Drug safety · 2026
Developing a Hierarchical Algorithm to Identify Pregnancies and Determine Gestational Age from Nationwide Linked Health Data in Taiwan.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Hsieh Miyuki Hsing-Chun, Chang Zoe Chi-Jui, Lin Chih-Wan, Li Brian Meng-Hsun, Lai Edward Chia-Cheng, Huang Wan-Ting
Аннотация
BACKGROUND: Administrative health data are vital for investigating medication safety during pregnancy. In Taiwan, while the National Health Insurance Research Database (NHIRD) and Birth Certificate Application (BCA) capture pregnancy data, no standardized approach exists for integrating these sources to identify episodes and assign gestational age (GA). This study aimed to develop a hierarchical pregnancy-identification algorithm tailored to Taiwan's linked claims and birth registry data.
METHODS: We adapted an ICD-10-CM/PCS-based algorithm to the Taiwanese coding environment, incorporating clinician input, local billing practices, and birth registry information. The algorithm refines pregnancy outcome classification, estimates pregnancy start dates, and assigns GA. As a proof of concept, we applied the final algorithm to 2016-2022 data to summarize outcome distributions.
RESULTS: We developed a seven-step hierarchical algorithm that uses claims codes to classify pregnancy outcomes, groups records into episodes using spacing rules, assigns pregnancy start dates from ordered prenatal services, validates and refines GA via linkage to the BCA database, determines the number of pregnancies and fetuses, and checks the plausibility of GA and inter-pregnancy intervals. A total of 1,696,229 pregnancies contributed by 1,169,779 women were identified; outcome distributions (69.3% live birth, 16.3% spontaneous abortion, 5.7% elective abortion, 6.0% ectopic pregnancy, 1.4% delivery with unknown outcome, 1.1% stillbirth, 0.2% trophoblastic and other abnormal products of conception) were consistent with national statistics and prior literature.
CONCLUSIONS: This hierarchical algorithm provides a transparent, reproducible framework for identifying pregnancies and estimating GA in Taiwan. It establishes a critical foundation for future real-world studies of medication safety during pregnancy.
METHODS: We adapted an ICD-10-CM/PCS-based algorithm to the Taiwanese coding environment, incorporating clinician input, local billing practices, and birth registry information. The algorithm refines pregnancy outcome classification, estimates pregnancy start dates, and assigns GA. As a proof of concept, we applied the final algorithm to 2016-2022 data to summarize outcome distributions.
RESULTS: We developed a seven-step hierarchical algorithm that uses claims codes to classify pregnancy outcomes, groups records into episodes using spacing rules, assigns pregnancy start dates from ordered prenatal services, validates and refines GA via linkage to the BCA database, determines the number of pregnancies and fetuses, and checks the plausibility of GA and inter-pregnancy intervals. A total of 1,696,229 pregnancies contributed by 1,169,779 women were identified; outcome distributions (69.3% live birth, 16.3% spontaneous abortion, 5.7% elective abortion, 6.0% ectopic pregnancy, 1.4% delivery with unknown outcome, 1.1% stillbirth, 0.2% trophoblastic and other abnormal products of conception) were consistent with national statistics and prior literature.
CONCLUSIONS: This hierarchical algorithm provides a transparent, reproducible framework for identifying pregnancies and estimating GA in Taiwan. It establishes a critical foundation for future real-world studies of medication safety during pregnancy.