Traditional social surveys and medical research have long treated sex and gender as equivalent concepts, typically relying on a single binary “male/female” item. This practice has led to the systemic underidentification and misclassification of transgender and gender-diverse populations in datasets. Furthermore, simply adding a “transgender” option to a single-item question may fail to identify transgender individuals who do not disclose their transgender status. To address this methodological challenge, the Two-Step Gender Measurement approach collects gender-related information through two independent measures: “sex assigned at birth” and “current gender identity.” Cross-tabulation of responses to these two measures can be used to identify and classify different gender groups, including transgender and nonbinary individuals. However, identifying individuals with intersex traits or differences of sex development (DSD) requires additional information. The two-step approach not only helps distinguish the respective associations of sex and gender with health outcomes but also provides more informative data for clinical trials and health policy development. This approach represents a crucial methodological advance in promoting gender inclusion and equity in medical research.