Abstract
Introduction:
Sexual orientation and gender identity data collection is necessary to address health inequities. This study examines sexual orientation and gender identity data reporting among community health centers.
Methods:
Using the 2016–2019 Uniform Data System for 1,381 community health centers, trends in reporting of sexual orientation and gender identity data were examined. Multivariable logistic regression was used to assess associations between community health center characteristics and whether sexual orientation and gender identity data were available for ≥75% of a community health center’s patients in 2019. Data were analyzed in 2021.
Results:
In 2016–2019, the percentage of community health centers with sexual orientation and gender identity data for ≥75% of patients increased from 14.9% to 53.0%. In 2019, community health centers were more likely to have this data for ≥75% of patients if they were in nonmetro counties (OR=1.48, 95% CI=1.04, 2.10 versus metro), were in the South (OR=2.27, 95% CI=1.57, 3.31) or West (OR=1.91, 95% CI=1.27, 2.88 versus the Northeast), and had more patients aged between 18 and 39 years (OR=1.04, 95% CI=1.02, 1.07), between 40 and 64 years (OR=1.04, 95% CI=1.02, 1.06 vs <18 years), or veterans (OR=1.10, 95% CI=1.01, 1.20). This was less likely among community health centers serving 10,000–20,000 patients (OR=0.70, 95% CI=0.52, 0.95) and >20,000 patients (OR=0.44, 95% CI=0.32, 0.61 vs <10,000) and community health centers with more patients of American Indian/Alaskan Native (OR=0.98, 95% CI=0.97, 0.99) or unknown race (OR=0.92, 95% CI=0.86, 0.97 versus White).
Conclusions:
Collection of sexual orientation and gender identity data by community health centers has increased substantially since 2016, although gaps remain.
INTRODUCTION
Systematic collection of sexual orientation and gender identity (SOGI) data in healthcare settings is integral to addressing health inequities among lesbian, gay, bisexual, transgender, and queer (LGBTQ) populations. The patient- and population-level health benefits of using evidence-based approaches for obtaining SOGI data are numerous and have been discussed widely.1–9 For example, evidence-based SOGI data collection and documentation can improve provider–patient rapport and increase opportunities for targeted preventive care, including for tobacco use and depression, and cervical, colon, or breast cancer screenings.1,4,10–12 In a meta-analysis of SO disclosure in healthcare settings, Ruben et al.12 found that disclosure was related to greater adherence to preventive care measures and more regular utilization of healthcare services. Systematic SOGI data collection also permits enhanced public health surveillance to improve LGBTQ health and address inequities, including in the context of the coronavirus disease 2019 (COVID-19) pandemic during which LGBTQ people have faced elevated risks of adverse health and economics effects of COVID-19.5,13–16
Beginning in 2016, the Health Resources and Services Administration (HRSA) required U.S. community health centers (CHCs) to report SOGI data for all adult patients. This requirement was consistent with many of the SOGI data policy priorities that had been highlighted across several federal agencies.17–20 For example, in 2011, the National Academy of Medicine released a landmark report on LGBTQ health that called for routine collection of SOGI data in healthcare settings and HHS established a timeline for improving collection of SOGI in national surveys in a manner consistent with Section 4302 of the Affordable Care Act.6,21 HRSA’s requirement for CHCs, which remains the only federal mandate to require SOGI data collection in any U.S. healthcare setting, is particularly important given CHCs’ role in providing safety net primary care for nearly 30 million patients in 2020. More than 60% of CHC patients are racial/ethnic minority individuals and >90% have incomes ≤200% of the federal poverty guidelines.22 Thus, SOGI data collection within CHCs provides opportunities to improve clinical care for and address health inequities experienced by individuals with multiple minoritized identities.
In 2016, the first year of the new HRSA requirement, Grasso and colleagues23 found SOGI data collection among CHCs was low: 77% of CHC patients were missing SO and 63% were missing GI information. The authors posit several explanations for these findings, including that the SOGI reporting requirement was communicated mid-year to CHCs and that many CHCs may have lacked electronic health records with the capacity to record structured SOGI information.23 There is limited information on changes in CHCs’ SOGI data reporting since 2016.
In this study, national CHC data from 2016 to 2019 are used to assess changes in SOGI data reporting and to examine associations between CHC characteristics and SOGI data reporting in 2019.
METHODS
Study Sample
This study used 2016–2019 data from the publicly available Uniform Data System (UDS). The UDS is aggregated to the CHC level and contains data on CHC grantee characteristics (including size, location, services provided) and CHC patients characteristics (including demographics, service use, and outcomes).18,24–26 The sample included 1,381 CHCs that received federal funds through Section 330 of the Public Health Service Act in 50 U.S. states and the District of Columbia in any year from 2016 to 2019.
During the study period, CHCs were encouraged to establish a routine data collection system for obtaining self-reported SOGI information from patients and were required to report SOGI data for all adult patients (aged ≥18 years) to HRSA yearly.18 Specifically, CHCs were instructed to report the number of patients in each of 6 SO groups and 6 GI groups (Appendix Table 1). SO groups were lesbian or gay, straight (not lesbian or gay), bisexual, something else, don’t know, and chose not to disclose. GI groups were male, female, transgender male/female-to-male, transgender female/male-to-female, other, and chose not to disclose. CHCs were instructed to include patients with missing SO data in the don’t know group and patients with missing gender identity data in the other group. HRSA supported but did not require collection of self-reported SOGI information from patients aged <18 years; if CHCs reported SOGI data for patients aged <18 years, these data were included in their general counts for each SOGI category.
Measures
The proportion of patients with SOGI data were calculated for each study year, including binary measures of whether CHCs had SO or GI data for ≥1 patient, ≥50% of patients, and ≥75% of patients and whether CHCs had both SO and GI data for ≥1 patient, ≥50% of patients, and ≥75% of patients. For each CHC, the number of patients with SO data was calculated as the sum of all CHC patients except those reported in the “don’t know” group; the number of patients with GI data was calculated as the sum of all CHC patients except those reported in the “other” group. All patients reported in the don’t know SO group or the other GI group were considered to be missing because it was not possible to distinguish between patients who self-reported an identity included in the don’t know or other groups and patients with missing data. Because the UDS reports aggregate CHC-level data, it was also not possible to determine whether patients with missing SOGI data were aged <18 years. In primary analyses, all CHC patients, regardless of age, were included in the denominator for these measures.
Changes in the proportion of patients reporting LGBTQ identities were also measured. These were calculated as the sum of patients reported in the lesbian or gay, bisexual, or something else SO groups and then, separately, the sum of the patients reported in the transgender male/female-to-male and transgender female/male-to-female GI groups. The denominator for these measures was the total number of CHC patients in the sample in each year.
The UDS contains health center characteristics, including census region, size (i.e., number of patients), whether the CHC provided certain types of services (i.e., mental health, substance use disorder, dental), and rurality based on the county-level Rural–Urban Continuum Code associated with the CHC grantee address.27
The UDS also provides patient characteristics, aggregated to the CHC level. These included the proportion of each CHC’s patients by age, race, ethnicity, income as percentage of federal poverty guidelines, insurance type, whether patients were best served in a language other than English, and whether patients belonged to specific groups that receive specialized federal funding (e.g., patients who lack housing or who are veterans).
Statistical Analysis
The proportion of patients with SOGI data and the proportion of patients reporting LGBTQ identities were calculated for CHCs in 2016–2019. Additionally, using multivariable logistic regression, associations between SOGI data reporting in 2019 and CHC and aggregate patient characteristics were studied. The outcome variable was whether CHCs had SOGI data for ≥75% of patients in 2019. This 75% threshold was selected to approximate the percentage of CHC patients who were aged ≥18 years and thus required by HRSA to have SOGI data.
Regression models were adjusted for 2019 CHC and aggregate patient characteristics that could be associated with SOGI data reporting, including characteristics used in previous literature.23 CHC characteristics were modeled as categorical or binary variables and included census region (Midwest, Northeast, South, West), whether the CHC provided certain types of services (mental health, substance use disorder, dental), size (<10,000 patients, 10,000–20,000 patients, ≥20,000 patients), and rurality (rural, non-metro, metro).23 Aggregate patient characteristics included age (0–17, 18–39, 40–64, ≥65 years), income as percentage of federal poverty guidelines (<100%, 100%–200%, >200%), insurance type (Medicaid or Children’s Health Insurance Program, Medicare or other public insurance, commercial insurance, uninsured), whether patients were best served in a language other than English, whether patients belonged to specific groups that receive specialized funding (patients who are agricultural workers, patients who lack housing, school-based health center patients, veteran patients), and race/ethnicity. The race/ethnicity variable was generated using separate race and ethnicity measures and included the following groups: Hispanic, non-Hispanic Black or African American, non-Hispanic American Indian or Alaska Native, non-Hispanic Asian, non-Hispanic White, non-Hispanic patients of another race (including Native Hawaiian, Other Pacific Islander, and >1 race), and non-Hispanic patients who did not report race. Because patient characteristics were aggregated to the CHC level, they were modeled as continuous variables representing the percentage of patients within a given CHC in each group (e.g., percentage of a CHC’s patients with commercial insurance).
Because CHCs were not required to report SOGI data for patients aged <18 years, sensitivity analyses were conducted in which the outcome of the proportion of a CHC’s patients with SOGI data was calculated using only the number of adult patients as the denominator; this likely represents an upper bound as some CHCs could be reporting SOGI data for some patients aged <18 years. An additional sensitivity analysis accounted for the percentage of patients who “chose not to disclose” SO or GI data by including these percentages at the CHC level as covariates in the model. This study was approved by the Mass General Brigham IRB. Analyses were conducted in 2021 using RStudio, version 3.5.3.28
RESULTS
The proportion of CHC patients in each of the SOGI groups changed over the study period (Table 1). As the percentage of patients in the missing/don’t know group for SO decreased from 2016 to 2019, the percentage of patients reported in the lesbian or gay, bisexual, or something else SO groups increased from 0.9% to 2.2%. Similarly, as the proportion of patients in the missing/other GI group decreased, the percentage reported in the transgender male/female-to-male and transgender female/male-to-female GI groups increased from 0.2% in 2016 to 0.3% in 2019.
Table 1.
Proportion of Patients Reported in SOGI Groups Among CHCs in 2016–2019
| 2016 | 2017 | 2018 | 2019 | |
|---|---|---|---|---|
| SOGI Groups | (n=25,413,089) | (n=26,719,145) | (n=27,890,141) | (n=29,309,889) |
| % | % | % | % | |
| Sexual orientation | ||||
| Lesbian or gay | 0.5 | 0.8 | 1.0 | 1.1 |
| Bisexual | 0.2 | 0.4 | 0.6 | 0.7 |
| Something else | 0.2 | 0.3 | 0.2 | 0.4 |
| Straight | 15.6 | 35.7 | 46.2 | 51.9 |
| Chose not to disclose | 6.3 | 8.8 | 9.3 | 9.1 |
| Missing/don’t know | 77.3 | 54.0 | 42.7 | 36.7 |
| Gender identity | ||||
| Transgender female | 0.1 | 0.1 | 0.1 | 0.1 |
| Transgender male | 0.1 | 0.1 | 0.2 | 0.2 |
| Female | 19.4 | 32.1 | 38.1 | 28.5 |
| Male | 13.8 | 22.1 | 26.6 | 41.0 |
| Chose not to disclose | 3.5 | 6.0 | 5.6 | 4.0 |
| Missing/other | 63.2 | 39.6 | 29.5 | 26.2 |
SOGI, sexual orientation and gender identity; CHC, community health center.
By 2019, nearly all (99.2%) of CHCs had both SO and GI data for ≥1 patient compared with 73.2% in 2016 (Figure 1). The percentage of CHCs with SOGI data for ≥75% of patients increased from 14.9% in 2016 to 53.0% in 2019. CHCs were more likely to have a higher percentage of patients with GI versus SO data. For example, in 2019, the percentage of CHCs with SO data for more than half of their patients was 77.7% compared with 85.8% with GI data. The proportion of CHCs with SOGI data for ≥50% and ≥75% of patients increased in sensitivity analyses where the percentage of patients with SOGI data was calculated only among patients aged ≥18 years, although trends were similar during the study period (Appendix Figure 1).
Figure 1.
Changes in proportion of CHCs with SOGI data for any, 50%, and 75% of patients in 2016‒2019.
CHC, community health center; SOGI, sexual orientation and gender identity.
In multivariable analyses, the odds of having SOGI data for ≥75% of patients in 2019 were lower for CHCs with 10,000–20,000 patients (OR=0.70, 95% CI=0.52, 0.95) and CHCs with >20,000 patients (OR=0.44, 95% CI=0.32, 0.61) versus CHCs with <10,000 patients (Table 2). CHCs located in non-metro versus metro counties were more likely to have SOGI data for ≥75% of patients (OR=1.48, 95% CI=1.04, 2.10); differences for CHCs in rural versus metro counties were not significant. Compared with CHCs in the Northeast, the odds of having SOGI data for ≥75% of patients were approximately 2 times higher among CHCs in the South (OR=2.27, 95% CI=1.57, 3.31) and West (OR=1.91, 95% CI=1.27, 2.88).
Table 2.
Association between SOGI Data Reporting for ≥75% of Patients and CHC Characteristics in 2019
| CHCs | ||||
|---|---|---|---|---|
| Variable | All | SOGI Data for <75% Patients | SOGI Data for ≥75% Patients | |
| N (%) | N (%) | N (%) | AOR (95% CI) | |
| CHC characteristics, n (%) | ||||
| Size | ||||
| <10,000 patients | 523 (100) | 182 (34.8) | 341 (65.2) | b |
| 10,000–20,000 patients | 381 (100) | 176 (46.2) | 205 (53.8) | 0.70 (0.52, 0.95) |
| >20,000 patients | 447 (100) | 277 (62.0) | 170 (38.0) | 0.44 (0.32, 0.61) |
| Rurality | ||||
| Metro | 980 (100) | 495 (50.5) | 485 (49.5) | b |
| Non-metro | 294 (100) | 109 (37.1) | 185 (62.9) | 1.48 (1.04, 2.10) |
| Rural | 77 (100) | 31 (40.3) | 46 (59.7) | 1.14 (0.64, 2.04) |
| Census region | ||||
| Northeast | 231 (100) | 122 (52.8) | 109 (47.2) | b |
| Midwest | 271 (100) | 159 (58.7) | 112 (41.3) | 1.44 (0.96, 2.18) |
| South | 458 (100) | 175 (38.2) | 283 (61.8) | 2.27 (1.57, 3.31) |
| West | 391 (100) | 179 (45.8) | 212 (54.2) | 1.91 (1.27, 2.88) |
| Service type offered | ||||
| Mental health | 1,311 (100) | 618 (47.1) | 693 (52.9) | 1.06 (0.52, 2.12) |
| Substance use disorder | 805 (100) | 370 (46.0) | 435 (54.0) | 1.09 (0.85, 1.39) |
| Dental | 1,149 (100) | 546 (47.5) | 603 (52.5) | 1.36 (0.95, 1.95) |
| Aggregate patient characteristics, mean % (SD)c | ||||
| Age in years | ||||
| 0–17 | 26.4 (12.7) | 30.2 (12.2) | 23.1 (12.2) | b |
| 18–39 | 29.7 (7.7) | 29.0 (6.2) | 30.4 (8.7) | 1.04 (1.02, 1.07) |
| 40–64 | 33.0 (8.6) | 30.8 (7.7) | 34.9 (8.9) | 1.04 (1.02, 1.06) |
| ≥65 | 10.9 (6.5) | 10.0 (6.0) | 11.6 (6.9) | 1.03 (0.98, 1.08) |
| Race/Ethnicity | ||||
| White | 41.9 (29.9) | 39.1 (29.2) | 44.4 (30.3) | b |
| American Indian/Alaskan Native | 2.1 (9.2) | 2.5 (10.7) | 1.7 (7.7) | 0.98 (0.97, 0.99) |
| Asian | 3.2 (9.2) | 3.2 (8.7) | 3.3 (9.6) | 1.01 (0.99, 1.03) |
| Black | 18.7 (22.6) | 17.9 (21.8) | 19.5 (23.3) | 1.00 (1.00, 1.01) |
| Hispanic | 26.5 (25.8) | 29.0 (26.4) | 24.2 (25.0) | 1.00 (0.99, 1.01) |
| Another race/ethnicity | 2.0 (4.4) | 2.0 (4.7) | 1.9 (4.2) | 1.01 (0.98, 1.04) |
| Unknown | 1.8 (2.4) | 2.1 (2.6) | 1.5 (2.1) | 0.92 (0.86, 0.97) |
| Specific populations | ||||
| Agricultural workersa | 2.5 (9.7) | 2.2 (7.6) | 2.8 (11.2) | 1.01 (0.99, 1.02) |
| Lack housing | 7.2 (17.4) | 6.2 (15.1) | 8.0 (19.2) | 0.99 (0.98, 1.00) |
| School-based | 2.8 (7.3) | 3.2 (7.5) | 2.3 (7.2) | 0.99 (0.98, 1.01) |
| Veteran | 1.8 (2.1) | 1.5 (1.8) | 2.0 (2.2) | 1.10 (1.01, 1.20) |
| Income | ||||
| >200% FPG | 7.3 (8.3) | 7.1 (8.0) | 7.4 (8.5) | b |
| 100%–200% FPG | 17.0 (10.1) | 16.8 (9.6) | 17.1 (10.6) | 1.01 (0.99, 1.04) |
| <100% FPG | 45.7 (22.8) | 44.7 (22.2) | 46.6 (23.2) | 1.02 (1.00, 1.04) |
| Unknown | 30.1 (25.2) | 31.5 (25.0) | 28.9 (25.4) | 1.01 (0.99, 1.03) |
| Insurance type | ||||
| Medicaid and CHIP | 42.5 (18.4) | 45.5 (17.6) | 39.9 (18.8) | b |
| Uninsured | 24.8 (17.7) | 23.3 (16.1) | 26.2 (18.8) | 1.00 (0.99, 1.01) |
| Medicare and other public | 11.8 (7.3) | 11.1 (6.7) | 12.5 (7.7) | 0.99 (0.95, 1.02) |
| Commercial | 20.8 (12.9) | 20.0 (12.1) | 21.5 (13.5) | 1.00 (0.99, 1.02) |
| Primary language besides English | 18.2 (20.1) | 20.0 (20.3) | 16.6 (19.9) | 1.00 (0.99, 1.01) |
Note: Sample size was n=1,351. One CHC with a missing rural-urban continuum code was dropped from this analysis. Boldface indicates that the 95% CI does not include 1.
Agricultural workers are migratory or seasonal agricultural workers and their family members, aged or elderly former migratory agricultural workers.
Indicates omitted group.
AOR correspond with a 1 percentage point change in the proportion of patients as variables were modeled as continuous percentages at the CHC level.
SOGI, sexual orientation and gender identity; CHC, community health center; CHIP, Children’s Health Insurance Program; FPG, Federal Poverty Guidelines.
As expected, the odds of having SOGI data for ≥75% of patients were greater among CHCs with more patients aged 18–39 years (OR=1.04, 95% CI=1.02, 1.07) and 40–64 years (OR=1.04, 95% CI=1.02, 1.06) versus <18 years. CHCs were also more likely to have SOGI data for ≥75% of patients if a higher percentage of their patients were veterans (OR=1.10, 95% CI=1.01, 1.20). The odds of having SOGI data for ≥75% were lower among CHCs with greater proportions of their patient population reporting American Indian or Alaskan Native race (OR=0.98, 95% CI=0.97, 0.99) or with unknown race (OR=0.92, 95% CI=0.86, 0.97).
These findings were similar in sensitivity analyses that accounted for the proportion of each CHC’s patients that were aged ≥18 years using an alternative outcome denominator and in an analysis accounting for the proportion of patients who chose not to disclose SO or GI (Appendix Table 2). Non-disclosure of SO was positively associated, and non-disclosure of GI was negatively associated with having SOGI data for ≥75% of patients.
DISCUSSION
This study finds that the proportion of CHC patients with SOGI data grew substantially over the first 4 years of mandated reporting. The largest increases occurred between 2016 and 2017, suggesting that many CHCs were able to adopt SOGI data collection procedures within the first year after the initial announcement in 2016. With greater SOGI reporting there were also increases in the proportion of patients reporting LGBTQ identities.
There were notable increases in the overall proportion of patients with SOGI data among CHCs the 4 years following the mandate, with the proportion of CHC patients with unknown SO dropping from 77% to 37% and GI from 63% to 26%. Trainings from the National LGBT Health Education Center were made available to CHCs in 2016 to support implementation of SOGI data collection, which could underscore the importance of pairing such mandates with resources for training and implementation.4,23,29,30 Understanding the extent to which CHCs utilized the available training and the implementation and infrastructure support that has facilitated uptake of SOGI data collection is critical for broader efforts to reduce remaining gaps in data collection within CHCs and in other healthcare settings.31
In 2019, the fourth year of mandated reporting just more than half of all CHCs had SOGI data for ≥75% of patients. Moreover, there was variation in SOGI data reporting by CHC and patient characteristics. For example, the odds of having SOGI data for ≥75% of patients differed by CHC size and location, even after adjusting for the percentage of patients aged <18 years (for whom SOGI data reporting is not required) as well rates of missingness for other types of data (i.e., race and income). Similar to the findings of Grasso et al.23 from the first year of the requirement, these results ease concerns that SOGI data collection is more difficult in CHCs that are smaller or located in rural settings.
In 2019, less than half of CHCs serving >20,000 patients and CHCs in metro counties had SOGI data for ≥75% of patients. These findings highlight a need to better understand barriers to SOGI data reporting among larger and urban CHCs such that interventions aimed to increase SOGI data collection can target these barriers. Cahill and colleagues19 have made recommendations for improving SOGI data collection across clinical settings, including training staff members, meeting with LGBTQ community members to communicate the importance of and privacy considerations for SOGI data collection, and ensuring that nondiscrimination policies are in place to protect LGBTQ patients. It is possible that larger or urban CHCs face additional challenges when implementing these recommendations, such as difficulty coordinate training across a wide range of clinical settings or service lines or communicating with more numerous and varied LGBTQ communities. Of note, this study did not account for municipal-level nondiscrimination policies, which were recently found to be associated with greater SOGI data reporting among CHCs.32
This study also identifies an association between race/ethnicity and SOGI data; specifically, CHCs with a greater proportion of American Indian or Alaska Native patients and patients with unknown race/ethnicity were less likely to have more complete SOGI data. Race/ethnicity data were missing for a much smaller proportion of CHC patients compared with SOGI data. However, the positive association between race/ethnicity and SOGI reporting could reflect broader barriers to collecting patient-reported data at the CHC level. Overlap in missingness of data on SOGI, race, and ethnicity could hinder efforts to mitigate inequities among communities with multiple marginalized identities.33–36 Relatedly, research focused on LGBTQ American Indian or Alaska Native populations has found substantial prevalence of depression and other chronic diseases, experiences of violence, and uninsurance, raising additional concerns about lower SOGI reporting among CHCs that serve greater shares of American Indian or Alaska Native patients.37
The CHCs that served a greater share of specific populations designated by HRSA, including agricultural workers and those lack housing, were similarly likely to collect SOGI data for ≥75% of patients. CHCs with a greater share of veteran patients were more likely to meet this threshold, consistent with previous work demonstrating that veterans were less likely than non-veterans to refuse to answer SO questions on a population-based survey.38
Limitations
The study has limitations. Because the UDS does not include patient-level data, it was not possible to determine the proportion of missingness that is due to age-related reporting requirements or assess patient-level factors associated with SOGI reporting. The associations between CHC characteristics and greater SOGI reporting were robust across analyses that accounted for CHC-level age distributions. However, this study was not able to precisely estimate the percentage of CHCs with SOGI data for ≥75% of adult patients, though findings suggest that this percentage falls between 53% and 75% of CHCs in 2019. This issue also complicates direct comparisons between the size of the LGBTQ population served by CHCs with the broader U.S. population. The requirement that CHCs combine missing SO and GI data with the “don’t know” and “other” groups, respectively, prevents capture of individuals reporting these SOs and GIs. For this reason, this study overestimates the proportion of individuals with missing SO and GI data. In 2020, HRSA updated its SOGI data guidelines, instructing CHCs to report patients with missing SOGI data into separate “unknown” SO and “unknown” GI groups, an important step toward addressing this limitation.39
CONCLUSIONS
This study evaluates SOGI data reporting among CHCs in the years following the introduction of HRSA’s reporting requirement in 2016. Findings highlight that SOGI collection is feasible in safety net care settings with large improvements in SOGI reporting over time. However, some persistent gaps in SOGI data collection remain, even after 4 years of mandated reporting, for larger and urban CHCs, suggesting the need for targeted efforts to increase SOGI data collection in such settings.
ACKNOWLEDGMENTS
This study was supported by a grant from the Agency for Healthcare Research and Quality (AHRQ, Principal Investigator: Fung, R01HS025378). The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of AHRQ. The study sponsors did not have any role in the study design, analysis and interpretation of data, writing the report, or the decision to submit the report for publication. No financial disclosures were reported by the authors of this paper.
Appendix Table 1. Instructions for Reporting SOGI Data for CHC Patients
| Variable | Description |
|---|---|
| Sexual orientation options | |
| Lesbian or Gay | A sexual orientation that describes a person who is emotionally and sexually attracted to people of their own gender. |
| Straight (not lesbian or gay) | A sexual orientation that describes a person who is emotionally and sexually attracted to people of the opposite gender. |
| Bisexual | A sexual orientation that describes a person who is emotionally and sexually attracted to people of their own gender and people of other genders. |
| Something else | Report patients who are emotionally and sexually attracted to people who identify themselves as queer, asexual, or pansexual or another sexual orientation not captured in previous options. |
| Don’t know | Report patients who self-report that they do not know their sexual orientation. Include patients for whom the health center does not know the sexual orientation (i.e., the health center did not implement systems to permit patients to state their sexual orientation). |
| Chose not to disclose | Report patients who chose not to disclose their sexual orientation. |
| Gender identity options | |
| Male | Report patients who identify themselves as a man/male. |
| Female | Report patients who identify themselves as a woman/female. |
| Transgender male/female-to-male | Report transgender patients who describe their gender identity as man/male. (Some may just use the term man). |
| Transgender female/male-to-female | Report transgender patients who describe their gender identity as woman/female. (Some may just use the term woman). |
| Other | Report patients who do not think that one of the four categories above adequately describes them. Include patients who identify themselves as genderqueer or non-binary. In addition, report patients where the health center does not know the patient’s gender identity (i.e., the health center did not implement systems to permit patients to state their gender identity). |
| Chose not to disclose | Report patients who chose not to disclose their gender. |
Note: Information in this table appears in Uniform Data System Reporting Instructions for 2016 Health Center Data. Health Resources & Services Administration, Bureau of Primary Health Care; 2016. https://bphc.hrsa.gov/datareporting/reporting/index.html.
SOGI, sexual orientation and gender identity; CHC, community health center.
Appendix Figure 1. Changes in proportion of CHCs with SOGI data for any, 50%, and 75% of patients in 2016‒2019, including only CHC patients ≥18 years in denominator. CHC, community health center; SOGI, sexual orientation and gender identity.
Appendix Table 2. Sensitivity Analyses for Association between SOGI Data Reporting for ≥75% of Patients and CHC Characteristics in 2019
| Variable | Including non-disclosure as covariate | SOGI data for ≥75% adult patients |
|---|---|---|
| AOR (95% CI) | AOR (95% CI) | |
| CHC characteristics | ||
| Size | ||
| <10,000 patients | b | b |
| 10,000–20,000 patients | 0.65 (0.47, 0.89) | 0.73 (0.51, 1.04) |
| >20,000 patients | 0.41 (0.29, 0.59) | 0.45 (0.31, 0.65) |
| Rurality | ||
| Metro | b | b |
| Non-metro | 1.80 (1.23, 2.64) | 1.09 (0.73, 1.65) |
| Rural | 1.18 (0.63, 2.24) | 0.90 (0.46, 1.83) |
| Census region | ||
| Northeast | b | b |
| Midwest | 1.36 (0.87, 2.13) | 1.01 (0.66, 1.54) |
| South | 2.19 (1.47, 3.27) | 2.12 (1.40, 3.21) |
| West | 1.42 (0.91, 2.22) | 1.45 (0.93, 2.27) |
| Service type offered | ||
| Mental health | 1.04 (0.48, 2.23) | 0.77 (0.29, 1.78) |
| Substance use disorder | 1.05 (0.80, 1.37) | 0.92 (0.70, 1.21) |
| Dental | 1.22 (0.83, 1.79) | 1.19 (0.80, 1.75) |
| Aggregate patient characteristics | ||
| Age, years | ||
| 0–17 | b | b |
| 18–39 | 1.06 (1.03, 1.08) | 0.99 (0.97, 1.01) |
| 40–64 | 1.06 (1.04, 1.08) | 1.01 (0.99, 1.03) |
| ≥65 | 1.02 (0.98, 1.08) | 1.00 (0.95, 1.05) |
| Race/Ethnicity | ||
| White | b | b |
| American Indian/Alaskan Native | 0.99 (0.97, 1.00) | 0.98 (0.96, 0.99) |
| Asian | 1.02 (1.00, 1.04) | 1.01 (0.99, 1.03) |
| Black | 1.01 (1.00, 1.01) | 1.00 (0.99, 1.01) |
| Hispanic | 1.00 (0.99, 1.01) | 1.00 (0.99, 1.01) |
| Another race/ethnicity | 1.00 (0.97, 1.03) | 1.02 (0.98, 1.06) |
| Not reported | 0.88 (0.82, 0.94) | 0.97 (0.92, 1.02) |
| Specific populations | ||
| Agricultural workersa | 1.01 (0.99, 1.02) | 1.01 (1.00, 1.04) |
| Lack housing | 0.98 (0.98, 0.99) | 0.99 (0.98, 1.00) |
| School-based | 1.00 (0.98, 1.02) | 1.01 (0.99, 1.03) |
| Veteran | 1.07 (0.98, 1.17) | 1.14 (1.03, 1.28) |
| Income | ||
| >200% FPL | b | b |
| 100%–200% FPL | 1.00 (0.98, 1.03) | 1.02 (1.00, 1.05) |
| <100% FPL | 1.01 (1.00, 1.03) | 1.02 (1.00, 1.04) |
| Not reported | 1.00 (0.98, 1.02) | 1.01 (0.99, 1.03) |
| Insurance type | ||
| Medicaid and CHIP | b | b |
| Uninsured | 1.00 (0.99, 1.01) | 1.00 (0.99, 1.01) |
| Medicare and other public | 1.00 (0.97, 1.04) | 0.97 (0.93, 1.01) |
| Commercial | 1.00 (0.98, 1.02) | 1.01 (1.00, 1.03) |
| Primary language besides English | 0.99 (0.98, 1.01) | 0.99 (0.98, 1.01) |
| Non-disclosurec | ||
| SO not disclosed | 1.11 (1.09, 1.14) | |
| GI not disclosed | 0.96 (0.94, 0.98) |
Note: Boldface indicates that the 95% CI does not include 1.
Agricultural workers are migratory or seasonal agricultural workers and their family members, aged or elderly former migratory agricultural workers.
Indicates omitted group.
Percentage of patients within a CHC who did not disclose SO or GI are modeled as continuous variables.
SOGI, sexual orientation and gender identity; CHC, community health center; CHIP, Children’s Health Insurance Program; FPG, Federal Poverty Guidelines.
Footnotes
All authors have met the following criteria: (1) the individual made a substantial contribution to conception and design of the study, to data acquisition, or to data analysis and interpretation; (2) the individual wrote or revised the article for important intellectual content; and (3) the individual read and approved the final version of the submitted manuscript.
Preliminary results from this analysis were presented at the 2021 AcademyHealth Annual Research Meeting.
REFERENCES
- 1.Streed CG, Grasso C, Reisner SL, Mayer KH. Sexual orientation and gender identity data collection: clinical and public health importance. Am J Public Health. 2020;110(7):991–993. 10.2105/ajph.2020.305722. [DOI] [Google Scholar]
- 2.Grasso C, McDowell MJ, Goldhammer H, Keuroghlian AS. Planning and implementing sexual orientation and gender identity data collection in electronic health records. J Am Med Inform Assoc. 2019;26(1):66–70. 10.1093/jamia/ocy137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Grasso C, Goldhammer H, Thompson J, Keuroghlian AS. Optimizing gender-affirming medical care through anatomical inventories, clinical decision support, and population health management in electronic health record systems. J Am Med Inform Assoc. 2021;28(11):2531–2535. 10.1093/jamia/ocab080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Maragh-Bass AC. The “Ask” Is the “Answer”: implementing routine documentation of sexual orientation and gender identity in health care. Am J Public Health. 2019;109(8):1071–1073. 10.2105/ajph.2019.305192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cahill S, Grasso C, Keuroghlian A, Sciortino C, Mayer K. Sexual and gender minority health in the COVID-19 pandemic: why data collection and combatting discrimination matter now more than ever. Am J Public Health. 2020;110(9):1360–1361. 10.2105/ajph.2020.305829. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Institute of Medicine. The Health of Lesbian, Gay, Bisexual, and Transgender People: Building a Foundation for Better Understanding. Washington, DC: The National Academies Press; 2011. 10.17226/13128. [DOI] [PubMed] [Google Scholar]
- 7.The Joint Commission. Advancing Effective Communication, Cultural Competence, and Patient- and Family-Centered Care for the Lesbian, Gay, Bisexual, and Transgender (LGBT) Community: A Field Guide.; 2011:99. https://www.jointcommission.org/-/media/tjc/documents/resources/patient-safety-topics/health-equity/lgbtfieldguide_web_linked_verpdf.pdf?db=web&hash=FD725DC02CFE6E4F21A35EBD839BBE97&hash=FD725DC02CFE6E4F21A35EBD839BBE97. Accessed January 18, 2022.
- 8.Lett E, Everhart A. Considerations for transgender population health research based on US national surveys. Ann Epidemiol. 2022;65:65–71. 10.1016/j.annepidem.2021.10.009. [DOI] [PubMed] [Google Scholar]
- 9.Thompson HM, Kronk CA, Feasley K, Pachwicewicz P, Karnik NS. Implementation of gender identity and assigned sex at birth data collection in electronic health records: where are we now? Int J Environ Res Public Health. 2021;18(12):6599. 10.3390/ijerph18126599. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Maragh-Bass AC, Torain M, Adler R, et al. Risks, benefits, and importance of collecting sexual orientation and gender identity data in healthcare settings: a multi-method analysis of patient and provider perspectives. LGBT Health. 2017;4(2):141–152. 10.1089/lgbt.2016.0107. [DOI] [PubMed] [Google Scholar]
- 11.Grasso C, Goldhammer H, Brown RJ, Furness BW. Using sexual orientation and gender identity data in electronic health records to assess for disparities in preventive health screening services. Int J Med Inform. 2020;142:104245. 10.1016/j.ijmedinf.2020.104245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ruben MA, Fullerton M. Proportion of patients who disclose their sexual orientation to healthcare providers and its relationship to patient outcomes: a meta-analysis and review. Patient Educ Couns. 2018;101(9):1549–1560. 10.1016/j.pec.2018.05.001. [DOI] [PubMed] [Google Scholar]
- 13.Sell RL, Krims EI. Structural transphobia, homophobia, and biphobia in public health practice: the example of COVID-19 surveillance. Am J Public Health. 2021;111(9):1620–1626. 10.2105/ajph.2021.306277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lynch KE, Shipherd JC, Gatsby E, Viernes B, DuVall SL, Blosnich JR. Sexual orientation-related disparities in health conditions that elevate COVID-19 severity. Ann Epidemiol. 2022;66:5–12. 10.1016/j.annepidem.2021.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ruprecht MM, Wang X, Johnson AK, et al. Evidence of social and structural COVID-19 disparities by sexual orientation, gender identity, and race/ethnicity in an urban environment. J Urban Health. 2021;98:27–40. 10.1007/s11524-020-00497-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Keuroghlian AS. Electronic health records as an equity tool for LGBTQIA+ people. Nat Med. 2021;27(12):2071–2073. 10.1038/s41591-021-01592-3. [DOI] [PubMed] [Google Scholar]
- 17.HHS. Program Assistance Letter (PAL 2016–02). Approved Uniform Data System Changes for Calendar Year 2016. Bureau of Primary Health Care, Health Resources and Services Administration; 2016. https://bphc.hrsa.gov/sites/default/files/bphc/datareporting/reporting/program-assistance-letter-2016-02.pdf. Accessed November 17, 2020. [Google Scholar]
- 18.Uniform Data System Reporting Instructions for 2016 Health Center Data. Health Resources & Services Administration, Bureau of Primary Health Care; 2016. https://bphc.hrsa.gov/datareporting/reporting/index.html. Accessed January 10, 2022.
- 19.Cahill SR, Baker K, Deutsch MB, Keatley J, Makadon HJ. Inclusion of sexual orientation and gender identity in Stage 3 Meaningful Use Guidelines: a huge step forward for LGBT health. LGBT Health. 2016;3(2):100–102. 10.1089/lgbt.2015.0136. [DOI] [PubMed] [Google Scholar]
- 20.Institute of Medicine (US) Board on the Health of Select Populations. Collecting Sexual Orientation and Gender Identity Data in Electronic Health Records: Workshop Summary. National Academies Press; (US: ); 2013. 10.17226/18260. [DOI] [PubMed] [Google Scholar]
- 21.HHS. Improving Data for the LGBT Community. https://www.minorityhealth.hhs.gov/assets/pdf/checked/1/Fact_Sheet_LGBT.pdf. Published 2011. Accessed August 20, 2021.
- 22.Health Resources and Services Administration. National Health Center Data. https://data.hrsa.gov/tools/data-reporting/program-data/national. Accessed July 19, 2021.
- 23.Grasso C, Goldhammer H, Funk D, et al. Required sexual orientation and gender identity reporting by US health centers: first-year data. Am J Public Health. 2019;109(8):1111–1118. 10.2105/ajph.2019.305130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Uniform Data System Reporting Instructions for 2017 Health Center Data. Health Resources & Services Administration, Bureau of Primary Health Care; 2017. https://bphc.hrsa.gov/datareporting/reporting/index.html. Accessed January 10, 2022.
- 25.Uniform Data System Reporting Instructions for 2018 Health Center Data. Health Resources & Services Administration, Bureau of Primary Health Care; 2018. https://bphc.hrsa.gov/datareporting/reporting/index.html. Accessed January 10, 2022.
- 26.Uniform Data System Reporting Instructions for 2019 Health Center Data. Health Resources & Services Administration, Bureau of Primary Health Care; 2019. https://bphc.hrsa.gov/datareporting/reporting/index.html. Accessed January 10, 2022.
- 27.U.S. Department of Agriculture Economic Research Services. Rural-Urban Continuum Codes. https://www.ers.usda.gov/data-products/rural-urban-continuum-codes/. Accessed July 16, 2021.
- 28.RStudio Team. RStudio: Integrated Development for R. RStudio, PBC; 2020. http://www.rstudio.com/. Accessed January 10, 2022.
- 29.Cahill S, Singal R, Grasso C, et al. Do ask, do tell: high levels of acceptability by patients of routine collection of sexual orientation and gender identity data in four diverse American community health centers. PLoS One. 2014;9(9):e107104. 10.1371/journal.pone.0107104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ard KL, Keuroghlian AS. Training in sexual and gender minority health - expanding education to reach all clinicians. N Engl J Med. 2018;379(25):2388–2391. 10.1056/nejmp1810522. [DOI] [PubMed] [Google Scholar]
- 31.Perry NS, Elwy AR. The role of implementation science in reducing sexual and gender minority mental health disparities. LGBT Health. 2021;8(3):169–172. 10.1089/lgbt.2020.0379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Almazan AN, King D, Grasso C, et al. City-level structural stigma and patient sexual orientation and gender identity data collection at US health centers, 2018. Am J Public Health. 2021;111(11):2059–2063. 10.2105/ajph.2021.306414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hernandez SM, Sparks PJ. Barriers to health care among adults with minoritized identities in the United States, 2013–2017. Am J Public Health. 2020;110(6):857–862. 10.2105/ajph.2020.305598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sholle ET, Pinheiro LC, Adekkanattu P, et al. Underserved populations with missing race ethnicity data differ significantly from those with structured race/ethnicity documentation. J Am Med Inform Assoc. 2019;26(8‒9):722–729. 10.1093/jamia/ocz040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Lett E, Dowshen NL, Baker KE. Intersectionality and health inequities for gender minority Blacks in the U.S. Am J Prev Med. 2020;59(5):639–647. 10.1016/j.amepre.2020.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Lytle MC, Blosnich JR, Kamen C. The association of multiple identities with self-directed violence and depression among transgender individuals. Suicide Life Threat Behav. 2016;46(5):535–544. 10.1111/sltb.12234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Wilson BDM, Bouton L, Mallory C. American Indian and Alaska Native LGBT adults in the US: LGBT well-being at the intersection of race. Williams Institute, UCLA School of Law; 2021:84. https://williamsinstitute.law.ucla.edu/publications/aian-lgbt-adults-us/. Accessed January 10, 2022. [Google Scholar]
- 38.Ruben MA, Blosnich JR, Dichter ME, Luscri L, Shipherd JC. Will veterans answer sexual orientation and gender identity questions? Med Care. 2017;55(9 suppl 2):S85–S89. 10.1097/mlr.0000000000000744. [DOI] [PubMed] [Google Scholar]
- 39.Uniform Data System Reporting Instructions for 2020 Health Center Data. Health Resources & Services Administration, Bureau of Primary Health Care; 2020. https://bphc.hrsa.gov/sites/default/files/bphc/datareporting/pdf/2020-uds-manual.pdf. Accessed August 6, 2021.