144 Chapter 8 METHODS Study population Transgirls who were referred to start treatment with GnRHa between June 2011 and August 2017 at the gender identity clinic Curium-Leiden University Medical Centre were eligible. Those who declined participation were excluded. The study population consisted of 35 transgirls. Before the start of treatment with GnRHa, all adolescents had a diagnostic evaluation to confirm the diagnosis gender dysphoria according to the DSM-5 criteria (Hembree et al., 2017) and to assess the presence of any psychiatric or psychosocial problems that might interfere with treatment. This evaluation included an IQ test and a psychiatric interview. Written informed consent for treatment with GnRHa was obtained from adolescents and for those aged younger than 16 years also from their parents/ guardians. Informed consent forms contained information about the risk of infertility. The option of FP was discussed by the paediatric endocrinologist. If adolescents wished to cryopreserve sperm and thought they would be able to produce a semen sample via masturbation, they were referred to a fertility clinic where semen samples were preserved, on three occasions if indicated. Data collection Extracted data from the medical files were age, IQ, Tanner stage, testicular volume, ethnicity, sexual orientation, psychiatric comorbidity (depression, anxiety disorder, posttraumatic stress disorder, and autism spectrum disorder), family situation (i.e., raised in an intact family, divorced parents or other [including adopted, living in a foster-home]) and information about the desire to have children. Ethics The study is part of an observational study on the effects of hormonal treatment in adolescents with gender dysphoria, which was approved by the medical ethical committee of the Leiden University Medical Centre. Statistics Statistical analyses were performed with SPSS version 23 (IBM) using the t-test for continuous variables or Manne Whitney U test if data were not normally distributed. Chi-square test or Fisher’s exact test was used for categorical variables. No Bonferroni correction was applied when testing the association between several variables and the use of FP because this was meant to explore rather than confirm factors that may be of influence.