Fracture Incidence and Long-Term Fracture Risk in Transgender Women on Feminizing Hormone Therapy

What the evidence says

The clearest signal in the sources is that fracture risk in trans women on feminizing hormone therapy (HT) depends on both age and which reference group is used for comparison. In trans women younger than 50, fracture incidence shows no measured difference from age-matched reference men (2.4% vs 3.0%, OR 0.78, 95% CI 0.51–1.19). The same under-50 cohort tends toward a higher fracture rate than age-matched reference women (1.6%, OR 1.49, 95% CI 0.96–2.32), though the confidence interval crosses 1.0, so this is a trend rather than a confirmed effect. From age 50 onward, the picture changes: fracture risk increases in trans women compared with age-matched reference men (4.4% vs 2.4%, OR 1.90, 95% CI 1.32–2.74), a confidence interval that does not cross 1.0. Against age-matched reference women, the ≥50 cohort again shows no measured difference (4.2%, OR 1.05, 95% CI 0.75–1.49).

One source addresses mechanism directly: the increased fracture risk observed in trans women is attributed to low bone mineral density (BMD) present before hormone treatment starts, not to an effect of the hormonal treatment itself. This is offered as an explanation for the age-related pattern above, though it is described as author interpretation built on prior findings about BMD trajectories, not a separately measured causal test in this claim set.

Beyond incidence rates, the composition of fractures also differs. Trans women on feminizing HT have a higher proportion of osteoporotic-type fractures — hip, spine, forearm, or humerus — than age-matched reference men: 41.8% of fractures in trans women fell into this category compared with 26.6% in reference men (p = 0.014). This is a distinct outcome from overall fracture incidence: it says something about which bones fracture, not how often fracture happens overall.

A separate review-level source characterizes the broader literature on estrogen-based and testosterone-based gender-affirming hormone therapy as having a "neutral to positive effect on bone health and fracture risk" in transgender and gender-diverse adults generally. This claim carries lower confidence in both direction and magnitude than the cohort data above, and the source claim itself notes it provides no specifics on sample sizes, follow-up duration, or effect estimates.

No claim in this set addresses fracture risk beyond these two age bands (under 50 and 50-plus), fracture risk in trans women not on feminizing HT, dose or duration of HT as a continuous variable, or fracture risk trajectories over time within the same individuals.

Where the evidence splits

The cohort data (Wiepjes 2019) find a measured increase in fracture odds specifically in trans women aged 50 and older compared with reference men (OR 1.90, CI excludes 1.0), while the review-level source (Jacobson 2025) characterizes the overall literature on estrogen-based HT and fracture risk as "neutral to positive." These are not claims about the same comparison — one is a specific age- and reference-group-stratified odds ratio, the other a general summary offered without sample size, follow-up duration, or effect estimates — and the claims here do not state what reconciles the two. The split is unexplained by the sources on this page.

Thresholds and dose-response

Age 50 marks a threshold in the cohort data: in trans women under 50, fracture risk shows no measured difference from reference men (OR 0.78); at age 50 and older, fracture risk increases relative to reference men (OR 1.90, CI 1.32–2.74). No claim states a threshold for hormone dose, duration of treatment, or degree of BMD loss at which fracture risk changes.

Open empirical questions

• What accounts for the age-50 threshold in fracture risk relative to reference men — is it cumulative HT duration, chronological aging, or something else — given the claims only describe the pattern, not its cause?
• Does fracture risk relative to reference men versus reference women converge or diverge further beyond the two age bands measured here (under 50, 50 and older)?
• What specific studies, sample sizes, and effect estimates underlie the Jacobson 2025 review's "neutral to positive" characterization, and do they include the Wiepjes 2019 cohort or independent data?
• Does the higher proportion of osteoporotic-type fractures in trans women correspond to a higher absolute rate of those fracture types, or only a different distribution among fractures that occur regardless?