蛋播视频一区,无码鲁丝一区二区,精品 久久 五月天,国产老熟女,五月草草在线观看,中文日韩欧美,情色一区二区三区,欧美日韩亚洲激情在线,亚洲制服在线香蕉

Regionalized Care onInter-hospital Transfer of Heart Attack Patients (Empirical and data analytics)

區(qū)域化醫(yī)療資源整合：

來(lái)自心臟病人轉(zhuǎn)院急救的數(shù)據(jù)驅(qū)動(dòng)研究

講座時(shí)間：12月20日9：30-11：30

講座地點(diǎn)：25樓A區(qū)3層C教室

主講人：盧峰

主講人簡(jiǎn)介：

副教授，美國(guó)西北大學(xué)凱洛格商學(xué)院博士，普度大學(xué)終身職，康奈爾大學(xué)未來(lái)健康研究院特邀學(xué)者，普度大學(xué)醫(yī)療工程研究院研究員，北京大學(xué)國(guó)家發(fā)展研究院訪問(wèn)學(xué)者。盧教授是《生產(chǎn)與運(yùn)作管理》(

講座內(nèi)容：

In the first paper, weempirically investigate the pattern of where heart attack patients aretransferred between hospitals. Using 2011 Florida State Emergency Departmentand Inpatient Databases, we demonstrate the relative importance of three keyfactors in determining transfer destinations: hospital relationship, distance,and quality. Our conditional logit analysis shows that the relationship ofbeing affiliated with the same multihospital system plays a dominant role inthe choice of transfer destinations, compared to distance and quality. Thisresult is robust to three alternative specifications of choice sets usingdistance ranking, radius circles, and Hospital Referral Regions. When using30-day readmission rate to evaluate the health outcome of transferred patients,we find that relationship-based transfers are associated with a much higherreadmission rate than distance-based and quality-based transfers. We also findthat nonprofit hospitals are more likely to conduct quality-based transfersthan their for-profit counterparts. Our study calls for a reevaluation ofcurrent practice in interhospital transfer of heart attack patients—a change ofprotocol for selecting transfer destinations may significantly decreasehospital readmission rates.

In the second paper, weintegrate our healthcare domain knowledge and structural modeling into datascience. We propose a two-stage support vector machine approach to identifythose patients whose survival chance would be increased significantly byinter-hospital transfer. Moreover, we designed a statistical approach toovercome the shortcoming of machine learning in prediction. Our recommendationshelp to increase the survival probability compared to doctors’ decisions.

封面圖制作 / 許林宇

圖文編輯 / 劉鑫