Sexual orientation and gender identity based disparities in colorectal, cervical, and breast cancer screening in the United States

Abstract

Background

Sexual orientation and gender identity (SOGI) minorities experience disparities in health care access and quality of care, however national evidence regarding adherence to recommended cancer screening remains limited. This study aimed to evaluate associations between SOGI status and adherence to US Preventive Services Task Force‐recommended colorectal, cervical, and breast cancer screening.

Methods

Adults eligible for colorectal (CRC; 45–74 years old), cervical (25–64 years old), or breast cancer (40–74 years old) screening were identified from the 2018–2022 Behavioral Risk Factor Surveillance System according to guideline‐concordant criteria. Respondents were categorized as sexual orientation minorities (SOM) versus heterosexual and as gender identity minorities (GIM) versus cisgender. Univariable results were reported as unweighted frequencies with survey‐weighted proportions. Survey‐weighted Poisson regression models were used to estimate adjusted prevalence ratios (aPRs) for screening adherence.

Results

Among 663,924 unweighted screening‐eligible respondents, weighted 1.2% (n = 8108) identified as SOM and 0.4% (n = 2315) as GIM; 33.5% (n = 223,065) were assigned male at birth and 67.9% (n = 522,845) identified as White. After adjustment for covariates, SOM was associated with lower adherence to cervical (aPR, 0.92; 95% CI, 0.87–0.97) and breast cancer screening (aPR, 0.84; 95% CI, 0.77–0.92) compared with heterosexual women. CRC screening adherence among SOM men was slightly higher (aPR, 1.10; 95% CI, 1.00–1.21). GIM status was associated with substantially lower adherence to cervical (aPR, 0.58; 95% CI, 0.37–0.90) and breast cancer screening (aPR, 0.24; 95% CI, 0.08–0.74) compared with cisgender respondents, with no difference in CRC screening (p > .05).

Conclusions

Disparities in cervical and breast cancer screening persist among SOGI minorities. These findings underscore prevalent structural and systemic barriers in preventive cancer care and highlight the need for inclusive, culturally sensitive interventions to promote equitable screening access.

Short abstract

Using Behavioral Risk Factor Surveillance System data (2018–2022) and survey‐weighted analyses, the authors characterize how adherence to US Preventive Services Task Force‐recommended colorectal, cervical, and breast cancer screening differs across sexual orientation and gender identity minority. These findings offer actionable insights to inform guideline development and targeted interventions to improve equitable access to preventive cancer care for these historically understudied populations.

INTRODUCTION

Despite advances in diagnostics and treatment, nearly 10 million people die from cancer worldwide each year. ¹ In the United States, the CDC reported 1.8 million new cancer diagnoses in 2022. In 2023, deaths included 53,779 from colorectal cancer, 42,213 from female breast cancer, and 4162 from cervical cancer. ² Screening can substantially reduce cancer morbidity and mortality through early detection and timely treatment. The US Preventive Services Task Force (USPSTF) recommends routine screening for colorectal, cervical, and breast cancers for eligible individuals because effective, accessible, and cost‐effective screening tests exist for these cancers. ³ Nonetheless, substantial disparities in cancer diagnosis persist across sociodemographic groups, driven by structural barriers, unequal access to care, and differences in social determinants of health. ⁴ , ⁵

Sexual orientation and gender identity (SOGI) minorities constitute approximately 5.5% of US adults and experience distinct health disparities that affect preventive and therapeutic care. ⁶ They frequently encounter discrimination, a shortage of culturally competent providers, and structural obstacles within health care systems, ⁷ they are more likely to rely on public insurance and to report difficulty accessing care compared with cisgender heterosexual, and up to 16% have reported discrimination in health care settings. ⁸ , ⁹ Anticipated discrimination further discourages care‐seeking as nearly one in six SOGI adults avoid health care because they expect mistreatment. ⁹ Additionally, previous studies have reported that SOGI populations face elevated cancer risk through distinct yet interconnected pathways. Structural and psychosocial stressors, including minority stress, driven by homophobia and transphobia, and history of sexual abuse having a downstream effect by predisposing to several behavioral (obesity, alcohol use, and smoking) and biological (HIV and human papillomavirus [HPV] infections) risk factors are known to increase the risk of cancer. ⁷ , ¹⁰ , ¹¹ , ¹² , ¹³ National analyses of SOGI differences in cancer prevalence and screening adherence have been limited, largely because most cancer registries do not collect SOGI data. ⁷ , ¹⁴

This study aimed to evaluate associations between SOGI status and adherence to USPSTF‐recommended colorectal, cervical, and breast cancer screening. Specifically, we examined screening adherence across specific gender identity subgroups and assessed differences in self‐reported cancer prevalence by sexual orientation and gender identity.

MATERIALS AND METHODS

Data source and study population

Data were extracted from the 2018–2022 Behavioral Risk Factor Surveillance System (BRFSS), a nationally representative, annual telephone survey of US adults conducted by the Centers for Disease Control and Prevention (CDC). ¹⁵ BRFSS collects information on health behaviors, chronic conditions, and preventive services via standardized telephone surveys. Data files were available at https://www.cdc.gov/brfss/annual_data/annual_data.htm and were accessed on February 24, 2026. Respondents provided verbal informed consent following a brief explanation of the survey objectives. ¹⁵ Because this study used publicly available, de‐identified BRFSS data, it was exempt from institutional review board approval. ¹⁵ The results of the study were reported following the STROBE checklist (Supporting Information S2).

For cancer screening analyses, respondents were included if they had data on sexual orientation or gender identity and met eligibility criteria for colorectal, cervical, or breast cancer screening. Respondents with missing data for age, screening items, or SOGI status were excluded. In BRFSS, sexual orientation was ascertained by the question “Which of the following best represents how you think of yourself?” Respondents answering “Gay/Lesbian,” “Bisexual,” or “Something else” were classified as sexual orientation minorities (SOM); individuals answering “Straight” were classified as heterosexual. Gender identity was ascertained by the question “Do you consider yourself to be transgender?” Respondents answering “Yes, transgender, male‐to‐female,” “Yes, transgender, female‐to‐male,” or “Yes, transgender, gender nonconforming” were grouped as gender identity minorities (GIM); respondents answering “No” were classified as cisgender.

For cancer prevalence analyses, respondents were included if they reported SOGI status and answered the question about history of nonskin cancers (“Ever told [you had] any other types of cancer?”). Respondents with missing data on age, nonskin cancer history, or SOGI status were excluded. Respondents with data only on nonmelanoma skin cancer were excluded.

Screening eligibility was defined per USPSTF guidelines:

• Colorectal cancer (CRC): age 50–74 for the 2018–2020 data sets (USPSTF 2016) and age 45–74 for the 2021–2022 data sets (USPSTF 2021), without prior CRC. ¹⁶ , ¹⁷

• Cervical cancer: individuals assigned female at birth aged 25–64 years without prior cervical cancer or hysterectomy (USPSTF 2018). Age 25 was used as the lower limit because BRFSS reports age in 5‐year categories (starting with 18–24 years), avoiding misclassification of 18‐ to 20‐year‐olds as eligible. ¹⁸

• Breast cancer: individuals assigned female at birth aged 40–74 years without prior breast cancer (USPSTF 2016) (Table S1). ¹⁹

Sex assigned at birth was ascertained with a binary question “Male” or “Female,” which is a harmonized variable derived on the answer from LANDSEX, COLGSEX, and CELLSEX.

Both the SOGI module and the cancer screening modules are optional in BRFSS and administered only by select states (e.g., SOGI module was administered by 35 states in 2018, 31 in 2019, 33 in 2020, 32 in 2021, and 32 in 2022), therefore the analytic samples were not nationally representative of all US adults eligible for screening. In addition, BRFSS does not use a two‐step measure (sex assigned at birth and current gender identity) to correctly identify transgender respondents, and the sexual orientation and gender identity questions are administered as separate items; respondents may have answered one but not the other and they are not mutually inclusive. Missingness in SOGI arose from state‐level nonadministration and individual nonresponse; the former led to exclusion, whereas the latter was treated as missing. The same approach was applied to the cancer screening modules. Respondents were included only if their state administered both the SOGI module and the relevant cancer screening module in a given year (Figure S1).

Exposure, outcomes, and covariates

Primary exposures were self‐reported sexual orientation (heterosexual vs. SOM) and self‐reported gender identity (cisgender vs. GIM). The primary outcomes were adherence to USPSTF‐cancer screening recommendations among eligible respondents. Secondary outcomes included prevalence of colorectal, cervical, and breast cancer among respondents reporting cancer history.

Screening adherence definitions were:

• CRC: reporting at least one of the following—colonoscopy within 10 years; sigmoidoscopy or CT colonography within 5 years; fecal immunochemical test (FIT) or fecal occult blood test (FOBT) within 1 year; or stool DNA test within 1–3 years. ¹⁶ , ¹⁷

• Cervical cancer: reporting at least one of the following — Pap test within 3 years; HPV testing alone; or Pap plus HPV cotesting within 5 years. ¹⁸

• Breast cancer: reporting a mammogram within the last 2 years (Table S1). ¹⁹

Covariates included age, sex assigned at birth (male vs. female), race (White vs. minority: Black, Asian, Hispanic, Native Hawaiian or Pacific Islander, American Indian or Alaska Native, and other), education (college graduate vs. less than college), employment status (employed vs. unemployed/retired/student), annual household income (<$75,000 vs. ≥$75,000), marital status (married, never married, or divorced/widowed/separated), residence (rural vs. urban), insurance type (uninsured, public, or private), and presence of one or more personal physicians versus none.

Statistical analyses

All analyses used BRFSS survey weights to produce nationally representative estimates. Baseline characteristics and outcomes were compared between SOM and heterosexual respondents and between GIM and cisgender respondents using χ2 tests; results are presented as unweighted counts with weighted proportions. Multivariable survey‐weighted Poisson regression with robust variance estimators estimated associations between SOM or GIM status and adherence to each USPSTF screening guideline among eligible respondents; results were reported as adjusted prevalence ratios (aPRs) with 95% confidence intervals (CIs). CRC screening analyses were stratified by sex assigned at birth; cervical and breast screening analyses included only respondents assigned female at birth. Multivariable models were adjusted for age, race, income, education, marital status, employment status, insurance type, and rural versus urban residence.

Cancer prevalence among SOM versus heterosexual and GIM versus cisgender groups was estimated in the cancer‐prevalence–eligible sample; multivariable models adjusted for age, race, education, and insurance status. For cervical cancer screening among females, a Blinder–Oaxaca decomposition quantified the portion of disparity between SOM and heterosexual respondents explained by observed characteristics versus unexplained.

Exploratory sub analyses examined screening adherence across gender identity subgroups (male‐to‐female, female‐to‐male, and gender nonconforming) compared with cisgender respondents. Because sample sizes and cancer event counts were small in transgender and gender nonconforming subgroups, prevalence analyses for these groups were descriptive and regression models were not estimated. Robust variance estimation accounted for overdispersion and heteroscedasticity. Variance inflation factors (VIFs) assessed multicollinearity (no substantial collinearity detected; all VIFs <8; mean VIF = 3.05). Two‐sided p values <.05 denoted statistical significance. Analyses were performed using Stata version 19 (StataCorp, College Station, Texas). ChatGPT (OpenAI, San Francisco, California), a large language model‐based tool, was used to assist in the creation of figures. All AI‐generated content was reviewed, verified, and approved by the authors, who take full responsibility for the integrity and accuracy of the work.

RESULTS

Baseline characteristics

Among 663,924 respondents included in the cancer‐screening analyses (weighted US adult population = 63,867,917), 1.2% (n = 8108) identified as SOM and 24.7% (n = 164,231) as heterosexual; 0.4% (n = 2315) identified as GIM and 99.6% (n = 661,124) as cisgender. Overall, 50.8% (n = 323,327) were assigned female at birth and 67.9% (n = 522,845) identified as White (Table S2). Compared with heterosexual respondents, SOM were younger (age 25–39: 29.1% vs. 13.4%), more likely to be female (68.4% vs. 61.1%), more frequently from racial/ethnic minority groups (41.4% vs. 36.1%), and had lower annual household income (<$75,000/year: 50.3% vs. 39.9%). SOM respondents were also more likely to be uninsured (11.8% vs. 7.2%) and to report no personal physician (18.0% vs. 12.9%), as well as report higher rates of prior HIV testing (59.8% vs. 36.8%) (all p < .001) (Table S3). Among respondents, only 25.9% (n = 172,339) provided a response to the sexual orientation question and constituted the analytic sample for sexual orientation‐stratified analyses. The remaining 74.1% (n = 491,585) were excluded due to missing sexual orientation data, arising from individual nonresponse within administering states. Importantly, the sexual orientation and gender identity questions are separate items, resulting in the different denominators for the SOM and GIM samples.

Compared with cisgender respondents, GIM individuals more often belonged to racial/ethnic minority groups (44.7% vs. 32.0%), had lower household income (<$75,000/year: 55.6% vs. 40.6%), and were more frequently unmarried (30.5% vs. 4.4%) (all p < .05). GIM respondents were less likely to report having a personal physician (17.9% vs. 2.2%) with no difference in insurance coverage (p > .05) and reported higher rates of prior HIV testing (41.6% vs. 36.7%; p = .025) (Table S4). For the cancer screening analysis, the analytic sample reflected only respondents from states and years in which both the SOGI module and the relevant cancer screening module were administered.

Among 1,205,096 respondents analyzed for cancer prevalence (weighted US adult population = 667,276,196), 6.9% (n = 68,866) identified as SOM and 90.1% (n = 1,103,836) as heterosexual; 0.6% (n = 5798) identified as GIM and 99.4% (n = 1,194,497) as cisgender. Overall, 42.0% (n = 522,401) were assigned female at birth and 64.2% (n = 920,604) identified as White (Table S2).

Cancer screening adherence and prevalence

Among 70,774 males with SOM data, 4.1% (n = 2873) SOM respondents were eligible for CRC screening; adherence did not differ between SOM and heterosexual groups (p = .275). Among 81,522 females, 3.8% (n = 3117) SOM respondents were eligible for CRC screening; adherence again did not differ (p = .291) (Table 1). For cervical cancer screening, 37,697 respondents were eligible; 7.7% (n = 2908) were SOM, and adherence was lower among SOM compared with heterosexual respondents (65.2% vs. 73.6%; p < .001). For breast cancer screening, among 88,262 eligible respondents (SOM, 4.0%, n = 3551), adherence was also lower among SOM (40.0% vs. 48.8%; p < .001) (Table 1).

TABLE 1.

Univariable comparison of cancer screening adherence by SOM and GIM status.

Screening adherence	Total, No. (%)	Heterosexual, No. (%)	SOM, No. (%)	p
Male
CRC a	25,887 (41.2)	24,847 (41.1)	1040 (44.0)	.275
Female
CRC	30,871 (42.2)	29,777 (42.3)	1094 (40.1)	.291
Cervical cancer	25,741 (72.9)	23,885 (73.6)	1856 (65.2)	<.001
Breast cancer	36,531 (48.4)	35,195 (48.8)	1336 (40.0)	<.001
		Cisgender	GIM
Male
CRC b	26,182 (41.1)	26,083 (41.1)	99 (46.7)	.530
Female
CRC	31,352 (42.1)	31,302 (42.2)	50 (14.1)	.018
Cervical cancer	26,036 (72.8)	25,967 (72.9)	69 (41.6)	.0001
Breast cancer	37,036 (48.2)	36,982 (48.4)	54 (12.5)	.0005

Abbreviations: CRC, colorectal cancer; GIM, gender identity minority; SOM, sexual orientation minority.

^a For male CRC screening by SOM, the total eligible analytic cohort was 70,774 (heterosexual = 67,901; SOM = 2873). For female CRC screening by SOM, the total eligible analytic cohort was 81,522 (heterosexual = 78,405; SOM = 3117). For cervical cancer screening by SOM, the total eligible analytic cohort was 37,697 (heterosexual = 34,789; SOM = 2908). For breast cancer screening by SOM, the total eligible analytic cohort was 88,262 (heterosexual = 84,711; SOM = 3551).

^b For male CRC screening by GIM, the total eligible analytic cohort was 71,799 (cisgender = 71,521; GIM = 278). For female CRC screening by GIM, the total eligible analytic cohort was 83,204 (cisgender = 83,023; GIM = 181). For cervical cancer screening by GIM, the total eligible analytic cohort was 38,220 (cisgender = 38,090; GIM = 130). For breast cancer screening by GIM, the total eligible analytic cohort was 89,875 (cisgender = 89,683; GIM = 192). All data are presented as unweighted sample frequencies and weighted percentages. Survey‐weighted analyses used subpopulation estimation to account for the complex sampling design.

Relative to gender identity, among 71,799 male respondents eligible for CRC screening, 0.39% (n = 278) were GIM and adherence was comparable to cisgender peers (p = .530). Among 83,204 eligible female respondents (GIM, 0.22%, n = 181), GIM status was associated with lower CRC screening adherence (14.1% vs. 42.2%; p = .018). For cervical cancer screening (n = 38,220; GIM, n = 130), adherence was lower among GIM than cisgender respondents (41.6% vs. 72.9%; p < .001). For breast cancer screening (n = 89,875; GIM, n = 192), adherence was lower among GIM than cisgender respondents (12.5% vs. 48.4%; p < .001) (Table 1). Among 143,322 males with SOM data, 5.4% (n = 7725) were SOM; CRC prevalence did not differ by SOM status. Among 163,175 females, 6.6% (n = 10,761) were SOM; CRC prevalence was lower among SOM than heterosexual respondents (0.02% vs. 0.07%; p = .011). In the cohort with cancer‐history data (n = 163,175; SOM, 6.6%), cervical cancer prevalence was slightly higher among SOM patients (0.20% vs. 0.11%; p = .03). Breast cancer prevalence in this cohort was lower among SOM than heterosexual respondents (0.14% vs. 0.37%; p < .001) (Table 2). In assessing gender identity, among 145,346 males, 0.53% (n = 759) were GIM; among 167,022 females, 0.43% (n = 719) were GIM. No differences in colorectal, cervical, or breast cancer prevalence were observed between cisgender and GIM respondents within either birth‐sex cohort.

TABLE 2.

Univariable comparison of cancer frequency by SOM and GIM status.

Cancer frequency	Total, No. (%)	Heterosexual, No. (%)	SOM, No. (%)	p
Male
CRC a	219 (0.07)	209 (0.07)	10 (0.04)	.19
Female
CRC	234 (0.07)	227 (0.07)	7 (0.02)	.011
Cervical cancer	292 (0.11)	266 (0.11)	26 (0.20)	.03
Breast cancer	1250 (0.35)	1209 (0.37)	41 (0.14)	<.001
		Cisgender	Gender identity minority
Male
CRC b	224 (0.07)	222 (0.07)	2 (0.08)	.903
Female
CRC	246 (0.07)	246 (0.07)	0 (0.00) c	.652
Cervical cancer	294 (0.11)	293 (0.11)	1 (0.15)	.755
Breast cancer	1288 (0.35)	1285 (0.36)	3 (0.15)	.239

Note: All data are presented as unweighted sample frequencies and weighted percentages.

Abbreviations: CRC, colorectal cancer; GIM, gender identity minority; SOM, sexual orientation minority.

^a For male colorectal cancer (CRC) frequency by sexual orientation, the total eligible analytic cohort (male respondents who were asked and answered the question on ever having colorectal cancer and had nonmissing sexual orientation) was 143,322 (heterosexual = 135,597; sexual orientation minority = 7725). For female CRC frequency by sexual orientation, the total eligible analytic cohort was 163,175 (heterosexual = 152,414; sexual orientation minority = 10,761); for cervical cancer frequency, 163,175 (heterosexual = 152,414; sexual orientation minority = 10,761); and for breast cancer frequency, 163,175 (heterosexual = 152,414; sexual orientation minority = 10,761).

^b For male colorectal cancer (CRC) frequency by gender identity, the total eligible analytic cohort (male respondents who were asked and answered the question on ever having colorectal cancer and had nonmissing gender identity) was 145,346 (cisgender = 144,587; gender identity minority = 759). For female CRC frequency by gender identity, the total eligible analytic cohort was 167,022 (cisgender = 166,303; gender identity minority = 719); for cervical cancer frequency, 167,022 (cisgender = 166,303; gender identity minority = 719); and for breast cancer frequency, 167,022 (cisgender = 166,303; gender identity minority = 719).

^c No unweighted colorectal cases where observed among female GIM.

In survey‐weighted Poisson regression, compared with heterosexual male respondents eligible for CRC screening, SOM status was associated with higher adjusted prevalence of CRC screening adherence (aPR, 1.10; 95% CI, 1.00–1.21); this association was not observed among females. Compared with heterosexual females, SOM was associated with lower adherence to cervical screening (aPR, 0.92; 95% CI, 0.87–0.97) and breast screening (aPR, 0.84; 95% CI, 0.77–0.92) (Figure 1). Relative to gender identity, GIM status was not associated with CRC screening adherence in either sex. However, among screening eligible females, GIM status was associated with substantially lower‐adherence to cervical screening (aPR, 0.58; 95% CI, 0.37–0.90) and breast screening (aPR, 0.24; 95% CI, 0.08–0.74) (Table 3). Multivariable analyses of cancer prevalence demonstrated no association between SOM or GIM status and cervical or breast cancer prevalence (p > .05) (Table 4). Sensitivity analyses among respondents with missing sex assigned at birth demonstrated no evidence of increased cancer prevalence among SOGI minority groups; however, estimates should be interpreted with caution given the limited sample (Table S5).

FIGURE 1.

Multivariable Poisson regression analyses for cancer screening adherence by sexual orientation and gender identity.

TABLE 3.

Association between SOM and GIM and adherence to cancer screening guidelines, stratified by sex assigned at birth: Survey‐weighted Poisson regression.

Screening adherence	Univariate PR (95% CI)	p	Multivariable aPR (95% CI)	p
SOM a
Male
CRC	1.07 (0.95–1.21)	.264	1.10 (1.00–1.21)	.043
Female
CRC	0.94 (0.86–1.04)	.299	0.98 (0.90–1.08)	.814
Cervical cancer	0.88 (0.83–0.93)	<.001	0.92 (0.87–0.97)	.005
Breast cancer	0.82 (0.75–0.89)	<.001	0.84 (0.77–0.92)	<.001
GIM b
Male
CRC	1.13 (0.77–1.67)	.509	1.17 (0.94–1.45)	.141
Female
CRC	0.33 (0.10–1.06)	.064	0.27 (0.07–1.00)	.051
Cervical cancer	0.57 (0.37–0.87)	.010	0.58 (0.37–0.90)	.015
Breast cancer	0.25 (0.08–0.74)	.013	0.24 (0.08–0.74)	.013

Note: Model adjusted for age, race, income, education, marital status, employment, insurance type, and rural/urban residence.

Abbreviations: aPR, adjusted prevalence ratio; CRC, colorectal cancer; GIM, gender identity minority; PR, prevalence ratio; SOM, sexual orientation minority.

^a For the SOM analysis: N = 172,339 (heterosexual, n = 164,231; SOM, n = 8108). For sexual orientation, the female cohort was eligible for CRC, breast, and cervical cancer screening, and the male cohort was eligible for CRC screening and had a nonmissing sexual orientation answer.

^b For the GIM analysis: N = 663,439 (cisgender, n = 661,124; GIM, n = 2315). For gender identity, the female cohort was eligible for CRC, breast, and cervical cancer screening and the male cohort was eligible for CRC screening and had a nonmissing gender identity answer.

TABLE 4.

Association between SOM and GIM status and ever diagnosis of colorectal, cervical, or breast cancer: Survey‐weighted Poisson regression models estimating PR and aPR.

Cancer prevalence	Univariate PR (95% CI)	p	Multivariable aPR (95% CI)	p
SOM
Male
CRC	0.57 (0.24–1.34)	.196	—	—
Female a
CRC	0.28 (0.10–0.80)	.017	0.66 (0.23–1.90)	.45
Cervical cancer	1.88 (1.05–3.37)	.032	1.81 (0.94–3.47)	.076
Breast cancer	0.36 (0.20–0.66)	.001	0.82 (0.45–1.51)	.53
GIM
Male
CRC	1.10 (0.24–4.95)	.903	—	—
Female
CRC	2.32e−09 (1.60e−09 to 3.34e−09) b	<.001	1.34e−08 (6.58e−09 to 2.74e−08) b	<.001
Cervical cancer	1.37 (0.19–10.03)	.756	1.24 (0.16–9.33)	.835
Breast cancer	0.41 (0.09–1.90)	.255	0.80 (0.15–4.19)	.794

Note: For the SOM analysis: N = 1,205,096 (heterosexual, n = 1,103,836; SOM, n = 68,866). For the GIM analysis: N = 1,205,096 (cisgender, n = 1,194,497; GIM, n = 5798).

Abbreviations: aPR, adjusted prevalence ratio; CRC, colorectal cancer; GIM, gender identity minority; PR, prevalence ratio; SOM, sexual orientation minority.

^a Multivariable models were adjusted for age, race/ethnicity, education, and insurance status.

^b No colorectal cancer cases were observed among gender minority females; regression estimates were unstable due to quasi‐complete separation.

Blinder–Oaxaca decomposition

On Blinder–Oaxaca decomposition analysis for the cervical cancer screening eligible cohort, overall adherence was 8.3 percentage points lower among SOM versus heterosexual respondents. Observed sociodemographic and health care access characteristics explained 5.3 percentage points (≈64%) of the disparity; the remaining 3.0 percentage points (36%) were unexplained (p = .093). Insurance status contributed the largest explained share (1.14 percentage points), followed by age (0.74), marital status (0.69), employment status (0.65), income (0.44), and education (0.36) (Figure 2).

FIGURE 2.

Blinder–Oaxaca decomposition for the disparity in cervical cancer screening adherence among sexual orientation minority compared to heterosexual females.

Compared with cisgender females, female‐to‐male (FtM) respondents were associated with lower CRC screening adherence (aPR 0.20; 95% CI, 0.04–0.92) and lower cervical screening adherence (aPR 0.50; 95% CI, 0.30–0.84). Compared with cisgender males, male‐to‐female and gender nonconforming respondents demonstrated no differences in CRC or breast screening adherence (Table S6). Overall cancer prevalence was low across all gender‐identity subgroups (Table S7).

DISCUSSION

Cancer is the second leading cause of death worldwide and cancer deaths are projected to rise globally to 18.5 million annually by 2050, underscoring the critical role of population‐based screening to reduce cancer‐related morbidity and mortality. ¹ , ²⁰ Despite growing recognition of health care disparities affecting SOGI minorities, national data on cancer screening behaviors remain limited. In the current study, we observed distinct patterns of screening, CRC screening adherence was higher among male SOM, whereas female SOM had substantially lower cervical and breast cancer screening rates. Similarly, female GIM had lower adherence to cervical and breast cancer screening but no difference in CRC screening adherence. Notably, FtM transgender had the lowest cervical cancer screening adherence. SOM/GIM status was not associated with meaningful differences in cancer prevalence. These findings highlight screening disparities among SOGI minority, historically understudied groups, and provide evidence to inform targeted interventions and more inclusive screening strategies to improve equitable access to preventive cancer care.

Evidence regarding CRC screening in SOM populations is mixed. Prior analyses reported that gay and bisexual men may be more likely to receive recommended CRC screening than heterosexual men, whereas differences among women were less consistent. ²¹ Heslin et al. ²² reported up to 6% higher odds of CRC screening adherence among SOM men. Our findings align with these observations: SOM men had a 10% higher adjusted prevalence ratio for CRC screening adherence. This finding may reflect distinct patterns, one plausible explanation is higher engagement with preventive sexual health services among gay and bisexual men, evidenced by increased HIV testing in the literature and in our cohort of SOM versus heterosexual individuals. ²³ Frequent attendance at sexual health clinics by men who have sex with men may increase opportunities to discuss diagnostic colonoscopy or sigmoidoscopy with providers. ⁸ By contrast, cervical and breast cancer screening adherence was substantially lower among female SOM, consistent with Baumann et al., ²⁴ who reported lower odds of cervical cancer screening among Black and White female SOM, as well as Zhang et al. ²⁵ who reported lower odds of breast cancer screening among female SOM compared with heterosexual women. Although insurance and sociodemographic differences contribute to the association between SOM status and screening, more than 35% of the disparity in cervical cancer screening in our sample remained unexplained, suggesting additional underlying factors including psychosocial factors, such as adverse childhood experiences and prior abuse. ²⁶ Like other vulnerable female cohorts (including survivors of child abuse), lesbian and bisexual women frequently report reduced access to reproductive health care compared with heterosexual women, ²⁷ potentially related to stronger perceived stigma and discrimination. ²⁸ Another contributor may be heterosexism and discrimination in health care settings or reluctance to disclose sexual orientation to providers. ²⁹

Gender identity minority respondents likewise had lower adherence to cervical (aPR, 0.58) and breast (aPR, 0.24) cancer screening, consistent with Kiran et al., ³⁰ who reported similar disparities among transgender individuals compared with cisgender populations. Sociodemographic factors, including minority race, lower education, lower household income, and lack of a usual source of care, partly explain these differences, but system‐level bias also likely contributes. Most cancer screening guidelines, including USPSTF recommendations, are sex‐specific; GIM individuals face documented challenges obtaining insurance coverage for gender‐affirming care and may be denied services if the sex assigned at birth differs from the sex for which a service is listed. ³¹ CRC screening, which is less embedded in gendered health care contexts, did not show the same disparity, suggesting fewer gender‐related barriers to access. Subgroup analysis revealed that FtM transgender individuals had a 50% lower likelihood of cervical cancer screening compared with cisgender women, whereas gender nonconforming individuals did not show the same difference. Gender dysphoria and discomfort in traditional women's health settings, combined with provider uncertainty, only 33% of providers reported feeling comfortable caring for FtM individuals in one study, further compound this barrier. ³²

Despite differences in screening adherence, cancer prevalence in the current sample was comparable across SOGI minority groups and their heterosexual/cisgender counterparts, consistent with Yang et al. ³³ Although screening uptake is strongly influenced by health care access, social determinants, and attitudes toward preventive care, cancer incidence is more closely linked to biological and genetic factors tied to birth sex. ³⁴ Nonetheless, disparities in screening may increase the risk of delayed diagnosis and treatment among SOGI individuals, potentially worsening outcomes.

Our findings have important clinical and public health implications. Health care systems should promote gender‐affirming environments through nonheteronormative language, inclusive intake forms, and clinician training to improve patient–provider communication and screening participation. ³⁵ Alternative screening approaches, such as self‐collected HPV testing, may increase cervical screening uptake among FtM individuals. ³⁶ At the policy level, federal and state health agencies should mandate the inclusion of SOGI questions in national surveys and cancer registries, using a two‐step measure capturing both sex assigned at birth and current gender identity, to enable systematic surveillance of disparities. ³⁷ , ³⁸ , ³⁹ Insurers should revise coverage policies to ensure screening recommendations are based on anatomy rather than administrative gender records, and expand coverage of gender‐affirming and culturally competent preventive services. National guideline should issue explicit addendums addressing SOGI minorities, public health campaigns and clinical messaging that de‐gender screening recommendations. ⁴⁰ These findings also point to several important directions for future research. Longitudinal designs to establish causal pathways between SOGI minority status, minority stress, and screening behavior, and studies with larger GIM samples to examine disparities across transgender and gender nonconforming subgroups, as well as the role of gender‐reaffirming care and hormone therapy on cancer risk and screening eligibility, are needed. Interventional studies evaluating the effectiveness of culturally affirming care models, patient navigation programs, and self‐collected screening modalities in improving uptake among SOGI minorities are critically needed.

The current study had several limitations. As a cross‐sectional analysis of survey data, the study was subject to inherent limitations including self‐report, recall bias, and potential misclassification of SOGI status, screening adherence, and cancer history. Highly disadvantaged individuals without reliable phone access may be underrepresented and variability in question delivery across interviewers cannot be excluded. Age was reported in grouped categories, limiting precise identification of women 21–24 years old eligible for cervical screening. The SOGI module was optional and administered only in selected states and years, resulting in a smaller analytic sample and reduced statistical power for some subgroup analyses, making the missingness unlikely to be random. ⁴¹ Moreover, the sexual orientation and gender identity questions were separate items, meaning respondents could have answered one but not the other, resulting in different denominators for the two analytic samples. The survey uses a single question on transgender identity without assessing sex assigned at birth, resulting in misclassification of some transgender people, and may lead to underestimation of findings in the GIM population. ³⁹ Finally, the absence of data on medical or surgical transition and hormone therapy use prevented analysis by transition status.

In conclusion, these results highlight persistent gaps in preventive cancer care for SOGI minority, as this populations grow, further research is needed to elucidate the drivers of these disparities and to inform targeted interventions that improve equitable access to preventive cancer services. Expanding culturally sensitive care and routinely collecting SOGI data in national health surveys will be critical for monitoring and addressing these disparities.

AUTHOR CONTRIBUTIONS

Lorenza Arena: Conceptualization; visualization; methodology; formal analysis; data curation; writing—original draft; writing—review and editing. Qaidar Alizai: Conceptualization; methodology; investigation; data curation; validation; writing—review and editing. Abdulaziz Elemosho: Investigation; data curation; validation; writing—review and editing. Odysseas P. Chatzipanagiotou: Data curation; validation; investigation; writing—review and editing. Timothy M. Pawlik: Conceptualization; methodology; supervision; resources; writing—review and editing and project administration.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Supplementary Material

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available in the Behavioral Risk Factor Surveillance System 2018–2022 data sets, publicly accessible from the Centers for Disease Control and Prevention at: https://www.cdc.gov/brfss/annual_data/annual_data.htm.