The Challenges Presented around Collection of Patient Sexual Orientation and Gender Identity Information for Reduction of Health Disparities

Research has well-established the physical and mental health and health care disparities (e.g., higher rates of substance use, depression, and unmet health needs) experienced by sexual and gender minority (SGM) populations.^1-5 SGM is a broad term that refers to populations whose sexual orientation (SO) or behavior is other than heterosexual or whose gender identity (GI) is other than cisgender (i.e., GI is concordant with the sex assigned at birth) or within the gender binary (male or female). Recent Gallup Poll findings estimated that > 10 million adults in the United States identify as lesbian, gay, bisexual, or transgender.⁶ This number, however, likely underestimates the number of all individuals across the SGM spectrum.

In an effort to obtain a more accurate depiction of SGM population health and to assess, track, and combat population-level health disparities, federal agencies and national organizations alike have strongly recommended that health care systems routinely collect and document patient SO and GI information, as we do with other demographic characteristics (eg, race).^7-10 Having information on patient SO/GI would clearly improve our ability for epidemiological study of SGM populations. In addition, provider knowledge of patient SO/GI may improve individual clinical care and patient experience. SGM populations and health care patient populations have, in general, indicated a willingness to respond to questions about their SO/GI.^11-14

Well-intentioned policies and practices that seem promising on face value, however, often come with unintended or unforeseen consequences or considerations that may limit the beneficial impact of such practices. Exploring questions about the potential impact of new practices can be helpful to refine implementation and evaluation. In the case of routine collection of patient SO/GI data for the purposes of tracking disparities in care access, experience, and health outcomes, unintended consequences may result from challenges to obtaining information that accurately reflects a patient’s identities.

Lack of complete and accurate information inhibits our abilities to track disparities. Further, although not wanting to silence individual identities, identifying individuals as having SGM status can increase opportunities for further discrimination and prejudice. The question, then, is how to capture and track patients’ SGM identities in ways that allow us to decrease, without exacerbating, disparities. Below, we discuss recent research findings that illuminate the challenges and offer considerations for innovative solutions for overcoming such challenges while advancing the goal of reducing disparities.

THE CHALLENGE

Although asking a patient about their SO/GI is the singular way to obtain this information, the inquiry might not result in an accurate reflection of the patient’s identities to: (a) barriers to disclosure of one’s SO and/or GI, (b) fluidity of and changes in identities over time, and (c) mismatch between label options and individual identities. That is, patients may not be willing to share their SO and/or GI, may vary in their SO/GI identities over time or context, and/or may not identify with categorization or labeling of their SO/GI.

Patients, especially those whose identities are stigmatized and discriminated against, may not feel safe or comfortable disclosing such identities to a particular provider or care team member and might have concerns about the privacy of their information documented in their medical record that limits their openness to sharing. Research on disclosure of SO/GI as well as other sensitive topics, such as the experience of intimate partner violence or risky health behaviors, has revealed reports of patients choosing not to disclose such sensitive information due to concerns about privacy, judgment, bias, and stigma.^14-16 In cases of discomfort with disclosing SO/GI, patients may instead conceal their true identities (ie, report a label that does not reflect their actual identity) or choose not to respond to the inquiry (ie, choose not to answer).¹⁶

Second, individual SO and GI may vary across time and/or context. Labels are used to convey affiliation, behavior, or internal identity thus may be used differently in different contexts. For example, one may identify as “queer” as a political or social label but select “heterosexual” or “bisexual” in a clinical context. Further, orientations and identities may be fluid over time. In focus groups with university students about SO/GI disclosure in health care, participants reflected on discomfort with checking SO/GI boxes because of the idea that doing so implies a static identity without the possibility of change over time.¹⁷ For some, even the idea of selecting a label was off-putting and felt in- authentic. Data systems that capture a label at one point in time may no longer be valid beyond that point or in a different con- text. A patient may even choose to use different labels for different providers for different purposes. Asking about SO/GI, even if asked over time, implies a trait that for some feels quite solid but denies the possibility that, for others, it is less so.

Finally, data and documentation systems may require a limited set of SO/GI labels or categories that might not reflect a particular individual’s identities. As found in focus group research,¹⁷ some individuals do not identify their SO and/or GI as having any category or label; others may identify with a category that is not offered as a selection within the data systems. Focus group participants reported at times selecting a label closest to their identities while also noting that it did not always fit exactly. Another alternative is selecting an option to not answer, or to choose an “other” or “none of the above” option, which may get coded as unknown information or missing data, again, failing to reflect the true diversity of sexual and GIs and likely missing the potentially more op- pressed or neglected populations.

These challenges to complete and accurate reporting of SO/GI identities can impact public health tracking of SO/GI disparities. If the data collected from medical record documentation are not accurate or complete, epidemiological analysis and conclusions based on these data may also be biased or misleading, potentially undercounting SGM populations and failing to identify the full scope of disparities in clinical care based on SO/GI. In fact, a recent analysis of the first year of SO/GI data collection in 1367 US health centers, found that a majority of the patients had missing data for SO (77.1%) and GI (62.8%). Of the patients with complete SO/ GI data, 27.5% of patients chose not to disclose their SO and 9.3% chose not to disclose GI.¹⁸ With a large proportion of missing or undisclosed information, data systems fail to include—and may further silence or hide—the individuals perhaps most vulnerable to health disparities.

In addition to the challenges of complete and accurate SO/ GI data collection, the attempt to collect such information raises questions about: (a) how the collection of information related to a particular individual may impact the individual’s care, and (b) the implications of focusing on SO and GI as categories and labels. Information on a patient’s SO/GI is not necessary for clinical care. In some cases, information on patient sexual behaviors may be relevant for clinical care, but this information cannot be deduced from patient SO (and knowledge of a patient’s SO may lead to erroneous assumptions about a patient’s sexual activity).¹⁹ Knowing a patient’s chosen name and ap- propriate pronouns is important for patient-centered care and to ensure appropriate use of gendered language, such as pronouns (e.g., she/he/they) or gendered titles (e.g., Mr., Ms., Mx.); however, this information is not necessarily known by disclosed GI and may be collected through asking patients for their pronouns rather than GI. Doing so allows the patient to select the way in which they want to be referred in that setting and thereby avoiding unintentional misgendering.

Further, in a time when protections against discrimination for SGM populations are being eroded in some cases (though expanded in others) and new policies enacted that establish license to refuse health care on the basis of a moral or religious objection to a patient’s identity or behavior,²⁰ tagging patients with labels may offer new opportunity for further discrimination. The newly expanded “conscience rule” in the United States may inhibit further disclosure of SGM status due to increased concerns about discrimination. Although silencing and closeting of SGM identities does not serve to improve individual health or societal acceptance — in fact, can contribute to further discrimination, stigma, and poor health and health care access — individuals in particular environments may need to be more guarded about sharing their personal information to protect against further discrimination.

INNOVATION TOWARD SOLUTION

The question posed by the challenges described above is how we can track accurate information on patient SO/GI to identify disparities by SGM status in health care access, experience, and health outcomes while minimizing potential negative impacts for individuals. A possible solution may be to provide a way for patients to self-report their SO/GI directly to a secure data collection system that links to the individual medical record would allow for collection of such information for tracking and analysis purposes while removing some of the barriers associated with disclosing to a particular individual or in a particular setting, as well as potential discrimination. The system would need to allow for patients to update their information as they see fit and offer data security protections such that not all who have access to the patient’s medical record would necessarily have access to the SO/GI information. Including options for patients to select labels appropriate to their identities would, further, be important (i.e., not limiting labeling options).

This kind of self-report system with data access protections would not, however, address all of the challenges inherent in SO/ GI data collection. There remains the challenge of rejection of the very idea of categorization and labeling of one’s SO or GI. Individuals, especially those who choose not to impose such labels, as well as those who continue to have privacy concerns despite added protections, may choose not to provide their SO/GI in- formation, continuing the challenge of missing data, though this would likely be minimized compared with current systems. Like all electronic systems, there would be a risk of data security breaches, but this secure system would offer some privacy protections that do not currently exist in the electronic medical records. Any security breaches would not likely to proliferate individual information as broadly as having the information displayed in the medical record accessible to all clinical staff and providers.

In addition to encouraging patients to self-report their SO/ GI to a secure system that offers privacy protections, health care providers and practice settings should also work to ensure a welcoming and supportive environment that allows patients to share information about their sexuality and gender in ways that feel comfortable, safe, and appropriate to the patient. Health care settings and systems should be prepared to respond to patient needs related to sexuality and GI and to adjust practices as necessary to reduce disparities and allow patients the autonomy to make decisions about whether and how information about their SO/GI is recorded and utilized in their clinical care.

CONCLUSIONS

Identifying and eliminating health disparities based on GI and SO is critical to improving population health and health equity. Although patient-level data on SO/GI could be useful for identifying population-level disparities and trends, we have identified ways in which current and emerging practices around collection of individual SO/GI as part of clinical care can come with unintended consequences that may even serve to further disparities and inequities in health and health care, and fail to reflect the true characteristics of the population. In asking about patient SO/GI, we need to be clear about why we are doing so, how the information is used, and whether the current approaches to doing so make the most sense to achieve health equity.

The findings of recent research and considerations noted above encourage reconsideration of how patient SO/GI in- formation is collected. Designing systems that allow for patients to disclose their SO/GI confidentially — that is, to have their SO/ GI identified and linked to their other medical records data but without this information appearing on their medical record or needing to disclose to a particular person — may provide some of the privacy protections that would facilitate data collection and tracking. At the same time, it is important that we work against stigma and discrimination so that individuals do not feel a need to conceal their identities for the sake of their safety and well-being. Tracking disparities with data that are confidential at the individual level may expose areas where further work against discrimination is needed. Clinical care settings should provide opportunities for, though not require, patients to share with individual members of their health care team information about themselves that they want to be known and to signal a welcoming environment for a broad diversity of GIs and SOs.

As we proceed with erasing prejudice, discrimination, and disparities on the basis of SO and GI, we may also seek to understand further the need for and implications of categorizing individuals by SO and GIs, and the benefits and challenges of doing so. Although we do not want to impose assumptions about an individual’s experiences, needs, or well-being on the basis of SO/ GI categories, neither do we want to necessarily erase the elements of one’s experience or identity tied to such categories. SO/GI identities may be relevant to an individual’s life experience and related to their health and health outcomes. And, yet, we do want to be careful not to presume any particular outcome or experience on the basis of an individual’s SO/GI.

To achieve improved patient-centered care and tracking population-level data we need to consider the evidence from research with both providers and patients that highlights the potential unintended challenges of routine patient SO/GI in- formation collection and compels us to tackle implementation strategies responsive to these concerns. If patients do not feel safe disclosing their SO/GI or do not find that their experience and identities fit within proscribed categories, data may fail to reflect the reality of the population’s identities. Further, the re- porting and documentation of patient SO/GI may increase individual risk for experience of discrimination, bias, prejudice, and poor care due to false assumptions. The combination of health care settings fostering welcoming and inclusive clinical environments, and providing patients with the opportunity to disclose on a system linked to their health care record that gives them control over their information, may mitigate these concerns and provide more complete and accurate SO/GI data for epidemiological purposes and combating health disparities. Ultimately, we will ideally reach a point at which we see no disparities on the basis of SGM status and at which people are equally freely, safely, and comfortably able to identify with a particular SO or GI, or not feel a need to select a category at all.