Influence of sex hormone use on sleep architecture in a transgender cohort 183 whereas objective sleep could more strongly affected by physiological and biological factors, such as sleep apnea. The current study has a number of strengths. First of all, it was the first prospective study to investigate the effects of both feminizing and masculinizing GAHT on sleep architecture. In our setup, we used a singleelectrode EEG device that could be used for ambulatory measurements, which means that participants were able to measure their sleep for multiple days at home. This strengthens the external validity of our sleep measurements, since participants slept in their normal home situations. Furthermore, our study design allows us to examine the effects of longer term hormone use compared to previous work. Many studies assessing effects of exogenous sex hormones on sleep architecture in healthy participants focus on the effects of acute- or very short-term hormone exposure (Friess et al., 1997; Söderpalm et al., 2004), which means results typically display the acute effects, but not the longer term effects of sex hormone use. Compared to these studies, 3 months of GAHT exposure shows the effects of longer-term hormone use. However, many transgender GAHT users will typically use GAHT for the rest of their lives (Cocchetti et al., 2022), and future work should focus on longer-term effects of GAHT. The current study did not include outcomes after 12 months of GAHT use, since data collection was still ongoing, but future research from the current cohort will also provide more insight into the 12-month trajectory of sleep architecture changes after starting GAHT use. There are also a number of limitations of this study. Firstly, it is possible that there is a form of selection bias in the participant group. Some participants reported that they found the sleep EEG device uncomfortable or that they forgot to charge or to wear the device, and opted out of the measurements. In the supplementary materials, we have supplied an overview of the participant demographics (e.g. age, psychotropic medication use and symptoms of depression, stress or insomnia) of participants who participated in the objective sleep measures compared to those who did not (see Table S6.3; supplementary materials). These data show no clinically significant differences in symptoms of insomnia, depression or perceived stress between the groups, indicating that the participants who opted in or out of the sleep EEG measurements were comparable. However, it is still possible that the participants with more poor sleep were less likely to
RkJQdWJsaXNoZXIy MTk4NDMw