2.2 Data collection
Patients were identified from the Netherlands Cancer Registry (NCR), which is a population-based registry with coverage of all newly diagnosed malignancies in the Netherlands since 1989. Dedicated registration clerks routinely extract patient information from medical records within the hospitals. Information on vital status and date of death were obtained from the municipal demography registries18. The NCR database was linked with patient files in the Pathological Anatomical National Automated Archive (PALGA), which is a national histo- and cytopathology register that archives all pathology reports in the Netherlands19. Patients’ age at time of diagnosis, treatment information, final surgery histopathology, and follow-up data were obtained for each patient. Treatment information consisted of date of surgery, type of surgery (e.g., with or without BSO), details about the hospital at which surgery was performed, and information regarding adjuvant treatment (e.g., radiotherapy, chemotherapy, hormone therapy). Final surgery histopathology data collected were histotype according to the International Classification of Diseases for Oncology (ICD-O), differentiation grade, FIGO stage 200920, maximum tumour diameter and presence of LVSI. Follow-up data was available on histologically confirmed recurrence of cancer, including date and site of recurrence, vital status, and date of death. In case of incoherent data, the PALGA histopathology records were leading for histotype and differentiation grade, and the NCR database was leading for disease stage.
National hospitals were contacted to retrieve data regarding hospital manipulator preference (i.e., IU or non-IU manipulator) between 2010 and 2020 by means of an online survey. The survey was formatted via Microsoft Forms and distributed to the appointed representative, a gynaecologist who performs minimally invasive hysterectomy for endometrial cancer, of each national hospital. Questions included in the survey are displayed in Table S1. In the absence of response, reminder emails were sent to those hospitals. Hospitals with disputable responses were contacted again for further clarification (by means of a phone call or an email). If the manipulator preference within a hospital changed between 2010 and 2020, these hospitals were analysed as independent institutions before and after the date of change in manipulator preference. Hospitals that used both IU and non-IU manipulators simultaneously between 2010 and 2020 and their corresponding patients were excluded from the analysis, as patients were categorized into IU group and non-IU group according to hospital manipulator preference rather than manipulator use at patient level.