EVIDENCE LIBRARY
How Can Gender-Affirming Hormone Treatment Affect a Calculated BAC?
A 2022 modeling study examined body-water assumptions in alcohol calculations after hormone treatment. Its proposed equations were estimates, not validated individual BAC measurements.
Hormone-related changes in body composition can affect the assumptions used to calculate an alcohol concentration from a drinking history. A 2022 forensic modeling study found that choosing different total-body-water equations could materially change estimated BAC for transgender people undergoing gender-affirming hormone treatment. It did not measure alcohol pharmacokinetics in a new group of transgender volunteers or validate a person-specific correction.
The central DUI question is about the basis of a calculation: does the selected equation reasonably describe the individual, and how uncertain is the resulting estimate? It is separate from interpreting a directly measured ethanol result.
Why total body water enters the calculation
A dose-based alcohol calculation distributes an assumed amount of absorbed ethanol across an estimated body-water volume and accounts for the water fraction of blood. Holding other assumptions constant, a smaller estimated distribution volume produces a larger calculated concentration.
Gender identity, sex assigned at birth and measured physiological characteristics are distinct information. An equation label is not a direct measurement of someone’s body water. Replacing one label with another without examining the model can change the numerical answer without making it more reliable.
What Maskell and colleagues actually analyzed
The researchers combined previously published body-composition and hematocrit findings during hormone treatment with existing body-water data from 582 cisgender men and 884 cisgender women. They applied estimated average changes to construct modeled body-water values and compared conventional and revised anthropometric calculations.
The model used approximately a 3% decrease in total body water for transgender women and a 10% increase for transgender men based on selected findings around 12 months of treatment. These were population-based assumptions derived from lean-mass changes, not measured changes for every individual receiving treatment.
The modeled blood-water fraction was also inferred from published hematocrit relationships. The authors explicitly noted the absence of direct validation of these blood-water estimates in the relevant transgender populations.
How much could equation choice change the answer?
The paper illustrates calculations for assumed ethanol doses of 16 and 80 grams. In its 80-gram example for transgender women, mean hypothetical time-zero concentrations were approximately 188 mg/100 mL using the conventional female equation and 156 mg/100 mL using the proposed revised model. For transgender men, the conventional female calculation produced approximately 208 mg/100 mL versus 182 mg/100 mL in the revised model.
Those comparisons describe constructed averages under the study assumptions. They are not observed BACs from people who consumed those doses. The difference between two calculated means also does not provide an individual prediction interval or prove that the revised value is correct.
Hypothetical time zero is not the time of driving
The study’s concentration term represents a hypothetical value before ethanol metabolism, rather than a measured concentration at an actual driving event. Moving from that model output to an incident-time estimate requires further assumptions about absorption, drinking duration, timing and elimination.
Whole-blood ethanol measurement, serum/plasma alcohol, breath readings, urine and biomarkers each have their own analytical and interpretive questions. The body-water model does not turn any of those results into another specimen type, and it does not establish behavioral impairment.
Why the proposed revisions need caution
The authors state that their proposed equations were not validated. The assumptions largely reflected average changes after approximately 12 months of treatment in populations that did not represent every age, body size, ancestry, treatment regimen or treatment duration.
The study did not establish personal elimination rates, experimentally measured absorption curves or an exact treatment effect on BAC for a particular person. It also did not establish that all people sharing an identity or treatment history have the same body composition. The model’s purpose was to expose potentially consequential assumptions and the need for better validation.
What should a DUI expert disclose?
A calculation should distinguish documented inputs from assumptions: the drinking history, amount of ethanol, body measurements, equation, relevant treatment information if available, and the basis for any adjustment. Competing plausible assumptions should be evaluated rather than concealed by a single precise-looking result.
This is a specific example of the broader problem addressed in retrograde BAC estimation and uncertainty: a modeled answer remains conditional on its inputs. This paper supports transparent physiological assumptions; it does not establish a universally valid correction, a legal disclosure obligation or a DUI conclusion.
Primary research source
Maskell PD, Sang KJC, Heymsfield SB, Shapses S, Dekorompay A. Forensic alcohol calculations in transgender individuals undergoing gender-affirming hormonal treatment. Journal of Forensic Sciences. 2022;67:1624–1631. https://doi.org/10.1111/1556-4029.15052.