Invisible Bodies: Representing Gender and Gender Variance in Medical Records and Health Data
Prior to 2010, there was virtually no population-based health data on trans and gender variant populations at the federal, state, or local level in California. The population was invisible in the data. This research project took the formation of such an odd silence in the data as the motivation to form a qualitative approach to studying how gender has been encoded in medical records and health data in California. The project focused on identifying which aspects of gender were able to be recorded and what aspects were not able to be represented within the affordances observed in the recordkeeping structures. The research materials documented practices in public health and medical recordkeeping contexts at a critical historical juncture, when practices were in flux, both locally and nationally, focusing on three research sites in California, in San Francisco, Los Angeles, and Sacramento. All of the sites were directly engaged with recordkeeping questions around how gender is encoded for trans and gender variant populations, working collaboratively to develop the information structures for representing gender and gender variance in their recordkeeping systems.
The analysis produced documentation of the information structures encoding gender at each of the sites and concluded that there are six (6) major functional information recordkeeping elements conflated under the current system of binary gender markers, including elements of gender identity, social gender (pronouns), medical gender, legal gender, organ lists and sex assigned at birth. Several of these elements operate in unrelated fields of policy discourse, such as those around identity documents (legal gender) versus medical diagnoses (medical gender). These overlapping policy environments often conflict in reality, yielding uncountable populations of trans and gender variant persons attempting to access care with unresolvable incongruence between the legal, medical and social records of gender. The information disparities in the functionality to record gender variance suggests that system designers need to be more concerned with the edge effects affecting populations who are poorly represented within data structures, especially as these recordkeeping systems are scaled up to include the entire general population writ large. Many patients still withhold gender identity information from healthcare systems, which suggests that conflicts in recordkeeping practices cannot be resolved without first addressing basic safety and data security issues for trans and gender variant populations.