner
relationship. Outcome: prevalence of intimate partner violence in
same-sex relationships. Design: Cross-sectional studies and baseline
cohort study measurements will be included. Studies evaluating interventions
(quasi-experimental and clinical trials) and review studies (narrative, scoping
review, systematic reviews) will be excluded. Setting: We will include
studies reported in Spanish and English. We will include studies from January
1, 1900, to March 1, 2024. Only studies published in peer-reviewed scientific
journals will be included, excluding books, theses, preprints, or other grey
literature documents.
Information Sources and Search Strategy
A search was conducted in four specialized databases (Scopus, Web of
Science, PsycINFO, and PubMed). The search strategy included specific terms for
"prevalence", "lesbian, gay, transgender, and bisexual",
and "intimate partner violence" (see Supplementary Material 2).
Selection process
The records obtained from the search were downloaded in RIS format, and
imported into the EndNoteX9 bibliographic manager, where automatic and manual
techniques were applied to eliminate duplicates. After this cleaning, they were
exported to an RIS file and loaded into ASReview.
This open-source tool uses artificial intelligence to optimize the study
selection process by prioritizing the most relevant records. This program was
used during the title and abstract review, with a criterion for completion of
the review being that more than half of the documents have been reviewed, and
no relevant record is identified among the last 200 analyzed. ASReview was used for title and abstract review only, with
the feature extraction technique of TF-IDF, the classifier Navie Bayes, with
the maximum query strategy, and the balancing strategy of dynamic resampling.
Full-text assessment was performed using Rayyan.ai, a free platform designed
for collaborative systematic reviews. Two independent reviewers performed a
title, abstract, and full-text review. In case of disagreement, they engaged in
a dialogue to reach a consensus; if disagreement persisted, a third reviewer
intervened to decide on the inclusion or exclusion of records. Those records
excluded after full-text review were detailed in the Supplementary Material 3,
indicating the reasons for their exclusion.
Data collection process
Once the records to be reviewed have been selected, they will be
exported to an Excel database to extract the relevant information. Two
independent reviewers perform this process. Once the review is complete, the
extracted relevant information is collated and consolidated into an Excel
document. In the event of discrepancies in the information collected, the
independent reviewers will discuss the matter until a consensus is reached. If
differences persist, a third reviewer will make the final decision.
Data items
Data extraction focused on collecting relevant information about the
characteristics of the included article, such as the first author's last name,
the first author's country of origin, the journal, and the year of publication.
Details of participants were also collected, including total number, mean age,
brief description, age group (adults), and type of participant (e.g., gay,
lesbian, bisexual, transgender, etc.). In addition, outcome information was
recorded, including the prevalence of intimate partner violence, confidence interval,
and type of violence (sexual, physical, psychological). The type of sampling
used in the study (probability and non-probability) and other relevant
characteristics of the study, such as interest and funding, were noted.
Study risk of bias assessment
To assess the risk of bias, we used the JBI Systematic Reviews
"Checklist for Prevalence Studies" tool (Munn et al., 2020). This
tool focuses on examining the methodological quality of studies and their
ability to minimize bias in design, conduct, and analysis through nine specific
domains. Each domain was assessed by considering whether it was at risk of
bias, not at risk, uncertain risk, or not applicable. Two independent reviewers
carefully performed the assessment. In the event of disagreement, consensus was
sought through discussion; if disagreement persisted, a third reviewer made the
final decision. We presented graphically the risk of bias analysis of each
study individually, as well as grouped by each of the risk of bias domains
identified in the Checklist for Prevalence Studies.
Synthesis methods
Description of studies
A narrative description of the included studies was provided, and a
table summarizing the characteristics of the included studies was presented,
e.g., countries with the highest number of articles, most common type of
participant, type of violence studied, and total number of participants.
Publication bias
When 10 or more studies were included in the meta-analysis, publication
bias analysis was performed using funnel plots with significance contours and
quantitative tests. We conducted publication bias assessments for analyses with
sufficient studies to ensure adequate statistical power for bias detection.
Peters' test was employed for each analysis, specifically designed for
proportion meta-analyses and with better type I error control than Egger's test
for this type of data (Peters et al., 2006). Publication bias was confirmed by
asymmetric distribution in the funnel plot and a significant Peters' test
(p-value <0.05).
Heterogeneity analysis
We assessed heterogeneity between studies using four indicators: a)
Cochran's Q statistic, which assumes significant heterogeneity with a p-value
< 0.05 (5%); b) Higgins' I2 statistic, which categorizes the degree of
heterogeneity as low (I2 > 25%), moderate (I2 > 50%), and high (I2 >
70%) (Higgins et al., 2023); c) the H2 index, which indicates no heterogeneity
with H2 = 1 or less (Higgins et al., 2023); and d) the between-study variance
(τ2), where τ2 = 0 assumes no true heterogeneity between effect estimates (Huedo-Medina et al., 2006). We used random effects models
because of the general assessment of heterogeneity.
Meta-regression analysis
To explore potential sources of heterogeneity, we conducted
meta-regression analyses examining study-level characteristics as potential
moderators. The variables to be examined included participant mean age,
publication year, and sample size. The publication year was centered on the
means for each analysis to improve interpretability. The sample size was
log-transformed to normalize the distribution. Meta-regression was performed
using random-effects models with Freeman-Tukey transformed proportions. For population
comparisons between subgroups with sufficient studies, mixed-effects
meta-regression was used, treating population type as a categorical moderator
variable.
Meta-analysis
As high heterogeneity was expected, a random effects model was used to
estimate the meta-analytic prevalence of IPV. First, we conducted a
meta-analysis of all forms of violence as a unified category across all LGBT
populations. Subsequently, we performed meta-analyses for each specific type of
violence (physical, psychological, and sexual) across all LGBT populations to
provide comprehensive prevalence estimates. We then conducted stratified
analyses by subgroups based on the type of LGBT participant, including only
populations with at least three studies per violence subtype to ensure adequate
statistical power, stable heterogeneity estimation, and reliable confidence
interval calculation.
For the effect size calculation, we used the Freeman-Tukey
transformation for proportions, which is particularly suitable for handling the
skewness common in prevalence data. We calculated 95% confidence intervals for
all effect estimates and included prediction intervals for the overall analysis
to provide an estimate of the dispersion of prevalences expected in future
studies.
Tests for subgroup differences were conducted to evaluate if prevalence
rates varied significantly according to violence type or population group, with
statistical significance set at p<0.05. All analyses were performed using R
(version 2024.12.1) with the metafor package. Forest
plots were created to visualize individual and pooled effect sizes along with
their corresponding confidence intervals.
RESULTS
Study selection
We initially
identified 2,807 records across different databases. After removing 1,086
(38.6%) duplicates, 1,721 (61.3%) records progressed to title and abstract
screening. Of these, we excluded 1,664 (96.7%), leaving 57 (3.3%) records for
full-text review. Subsequently, 31 (54.4%) were excluded, resulting in 26
(45.6%) articles selected for qualitative systematic review. Figure 1 shows the
complete review process, and supplementary material 3 lists the articles
excluded.
For
quantitative analyses, we applied additional eligibility criteria. Of the 26
studies included in the systematic review, 17 were included in the
meta-analysis of all forms of intimate partner violence. For specific violence
type analyses, we included only subpopulations with at least three studies per
violence subtype (physical, psychological, or sexual), a criterion met only by
MSM and transgender populations.
Figure
1. Flowchart.
Characteristics
of Included Studies
Among the 26
studies identified for the systematic review, the United States had the highest
number of publications (14/26, 53.84%), followed by China (5/26, 19.23%) and
other countries with lower representation. Most studies were published in 2021
(7/26, 26.92%). Regarding assessment instruments, 17 studies (65.38%) employed
validated psychometric scales, while 9 (34.61%) used surveys specifically
designed for this population. Table 1 presents the characteristics of studies
included in the systematic review, specifying which were used in the different
meta-analyses.
Table 1.
Characteristics of included studies (n=26).
|
Author
(Year) |
Country |
Study
Population |
Sample Size |
Mean Age
(SD) |
Measurement
Instrument |
Violence
Types* |
|
Greenwood
(2002) |
USA |
MSM |
2,881 |
NR |
Conflict
Tactics Scale |
P, S |
|
Wong (2020) |
USA |
LGBT |
- |
30.5 (NR) |
National
Intimate Partner and Sexual Violence Survey |
P, Ph, S |
|
Sabidó (2015) |
Brazil |
MSM |
3,745 |
30.3 (NR) |
Unspecified
psychometric scale |
S |
|
Owen (2004) |
USA |
Gay |
66 |
NR |
National
Violence Against Women Survey |
AFV |
|
Wall (2014) |
USA |
Gay,
Bisexual, |
190, 86 |
33.5 (NR) |
Conflict
Tactics Scale |
AFV, Ph, S |
|
MSM |
||||||
|
Davis
(2016) |
USA |
MSM |
189 |
31.8 (NR) |
Partner
Violence Scale-GBM |
AFV |
|
Stephenson
(2011) |
USA |
MSM |
528 |
27.0 (NR) |
Conflict
Tactics Scale Revised |
AFV, Ph, P,
S |
|
Shufang
(2022) |
China |
MSM |
413 |
32.4 (NR) |
Unspecified
psychometric scale |
Ph, P, S |
|
King (2021) |
USA |
Transgender |
23,999 |
NR |
U.S.
Transgender Survey |
P, S |
|
Valentine
(2017) |
USA |
Transgender |
324 |
37.5 (13.6) |
Questionnaire
based on Abuse Assessment Screen |
Ph, S |
|
Murphy
(2019) |
Peru |
Transgender |
389 |
26.0 (NR) |
Computer-Assisted
Self-Interview Survey |
AFV, Ph, S |
|
Walsh
(2021) |
USA |
MSM |
214 |
36.02
(9.24) |
Gay and
Bisexual Men Intimate Partner Violence scale |
P |
|
Stults
(2023) |
USA |
Transgender |
200 |
24.4 (3.2) |
Modified
Conflict Tactics Scale, Transgender-related IPV Scale, Identity Abuse Scale |
AFV, Ph, P |
|
Stults
(2015) |
USA |
MSM |
598 |
NR |
Unspecified
psychometric scale |
AFV |
|
Li (2021) |
China |
MSM |
272 |
24.87
(6.53) |
Conflict
Tactics Scales (CTS2) |
AFV |
|
Finneran
(2014) |
USA |
MSM |
1,575 |
NR |
Unspecified
psychometric scale |
S |
|
Miltz
(2019) |
UK |
MSM |
410 |
NR |
Health and
Relationships survey |
AFV, Ph, P,
S |
|
Hong (2022) |
Multicenter |
MSM |
9,420 |
36.4
(11.25) |
Unspecified
psychometric scale |
AFV |
|
Dunkle
(2013) |
China |
MSM |
404 |
29.6 (10.4) |
Unspecified
survey |
AFV |
|
Hillman
(2021) |
USA |
Transgender |
3,462 |
59 (6.41) |
National
Intimate Partner and Sexual Violence Survey |
AFV, P, S |
|
Longares (2017) |
Spain |
Gay,
Lesbian |
- |
29.43
(9.78) |
Psychological
Abuse in Couple Scale |
Ph, P, S |
|
Wei (2021) |
China |
MSM |
431 |
27.6 (8.2) |
IPV-GBM
scale |
Ph, P, S |
|
Zhu (2021) |
China |
MSM |
578 |
NR |
IPV-GBM
scale |
AFV, Ph, P,
S |
|
Thirunavukkarasu
(2021) |
India |
MSM |
235 |
25.5 (6.6) |
Unspecified
survey |
AFV |
|
Miller
(2024) |
USA |
MSM |
557 |
33 (33.3) |
Unspecified
psychometric scale |
S |
|
Wu (2015) |
USA |
MSM |
74 |
41.8 (8.4) |
Conflict
Tactics Scale (CTS2) |
AFV, Ph, P,
S |
Note: *AFV =
All forms of violence; Ph = Physical violence; P = Psychological violence; S =
Sexual violence; NR = Not reported; **AFV = Meta-analysis of all forms of
violence; Ph-MSM = Physical violence in MSM; P-MSM = Psychological violence in
MSM; S-MSM = Sexual violence in MSM; Ph-Trans = Physical violence in
transgender; P-Trans = Psychological violence in transgender; S-Trans = Sexual
violence in transgender; §Only data corresponding to MSM were included; data
from gay and bisexual populations were not included because these
subpopulations did not meet the minimum criterion of 3 studies per violence
subtype.
Risk of Bias
Assessment
The risk of
bias assessment was conducted on 26 studies reporting 55 prevalence
measurements using the Joanna Briggs Institute Critical Appraisal Checklist for
Studies Reporting Prevalence Data (Figure 2A). The overall risk of bias
assessment revealed that 43 measurements (78.2%) were classified as having high
risk of bias, while only 5 measurements (9.1%) demonstrated low risk of bias,
and 7 measurements (12.7%) showed some concerns.
Domain-specific
analysis identified response rate adequacy and management (RoB9) as the most
problematic area, with 33 measurements (60.0%) showing a high risk of bias.
Sampling methods (RoB2) presented concerns in 26 measurements (47.3%), followed
by sample frame appropriateness (RoB1) in 18 measurements (32.7%). The
distribution of risk across domains is presented in Figure 2B, showing the
proportion of low-risk, some concerns, and high-risk
assessments for each methodological domain and the overall evaluation.
Figure
2. Risk of bias assessment for prevalence studies. (A)
Individual study risk of bias assessment across nine domains using the Joanna
Briggs Institute Critical Appraisal Checklist. (B) Summary of the risk of bias
proportions across all domains and overall assessment.
Meta-analysis
of Overall Violence Prevalence
The
meta-analysis of all forms of violence (Figure 3) included 17 studies with a
total sample of 17,144 participants from various LGBT subpopulations. The
pooled prevalence using the random-effects model was 0.295 (95% CI:
0.208–0.390), with extremely high heterogeneity between studies (I² = 99.2%, Q
= 1,376.5, p < 0.001). The wide prediction interval (0.023 to 0.702)
reflects the considerable variability in population estimates, indicating that
true prevalence in individual settings may range substantially beyond the
pooled estimate.
The analysis
encompassed diverse subpopulations: 11 studies in MSM populations, 3 studies in
transgender populations, 2 studies in gay populations, 1 study in the bisexual
population, and 1 study in the lesbian population. Individual study prevalences
ranged from 0.081 (Thirunavukkarasu, 2021) to 0.545 (Davis, 2016) in MSM
studies, and from 0.152 (Murphy, 2019) to 0.570 (Stults, 2023) in transgender
studies.
Figure
3. Forest plot of intimate
partner violence prevalence for all forms of violence across LGBT populations.
Meta-analysis
of Prevalence by Specific Violence Types Across All LGBT Populations
We conducted
comprehensive meta-analyses for each specific type of violence across all LGBT
populations. The composition of studies for each analysis varies according to
the violence types assessed in individual investigations, with some studies
contributing to multiple analyses.
Physical
violence across all LGBT populations (Figure 4) included 10 studies with a
total sample of 3,537 participants, yielding a pooled prevalence of 0.138 (95%
CI: 0.078-0.212) with extremely high heterogeneity (I² = 97.0%, Q = 248.2, p
< 0.001). Individual study estimates ranged from 0.034 (Shufang, 2022) to
0.385 (Stults, 2023).
Figure
4. Forest plot of physical
violence prevalence across all LGBT populations.
Psychological
violence across all LGBT populations (Figure 5) demonstrated the highest
prevalence among specific violence types. Analysis of 13 studies (n = 33,404)
showed a pooled prevalence of 0.293 (95% CI: 0.210-0.384) with extremely high
heterogeneity (I² = 99.4%, Q = 886.5, p < 0.001). Individual study
prevalences ranged from 0.127 (Shufang, 2022) to 0.561 (Longares,
2017).
Figure
5. Forest plot of
psychological violence prevalence across all LGBT populations.
Sexual
violence across all LGBT populations (Figure 6) included 15 studies with a
total sample of 39,556 participants, showing a pooled prevalence of 0.078 (95%
CI: 0.052-0.109) with extremely high heterogeneity (I² = 98.6%, Q = 2,595.0, p
< 0.001). Individual estimates varied from 0.021 (Pando, 2014) to 0.219
(King, 2021).
Figure
6. Forest plot of sexual
violence prevalence across all LGBT populations.
Subgroup
Analysis by Population
Subgroup
analyses were conducted for populations with sufficient studies to ensure
statistical reliability and explore potential sources of heterogeneity. We
included only populations with at least three studies per violence subtype, a
criterion established to ensure adequate statistical power, stable
heterogeneity estimation, and reliable confidence interval calculation for each
specific violence type within each population. This criterion was met by MSM
populations (7-11 studies per violence type) and transgender populations (3-4
studies per violence type), while gay (2 studies), lesbian (1 study), and
bisexual populations (1 study) had insufficient study numbers for reliable
subgroup meta-analysis across violence types.
Meta-analysis
of Prevalence by Violence Type in Men Who Have Sex with Men (MSM)
We conducted
detailed subgroup analyses for the MSM population across all violence types (Supplementary
material 4). The analysis of any form of violence in MSM included 11 studies (n
= 12,825) with a pooled prevalence of 0.306 (95% CI: 0.212-0.408) and extremely
high heterogeneity (I² = 98.6%, p < 0.001). Physical violence in MSM
included 7 studies (n = 2,624) with a pooled prevalence of 0.130 (95% CI:
0.068-0.208) and extremely high heterogeneity (I² = 96.4%, Q = 152.8, p <
0.001). Psychological violence was analyzed in 8 studies (n = 5,529) with a
pooled prevalence of 0.219 (95% CI: 0.138-0.313) and extremely high
heterogeneity (I² = 98.0%, Q = 341.3, p < 0.001). Sexual violence, evaluated
in 11 studies (n = 11,382), showed a pooled prevalence of 0.074 (95% CI:
0.048-0.106) with extremely high heterogeneity (I² = 96.6%, Q = 254.9, p <
0.001). The test for subgroup differences confirmed significant variation
between violence types within the MSM population (Q[df=2]
= 12.58, p = 0.002), with psychological violence demonstrating higher
prevalence than sexual violence among specific types.
Meta-analysis
of Prevalence by Violence Type in Transgender Population
We conducted
comprehensive analyses for the transgender population (Supplementary material 5)
across available violence types, with study composition varying according to
the specific violence types assessed in each investigation. The analysis of any
form of violence in transgender individuals included 3 studies (n = 4,051) with
a pooled prevalence of 0.364 (95% CI: 0.143-0.621) and extremely high
heterogeneity (I² = 99.1%, p < 0.001). Physical violence in transgender
individuals included 3 studies (n = 913) with a pooled prevalence of 0.157 (95%
CI: 0.022-0.379) and extremely high heterogeneity (I² = 98.3%, Q = 95.4, p <
0.001). Psychological violence showed a pooled prevalence of 0.345 (95% CI:
0.271-0.422) from 3 studies (n = 27,661) with extremely high heterogeneity (I²
= 98.3%, Q = 201.1, p < 0.001). Sexual violence was analyzed in 4 studies (n
= 28,174) with a pooled prevalence of 0.087 (95% CI: 0.025-0.181) and extremely
high heterogeneity (I² = 99.5%, Q = 569.3, p < 0.001). Statistical testing
identified significant differences between violence types within the
transgender population (Q[df=2] = 7.73, p = 0.021).
Comparison
Between MSM and Transgender Populations
Statistical
comparisons between MSM and transgender populations were conducted using
mixed-effects meta-regression, which treats population type as a categorical
moderator variable to evaluate differences in prevalence estimates between
groups. Table 2 presents the comparison of prevalences between MSM and
transgender populations by violence type. For any form of violence, transgender
individuals showed a numerically higher pooled prevalence compared to MSM
(0.364 vs 0.306), although this difference was not statistically significant
(Q[df=1] = 0.25, p = 0.616). Similarly, comparisons
for specific violence types showed no statistically significant differences
between populations: physical violence (Q[df=1] =
0.05, p = 0.82), psychological violence (Q[df=1] =
2.84, p = 0.09), and sexual violence (Q[df=1] = 0.12,
p = 0.73). Analyses for gay, lesbian, and bisexual populations were limited to
inclusion in overall analyses due to insufficient study numbers per violence
type to meet meta-analysis requirements (fewer than 3 studies per subgroup for
most violence types).
Table 2. Meta-analysis
of prevalence by violence type and population.
|
Violence
Type |
All LGBT
[95% CI] |
MSM [95%
CI] |
Transgender
[95% CI] |
MSM vs
Transgender |
|
Any
violence (All forms) |
0.295
[0.208-0.390] |
0.306
[0.212-0.408] |
0.364
[0.143-0.621] |
Q=0.25,
p=0.616 |
|
Studies (n) |
17 |
11 |
3 |
|
|
Sample size |
17,144 |
12,825 |
4,051 |
|
|
I² (%) |
99.2 |
98.6 |
99.1 |
|
|
Physical
violence |
0.138
[0.078-0.212] |
0.130
[0.068-0.208] |
0.157
[0.022-0.379] |
Q=0.05,
p=0.82 |
|
Studies (n) |
10 |
7 |
3 |
|
|
Sample size |
3,537 |
2,624 |
913 |
|
|
I² (%) |
97 |
96.4 |
98.3 |
|
|
Psychological
violence |
0.293
[0.210-0.384] |
0.219
[0.138-0.313] |
0.345
[0.271-0.422] |
Q=2.84,
p=0.09 |
|
Studies (n) |
13 |
8 |
3 |
|
|
Sample size |
33,404 |
5,529 |
27,661 |
|
|
I² (%) |
99.4 |
98 |
98.3 |
|
|
Sexual
violence |
0.078
[0.052-0.109] |
0.074
[0.048-0.106] |
0.087
[0.025-0.181] |
Q=0.12,
p=0.73 |
|
Studies (n) |
15 |
11 |
4 |
|
|
Sample size |
39,556 |
11,382 |
28,174 |
|
|
I² (%) |
98.6 |
96.6 |
99.5 |
|
Note: All
meta-analyses performed using random-effects model. "Any violence (All
forms)" represents studies that measured experience of any type of
intimate partner violence as a unified category. MSM vs Transgender comparisons
conducted using mixed-effects meta-regression treating population type as a
categorical moderator. Tests for subgroup differences within populations:
Violence types within MSM: Q(df=2) = 12.58, p =
0.002; Violence types within Transgender: Q(df=2) =
7.73, p = 0.021.
Meta-regression
Analysis
To explore
potential sources of the observed high heterogeneity across all analyses, we
conducted meta-regression analyses examining participant mean age, publication
year, and sample size as potential moderators (Supplementary material 6). The
extremely high heterogeneity observed across all violence types (I² > 96% in
all analyses) suggests substantial variability between studies that may reflect
differences in study populations, measurement instruments, cultural contexts,
or other unmeasured factors.
None of the
examined variables showed significant associations with prevalence estimates
across any violence type (all p > 0.05). Mean age showed no significant
association across any violence type (β ranging from -0.005 to 0.001, all p
> 0.05). Publication year showed minimal association with sexual violence
prevalence (R² = 4.0%) but remained non-significant (p = 0.207). The sample
size showed no meaningful association with any violence type. These findings
indicate that the substantial between-study heterogeneity remains largely unexplained
by the examined study-level characteristics.
Publication
Bias
We assessed
publication bias using visual inspection of funnel plots and Peters' test for
analyses with sufficient studies (≥10 studies) to ensure adequate statistical
power for bias detection.
<Supplementary material 7 displays the
funnel plot for all forms of violence across LGBT populations (17 studies).
Peters' test showed no significant asymmetry (p = 0.222), suggesting no
evidence of publication bias in the primary analysis. Supplementary material 8
shows the funnel plot for physical violence across LGBT populations (10
studies). Peters' test indicated no significant asymmetry (p = 0.557),
suggesting no evidence of publication bias for this violence type.
However,
statistical tests detected significant asymmetry for other violence types. Supplementary material 9 displays the funnel plot for
psychological violence across LGBT populations (13 studies), with Peters' test
detecting significant asymmetry (p = 0.006). Similarly, Supplementary material
10 shows the funnel plot for sexual violence across LGBT populations (15
studies), with Peters' test indicating significant asymmetry (p = 0.022). These
findings suggest potential publication bias for psychological and sexual
violence, with possible underrepresentation of studies reporting lower
prevalence estimates.
DISCUSSION
Main findings
This meta-analysis represents the first quantitative
synthesis examining IPV prevalence across multiple LGBT populations worldwide.
Among the 26 studies included in the systematic review, 17 provided data
suitable for meta-analysis of all forms of violence, encompassing 17,144
participants and revealing a pooled prevalence of 29.5% (95% CI: 20.8%–39.0%).
This finding indicates that nearly one in three individuals in LGBT
relationships experience some form of intimate partner violence, representing a
substantial public health burden. However, this estimate is characterized by
extremely high heterogeneity (I² = 99.2%) and should be interpreted cautiously
given the geographical concentration of studies primarily in North American
contexts (53.84% from the United States).
When examining specific violence types across all LGBT
populations, psychological violence emerged as the most prevalent form,
affecting 29.3% (95% CI: 21.0%–38.4%) of individuals based on 13 studies. This
was followed by physical violence at 13.8% (95% CI: 7.8%–21.2%) from 10
studies, and sexual violence at 7.8% (95% CI: 5.2%–10.9%) from 15 studies. The
predominance of psychological violence aligns with emerging understanding of
IPV as a multifaceted phenomenon extending beyond physical manifestations. However,
publication bias was statistically confirmed for psychological (Peters' test, p
= 0.006) and sexual violence (Peters' test, p = 0.022), suggesting these
prevalence rates may be overestimated due to underrepresentation of studies
reporting lower estimates in the published literature.
Although our initial objective encompassed the entire
LGBT community, the limited availability of studies meeting our methodological
criteria restricted detailed subgroup analyses to MSM and transgender
populations only. Among MSM, overall IPV prevalence was 30.6%, with the pattern
of psychological violence predominating (21.9%), followed by physical (13.0%)
and sexual violence (7.4%). Transgender individuals demonstrated numerically
higher prevalence across all categories (36.4% overall), with psychological
violence again being most prevalent (34.5%), followed by physical (15.7%) and
sexual violence (8.7%). Despite these numerical differences, statistical
comparisons revealed no significant differences between MSM and transgender
populations across violence types (all p > 0.05), suggesting that while
effect sizes may vary, the statistical evidence does not support differential
prevalence patterns between these populations.
The substantial heterogeneity observed across all
analyses, with meta-regression examining participant age, publication year, and
sample size explaining less than 4% of between-study variance, underscores the
complexity of synthesizing IPV prevalence data across diverse LGBT populations
and contexts. This unexplained variability likely reflects differences in study
methodology, cultural contexts, legal frameworks, and measurement approaches
that were not captured in our available moderator variables.
Comparison with other studies
Previous systematic reviews have predominantly focused
on individual LGBT subpopulations rather than multi-population synthesis,
examining lesbian women (Badenes-Ribera et al.,
2016), MSM (Finneran & Stephenson, 2012; Liu et al., 2021), or transgender
individuals (Peitzmeier et al., 2020) separately. Our approach differs
methodologically by enabling direct comparisons across populations while
maintaining quality standards through exclusive inclusion of peer-reviewed
publications, contrasting with some previous reviews that incorporated grey
literature (Otero et al., 2015; Peitzmeier et al., 2020).
The observed discrepancy in psychological violence
prevalence among MSM compared to previous meta-analyses warrants examination.
While Liu et al. (2021) reported an emotional violence prevalence of 33% among
MSM, our analysis found 21.9%. This difference may reflect several
methodological and contextual factors. First, variations in recall periods and
violence definitions across studies may contribute to different prevalence
estimates. Second, underreporting of IPV by MSM may occur due to internalized
homophobia and concerns about reinforcing negative LGBT stereotypes, as
same-sex partner violence is often overlooked in both research and clinical
contexts (Rojas-Solís et al., 2020; Rojas-Solís et al., 2021). Third,
institutional prejudice within healthcare, religious, and law enforcement
contexts may lead to differential reporting patterns across geographic regions
and study periods. Finally, the predominant use of convenience sampling in both
our included studies and previous research limits the representativeness of all
estimates.
For transgender populations, our findings showed
greater consistency with existing literature. The prevalence of psychological
violence of 34.5% aligns reasonably with Peitzmeier et al. (2020) estimates
around 25%, particularly considering the limited number of studies and
methodological variations. This convergence may reflect the particularly severe
and visible nature of violence experienced by transgender individuals, making
prevalence estimates less susceptible to the reporting variations observed in
other populations.
The challenge of high heterogeneity appears to be a
consistent pattern across IPV research in LGBT populations. Previous
meta-analyses have documented similar difficulties: Buller et al. (2014)
reported I² = 95.7% in their analysis of six MSM studies, while Liu et al.
(2021) found I² = 98.6% across 34 studies examining MSM victimization. Even
after excluding outlier studies, heterogeneity remained problematically high in
previous analyses. This persistent pattern across multiple independent
meta-analyses suggests that high heterogeneity reflects inherent complexity in
IPV measurement across diverse LGBT populations and contexts, rather than
methodological limitations specific to any single review approach.
Public health implications
The pattern of violence identified in this analysis
has direct implications for healthcare practice and policy development. The
predominance of psychological over physical violence suggests that current IPV
screening protocols, which often focus on identifying physical injuries, may
systematically miss most LGBT individuals experiencing partner violence.
Healthcare systems should prioritize developing and implementing screening
tools that specifically assess for controlling behaviors, threats related to
sexual orientation or gender identity disclosure, and forms of economic or
social abuse that may be particularly relevant in LGBT relationships. Training
programs for healthcare providers must emphasize recognition of non-physical
forms of abuse and their serious health consequences.
The significant research gaps identified for lesbian,
gay, and bisexual populations represent a critical challenge for evidence-based
public health planning. The absence of sufficient data for these communities
limits the development of comprehensive prevention programs and may result in
resource allocation decisions that inadequately serve the full spectrum of LGBT
individuals. Public health authorities should prioritize funding research
initiatives that specifically target underrepresented LGBT populations to
ensure that prevention strategies and clinical guidelines are informed by
appropriate evidence rather than extrapolations from MSM and transgender data
alone.
Strengths and limitations
This systematic review and meta-analysis present
several methodological strengths. First, we employed rigorous statistical
approaches specifically designed for prevalence data, including Freeman-Tukey
transformation for proportions and Peters' test for publication bias
assessment, which are particularly suitable for handling the distributional
characteristics inherent in prevalence studies. Second, our analysis
incorporated comprehensive heterogeneity exploration through multiple
statistical indicators (I², Q-statistic, H², and τ²) combined with
meta-regression analyses to identify potential sources of between-study
variability. Third, we implemented a multi-dimensional analytical framework
that simultaneously categorized studies by both violence type and population
characteristics, enabling a more nuanced understanding of IPV patterns across
different LGBT subgroups. Fourth, we established priori methodological criteria
for subgroup analyses (minimum three studies per violence type), following
established recommendations for adequate statistical power and reliable
confidence interval estimation, thereby minimizing data-driven analytical
decisions.
However, this study presents several important
limitations that must be considered when interpreting the findings. First,
publication bias was statistically confirmed for psychological violence
(Peters' test, p = 0.006) and sexual violence (Peters' test, p = 0.022),
suggesting potential overestimation of prevalence rates for these violence
types and raising concerns about the representativeness of available
literature. Second, extremely high heterogeneity was observed across all
analyses (I² > 95%), with meta-regression explaining less than 4% of
between-study variance, indicating that the sources of variability remain
largely unexplained and challenging the interpretability of pooled estimates.
Third, the geographical distribution of included studies was heavily
concentrated in developed countries, particularly the United States (53.84% of
studies), which limits the generalizability of findings to diverse cultural,
legal, and social contexts globally. Fourth, the overall quality of evidence
was limited, with 78.2% of prevalence measurements classified as having high
risk of bias, primarily due to inadequate sampling methods and response rate
management, which affects the reliability of the synthesized estimates. Fifth,
a substantial proportion of studies (34.61%) employed measurement instruments
lacking sufficient psychometric validation specifically for LGBT populations,
potentially compromising the comparability and accuracy of prevalence
assessments across studies. Finally, the predominant use of non-probabilistic
sampling methods in included studies limits the capacity for population-level
inference, as convenience samples may not adequately represent the broader LGBT
communities from which prevalence estimates are derived.
Conclusions
This systematic review demonstrates that intimate
partner violence represents a significant public health concern across LGBT
populations, with psychological violence emerging as the predominant form
affecting both MSM and transgender individuals. However, confirmed publication
bias, extremely high unexplained heterogeneity (I² > 95%), and geographical
concentration primarily in North American studies substantially limit the
reliability and generalizability of prevalence estimates.
The analysis revealed critical research gaps,
particularly the underrepresentation of lesbian, gay, and bisexual populations,
which prevented comprehensive analysis across the full spectrum of LGBT
identities. Combined with the predominant use of convenience sampling and high
risk of bias in 78.2% of included studies, these limitations underscore the
urgent need for methodologically rigorous, geographically diverse research
employing validated instruments specific to LGBT relationships to better inform
evidence-based prevention strategies and policy development.
ORCID
Juan Trujillo-Guablocho: https://orcid.org/0009-0001-1459-911X
Cristian Mosquera Minaya: https://orcid.org/0000-0002-3242-046X
Gianfranco Centeno-Terrazas: https://orcid.org/0000-0002-0773-9866
AUTHORS’
CONTRIBUTION
Juan Trujillo-Guablocho:
Conceptualization, Investigation, Data curation, Project administration
Cristian Mosquera Minaya:
Conceptualization, Investigation, Data curation, Project administration
Gianfranco Centeno-Terrazas: Formal
Analysis, Methodology, Data Curation, Software, Supervision, Writing- original
draft
FUNDING SOURCE
This study has not been funded by any institution.
CONFLICT OF INTEREST
The author declares no conflict of
interest.
ACKNOWLEDGMENTS
Not applicable.
REVIEW PROCESS
This study has been reviewed by external peers in a double-blind mode.
The editor in charge was David Villarreal-Zegarra. The review process is
included as supplementary material 11.
DATA AVAILABILITY STATEMENT
Not applicable.
DECLARATION OF THE USE OF GENERATIVE ARTIFICIAL
INTELLIGENCE
We used DeepL to translate specific sections
of the manuscript and Grammarly to improve the wording of certain sections. The
final version of the manuscript was reviewed and approved by all authors.
DISCLAIMER
The authors are responsible for all statements made in this article.
REFERENCES
Badenes-Ribera, L., Bonilla-Campos, A.,
Frias-Navarro, D., Pons-Salvador, G., & Monterde-i-Bort, H. (2016). Intimate Partner Violence in Self-Identified Lesbians:
A Systematic Review of Its Prevalence and Correlates. Trauma, Violence, & Abuse, 17(3), 284–297.
https://doi.org/10.1177/1524838015584363
Badenes-Ribera, L., Frias-Navarro, D.,
Bonilla-Campos, A., Pons-Salvador, G., & Monterde-i-Bort, H. (2014). Intimate Partner Violence in Self-identified Lesbians:
A Meta-analysis of its Prevalence. Sexuality Research and Social Policy, 12(1),
47–59. https://doi.org/10.1007/s13178-014-0164-7
Badenes-Ribera, L., Sánchez-Meca, J., & Longobardi, C.
(2019). The relationship between internalized homophobia and intimate partner
violence in same-sex relationships: A meta-analysis. Trauma Violence &
Abuse, 20(3), 331–343. https://doi.org/10.1177/1524838017708781
Buller, A. M., Devries, K. M., Howard, L. M., &
Bacchus, L. J. (2014). Associations between intimate partner violence and
health among men who have sex with men: A systematic review and meta-analysis. PLoS
Medicine, 11(3), e1001609. https://doi.org/10.1371/journal.pmed.1001609
Callan, A., Corbally, M., & McElvaney, R. (2021).
A scoping review of intimate partner violence as it relates to the experiences
of gay and bisexual men. Trauma Violence & Abuse, 22(2), 233–248.
https://doi.org/10.1177/1524838020970898
Duval, S., & Tweedie, R. (2000). Trim and Fill: A
Simple Funnel-Plot–Based Method of Testing and Adjusting for Publication Bias
in Meta-Analysis. Biometrics, 56(2), 455–463.
https://doi.org/10.1111/j.0006-341X.2000.00455.x
Edwards, K. M., Sylaska, K. M., Barry, J. E.,
Moynihan, M. M., Banyard, V. L., Cohn, E. S., Walsh, W. A., & Ward, S. K.
(2014). Physical dating violence, sexual violence, and unwanted pursuit
victimization: A comparison of incidence rates among sexual-minority and
heterosexual college students. Journal of Interpersonal Violence, 30(4),
580–600. https://doi.org/10.1177/0886260514535260
Edwards, K. M., Sylaska, K. M., & Neal, A. M.
(2015). Intimate partner violence among sexual minority populations: A critical
review of the literature and agenda for future research. Psychology of
Violence, 5(2), 112–121. https://doi.org/10.1037/a0038656
Falluji, E., Li, N., Newman, Z., & Carpino, T. (2024).
Intimate partner violence in sexual minority women in same-sex relationships: A
review. HPHR Journal, 82. https://doi.org/10.54111/0001/DDDD8
Finneran, C., & Stephenson, R. (2012). Intimate
Partner Violence among Men Who Have Sex with Men: A Systematic Review. Trauma,
Violence, & Abuse, 14(2), 168–185.
https://doi.org/10.1177/1524838012470034
Finneran, C., & Stephenson, R. (2013). Intimate
partner violence among men who have sex with men: A systematic review. Trauma
Violence & Abuse, 14(2), 168–185.
https://doi.org/10.1177/1524838012470034
Higgins, J., Thomas, J., Chandler, J., Cumpston, M.,
Tianjing, L., Page, M., & Welch, V. (2023). Cochrane Handbook for
Systematic Reviews of Interventions version 6.4. Cochrane.
https://training.cochrane.org/handbook/current
Huedo-Medina, T. B., Sánchez-Meca, J.,
Marín-Martínez, F., & Botella, J. (2006). Assessing heterogeneity in meta-analysis: Q statistic or I2 index? Psychological
Methods, 11(2), 193–206. https://doi.org/10.1037/1082-989X.11.2.193
Liu, M., Cai, X., Hao, G., Li, W., Chen, Q., Chen, Y.,
& Xiong, P. (2021). Prevalence of intimate partner violence among men who
have sex with men: An updated systematic review and meta-analysis. Sexual Medicine, 9(6), 100433.
https://doi.org/10.1016/j.esxm.2021.100433
Longobardi, C., & Badenes-Ribera, L. (2017). Intimate partner violence in same-sex relationships
and the role of sexual minority stressors: A systematic review of the past 10
years. Journal of Child and Family Studies, 26(8), 2039–2049.
https://doi.org/10.1007/s10826-017-0734-4
Munn, Z., Moolla,
S., Lisy, K., Riitano, D.,
& Tufanaru, C. (2020). Chapter 5: Systematic reviews of prevalence and
incidence. In Aromataris E, Munn Z (Editors). JBI Manual for Evidence
Synthesis. JBI.
Otero, L., Fernández, M., Fernández, M., & Castro,
Y. (2015). Violence in transsexual, transgender and intersex couples: A
systematic review. Saude e sociedade, 24(3), 914–935. https://doi.org/10.1590/S0104-12902015134224
Peitzmeier, S. M., Malik, M., Kattari, S. K., Marrow,
E., Stephenson, R., Agénor, M., & Reisner, S. L. (2020). Intimate partner
violence in transgender populations: Systematic review and meta-analysis of
prevalence and correlates. American Journal of Public Health, 110(9),
e1–e14. https://doi.org/10.2105/AJPH.2020.305774
Peters, J.L., Sutton, A.J.; Jones, D.R.; Abrams, K.R. & Rushton L. (2006). Comparison of Two Methods to Detect Publication Bias
in Meta-analysis. JAMA, 295(6), 676–680. doi:10.1001/jama.295.6.676
Porsch, L. M., Xu, M., Veldhuis, C. B., Bochicchio, L.
A., Zollweg, S. S., & Hughes, T. L. (2023). Intimate partner violence among
sexual minority women: A scoping review. Trauma Violence & Abuse, 24(5),
3014–3036. https://doi.org/10.1177/15248380221122815
Riley, R. D., Higgins, J. P. T., & Deeks, J. J.
(2011). Interpretation of random effects meta-analyses. BMJ, 342, d549.
https://doi.org/10.1136/bmj.d549
Rojas-Solís, J. L., Meza-Marín, R. N.,
Villalobos-Raygoza, A., & Rojas-Alonso, I. (2020). Revisión sistemática
sobre características metodológicas en el estudio de la violencia de pareja en
hombres que tienen sexo con hombres. Revista Logos, Ciencia &
Tecnología, 13(1), 144–159. https://doi.org/10.22335/rlct.v13i1.1312
Rojas-Solís, J. L., Rojas Alonso, I., Meza
Marín, R. N., & Villalobos Raygoza, A. (2021). Violencia de parejas gays y en hombres que tienen sexo con hombres: una revisión
sistemática exploratoria. Revista Criminalidad, 63(1), 173-186.
Rollè, L., Giardina, G., Caldarera, A. M.,
Gerino, E., & Brustia, P. (2018). When intimate partner violence meets same sex couples: A review of same
sex intimate partner violence. Frontiers in Psychology, 9, 1506.
https://doi.org/10.3389/fpsyg.2018.01506
Stephenson, R., & Finneran, C. (2013). The IPV-GBM
scale: A new scale to measure intimate partner violence among gay and bisexual
men. PLoS One, 8(6), e62592. https://doi.org/10.1371/journal.pone.0062592
Thompson, S. G., & Higgins, J. P. T. (2002). How
should meta-regression analyses be undertaken and interpreted? Statistics in
Medicine, 21(11), 1559–1573. https://doi.org/10.1002/sim.1187
West, C. M. (2012). Partner abuse in ethnic minority
and gay, lesbian, bisexual, and transgender populations. Partner Abuse, 3(3),
336–357. https://doi.org/10.1891/1946-6560.3.3.336
World Health Organization. (2021). Violence against
women prevalence estimates, 2018: Global, regional and national prevalence
estimates for intimate partner violence against women and global and regional
prevalence estimates for non-partner sexual violence against women. World
Health Organization. https://www.who.int/publications/i/item/9789240022256
Violencia de pareja en personas lesbianas, gais, transgénero, hombres
que tienen sexo con hombres, mujeres que tienen sexo con mujeres y personas
bisexuales: Revisión sistemática y metaanálisis de prevalencia
RESUMEN
Introducción: La violencia de pareja (VP) en poblaciones
LGBT representa un problema relevante de salud pública. Aunque las
investigaciones al respecto han aumentado, el conocimiento sigue siendo
limitado, ya que las revisiones existentes se han centrado en poblaciones
específicas. La prevalencia de VP en algunos estudios alcanza hasta un 48 % en
mujeres lesbianas y un 33 % en hombres que tienen sexo con hombres (HSH),
mientras que, entre personas transgénero, un 37.5 % ha experimentado violencia
física y un 25 % violencia sexual. Diversos factores agravan el impacto y
dificultan la recolección de datos.
Objetivo: Esta revisión sistemática y metaanálisis
tuvo como objetivo sintetizar la evidencia disponible sobre la prevalencia de
violencia de pareja en personas adultas lesbianas, gais, bisexuales y
transgénero, considerando las distintas formas de VP y proporcionando
estimaciones más precisas para orientar futuras intervenciones y políticas.
Métodos: Este estudio es una revisión sistemática.
Se realizaron búsquedas en cuatro bases de datos especializadas en artículos
científicos: Scopus, Web of
Science, PsycINFO y PubMed.
Se incluyeron estudios con población adulta entre 18 y 65 años, en relaciones
de pareja formales o casuales del mismo sexo o género. Se consideraron estudios
transversales y mediciones basales de cohortes. Se utilizó la herramienta
"Checklist for Prevalence Studies" de JBI Systematic Reviews para evaluar
el riesgo de sesgo en cada estudio. El protocolo fue registrado en PROSPERO
(CRD42024529982).
Resultados: Veintiséis estudios cumplieron con los
criterios de inclusión; 17 fueron incluidos en el metaanálisis, con un total de
17,144 participantes pertenecientes a distintos subgrupos LGBT. La prevalencia
agrupada fue de 29.5 % (IC 95 %: 20.8 %–39.0 %), con alta heterogeneidad entre
estudios (I² = 99.2 %). Las tasas de prevalencia variaron considerablemente,
especialmente entre los HSH (8.1 % a 54.5 %) y personas transgénero (15.2 % a
57.0 %), lo que evidencia una notable variabilidad según el subgrupo analizado.
Conclusiones: El estudio concluye que la violencia de
pareja constituye un problema relevante a nivel global para los HSH y personas
transgénero. Destaca que la violencia psicológica y emocional es la forma más
prevalente en ambos grupos. Asimismo, se resalta la necesidad de
investigaciones más inclusivas que reflejen la diversidad cultural y social.
Palabras claves: Revisión sistemática, Metaanálisis, Minorías sexuales y de género,
Violencia de pareja, Prevalencia.