Sex trouble: Sex/gender slippage, sex confusion, and sex obsession in machine learning using electronic health records

Conducting inclusive research in genetics for transgender, gender-diverse, and sex-diverse individuals: Case analyses and recommendations from a clinical genomics study

A person's phenotypic sex (i.e., endogenous expression of primary, secondary, and endocrinological sex characteristics) can impact crucial aspects of genetic assessment and resulting clinical care recommendations. In studies with genetics components, it is critical to collect phenotypic sex, information about current organ/tissue inventory and hormonal milieu, and gender identity. If researchers do not carefully construct data models, transgender, gender diverse, and sex diverse (TGSD) individuals may be given inappropriate care recommendations and/or be subjected to misgendering, inflicting medical and psychosocial harms. The recognized need for an inclusive care experience should not be limited to clinical practice but should extend to the research setting, where researchers must build an inclusive experience for TGSD participants. Here, we review three TGSD participants in the Family History and Cancer Risk Study (FOREST) to critically evaluate sex- and gender-related survey measures and associated data models in a study seeking to identify patients at risk for hereditary cancer syndromes. Furthermore, we leverage these participants' responses to sex- and gender identity-related questions in FOREST to inform needed changes to the FOREST data model and to make recommendations for TGSD-inclusive genetics research design, data models, and processes.

Improving radiology information systems for inclusivity of transgender and gender-diverse patients: what are the problems and what are the solutions? A systematic review

INTRODUCTION

In medical radiation science (MRS), radiology information systems (RISs) record patient information such as name, gender and birthdate. The purpose of RISs is to ensure the safety and well-being of patients by recording patient data accurately. However, not all RISs appropriately capture gender, sex or other related information of transgender and gender-diverse (TGD) patients, resulting in non-inclusive and discriminatory care. This review synthesises the research surrounding the limitations of RISs preventing inclusivity and the features required to support inclusivity and improve health outcomes.

METHODS

Studies were retrieved from three electronic databases (Scopus, PubMed and Embase). A quality assessment was performed using the Johns Hopkins Nursing Evidence-Based Practice Research and Non-Research Evidence Appraisal Tools. A thematic analysis approach was used to synthesise the included articles.

RESULTS

Eighteen articles were included based on the predetermined eligibility criteria. The pool of studies included in this review comprised primarily of non-research evidence and reflected the infancy of this research field and the need for further empirical evidence. The key findings of this review emphasise how current systems do not record the patient's name and pronouns appropriately, conflate sex and gender and treat sex and gender as a binary concept.

CONCLUSION

For current systems to facilitate inclusivity, they must implement more comprehensive information and data models incorporating sex and gender and be more flexible to accommodate the transient and fluid nature of gender. However, implementation of these recommendations is not without challenges. Additionally, further research focused on RISs is required to address the unique challenges MRS settings present to TGD patients.

Queering genomics: How cisnormativity undermines genomic science

Over the past century, genetics and genomics ("genomics") have contributed significantly to our knowledge of human biology and disease. Genomics has also bolstered inaccurate and harmful arguments about "essential" differences between socially defined groups. These purported differences have reinforced class hierarchies and justified the mistreatment of groups such as Black people, Indigenous people, and other people of color and/or people with disabilities. With this history in mind, we explore how genomics is used to reinforce scientifically unsound understandings of the relationship between two fundamental aspects of the human experience: sex and gender. We argue that imprecise, inaccurate practices for collecting data and conducting genomic research have adversely influenced genomic science and can contribute to the stigmatization of people whose sex and/or gender challenge binary expectations. The results have been to preclude transgender and intersex people from accessing high-quality, evidence-based healthcare and to hinder their participation in scientifically sound research. In this perspective, we use the lens of queer theory to render this situation more visible. First, we highlight the theoretical contributions queer theory can make to genomic science. Second, we examine practices in research and clinical genomics that exclude and stigmatize transgender and intersex people. Third, we highlight the ways that many current genomic research practices generate false conclusions that are used to support unjust public policies. We conclude by recommending ways that clinicians and researchers can-and should-harness the scientific, social, and cultural power of genomics to advance knowledge and improve lives across the spectra of sex and gender.

Two researchers share how their cross disciplinary collaboration enables work to guide the future of data science

In their recent perspective published in Patterns, Maggie Delano and Kendra Albert highlight the limitations of sex and gender data classification in health systems and show how this contributes to the marginalization of trans and non-binary individuals. They provide recommendations to improve incorporating gender data into healthcare algorithms. Here they discuss their collaboration and how it enabled this cross-disciplinary research.