Annals of surgical oncology · 2026
Gravity Modeling to Predict Patient Choice of Hospital for Pancreaticoduodenectomy.
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.
Ross-Driscoll Katie, Huber Sarah, Hughes Alexa J, Yang Kelly, Wilk Adam S, Bilimoria Karl Y, Patzer Rachel E, Ellis Ryan J
Аннотация
BACKGROUND: Prior studies simulating how regionalization of complex surgeries such as pancreatoduodenectomy (PD) would affect access to care assumed that patients travel to their closest hospital. The validity of this assumption had not been evaluated. Our objectives were to assess the accuracy of a distance-only model in predicting where a patient received PD and determine whether incorporating hospital characteristics improved accuracy.
METHODS: Data on patients undergoing PD from 2016 to 2017 in New York or Florida were obtained from the Statewide Inpatient Databases. We assessed three models: distance alone (model 1), distance and hospital volume (model 2), and distance, hospital volume, rurality, cancer accreditation, medical school affiliation, and centralization taxonomy of the health system (model 3). Model accuracy was determined by comparing the predicted hospital and the actual hospital where the patient received care.
RESULTS: Among the 2949 included patients, model 1 was accurate for only 16.3%; model 2 was accurate for 32.9%, and model 3 was accurate for 24.3%. Patients who were accurately classified with models 2 or 3 but not 1 were more likely to live in major metropolitan areas (74.7% vs. 61.0%, p < 0.001), more likely to be non-Hispanic white (71.3% vs. 68.3%, p = 0.04), more likely to have private insurance (35.0% vs. 27.8%, p < 0.001), and had lower in-hospital mortality (1.3% vs. 3.3%, p = 0.03).
CONCLUSIONS: Distance-only models were largely inaccurate for predicting where patients received PD. Simulations of patient redistribution should incorporate volume, distance, and additional patient- and provider-level factors.
METHODS: Data on patients undergoing PD from 2016 to 2017 in New York or Florida were obtained from the Statewide Inpatient Databases. We assessed three models: distance alone (model 1), distance and hospital volume (model 2), and distance, hospital volume, rurality, cancer accreditation, medical school affiliation, and centralization taxonomy of the health system (model 3). Model accuracy was determined by comparing the predicted hospital and the actual hospital where the patient received care.
RESULTS: Among the 2949 included patients, model 1 was accurate for only 16.3%; model 2 was accurate for 32.9%, and model 3 was accurate for 24.3%. Patients who were accurately classified with models 2 or 3 but not 1 were more likely to live in major metropolitan areas (74.7% vs. 61.0%, p < 0.001), more likely to be non-Hispanic white (71.3% vs. 68.3%, p = 0.04), more likely to have private insurance (35.0% vs. 27.8%, p < 0.001), and had lower in-hospital mortality (1.3% vs. 3.3%, p = 0.03).
CONCLUSIONS: Distance-only models were largely inaccurate for predicting where patients received PD. Simulations of patient redistribution should incorporate volume, distance, and additional patient- and provider-level factors.