World journal of urology · 2025
Enhancing surgical efficiency: predicting same-day cancellations in urologic procedures.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
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.
Suarez Pablo A, Srirangapatanam Sudarshan, Leng Lynn, Momodu Mubarak M, Neuhaus John, Bayne David B
Аннотация
PURPOSE: Day-of-surgery (DOS) cancellations negatively impact healthcare efficiency and patient outcomes. Early identification of at-risk patients is essential for timely perioperative interventions. This study aims to develop and evaluate a predictive model for DOS cancellations in urologic surgeries using a comprehensive set of electronic health record (EHR) derived variables, including social determinants of health (SDOH).
METHODS: This retrospective cohort study examined adult patients scheduled for non-emergent urologic procedures at Zuckerberg San Francisco General Hospital (ZSFGH), a major safety-net institution. EHR data from January 2021 to June 2023 included demographics, social determinants of health, clinical history, and healthcare utilization. A lasso-optimized logistic regression model was used to predict DOS cancellations, with performance evaluated via area under the curve (AUC) of the receiver operating characteristic curve (ROC), accuracy, predictive values, sensitivity, specificity, and kappa at a 0.4 threshold to prioritize high-risk identification. Internal validation was conducted using cross-validation.
RESULTS: Of the 778 patients identified as undergoing elective urologic procedures, 136 (17.5%) had a DOS cancellation. Our prediction model consisted of 13 variables selected by optimizing the penalty (λ) of the lasso to achieve maximal accuracy. When evaluating the model on the validation data, it performed with an AUC of 0.80 [CI: 0.70-0.89], 83% accuracy, and a kappa of 0.20. The model also had 81% sensitivity and 55% specificity with a cut-off threshold of 0.4.
CONCLUSION: This is the first study to integrate comprehensive SDOH data into a risk score model for predicting DOS cancellations. With an AUC of 0.80, the model offers a novel tool to help identify high-risk patients of DOS cancellation and guide preoperative interventions. However, further studies assessing the external validity of our model at other clinical sites are necessary.
METHODS: This retrospective cohort study examined adult patients scheduled for non-emergent urologic procedures at Zuckerberg San Francisco General Hospital (ZSFGH), a major safety-net institution. EHR data from January 2021 to June 2023 included demographics, social determinants of health, clinical history, and healthcare utilization. A lasso-optimized logistic regression model was used to predict DOS cancellations, with performance evaluated via area under the curve (AUC) of the receiver operating characteristic curve (ROC), accuracy, predictive values, sensitivity, specificity, and kappa at a 0.4 threshold to prioritize high-risk identification. Internal validation was conducted using cross-validation.
RESULTS: Of the 778 patients identified as undergoing elective urologic procedures, 136 (17.5%) had a DOS cancellation. Our prediction model consisted of 13 variables selected by optimizing the penalty (λ) of the lasso to achieve maximal accuracy. When evaluating the model on the validation data, it performed with an AUC of 0.80 [CI: 0.70-0.89], 83% accuracy, and a kappa of 0.20. The model also had 81% sensitivity and 55% specificity with a cut-off threshold of 0.4.
CONCLUSION: This is the first study to integrate comprehensive SDOH data into a risk score model for predicting DOS cancellations. With an AUC of 0.80, the model offers a novel tool to help identify high-risk patients of DOS cancellation and guide preoperative interventions. However, further studies assessing the external validity of our model at other clinical sites are necessary.