JAMIA open · 2026
Spatial bias in quality of estimated delivery date records in the electronic health record.
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.
Lewis Abigail, Foraker Randi, Lian Min, Zhao Min, Oh Inez, Gupta Aditi, Frolova Antonina, Payne Philip R O
Аннотация
OBJECTIVES: This study evaluates the quality of estimated delivery dates (EDD) captured in electronic health record (EHR) systems, focusing on identifying and quantifying bias in these estimates for patients seeking prenatal care.
MATERIALS AND METHODS: We measured spatial and racial biases in the completeness and concordance of EDDs recorded in the EHR for patients seeking prenatal care. Element presence and agreement were compared across the catchment area to identify spatial patterns of bias in EDD records.
RESULTS: We obtained 38 477 unique pregnancy episodes in the EHR representing 33 081 unique patients between 2018 and 2022. Clear spatial and racial biases emerged in the completeness and concordance of EDDs (P < .05 across tests): Black patients had higher levels of missing EDDs than White patients (8.43% vs 5.11%). Moreover, incompleteness and discordance in EDDs tended to occur in the same locations (P < .05 for all global tests), indicating a potential compounding effect on prenatal care access for the patients with implications for data reuse.
DISCUSSION: This work demonstrates spatial and racial biases in quality of EDDs recorded in the EHR and, therefore, the need to prioritize bias assessment within EHR data quality (DQ) frameworks. Continued efforts would support equitable improvements in EHR-based research by improving the reliability of current health service analyses and informing the development of more inclusive and effective EHR-based technologies.
CONCLUSION: Including bias related to socioeconomic characteristics as a core axis of EHR DQ would extend existing frameworks to address potential inequities embedded in EHR data.
MATERIALS AND METHODS: We measured spatial and racial biases in the completeness and concordance of EDDs recorded in the EHR for patients seeking prenatal care. Element presence and agreement were compared across the catchment area to identify spatial patterns of bias in EDD records.
RESULTS: We obtained 38 477 unique pregnancy episodes in the EHR representing 33 081 unique patients between 2018 and 2022. Clear spatial and racial biases emerged in the completeness and concordance of EDDs (P < .05 across tests): Black patients had higher levels of missing EDDs than White patients (8.43% vs 5.11%). Moreover, incompleteness and discordance in EDDs tended to occur in the same locations (P < .05 for all global tests), indicating a potential compounding effect on prenatal care access for the patients with implications for data reuse.
DISCUSSION: This work demonstrates spatial and racial biases in quality of EDDs recorded in the EHR and, therefore, the need to prioritize bias assessment within EHR data quality (DQ) frameworks. Continued efforts would support equitable improvements in EHR-based research by improving the reliability of current health service analyses and informing the development of more inclusive and effective EHR-based technologies.
CONCLUSION: Including bias related to socioeconomic characteristics as a core axis of EHR DQ would extend existing frameworks to address potential inequities embedded in EHR data.