[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127404-en":3,"doc-seo-127404-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},127404,962085571259,"Theodora","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Machine Learning-based Predictions of Healthcare Contacts Following Emergency Hospitalisation Using Electronic Health Records - Supplement","Machine learning models predict healthcare contacts after emergency hospitalisation using electronic health records. The supplement provides definitions of predictive features, describes missing-data patterns across outcomes, and reports baseline patient characteristics stratified by key endpoints such as in-hospital death, extended stay, home discharge, and geriatric medicine admission. It also details model configuration, including fine-tuned hyperparameters, comparisons of linear and non-linear estimators, stratified training/validation setup, and performance assessed with stratified validation and calibration-style analyses.","SUPPLEMENT  \nMachine Learning-based Predictions of Healthcare Contacts Following Emergency Hospitalisation Using Electronic  \nHealth Records  \nKonstantin Georgiev 1, Dimitrios Doudesis1, Joanne McPeake3, Nicholas L Mills1, Susan D  \nShenkin2, Jacques D Fleuriot4, Atul Anand 1  \n1 Institute of Neuroscience and Cardiovascular Research, Queen's Medical Research Institute, University of Edinburgh, EH16 4TJ, UK  \n2 Ageing and Health Research Group and Advanced Care Research Centre, Usher Institute, Edinburgh BioQuarter, University of Edinburgh, EH16 4UX, UK  \n3 The Healthcare Improvement Studies Institute, Department of Public Health and Primary Care, University of Cambridge, CB1 8RN, UK  \n4 Artificial Intelligence and its Applications Institute, School of Informatics, University of Edinburgh, EH8 9AB, UK  \nCorrespondence to: Konstantin Georgiev, University of Edinburgh, Institute of Neuroscience and Cardiovascular Research, Chancellor’s Building, 49 Little France Crescent, Edinburgh,  \nEH16 4SA, United Kingdom, [Email:](Email: K.S.Georgiev@sms.ed.ac.uk)[ ](Email: K.S.Georgiev@sms.ed.ac.uk)[K.S.Georgiev@sms.ed.ac.uk](Email: K.S.Georgiev@sms.ed.ac.uk)  \nTable and Figure legends:  \nSupplementary Table 1. Feature definitions for prediction of hospital outcomes and healthcare contacts. Supplementary Table 2. Summary of missingness across specialist outcomes within the top-ranking predictors of healthcare contacts.  \nSupplementary Table 3. Patient characteristics at baseline grouped by in-hospital death. Supplementary Supplementary Table 4. Patient characteristics at baseline grouped by extended stay.  \nSupplementary Table 5. Patient characteristics at baseline grouped by home discharge. Supplementary Supplementary Table 6. Patient characteristics at baseline grouped by admission to geriatric medicine services.  \nSupplementary Table 7. Summary of healthcare contacts distribution by each secondary outcome. Supplementary Table 8. Performance comparison across different linear and non-linear regression estimators for predictions of healthcare contacts at point of ED attendance.  \nSupplementary Table 9. Summary of stratified training and validation set characteristics for healthcare contacts prediction at point of ED attendance.  \nSupplementary Table 10. Summary of fine-tuned model hyperparameters across all outcomes for prediction models at point of ED attendance.  \nSupplementary Table 11. Performance comparison using stratified 10-fold validation for the healthcare contacts prediction model at point of ED attendance.  \nSupplementary Fig. 1. The annual distribution of secondary outcomes over the full data collection window.  \nSupplementary Fig. 2. Socio-demographic characteristics categorised by age and deprivation in patients with each secondary hospital outcome.  \nSupplementary Fig. 3. Box-plot showing the spread of log-transformed healthcare contacts across patients with and without each secondary outcome.  \nSupplementary Fig. 4. Violin plot showing the spread of log-transformed contacts and length of stay per individual across each hospital site.  \nSupplementary Fig. 5. Violin plot showing the spread of log-transformed contacts and length of stay per individual grouped by season of ED attendance.  \nSupplementary Fig. 6. Confusion Matrix summary showing the percentage of correctly captured and misclassified examples after quintile-based discretisation of the predicted contacts.  \nSupplementary Fig. 7. Performance trajectory curves for healthcare contact prediction, stratified by age group.  \nSupplementary Fig. 8. Performance trajectory curves for healthcare contact prediction, stratified by deprivation level.  \nSupplementary Fig. 9. Aalen-Johansen cumulative incidence function of in-hospital death stratified by healthcare contact level, adjusted for non-home discharge outcomes.  \nSupplementary Fig. 10. Performance trajectory curves for healthcare contact prediction measured in survivors to discharge.  \nSupplementary Fig. 11. Performance","cbCaibNkWK0V2Wuh","https://ap.wps.com/l/cbCaibNkWK0V2Wuh","pdf",4340773,1,35,"English","en",105,"# Supplementary tables and figures\n## Feature definitions and missingness\n## Patient characteristics by outcomes\n## Model performance and estimator comparison\n## Stratified training/validation and hyperparameters\n## Validation under COVID-19 lockdown\n## Visual summaries (confusion matrix, trajectories, incidence, distributions)","[{\"question\":\"What data source supports the prediction models in this supplement?\",\"answer\":\"Electronic health records are used to derive features and predict healthcare contacts following emergency hospitalisation.\"},{\"question\":\"Which supplementary elements explain how features and missingness are handled?\",\"answer\":\"Supplementary Table 1 defines predictive features, while Supplementary Table 2 summarizes missingness across specialist outcomes within top-ranking predictors.\"},{\"question\":\"How is model performance evaluated for healthcare contacts prediction?\",\"answer\":\"Performance is compared across different linear and non-linear regression estimators, assessed with stratified training/validation setups, and reported using stratified 10-fold validation, including performance trajectories and confusion-matrix summaries.\"}]","Machine Learning-based Predictions of Healthcare Contacts Following Emergency Hospitalisation Using Electronic Health Records - Supplement | PDF",1785938709,88,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-based-predictions-of-healthcare-contacts-following-emergency-hospitalisation-using-electronic-health-records-supplement","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-based-predictions-of-healthcare-contacts-following-emergency-hospitalisation-using-electronic-health-records-supplement/127404/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What data source supports the prediction models in this supplement?","Question",{"text":75,"@type":76},"Electronic health records are used to derive features and predict healthcare contacts following emergency hospitalisation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which supplementary elements explain how features and missingness are handled?",{"text":80,"@type":76},"Supplementary Table 1 defines predictive features, while Supplementary Table 2 summarizes missingness across specialist outcomes within top-ranking predictors.",{"name":82,"@type":73,"acceptedAnswer":83},"How is model performance evaluated for healthcare contacts prediction?",{"text":84,"@type":76},"Performance is compared across different linear and non-linear regression estimators, assessed with stratified training/validation setups, and reported using stratified 10-fold validation, including performance trajectories and confusion-matrix summaries.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]