Prevalence of Gender Dysphoria by Gender and Age in Japan: A Population-Based Internet Survey Using the Utrecht Gender Dysphoria Scale

Abstract

ABSTRACT Background Several studies have attempted to estimate the prevalence of gender dysphoria (GD) from a general population sample. However, no previous studies used reliable questionnaires. Aim To estimate the prevalence of GD in Japan by gender and age using the Utrecht Gender Dysphoria Scale (UGDS). Methods A cross-sectional observational study was conducted with 20,000 respondents between the ages of 20 and 69 who were registered with an internet research company. The study consisted of two phases. First, the participants were asked to self-identify their gender on two 5-point Likert scales. Second, the screened participants completed the UGDS. Outcomes Self-identified gender and GD were defined as follows: ambivalent gender (equally feeling like the birth gender and another gender), incongruent gender (a stronger sense of the latter vs the former), narrow GD (incongruent gender + UGDS score ≥ 41), and broad GD (ambivalent or incongruent gender + UGDS score ≥ 41). Results Among the eligible participants, the age-adjusted proportions of those classified as male (n = 7827) and female (n = 8903) at birth were 6.0% and 5.9%, respectively, for ambivalent gender, and 0.93% and 1.0%, respectively, for incongruent gender. The age-adjusted prevalence of GD was 0.27% (95% confidence interval, 0.18–0.42) and 0.35% (95% confidence interval, 0.25–0.50) for narrow GD and 0.87% (95% confidence interval, 0.69–1.1) and 1.1% (95% confidence interval, 0.86–1.3) for broad GD, respectively. No significant gender differences were found within the age groups, except for broad GD in respondents in their 50s (P = .016). However, for both genders, significant differences were found between age groups such that GD was more prevalent in younger vs older respondents, except for broad GD in respondents classified as female at birth (P = .063). Clinical Implications Clinicians should be aware that the prevalence of GD is not negligible and that it varies with age. GD should be assessed in detail from various perspectives in addition to self-identified gender. Strengths & Limitations This study used a reliable questionnaire to examine the prevalence of GD in a large population. However, the participants did not represent the general population because this was an internet survey. Conclusion The prevalence of GD was much higher than previously estimated by clinic-based studies, and was more frequently associated with participant age vs gender.