Charna Albert
November 2026—The first broad set of recommendations for gender-diverse patient care, endorsed by the ADLM and published in July, spell out how to interpret routine testing in this population—and go far beyond. They cover not only laboratory medicine but also anatomic pathology and informatics (Thomas TA, et al. J Appl Lab Med. 2026;11[4]:887–926).
Gender-affirming hormone therapy affects a relatively small group of laboratory tests, and decades of research have shown the test results can be interpreted in accordance with the patient’s affirmed gender, with a few exceptions. Tiffany Thomas, PhD, coauthor of the guidance and assistant professor in the Department of Pathology and Cell Biology at Columbia University, uses a flowchart-like way to describe the logic: “Is somebody on gender-affirming hormone therapy [GAHT]? If yes, know what few lab tests are affected by that. If it’s based on the affirmed gender, then you generally go with that.”
Simple enough, but go deeper into laboratory processes and there are complex decisions to be made, some of which don’t have a best practice to follow.
Take informatics, for example. A patient’s legal sex is recorded in the electronic health record and drives the reference ranges that populate for tests with sex-specific ranges. A patient who identifies as transgender might decide to change their legal sex, if local policy allows. If they do, nothing in the system will indicate the patient is transgender, unless the EHR includes sexual orientation and gender identity (SOGI) fields for sex assigned at birth, pronouns, gender identity, and other contextual information. Even so, SOGI fields won’t indicate if the patient is or is not on GAHT. And for patients not listed as either sex, reference intervals must be chosen.
Many of the pathology- and laboratory medicine-specific recommendations offered in the guidance have been published previously, but having all recommendations in one place, its authors hope, will help their colleagues put them into action.
“It’s so important to have this unified document that provides the guidance so that folks don’t have to approach these problems on their own,” says Gabrielle Winston-McPherson, PhD, D(ABCC), a coauthor and assistant professor in the Department of Pathology and Laboratory Medicine and director of clinical chemistry at the University of Wisconsin School of Medicine.
Dr. Thomas says it doesn’t all have to happen at the same time. “It can be an iterative process, and we don’t have to fix this complicated issue tomorrow,” she says. “But we should be moving toward it.”
The hematologic parameters are the prime example of a group of tests that undergo clear and clinically significant changes with GAHT. “The values absolutely shift, and you need to use the reference interval consistent with the affirmed gender” so as not to risk misinterpretation, Dr. Winston-McPherson says.
Testosterone therapy increases hemoglobin, hematocrit, and RBC counts to values that align with those of cisgender men, and the reference intervals for cisgender men should be used to interpret these tests. For people on feminizing estradiol therapy, testosterone suppression leads to lower hemoglobin, hematocrit, and RBC counts, and reference intervals for cisgender women can be used to interpret results.
For some other analytes, “we may see statistically significant shifts that are not clinically significant, where the interpretation is a bit more nuanced,” she says. Lipoprotein concentrations, for example, often shift in the direction of the affirmed gender, with estrogen therapy in transgender women potentially increasing HDL-C and slightly decreasing LDL cholesterol and triglycerides, and the opposite occurring in transgender men on testosterone. But clinical management should focus on overall cardiovascular risk and target lipid concentrations, rather than sex-based reference intervals. “You need to look at that patient holistically to interpret correctly,” Dr. Winston-McPherson says.
The sex hormones are another example of a clear shift. Masculinizing hormone therapy increases testosterone concentration, and published literature supports using cisgender male reference intervals for transgender men. One important note: In transgender men, estradiol concentrations show a higher upper reference limit than those derived from cisgender men. “Sometimes testosterone doesn’t completely suppress estradiol into the male reference range in transgender men,” Dr. Thomas says.

The guidance breaks with the Endocrine Society’s 2017 recommendation to maintain estradiol in a range of 100 to 200 pg/mL in transgender women on feminizing therapy. Nearly 10 years later, “there’s no evidence to support that [range],” says Dina Greene, PhD, D(ABCC), a coauthor and clinical associate professor in the Department of Pathology and Laboratory Medicine at the University of Washington. She and her coauthors cite studies supporting a wider range (Winston-McPherson GN, et al. LGBT Health. 2025;12[7]:477–489).
In addition, they say estradiol concentrations must be interpreted in the context of therapeutic regimen and other preanalytical factors, including the individualized goals of feminizing therapy. “Some women will want a lower concentration, and some will want a higher concentration,” Dr. Greene says. There’s no evidence to show negative outcomes for either, “as long as people are taking some amount of hormone, particularly agonadal people,” who are at risk for bone deterioration and other health problems in the absence of sex hormones.
The Endocrine Society’s 2017 guidelines say that for transgender women, total testosterone concentrations should be suppressed to less than 50 ng/dL. But antiandrogen therapy and history of orchiectomy are important factors in interpretation, Dr. Thomas says. For patients on spironolactone, an androgen receptor antagonist, “you can have a testosterone level that’s elevated and still have appropriate feminization of these patients because it’s blocking the activity of testosterone itself,” rather than suppressing testosterone secretion. Instead of increasing spironolactone, which is not a benign drug, the estradiol level can be increased slightly if the patient is not hitting their feminization goals, she says.
For some tests with sex-specific components, there is no consensus on whether to use sex assigned at birth or affirmed gender. For example, recommended equations for estimating glomerular filtration rate (eGFR) use sex-specific coefficients, so selecting male or female changes the calculated eGFR. For patients receiving GAHT, however, it remains unclear which selection produces the more accurate estimate.
For the creatinine-based eGFR equation, physicians may not be aware of the magnitude of the sex variable, Dr. Thomas says. “It changes estimated GFR by about 30 percent, so the time it takes for somebody to get on the transplant list could be a difference of six months, depending on which sex modifier one uses,” she says.
When the creatinine-based equation is used, the guidance recommends calculating eGFR twice—once with each sex modifier—and dual reporting the values, permitting physicians to observe GAHT-related changes and allow flagging of abnormal eGFR regardless of sex. “It’s complicated informatically because most systems only allow you to use one sex modifier,” and allow only for the equation to be reported once, Dr. Thomas says. “If you can only choose one, I would choose to use female as a modifier for both transgender males and females,” she says. Better than the creatinine equation would be to use a cystatin C-based equation, she adds, in which the sex modifier has only about a seven percent bias.
Algorithms for assessing acute myocardial injury have similar unknowns.
Although GAHT affects basal levels of cardiac troponin, its clinical significance for risk assessment of AMI or future events is unclear. No prospective clinical studies have evaluated the influence of sex-specific 99th percentile URLs on long-term cardiac outcomes in patients on GAHT, according to the authors of the guideline. They emphasize the importance of serial measurements of high-sensitivity cardiac troponin for patients on GAHT, and provisionally recommend the use of the 99th percentile that aligns with affirmed gender if the institution uses sex-specific hs-cTn URLs.
What’s to be done when flagging and alerts in the EHR are based on sex-specific URLs?
Dr. Winston-McPherson’s institution uses the male upper reference limit for all patients with a legal sex designated as male. “Everybody else gets the female upper limit of normal, which is lower,” she says. Some patients have their legal sex listed as nonbinary and others have an unknown sex marker, and “using a lower upper limit of normal in these instances is how we err on the side of safety,” she says.
For nonbinary patients in particular, “we said, ‘It’s not appropriate to ever release results with no reference interval or no upper limit of normal and not have any flagging, so how are we going to resolve that?’”
“This population deserves high-quality care,” she adds, “and we have to do our best even though we may not have a solid recommendation for what to do yet.”
Given the complex informatics considerations at play, the document’s authors expect a sizable number of questions on IT.
Michelle Stoffel, MD, PhD, a coauthor and associate chief medical informatics officer for laboratory medicine and pathology at M Health Fairview in Minnesota, names a few of the common questions she gets from others: How do software systems handle sex and gender demographic data? How does legal sex change reference interval assignment? How do EHRs, laboratory information systems, and middleware systems communicate SOGI data, and how does this information influence downstream decision-making?
“The most important thing I hope readers take away from the ADLM guidance,” she says, “is to ensure they know the capabilities of their LIS in the context of their EHR.”
Not every system handles SOGI data in the same way, or at all. And for those that do, whether the laboratory can leverage those capabilities depends on the institution and the way it has built out its systems. There can also be mismatches between the capabilities of the EHR and the LIS, Dr. Winston-McPherson says. “It gets complicated when you start to peel back the layers of that onion, which makes it difficult for us to make a single recommendation,” she says.
Dr. Stoffel says the guidance provides a framework for the readers to become aware, if they are not already, of what capabilities their own systems have and to define what they need. “Starting with a clear understanding of your system’s boundaries is the first step to applying best practices guidance,” she says.
For example, creating custom gender identifiers to drive reference interval logic can significantly alter automated reference range assignments and clinical decision support. “If your system relies on additional metadata, comments, or labels to be able to accurately interpret custom reference intervals or results values populated in reports or trendline views, it may be safer in some cases to not calculate an interval or a value and instead use an interpretive comment to call out the need for clinical guidance,” she says.
With more patients accessing and reviewing their laboratory data than previously, she says, “designing reports with both the patient and their health team in mind is thankfully getting more attention, and this guidance document is intended to support that patient-focused approach.”
That means, for one thing, ensuring the interpretive comments are respectful, Dr. Thomas says. When a comment outs a patient as transgender, “they may feel othered,” she explains. Or if a transgender woman sees she’s been given a male reference range for a test, she says, “that’s enough for a lot of people to not go back to that health care practitioner, unless that is their only option.”
SOGI fields offer advantages, such as personalizing reference ranges for patients on GAHT, but they also introduce privacy and confidentiality concerns and the risk of clinical misinterpretation if the data fails to reflect surgical history or hormone therapy status. And it’s important to be aware that SOGI demographic data can typically be modified directly within the EHR by patients, clinical staff, or health system providers. “Updates to these fields propagate across clinical decision support tools and automated workflows in ways that may not be immediately visible to care teams,” Dr. Stoffel cautions. Updating a transgender man’s legal sex to male may suppress automated clinical reminders for routine cervical cancer screenings or procedure-related pregnancy testing. “Conversely, updating a transgender woman’s sex to female can lead to changed baseline reference ranges for serum creatinine and eGFR or suppressing prostate cancer screening alerts.” The laboratory, she says, may want to use alerts or other actions to call attention to a change in SOGI data.
Gender-care–related policy differences from state to state can cause patients anxiety about how their data will be used, and thus “transparent, intuitive processes for patients to review their health records and request needed updates are essential,” she says.
A uniform protocol for collecting SOGI information would help, Dr. Thomas says. Often the fields are filled out when someone first enters the hospital system and then never updated. Sometimes they’re updated without consent, she says, and sometimes the patients themselves update them. In addition, patients may not know Epic has portability between hospital systems. “If you opt into some of the SOGI fields, they can then populate into other hospital systems where you potentially might have bias by being outed as a transgender person.”
Importantly, not all transgender and gender-diverse patients elect to have GAHT, and SOGI fields don’t capture the data. “I think that’s one of the biggest weaknesses that we have informatically with treating transgender patients,” she says.
Current EHR systems, Dr. Stoffel says, lack standardized tooling to track and visualize demographic status transitions over time. “Because updating demographic identifiers can overwrite baseline values in some systems, clinicians can lack clear historical context.” And though summary pages, if they exist, may contain previous legal and chosen names, data entry is nonstandardized. A visible and intuitive EHR/LIS “storyline” of a patient’s identity and treatment journey would support more equitable care, she says.
One feature that can help keep track of patient transition over time is an organ inventory—a table or checklist within the EHR that documents the presence or absence of specific organs and dates of surgical procedures, reflecting the patient’s current anatomy regardless of legal sex designation or gender identity. It can also help the care team select and keep track of the appropriate cancer screenings for such patients.
A transgender woman who has had breast augmentation, for example, needs regular mammograms. Patients who have had sigmoid colon vaginoplasty should have colon-specific screenings every 10 years. Cervical cancer screening rates in the U.S. are low in transgender men with cervixes, who have a high rate of unsatisfactory Pap tests and may have hormone-related tissue atrophy, causing difficulties in interpreting abnormal cytology related to HPV (primary HPV screening with self-collection protocols is a good option for this population).
“This is probably one of the most useful tools for health care practitioners,” Dr. Thomas says of the organ inventory. “It helps alleviate some of the awkward conversations about which screenings should be done.”
Prostate cancer screening for transgender women, who retain their prostates after orchiectomy and vaginoplasty, should follow the local guidelines for cisgender men. Evidence suggests, however, that transgender women on GAHT have lower baseline PSA values than cisgender men, and thus a PSA cutoff of 4.0 ng/mL may not be sufficiently sensitive.
Testosterone suppression diminishes prostate size, Dr. Thomas says. “You have prostatic atrophy of that tissue, so the threshold for action is probably lower. Having a baseline and a delta is probably the appropriate approach,” she adds, until the threshold level is better understood. In Europe, she notes, for the past 15-plus years, a research group has been collecting longitudinal data in transgender women, though it’s slow work. “The signal is going to be small because the number of [transgender] patients that develop prostate cancer will be exceedingly small.”
In transfusion medicine, information about sex assigned at birth can be critical.
Dr. Winston-McPherson recounts a case example in which a transgender man whose legal sex was male had retained reproductive capacity (Mays JA, et al. Transfusion. 2018;58[3]:823–825). “The lab was going to deploy O-positive units in this case because the sex was appropriately listed as male. There was no information to indicate this was a transgender man,” she says, and therefore no reason for the laboratory to administer the O-negative units that would normally be given to a patient with an unknown ABO blood type and the potential for pregnancy.
Says Dr. Thomas, “From the transfusion standpoint, they would like to have the sex assigned at birth on the label itself because it’s useful for them.” For transgender patients, “it’s potentially extremely harmful and could cause mistrust.”
Better and more appropriate documentation—most blood establishment information systems cannot separately record gender identity and sex assigned at birth, for example—would help balance the needs of patients with those of the health care team in transfusion settings.
In providing optimal care to all patient populations including the gender diverse, there’s much to bear in mind and many unanswered questions (the authors note several), among them: How should lipid markers and other data to estimate cardiovascular risk be used in the population on GAHT? For now, it’s helpful to recall Dr. Thomas’ assurance that not everything needs to be fixed tomorrow.
She offers other advice, too: “Kindness and respect go a long way.”
Charna Albert is CAP TODAY senior editor.