HL7 Cross Paradigm Implementation Guide: Gender Harmony - Sex and Gender Representation, Edition 1
1.0.0 - release

HL7 Cross Paradigm Implementation Guide: Gender Harmony - Sex and Gender Representation, Edition 1, published by HL7 Terminology Infrastruture Work Group. This is not an authorized publication; it is the continuous build for version 1.0.0. This version is based on the current content of https://github.com/HL7/fhir-gender-harmony/ and changes regularly. See the Directory of published versions


Impact of Sex and Gender on Clinical Care

In the HL7 Informative Document: Gender Harmony - Modeling Sex and Gender Representation, Release 1, excerpted below, we noted the following (The original references are replicated here in our Bibliography.)

In 2016, The Report of the 2015 U.S. Transgender Survey which asked 27,715 self-identified transgender persons about their experiences in health care, noted that 33% had at least one negative experience with a health care provider related to being transgender, that 23% did not seek the health care they needed due to fear of mistreatment, and 33% did not go to a health care provider when they really needed to because they could not afford it (26,75).

As National LGBTQ Task Force Director Rea Carey noted:

It is outrageous that basic health care is being denied to transgender and gender non-conforming people and that so much additional trauma is being caused by doctors instead of being resolved by doctors. The medical profession must take these data seriously and ensure that everyone in the medical care system knows how to provide transgender-sensitive medical care.

This pattern is not exclusive to the United States. Medical mistreatment and/or malpractice, as well as violence against transgender and gender-diverse persons is well documented in many countries. The situation for intersex individuals is not much better. Despite worldwide condemnation of intersex genital mutilation, the practice is still widespread and even hurts non-intersex (dyadic) persons, such as in the case of David Reimer. Reimer, known in medical circles as the patient involved in the “John/Joan” case, committed suicide in 2004 following years of abuse at the hands of medical professionals. Clinicians routinely lie to patients about their status as intersex, obfuscate clinical records, and promote practices which have been shown to have negative effects on patient mental health (46,47,50).

Despite the “depathologization” of homosexuality in the Diagnostic and Statistical Manual of Mental Disorders (DSM) in 1973, the movement to depathologize transgender persons has been slow, and receives pushback even today. Medical providers continue to add being transgender to problem lists and as diagnoses in multiple jurisdictions, under labels such as “gender identity disorder” (GID). GID was removed from the World Health Organization’s ICD listing in 2019. The LGBT rights director at Human Rights Watch, Graeme Reid noted that, “the WHO’s removal of ‘gender identity disorder’ from its diagnostic manual will have a liberating effect on transgender people worldwide. Governments should swiftly reform national medical systems and laws that require that now officially outdated diagnosis.”

However, despite these calls to action, in the last ten years, transgender and intersex persons continue to report harassment and mistreatment in all areas of medical care: oncology, cardiology, geriatrics, pediatrics, psychiatry, primary care, emergency care, radiology, internal medicine,neurology, obstetrics, gynecology, pathology, surgery, and urology, to name a few (72,74). Transness still appears in problem lists, coded using antiquated terminology from ICD-9 and ICD-10 (and sometimes terminology which is even more antiquated). In some cases, this is performed because providers feel that there is nowhere else in the health record to codify that a patient is transgender, but other cases persevere even when patients specifically ask for it to be removed. This can seriously affect a transgender person’s life, as many jobs require that health records be released to employers or even publicly available for anyone to access. Many of these issues have led to significant morbidity and even patient death.

In 2016, the United Nations Programme on HIV/AIDS, alongside the World Health Organization (WHO) Global Health Workforce Alliance, launched the Agenda for Zero Discrimination in Health Care, highlighting the importance of respectful care which takes into account the values and concerns of marginalized groups, including those of transgender people (43). The agenda prioritized creation of health care frameworks for adequate monitoring and evaluation of health care systems to ensure accountability and monitor progress of health disparities in vulnerable populations, including transgender, gender-diverse, and intersex persons (47). US organizations like Fenway Health in Boston (84), Callen-Lorde in New York (85) and UCSF Trans Care in California (86) have developed programs and services for LGBTQIA+ communities, as Trans Care BC (87) and Rainbow Health in Ontario (88) have similarly in Canada.

Achievement of these goals in affected populations is not possible without an underlying, consistent structure for data collection. Such data are essential to addressing health disparities and have catalyzed initiatives to improve SOGI (sexual orientation and gender identity) data collection at the Institute of Medicine, the American Medical Association, the U.S. Centers for Disease Control and Prevention (CDC) (4), the Council of Europe (56), the United Nations Human Rights Office of the High Commissioner (57), ILGA-Europe (58), Stonewall (59), and numerous other organizations worldwide. Third gender categories have even been added in national censuses in India and Nepal (both starting in 2011), despite their removal from the 2020 U.S. Census.

Given that gender and sex-related data greatly impact individual care, it is critical that the health care community create data models that enhance understanding and collection of critical data such as organ inventories; culturally-specific health factors; surgical histories; hormonal histories; chromosomal makeup; interlocking demographic factors; individual experiences of discrimination (such as deadnaming or misgendering), sexual assault, and physical violence; differences in sexually transmitted diseases, cancers, cardiovascular diseases, and mental health conditions, all of which directly impact the health outcomes of transgender, intersex, and gender-diverse populations. Mixing up these data can result in situations of life or death. Studies indicate that around 40% of transgender persons in the U.S. attempt suicide in their lifetime (26,103). As former president of the World Professional Association for Transgender Health (WPATH), Jamison Green said: “By not actually addressing what kinds of variabilities exist in people’s sexual behaviors, we blind ourselves to potential solutions to these problems” (62).

Sex and Gender in Quality Measurement

Clinical quality measures traditionally evaluate performance using manually abstracted clinical and administrative data. Electronic clinical quality measures (eCQMs) evaluate performance using data extracted from electronic health records and/or digital health information technology (HIT) systems. Patient demographic information is often used simply to specify eCQM inclusion and/or exclusion criteria. As described in the Current State section, Gender and Sex are often represented in a single data element utilizing various coding standards and that inconsistency in data capture and implementation leads to downstream issues for quality measurement instruments and outcomes.

For example, the National Committee for Quality Assurance (NCQA) produces the Healthcare Effectiveness Data and Information Set (HEDIS). Approximately 191 million people are enrolled in plans that report HEDIS results, making it one of health care’s most widely used performance improvement tools. HEDIS covers six (6) domains of care utilization measures, including Child and Adolescent Well-Care Visits, Frequency of Selected Procedures, Identification of Alcohol and Other Drug Services, Mental Health Utilization, Antibiotic Utilization and Healthcare-Associated Infection (HAI) Standard Ratio. For all these measures, individuals with nonbinary gender are excluded from HEDIS utilization measures that currently require a specific gender (male or female). NCQA recognizes this as an issue and has stated that it “continues to track industry standards for nonbinary gender”. In other words, the guidance provided in this specification can help improve representation of nonbinary gender and therefore measurement.

Sex and Gender Reporting in Payment for Care

Some EHR systems have developed features that may suggest tests or workflows based on sex or gender identity data which is often inaccurate and/or inadequate to address the needs of transgender, gender-diverse, and intersex persons. For instance, a patient may need to switch their insurance “sex” for a procedure to avoid denial of coverage or to even be offered a procedure or test in the first place. Pharmacies may also have to administratively change “sex” for approvals for particular medications and then switch the “sex” back to avoid denial of coverage. In addition, providers may have to address dozens of automatically flagged lab results which are irrelevant to the patient but are nonetheless required due to compliance regulations (63).

Switching “sex” fields back and forth may trigger hundreds of new results or diagnostic warnings or messages, adding to the already significant issue of alert fatigue among medical providers. Further, clinicians may miss proper risk assessments based on whether the “correct” sex field is provided. For instance, a transgender woman who is marked as “male” may miss crucial breast cancer screenings, but a transgender woman who is marked as “female” may miss prostate cancer screenings. Only by including contextual data about gender identity, sex assigned at birth, organ inventories, hormone levels, and chromosomal makeup can these issues be sufficiently avoided.

Our recommendation is that clinical decision support should be based on specific relevant information about the patient and not sex characterizations. There are existing decision support rules that do rely on sex categorization. Where this occurs, the decision support rules should provide guidance on how to determine the proper sex characterization. In these cases a Sex Parameter for Clinical Use (SPCU) based on that guidance would be appropriate.

Sex and Gender Uses in Data Analysis

Storing and exchanging data in structured formats ensures that EHR and HIT systems are better equipped to notify health care teams of appropriate and preventive services, but this is not an end itself. It is critical to have standardized high-quality data in order to conduct data analysis to address health inequities. While there have been some scientific advancements, there continues to be a dearth of data and literature on health outcomes and experiences for transgender and gender diverse people. Many transgender and gender diverse people remain largely invisible to their care providers, face stigma, barriers to accessing care and related health disparities. Standardized data will facilitate information-sharing for clinical care, research, and public health interventions that can further reduce health care disparities in this underserved population.

There are striking disparities in accessing health services that correlate with gender identity, as well as race/ethnicity and other factors. It is important to consider the intersecting identities and experiences of transgender and gender diverse people, to understand the cumulative discrimination and health inequities this community faces. Intersectionality is a framework for describing the disparities and culminating impact a person or group of people are affected by. Sex Parameter for Clinical Use (SPCU) and gender identity with other demographic data can be used to evaluate service utilization and health outcomes for subpopulations.

Currently, data resides in disparate systems with varying degrees of accuracy, preventing information from being used to support meaningful analysis, data visualization and insight generation. The goal is to not only answer questions but advance insights that drive action. Combining self-reported data from patients with clinical observations around the notions of gender and sex allows for rapid and efficient evaluations of emerging trends with more detail than might otherwise be possible. Creating opportunities for increased data analysis can contribute to new approaches to managing care, informing policy and future interventions.