[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122184-en":3,"doc-seo-122184-105":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":20,"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},122184,7971461740909,"Levi","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Evaluating Machine Learning for Predicting Youth Suicidal Behavior Up to One Year after Contact with Mental Health Specialty Care","This study evaluated multiple predictive modeling algorithms for suicide attempts resulting in inpatient hospitalization or suicide among youth aged 9–18 (n=34,528) following 6–12 months of contact with a mental health specialist in Stockholm, Sweden from 2006–2012. A total of 209 predictors across clinical, demographic, family, neighborhood, and social domains were derived from national registers. Standard logistic regression and random forest achieved the highest AUCs, while sensitivities varied widely across methods. Despite limited power due to low outcome prevalence, overall performance was broadly similar. The findings support continued research on how prediction models can augment clinical decision-making.","Evaluating Machine Learning for Predicting Youth Suicidal Behavior Up to One Year after  \nContact with Mental Health Specialty Care  \nLauren M. O’Reilly, BS 1 *  \nSeena Fazel, MBChB, MD, FRCPsych2 Martin E. Rickert, PhD3  \nRalf Kuja-Halkola, PhD4 Martin Cederlof, PhD4,5 Clara Hellner, MD, PhD6 Henrik Larsson, PhD4,5 Paul Lichtenstein, PhD4 Brian M. D’Onofrio, PhD3,4  \n1 Indiana University School of Medicine, Indianapolis, IN, USA  \n2 Department of Psychiatry, University of Oxford, Oxford, UK  \n3 Department of Psychological and Brain Sciences, Indiana University, Bloomington, IN, USA 4 Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden;  \n5 School of Medical Sciences, Örebro University, Örebro, Sweden  \n6 Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden  \n*Corresponding Author 410 W. 10th St.  \nIndianapolis, IN 46202 [loreilly@iu.edu](loreilly@iu.edu)  \nAbstract  \nThis paper assessed the performance of several predictive modeling algorithms of suicide attempt resulting in inpatient hospitalization or suicide among youth aged 9-18 (n=34,528) after contact (6-12 months) with a mental health specialist in Stockholm, Sweden from 2006-2012. Using 209 predictors across domains (e.g., clinical, demographic, family, neighborhood, social) identified from national registers, we applied standard logistic regression, regularized logistic regression, and machine learning algorithms (i.e., random forests, gradient boosting, support vector machines) . Standard logistic regression (0.77 [95% CI, 0.72-0.82]) and random forest models (0.80 [95% CI, 0.74-0.86]) demonstrated the highest AUCs. Sensitivities ranged from 0.33 (support vector machines)-0.91 (standard logistic regression) . While the study was underpowered to detect a difference between logistic regression and machine learning algorithms (outcome prevalence=0.7%), performance metrics were similar across models. Logistic regression is not clearly worse than machine learning approaches. Ongoing research is needed to  \nexamine how prediction models can augment clinical decision making.  \nThe rate of suicide have been trending upwards over the past 20 years in the United  \nStates (Center for Disease Control, 2018) . In 2021, the US Surgeon General and National Action Alliance for Suicide Prevention (NAASP) jointly released a Call to Action on the 20th  \nanniversary of the seminal report recognizing suicide as a major public health problem in the US.  \nIn the 2021 Call to Action, the Surgeon General and NAASP emphasized the importance of improving risk identification in health care settings (US Surgeon General & National Action Alliance for Suicide Prevention, 2021), which has been challenging for suicide research for two primary reasons: 1) the base rate of suicidal behavior is low in community samples (Nock et al., 2008), and 2) a large body of research informing our understanding of risk prediction includes long follow-up periods (i.e., 5-10 years) . Research suggests that suicide risk can vary across  \nshort time periods (e.g., hours or days) (Witte et al., 2006) . To aid suicidal behavior prediction, some researchers have shifted focus from predicting lifetime risk in the general population to acute periods of risk within higher-risk populations (Glenn & Nock, 2014) .  \nResearch has consistently demonstrated that adults (Stene-Larsen & Reneflot, 2019) and adolescents (Braciszewski et al., 2023) who die by suicide are often in contact with health care professionals within a proximal time frame prior to their death. It is important to aid suicidal prediction among those who contact the health care system to identify elevated risk, intervene, and triage care (Asarnow et al., 2016; Nock, 2012) . Research commonly refers to the period of heighted risk after contact with healthcare system as short-term risk. While this definition is poorly defined (Simon, 2006), it represents an attempt to shorten follow-up windows rela","cbCaias25YcPlf5F","https://ap.wps.com/l/cbCaias25YcPlf5F","pdf",195972,1,40,"English","en",105,"# Abstract\n## Predictive Modeling Approach\n## Study Design and Data Sources\n## Results and Model Performance\n## Implications and Future Research","[{\"question\":\"What population and time window were analyzed in this study?\",\"answer\":\"Youth aged 9–18 were studied after 6–12 months of contact with a mental health specialist, with follow-up outcomes including suicide attempt leading to inpatient hospitalization or suicide.\"},{\"question\":\"Which types of models were compared?\",\"answer\":\"The study compared standard logistic regression, regularized logistic regression, and several machine learning algorithms including random forests, gradient boosting, and support vector machines.\"},{\"question\":\"What were the main performance findings across models?\",\"answer\":\"Standard logistic regression and random forest showed the highest AUC values, while sensitivities differed substantially by model. Overall metrics were broadly similar across models, and the study lacked power to confirm differences.\"}]","Evaluating Machine Learning for Predicting Youth Suicidal Behavior Up to One Year after Contact with Mental Health Specialty Care | PDF",1785809239,101,{"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},"evaluating-machine-learning-for-predicting-youth-suicidal-behavior-up-to-one-year-after-contact-with-mental-health-specialty-care","",{"@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/evaluating-machine-learning-for-predicting-youth-suicidal-behavior-up-to-one-year-after-contact-with-mental-health-specialty-care/122184/",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","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What population and time window were analyzed in this study?","Question",{"text":75,"@type":76},"Youth aged 9–18 were studied after 6–12 months of contact with a mental health specialist, with follow-up outcomes including suicide attempt leading to inpatient hospitalization or suicide.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which types of models were compared?",{"text":80,"@type":76},"The study compared standard logistic regression, regularized logistic regression, and several machine learning algorithms including random forests, gradient boosting, and support vector machines.",{"name":82,"@type":73,"acceptedAnswer":83},"What were the main performance findings across models?",{"text":84,"@type":76},"Standard logistic regression and random forest showed the highest AUC values, while sensitivities differed substantially by model. Overall metrics were broadly similar across models, and the study lacked power to confirm differences.","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},[93,97,101,105,110,115,119,122,127,130,134],{"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":21,"slug":118},7,"Healthcare","healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]