Headache Prevalence in Transgender and Gender Diverse Youth: a Single-Center Case-Control Study

Abstract

Objective:

Assess the prevalence of headache in transgender and gender-diverse adolescents, comparing prevalence with and without exposure to gender-affirming hormone therapy.

Background:

Transgender and gender-diverse youth are an under-studied group in whom we can study effects of sex steroids on adolescents’ development of headache. We hypothesized that transfeminine adolescents treated with estrogen would have higher odds of headache than those not treated, and that transmasculine adolescents treated with testosterone would have lower odds of headache than those not treated.

Methods:

This retrospective case-control study analyzed all patients seen at the Boston Children’s Hospital Gender Multispecialty Service clinic from 2007 to 2017. Cases were defined as patients with headache, controls as those without headache, and exposure as treatment with gender-affirming hormone therapy (i.e., estrogen or testosterone). A computerized search identified cases that were then validated by chart review.

Results:

52 of the 763 transgender and gender-diverse patients seen were confirmed to have headache. Of 273 transfeminine patients 45% (123/273) received estrogen treatment. Transfeminine patients receiving estrogen were more likely to have headache than those not receiving estrogen (7% (9/123) vs. 1% (2/150); OR 5.84 (95% CI 1.24–27.6) p = 0.026). Of 490 transmasculine patients, 46% (227/490) received testosterone. Transmasculine patients receiving testosterone were more likely to have headache than those not receiving testosterone 12% (28/227) vs. 5% (13/263); OR 2.71 (95% CI 1.37–5.4), p = 0.005).

Conclusion:

Among transfeminine and transmasculine youth, those who received gender-affirming hormone therapy had higher odds of headache compared to those not taking gender-affirming hormone therapy. Further prospective studies to guide headache care of transgender and gender-diverse youth and adults are needed. Our results could be generalizable to other pediatric gender management clinics and may be worth discussing with patients considering treatment.

Introduction:

Headache is ranked as the second highest cause of years lived with disability worldwide.¹ At its peak, lifetime prevalence of this highly disabling condition is at least 20% higher in adult women than men, a sex difference widely attributed to the effects of estrogen and testosterone.^{2, 3} This higher headache prevalence in women as compared to men emerges during adolescence, a time of rapidly increasing levels of estrogen and testosterone and of sex-specific divergence in brain development.^{4 5, 6} Although the timing is suggestive, evidence of pubertal sex steroid exposure causing this development of a sex difference in headache prevalence at puberty is lacking.⁷ One group of adolescents in whom we can study the effects of sex steroid exposure at puberty is transgender and gender-diverse (TGD) youth, whose sex designated at birth differs from their gender identity (see Table 1 for terminology). TGD youth may receive gender-affirming hormone therapy (GAHT, i.e., estrogen for transfeminine individuals and testosterone for transmasculine individuals) as part of their gender-affirming care.⁸ Understanding the prevalence of headache in this group may help us to understand the known sex difference in headache prevalence found in the general population.

Table 1.

Terminology and definitions^{7, 8, 38, 39}

Current Terms	Definition
Transgender and gender-diverse (TGD) (adjective)	A person in whom gender identity and expression may differ from their sex designated at birth.
Cisgender (adjective)	A person whose gender identity is the same as their sex designated at birth
Transmasculine (adjective)	A transgender person who identifies as more masculine than feminine.
Transfeminine (adjective)	A transgender person who identifies as more feminine than masculine.
Sex designated at birth (noun)	Can be male or female, usually based on genitalia present at birth.
Gender dysphoria (noun)	Distress that is felt when a person’s gender identity is not congruent with their sex designated at birth.
Gender affirmation (noun)	Changes such as medical, social, and/or legal made to affirm gender identity.
Gender-affirming hormone therapy (GAHT) (noun)	Sex steroids taken to develop physical attributes congruent with gender identity (i.e., testosterone in transmasculine individuals and estrogen in transfeminine individuals).

During the middle years of life, estrogen appears to have a hyperalgesic effect;⁹ however, it is unknown whether the hyperalgesic effect of estrogen is secondary to estrogen withdrawal such as during menstrual cycling or secondary to estrogen exposure itself. Estrogen withdrawal, as during menses, is a known trigger for migraine in many women.^{4, 10} Transfeminine youth receiving GAHT are given the sex steroid of their gender identity, estrogen, at steady state rather than cyclically, allowing for the study of the effects of estrogen without withdrawal on the development of headache.¹¹ The TGD population is understudied.⁷ In addition to its role in addressing questions about the effects of sex steroids on headache, this study can also address the effect of GAHT on headache in TGD youth.¹²

The objective of this retrospective case-control study is to compare the prevalence of headache in TGD adolescents who did not receive GAHT to those who did receive GAHT. Although there are no previous studies of headache in TGD youth, there are available studies on headache and pain in TGD adults, sex steroid effects on headache in cisgender adults, and pre-clinical studies on the effects of sex steroids on pain.⁷ As stated above, studies seem to indicate that estrogen has a hyperalgesic effect, and testosterone has an analgesic effect.⁷ From these, we formed the hypothesis that transfeminine adolescents treated with estrogen would have higher odds of having headache than those who did not receive estrogen, and that transmasculine adolescents treated with testosterone would have lower odds of headache than those not treated with testosterone.

Methods

This study is a retrospective case-control study of patients seen at the Boston Children’s Hospital Gender Multispecialty Service (GeMS) clinic from 2007 to 2017. Cases were defined as patients with headache and controls as those without headache. The exposure was treatment with GAHT. Patients with headache were determined via diagnostic codes for headache or migraine or prescriptions for sumatriptan or amitriptyline. JAH then performed a retrospective review of medical documentation to validate the mention of headache for these cases using full text chart search for these terms; see Figure 1 for flow diagram of chart review. Patients whose headache was not confirmed by chart review were re-allocated to controls. Patients not treated with GAHT or who were not transgender or gender-diverse were excluded from the study based on chart review. Study data were collected and managed using REDCap (Research Electronic Data Capture) electronic data capture tools.^{13, 14} GAHT regimen was per standard protocol which varied over this 10-year period and in general adhered to The World Professional Association for Transgender Health guidelines.¹⁵ Descriptive statistics and measures of association were calculated for all study measures. Approval was obtained from the Institutional Review Board of Boston Children’s Hospital and did not require written consent.

Figure 1.

Study flow diagram. Tint shows all patients with headache; identification made by automated review of electronic medical record search for diagnostic codes for headache or migraine as well as full text search for those terms and sumatriptan or amitriptyline.

Statistical Analyses

This is a primary analysis of these data based on the a-priori hypothesis given in the Introduction section and the data in this study have not been previously reported elsewhere. The criterion for statistical significance of p < 0.05 was chosen. No statistical power calculation was conducted prior to the study and due to the rarity of the population, the data represent a convenience sample. Given the small number of cases identified in the study population, the 2-sided Fisher Exact test was used to test the association between patients on GAHT or not on GAHT and the diagnosis of headache. Odds ratios and confidence intervals were calculated with Woolf exact method. Frequency of gender, frequency of treatment with estrogen or testosterone, and frequency of headache were also calculated. All analyses were performed with STATA/IC 15.1 (StataCorp LLC, College Station, TX).

Data Availability

Anonymized data not published within this article will be made available after publication and for up to 4 years after publication. The authors will share the data with qualified investigators whose proposal of data use has been approved by an independent review committee.

Results:

766 patients were seen in GeMS during the study period (Figure 1). Three of these patients were excluded (1 with panhypopituitarism, 1 who was not transgender or gender-diverse, and 1 who did not receive estrogen or testosterone), leaving 763 patients analyzed for this study. Of these, 63 were initially identified as having headache, of whom 52 were confirmed to have headache via chart review.

In total, of the 763 adolescents analyzed for this study, 273 (36%) were transfeminine and 490 (64%) were transmasculine. Of the transfeminine patients, 45% (123/273) received estrogen. Seven percent (9/123) of the transfeminine adolescents who received GAHT had headache compared to 1% (2/150) of the transfeminine adolescents who did not receive GAHT; odds ratio of 5.8 (95% CI 1.24–27.6) p = 0.026 (Table 2).

Table 2.

Association between gender-affirming hormone therapy and headache status among 273 transfeminine adolescents and 490 transmasculine adolescents.

Group:	with headache	without headache	overall sample	Odds Ratio
Transfeminine, n (%)
treated with estrogen	9 (7)	114 (93)	123 (45)	5.8 (1.24–27.6)*
untreated	2 (1)	148 (99)	150 (55)
Total	11 (4)	262 (96)	273 (100)
Transmasculine n (%)
treated with testosterone	28 (12)	199 (88)	227 (46)	2.71 (1.37–5.4)**
untreated	13 (5)	250 (95)	263 (54)
Total	41 (8)	449 (92)	490 (100)

95% confidence interval Woolf exact method, significance 2‐sided Fisher’s exact test.

^*p-value is 0.026

^**p-value is 0.005.

Of the transmasculine patients, 46% (227/490) were treated with testosterone. Of those transmasculine adolescents treated with testosterone, 12% (28/227) were confirmed to have headache compared to 5% (13/263) of the transmasculine adolescents not treated with testosterone; odds ratio of 2.71 (95% CI 1.37–5.4), p = 0.005.

The majority of the instances of headache confirmed in charts were non-specific. For example, mention in-text in a non-headache provider’s note that patient was experiencing headache, or a positive item on a review of systems, without further detail. Those patients who were given a specific diagnosis by a provider or whose chart included suggestive information toward a secondary cause included: familial hemiplegic migraine, probable migraine with aura, chronic daily headache, concussion, “non-migraine”, vestibular migraine, and headache type unspecified by provider but noted on JAH’s chart review to occur after patient experienced whiplash.

Discussion:

This single-center case-control study showed higher odds of headache in both transfeminine and transmasculine youth receiving GAHT than in those who did not receive GAHT. The result that transmasculine youth receiving testosterone had higher odds of headache than those who did not differed from our hypothesis for this group. Although not part of our primary analysis, transmasculine youth also appeared to have a higher prevalence of headache than the transfeminine youth in our study. There are no prior studies of headache in TGD youth, although studies of TGD adults are available.⁷ With respect to studies on the effects of estrogen, a study of transfeminine adults in the Netherlands found that post-GAHT migraine prevalence was similar to the population prevalence in cisgender Dutch women (26% in both groups), but higher than the prevalence in cisgender Dutch men (7.5%).¹⁶ One caveat is that European GAHT has historically been more likely to include estrogen given at higher doses and in a different formulation, and may include progesterone.¹⁷ Another small study examining pain conditions in adult TGD patients found that most transfeminine adults with chronic headache had either developed headache or had a worsening of headache after initiation of GAHT.¹⁸ In cisgender men, a diagnosis of migraine has been associated with elevated estrogen levels.⁹ With respect to studies on the effects of testosterone, the same small study of pain conditions showed that adult transmasculine patients with chronic headache reported that headache pain improved after initiation of testosterone therapy, though their rate of headache was higher (50%) than in transfeminine patients (11%), consistent with the higher rates of headache in transmasculine patients than transfeminine patients in our study.¹⁸ Additionally, treatment of adult cisgender women with testosterone resulted in improvement or remission of headache.^{19, 20} Animal studies also showed a sex difference in pain response, with female animals having a greater pain response than male animals, and an analgesic effect of testosterone.^{21 22} Finally, for comparison with our own findings, the overall rate of headache in a Turkish study of adult transmasculine patients (including with and without exposure to GAHT) was 77%.²³

One explanation for the higher odds of headache in transmasculine patients who received GAHT in our results could be that the initiation of GAHT in adolescence has a different effect than GAHT initiated in adulthood. During puberty, sex steroids such as estrogen and testosterone appear to have an “organizing” effect on brain structure and function, causing permanent and lasting changes in brain function at key windows of development.^{24–27 28, 29} In rodent studies, an organizing and persistent effect on neural tissue and pain behavior has been demonstrated when manipulation of the hormonal milieu occurs prior to puberty.³⁰ Puberty is a critical period when sex specific divergence in brain development occurs and correlates with pubertal development rather than age.⁵ Thus, the effect of GAHT in TGD adults who have already completed this critical stage under the influence of endogenous sex steroids may be different than in TGD adolescents. This in turn may make a functional difference in pain and could explain the difference between adult transgender headache data and our findings in adolescents.

The benefits of research on headache in TGD youth are two-fold. First, there is an intrinsic importance in adding to our knowledge about the effects of GAHT in this under-studied population.⁷ TGD patients themselves desire more research in this area; a Patient Centered Outcomes Research Institute-funded study showed that TGD individuals want to know more about the incidence of migraine during GAHT.³¹ Further, the NIH has made research in this area a priority through the establishment of the Sexual and Gender Minority Research Office in 2015.³² The second benefit of this research is its potential to add to our understanding of the effects of sex steroids on headache in cisgender patients. Although studies in TGD patients are observational, GAHT in many ways resembles the interventional pre-clinical studies on the effects of testosterone and estrogen exposure on pain. This is key in understanding how results in a clinical population may be different from or similar to pre-clinical studies showing hyperalgesic effects of estrogen and analgesic effects of testosterone, particularly their effects when exposure occurs pre- as opposed to post-pubertally.

Limitations to our study include its reliance on incidental reports of headache in medical records. As a result, headache is not determined according to ICHD-3 criteria and there is a heterogeneity of headache type and severity reported. Further, the population prevalence in our study likely under-reports the prevalence in comparison to adolescent cisgender epidemiologic studies, which rely on questionnaires, surveys, and face-to face-interviews, and report prevalence in adolescents ages 10–20 years old from 8% to 79%.³³ In the clinic population studied from this regional specialty clinic, adolescents who received GAHT were subsequently followed by the GeMS clinic over time and so tended to have more extensive chart coverage with more opportunities for headache to be noted, leading to a risk of both transfeminine and transmasculine adolescents who received GAHT having more documentation of headache. Patients receiving GAHT may also be self-monitoring for or asked about the potential side effects of GAHT, which may also lead to over-reporting and over-documentation of headache symptoms. Another limitation attributable to reliance on the incidental report of headache is the difficulty determining the true timing of development of headache for comparison to initiation of GAHT. Similarly, age was not controlled for in this study. Further, patients receiving GAHT received varying GAHT protocols and duration of treatment. Finally, the overall number of patients receiving GAHT who had incidental report of headache was small in both the transmasculine and transfeminine adolescent populations and further research on a larger population of youth on GAHT with a headache diagnosis is warranted to confirm our findings. A strength of our study is our selection of TGD adolescents who had not received GAHT as our control group. This was done to control for the higher rates of anxiety and depression in TGD youth than in the general population.³⁴ There has been a call for this study design for neurodevelopmental studies of the effects of pubertal suppression in a recent consensus paper of international experts.²⁵ Another strength is that the population presented here is similar in its distribution of transmasculine and transfeminine youth to another large North American cohort from this time period.³⁵ Another strength is its use of manual retrospective chart review; although natural language processing is on the rise, manual chart review typically remains the gold standard for validation of these methods and of reviews based on medical coding.^{36, 37}

Our findings suggest an association between GAHT and increased odds of headache in both transmasculine and transfeminine youth. As the number of adolescents treated with GAHT increases, further prospective studies are needed to evaluate and quantify the magnitude of any association between GAHT and headache, the directionality of its effect, and whether it is attributable to treatment or comorbidities, or is reflective of ascertainment bias. Author JAH has been funded to study effects of GAHT on headache in transfeminine adolescents in a prospective observational trial involving a greater number of cases and improved ascertainment of headache which will ideally begin to address these questions. Given our selection of transgender and gender-diverse patients for our control group, the GeMS Clinic’s general adherence to The World Professional Association for Transgender Health guidelines, and the comparable makeup of this clinic’s gender balance to that of others, our results could be generalizable to other pediatric gender management clinics and may be worth discussing with patients considering GAHT.

Abbreviations:

TGD

transgender and gender diverse

GAHT

gender affirming hormone therapy

GeMS

Boston Children’s Hospital Gender Multispecialty Service