Diabetes Risk Higher for Transwomen vs Cisgender Females

Transwomen may face a higher risk for developing type 2 diabetes than cisgender women, a new study suggested.

In a review of new data from the Study of Transition Outcomes and Gender (STRONG) cohort, transwomen had a 30% higher odds of prevalent type 2 diabetes compared with the cisgender female population in an adjusted model (OR 1.3, 95% CI 1.1-1.5), reported Noreen Islam, MD, MPH, of Emory University School of Medicine in Atlanta, and colleagues.

Among those already diagnosed with type 2 diabetes at baseline, a total of 32% of transwomen were on gender-affirming hormone therapy, the group reported in the Journal of Clinical Endocrinology & Metabolism.

Transwomen also saw a 40% higher risk of developing incident type 2 diabetes during the average 3.1 years of follow-up compared with cisgender females (HR 1.4, 95% CI 1.1-1.8).

However, transwomen didn’t have any excess risk for developing diabetes when compared with cisgender men (HR 1.2, 95% CI 0.9-1.5), which the researchers said “likely reflects the known gender disparity in [type 2 diabetes] risk in the general population.”

And in an analysis restricted only to transgender and gender-diverse people receiving gender-affirming hormone therapy, transwomen didn’t see a significantly higher prevalence of type 2 diabetes (OR 1.0, 95% 0.7-1.3) nor risk for incident diabetes (HR 1.4, 95% CI 0.8-2.4) versus cisgender females. This suggests that the excess diabetes risk for this population wasn’t driven by hormonal therapy, the researchers said.

For these transwomen individuals diagnosed with incident diabetes after baseline, only 18% were diagnosed after initiation gender-affirming hormone therapy. Overall, there was a 9.3 per 1,000 person-years incident rate of type 2 diabetes among the entire transwoman population, with only a 5.9 per 1,000 person-years rate among those who initialed gender-affirming hormone therapy.

“Our study findings provide some reassurance that gender-affirming therapy does not increase the risk of type 2 diabetes, but our analysis was not designed to evaluate more subtle subclinical changes,” Islam pointed out in a statement. “For this reason, health care providers should continue monitoring the metabolic status of individuals receiving gender-affirming therapy.”

“Although more research is needed, there is little evidence that type 2 diabetes occurrence in either transgender women or transgender men is attributable to gender-affirming hormone therapy, at least in the short term,” she added.

The researchers noted that a few other studies have shown mixed findings on these potential subclinical effects of hormonal therapy. However, Islam and co-authors wrote, “[t]aken together, the published literature indicates that measures of glucose metabolism and insulin resistance do not appear to be adversely influenced by testosterone in transmasculine individuals but may be affected by estradiol therapy among transfeminine individuals.”

The team said the actual clinical significance of the reported associations is not entirely clear, and it is also possible that these small changes in laboratory parameters tied to hormonal therapy don’t actually translate into a significantly higher risk for a type 2 diabetes diagnosis.

For the current analysis, a total of 2,869 transwomen — i.e., women who were assigned male gender at birth — were matched with 28,300 cisgender women and 28,258 cisgender men based on age, race, ethnicity, year, and location. Then, 2,133 transmen — men who were assigned female gender at birth — were matched with 20,997 cisgender women and 20,964 cisgender men from the STRONG electronic health record (EHR) system of patients at three Kaiser Permanente health systems in Northern California, Southern California, and Georgia.

Prior to the index date, 32% of transwomen and 24% of transmen were on gender-affirming hormone therapy.

Study limitations, Islam and co-authors said, included that (1) unlike clinic-based cohorts, the EHR data in STRONG were not collected at specified intervals and the number and frequency of clinical encounters differed across participants; (2) the algorithm did not explicitly distinguish between type 1 and type 2 diabetes; and (3) there was a lack of data on other type 2 diabetes risk factors, including family history, socioeconomic status, and adverse childhood experiences and gender minority stress.