Sexual Orientation and Cervical Cancer Screening Among Cisgender Women

Key Points Question Are lesbian, gay, and bisexual (LGB) cisgender women less likely than heterosexual cisgender women to be screened for cervical cancer? Findings In this cross-sectional study that included 5167 cisgender women in Chicago, Illinois, 71.14% of LGB cisgender women reported up-to-date cervical cancer screening, compared with 76.95% of heterosexual cisgender women. LGB cisgender women with a primary care practitioner (PCP) were 93% more likely to be up-to-date per cervical cancer screening recommendations than LGB cisgender women without a PCP. Meaning These findings suggest that improving access to PCPs may reduce the disparity in cervical cancer screening between LGB cisgender women compared with heterosexual cisgender women.

(LGB) cisgender women less likely than heterosexual cisgender women to be screened for cervical cancer?Findings In this cross-sectional study that included 5167 cisgender women in Chicago, Illinois, 71.14% of LGB cisgender women reported up-to-date cervical cancer screening, compared with 76.95% of heterosexual cisgender women.LGB cisgender women with a primary care practitioner (PCP) were 93% more likely to be up-to-date per cervical cancer screening recommendations than LGB cisgender women without a PCP.
Meaning These findings suggest that improving access to PCPs may reduce the disparity in cervical cancer screening between LGB cisgender women compared with heterosexual cisgender women.

Introduction
Cervical cancer screening (CCS) in the US has drastically reduced cervical cancer incidence and mortality and is recommended by the US Preventive Services Task Force. 1 Sixty-four percent of eligible individuals residing in Chicago, Illinois, were screened for cervical cancer from 2021 to 2022. 2 Systemic racism and discrimination in the health care system, poor CCS insurance coverage, inequitable health care access, and stigma against lesbian, gay, bisexual, transgender, queer, asexual, and intersex (LGBTQAI) individuals have caused CCS services to be underutilized by marginalized populations. 3,46][7] Although sexually minoritized populations have poorer health care access in Chicago, 8 there is limited research on CCS disparities for LGBTQAI patients.This study sought to describe the prevalence of up-to-date CCS among lesbian, gay, and bisexual (LGB) cisgender women compared with heterosexual cisgender women in the US and Chicago in particular.We used the Health Equity Promotion Model, a framework to promote LGBTQAI health that recognizes intersectionality, structural context, and health-promoting and hindering factors, to inform variable selection and interpretation of the data. 9

Methods
Data from a retrospective, cross-sectional, population-based study of adults residing in Chicago between 2020 and 2022 were analyzed.Data were collected in the Healthy Chicago Survey (pooled years 2020-2022), an annual survey of approximately 4500 randomly selected adults conducted by the Chicago Department of Public Health (see the eAppendix in Supplement 1). 10 All data, including race and ethnicity, were self-reported by survey respondents via web or paper interview.
This study was approved by the University of Illinois Chicago institutional review board.Participants provided consent through the survey.The data report followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines for cross-sectional studies. 11riables were identified a priori as confounders (age) and independent variables potentially associated with the CCS outcome (income level, 12 race, 4 Hispanic/Latina ethnicity, 4 having a primary care practitioner [PCP], 13 and insurance coverage 3,8 ) according to their importance in the Health Equity Promotion Model and the relationship between sexual orientation and up-to-date CCS. 9 Age was included as a confounder because of its association with both identifying with the LGBTQAI community and seeking CCS.Sexual orientation, Hispanic/Latina ethnicity, having a PCP, and insurance coverage are binary variables.Income level categories include less than 199%, 200% to 399%, and greater than or equal to 400% of the Federal Poverty Level.Race categories include Asian American, Pacific Islander or Native Hawaiian, Black or African American, White, or some other race (includes American Indian or Alaska Native).
Respondents who self-identified as lesbian, gay, or bisexual or other than straight, lesbian, or bisexual were coded as LGB.Respondents who self-identified as straight were coded as heterosexual.Up-to-date CCS was measured by the survey question, "How long has it been since your last Pap test?"Those who had a Papanicolaou test within the past 3 years were considered up-to-date.Although this meets current recommendations for CCS, this measure may have underestimated up-to-date CCS for women aged 30 to 64, for whom cotesting with an HPV DNA test every 5 years is commonly recommended and used. 1 Having a PCP was measured by the survey question, "Do you have at least one person you think of as your personal doctor or health care provider?"The survey did not specify specialty or differentiate between physicians and advanced practice practitioners.

Statistical Analysis
Data analysis was performed from June to October 2023.Stratified prevalence ratios (PRs), logistic regression models, and interaction analysis were used to describe the association of sexual orientation with up-to-date CCS.Data were weighted using pooled weights to adjust for unequal probability of selection and nonresponse.Interaction was assessed via logistic regression; relative excess proportion due to interaction and a ratio of PRs were calculated according to methods described by Knol et al. 14 Backward elimination (threshold P < .05)was used to identify significant independent variables associated with up-to-date CCS.Variables were retained where at least 1 level met this threshold.Data analysis was performed in SAS Studio software version 3.81 using SAS survey procedures (SAS Institute).All P values were from 2-sided tests and results were statistically significant at P < .05.

Results
The total cohort of respondents eligible for CCS was 5399.We were unable to include cisgender women aged 21 to 24 years because the survey instrument grouped these observations with those aged 18 to 20 years, who are ineligible for CCS. 1 Observations missing sexual orientation (112 In logistic regression models, observations were removed that were missing at least 1 variable. The analytical sample included 5167 cisgender women aged 25 to 64 years with no history of hysterectomy.More than 90% of the sample (4720 participants) identified as heterosexual, and 447 identified as LGB.Among LGB cisgender women, 318 (71.14%) reported up-to-date CCS compared a Some other race means the participant selfidentified as "some other race" or American Indian or Alaska Native.e Some other race means the participant selfidentified as "some other race" or American Indian or Alaska Native.

JAMA Network Open | Oncology
Sexual Orientation and Cervical Cancer Screening Among Cisgender Women  3).
In further interaction analysis, there was no significant multiplicative interaction for any covariate, consistent with the regression model.However, the interaction between LGB sexual orientation and having a PCP is reported because of its importance in the Health Equity Promotion Model and significance on an additive scale. 9LGB cisgender women with a PCP were 93% more likely to be up-to-date on CCS compared with LGB cisgender women without a PCP (PR, 1.93; 95% CI, 1.37-2.72).In addition, heterosexual cisgender women with a PCP were 47% more likely to be up-todate on CCS compared with heterosexual cisgender women without a PCP (PR, 1.47; 95% CI, 1.31-1.64)(Table 4).The relative excess proportion of up-to-date CCS due to interaction was 0.18 (95% CI, 0.13-0.23).

Discussion
This cross-sectional study found that LGB cisgender women are less likely to be up-to-date on CCS than heterosexual cisgender women.Also, having a PCP is an important factor associated with up-todate CCS in Chicago-regardless of sexual orientation-but the association of having a PCP is greater Abbreviations: PCP, primary care practitioner; PR, prevalence ratio.
a PRs were weighted and adjusted for age.The measure of interaction on an additive scale (ie, the relative excess proportion due to interaction) was 0.18 (95% CI, 0.13-0.23).The measure of interaction on multiplicative scale (ie, the ratio of PRs) was 1.32 (95% CI, 0.01-153.29).

Table 2 .
Stratified PRs of Up-to-Date Cervical Cancer Screening, Lesbian, Gay, or Bisexual vs Heterosexual Participants (N = 5167)

Table 4 .
Additive Interaction Effect of Having a PCP on Up-to-Date Cervical Cancer Screening by Sexual Orientation (n = 5107)