General Discussion 211 8 essential. Integrating social theory grounds gender-related findings in a solid foundation of knowledge about ‘gender as a performative act’ and supports consistent reasoning by being more explicit about the underlying principles, concepts and assumptions that guide the interpretation of research results. Inclusive language recognizes individuals as human beings first rather defining them by characteristics or traits (e.g. “women who endorse traditional feminine traits” in instead of “feminine women” or “people with Parkinson’s disease” instead of “Parkinson’s patients”).7 Inclusive language is more precise and informative, conveys respect, dignity and empathy towards research participants, and is a powerful way to avoid the reproduction of stereotypes and promote equity in scholarly communications. This is also particularly relevant for bias mitigation in the increasing use and development of artificial intelligence and machine learning in medicine that is modelled on word embeddings in available medical scientific publications.8,9 Additionally, in Chapter 3, I performed a backwards regression analysis to deconstruct the masculinity and femininity scores of the participants to investigate which gendered characteristics were specifically contributing to better HRQoL of people with PD. Results of this backwards regression showed that ‘de-gendered’ characteristics such as ‘Athletic’, ‘Assertive’, ‘Self-sufficient’ and ‘Happy’ were specific contributors to better overall HRQoL. These characteristics could potentially be used as more direct predictors of HRQoL rather than as components of a specific gender role. This raises questions about the usefulness of the construct’s ‘masculinity’ and ‘femininity’ in medical research, operationalized as a set of psychological characteristics. However, while the singular underlying characteristics that are associated with masculinity and femininity can be useful to understand the direct relation between specific psychological traits and health outcomes regardless of gender, they do not capture the complex ways in which gender roles operate in society at large and shape people’s social positions and experiences. In other words, when we deconstruct the constructs of ‘masculinity’ and ‘femininity’ into their singular underlying components, we are no longer examining masculine or feminine gender roles because those concepts are then stripped of their combined cultural meaning. An appropriate measure for gender roles should reflect contextual societal norms and expectations about gender roles and informs research whether these constructs as social positions and experiences influences health outcomes. The reality is that very few quantitative analyses of the impact of gender norms and roles on health outcomes are available because direct measures of gender norms are absent in standard survey data.10 However, several studies have showed that the impact of gender norms on health outcomes can be assessed by creating proxy measures for norms in existing data, for example Ballering et al (2020).11,12 It
RkJQdWJsaXNoZXIy MTk4NDMw