{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T23:24:17Z","timestamp":1787268257980,"version":"build-2736575974"},"reference-count":26,"publisher":"Georg Thieme Verlag KG","issue":"04","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Rehabilitation (Stuttg)"],"published-print":{"date-parts":[[2026,8]]},"abstract":"<jats:title>Zusammenfassung<\/jats:title>\n                  <jats:p>Auf Patient*innencharakteristika beruhende Prognosemodelle von individuellen\npsychosomatischen Rehabilitationserfolgen k\u00f6nnten dazu beitragen,\npotenzielle Non-Responder zu identifizieren und die Passung zwischen\nPatient*innen und Behandlung zu optimieren. Die vorliegende Studie\nuntersucht erstmalig das diesbez\u00fcgliche Potenzial von maschinellem Lernen.\nHierf\u00fcr werden Random Forest-Modelle zur Prognose von Behandlungserfolg\nhinsichtlich der Dimensionen Symptomschwere, gesundheitsbezogene\nLebensqualit\u00e4t, interpersonelle Beziehungsgestaltung und Selbstwirksamkeit\ntrainiert sowie die Wichtigkeit der aufgenommenen Pr\u00e4diktoren analysiert. Da\nRandom Forest neben der Aufnahme von Skalenwerten auch die Aufnahme von\nEinzelitems als eigenst\u00e4ndige Pr\u00e4diktoren erlaubt, konnten in der\nvorliegenden Studie aus 16 Selbstbeurteilungsinstrumenten sowie 11\nPersonenmerkmale p=555 potenzielle Pr\u00e4diktoren genutzt werden.<\/jats:p>\n                  <jats:p>Es handelt sich um eine Sekund\u00e4rdatenanalyse eines im naturalistischen\nEingruppen-Pr\u00e4-Post-Designs erhobenen Datensatzes. F\u00fcr jede Dimension wurde\nein Random Forest-Modell mittels verschachtelter Kreuzvalidierung erstellt.\nIn der \u00e4u\u00dferen Kreuzvalidierung wurde das finale Modell validiert sowie die\naufgenommenen Pr\u00e4diktoren hinsichtlich der jeweiligen\nPermutationswichtigkeit analysiert.<\/jats:p>\n                  <jats:p>Je nach Dimension konnten zwischen 30,02% und 38,94% der Varianz des\nindividuellen Behandlungserfolges erkl\u00e4ren werden. In der \u00e4u\u00dferen\nKreuzvalidierung konnten zwischen 29,33% und 37,83% der Varianz erkl\u00e4rt\nwerden. \u00dcber alle Dimensionen hinweg wurden 45 Pr\u00e4diktoren \u2013 davon 17\nSkalenwerte und 28 Einzelitems \u2013 aufgenommen. Lediglich zwei Pr\u00e4diktoren\nwaren f\u00fcr alle Dimensionen relevant.<\/jats:p>\n                  <jats:p>Mittels Random Forest-Modellen konnten f\u00fcr alle untersuchten Dimensionen\nindividuellen Behandlungserfolges hohe Varianzanteile erkl\u00e4rt werden. In der\nKreuzvalidierung konnten diese best\u00e4tigt werden, was auf die\nGeneralisierbarkeit der Ergebnisse hindeutet. Die Analyse der relevanten\nPr\u00e4diktoren erbringt zus\u00e4tzliche Evidenz f\u00fcr bereits bekannte Pr\u00e4diktoren\nwie Rentenbegehren und verdeutlicht gleichzeitig die Notwendigkeit einer\ndifferenzierten Betrachtung von psychosomatischen Rehabilitationserfolgen.\nEine praktische Implementierung von personalisierten Behandlungen im Sinne\nder personalisierten Medizin scheint jedoch in Folge zahlreicher\nLimitationen zeitnah nicht umsetzbar.<\/jats:p>","DOI":"10.1055\/a-2822-5523","type":"journal-article","created":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T22:39:47Z","timestamp":1787265587000},"page":"207-215","source":"Crossref","is-referenced-by-count":0,"title":["Maschinelles Lernen zur Prognose des Behandlungserfolges in der\nstation\u00e4ren psychosomatischen Rehabilitation \u2013 Maschinelles Lernen im Kontext\nder psychosomatischen Rehabilitation"],"prefix":"10.1055","volume":"65","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-3653-4333","authenticated-orcid":false,"given":"Paul-Gerrit","family":"Velthuysen","sequence":"additional","affiliation":[{"name":"Klinische Psychologie und Psychotherapie, Universit\u00e4t Hildesheim,\nInstitut f\u00fcr Psychologie, Hildesheim, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christoph","family":"Kr\u00f6ger","sequence":"additional","affiliation":[{"name":"Klinische Psychologie und Psychotherapie, Universit\u00e4t Hildesheim,\nInstitut f\u00fcr Psychologie, Hildesheim, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Axel","family":"Kobelt-Poenicke","sequence":"additional","affiliation":[{"name":"Klinische Psychologie und Psychotherapie, Universit\u00e4t Hildesheim,\nInstitut f\u00fcr Psychologie, Hildesheim, Germany"},{"name":"Referat Rehastrategie, Deutsche Rentenversicherung\nBraunschweig-Hannover, Laatzen, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"194","published-online":{"date-parts":[[2026,4,20]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"s-2005","DOI":"10.1055\/s-2005-863397","article-title":"Metaanalyse der Effekte station\u00e4rer psychosomatischer Rehabilitation \u2013\nMESTA-Studie","volume":"55","author":"C L\u00f6schmann","year":"2005","journal-title":"Psychother Psych Med"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1007\/s40211-018-0290-1","article-title":"Meta-Analyse der Evaluationsergebnisse psychiatrischer-psychosomatischer\nRehabilitation in \u00d6sterreich","volume":"33","author":"M Sprung","year":"2019","journal-title":"neuropsychiatrie"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"121","DOI":"10.13109\/zptm.2014.60.2.121","article-title":"Therapie-Non-Response in der psychosomatischen Krankenhausbehandlung und\nRehabilitation \u2013 Eine systematische \u00dcbersicht","volume":"60","author":"L Reuter","year":"2014","journal-title":"Z F\u00fcr Psychosom Med Psychother"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"101810","DOI":"10.1016\/j.cpr.2019.101810","article-title":"Treating treatment non-responders: A meta-analysis of randomized controlled\npsychotherapy trials","volume":"75","author":"A T Gloster","year":"2020","journal-title":"Clin Psychol Rev"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"2333392818758523","DOI":"10.1177\/2333392818758523","article-title":"Treatment Changes in General Practice Patients With Chronic Mental Disorders\nFollowing a Psychiatric\u2013Psychosomatic Consultation","volume":"5","author":"M Linden","year":"2018","journal-title":"Health Serv Res Manag Epidemiol"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"671","DOI":"10.1037\/ccp0000904","article-title":"Data-informed psychological therapy, measurement-based care, and precision\nmental health","volume":"92","author":"W Lutz","year":"2024","journal-title":"J Consult Clin Psychol"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"525","DOI":"10.1037\/a0014693","article-title":"Survey on the use of clinical and mechanical prediction methods in clinical\npsychology","volume":"40","author":"S I Vrieze","year":"2009","journal-title":"Prof Psychol Res Pract"},{"key":"ref8","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-662-61643-7","volume-title":"Hrsg. Psychologische Diagnostik","author":"L Schmidt-Atzert","year":"2021"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1055\/s-0029-1214411","article-title":"\u201eWer profitiert?\u201d\u2013 Patientenmerkmale als Erfolgspr\u00e4diktoren in der\npsychosomatischen","volume":"48","author":"J Oster","year":"2009","journal-title":"Rehabilitation. Rehabilitation"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"392","DOI":"10.1055\/s-0032-1304612","article-title":"Wer profitiert nicht von der psychosomatischen Rehabilitation?","volume":"51","author":"M Lange","year":"2012","journal-title":"Rehabilitation"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"103530","DOI":"10.1016\/j.brat.2019.103530","article-title":"Predicting cognitive behavioral therapy outcome in the outpatient sector based\non clinical routine data: A machine learning approach","volume":"124","author":"K Hilbert","year":"2020","journal-title":"Behav Res Ther"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"100598","DOI":"10.1016\/j.imu.2021.100598","article-title":"Machine learning approaches to predict rehabilitation success based on clinical\nand patient-reported outcome measures","volume":"24","author":"M Tschuggnall","year":"2021","journal-title":"Inform Med Unlocked"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1002\/mpr.1463","article-title":"Prediction of remission in obsessive compulsive disorder using a novel machine\nlearning strategy","volume":"24","author":"K D Askland","year":"2015","journal-title":"Int J Methods Psychiatr Res"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Breiman and Cutlers Random Forests for Classification and Regression","volume":"45","author":"L Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"ref15","first-page":"3133","article-title":"Do we Need Hundreds of Classifiers to Solve Real World Classification\nProblems?","volume":"15","author":"M Fern\u00e1ndez-Delgado","year":"2014","journal-title":"J Mach Learn Res"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"323","DOI":"10.1037\/a0016973","article-title":"An introduction to recursive partitioning: Rationale, application, and\ncharacteristics of classification and regression trees, bagging, and random\nforests","volume":"14","author":"C Strobl","year":"2009","journal-title":"Psychol Methods"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"197","DOI":"10.1007\/s11749-016-0481-7","article-title":"A random forest guided tour","volume":"25","author":"G Biau","year":"2016","journal-title":"TEST"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"303","DOI":"10.1055\/a-0896-7379","article-title":"Mitteilungen der Deutschen Gesellschaft f\u00fcr Psychosomatische Medizin (DKPM)","volume":"69","author":"S Rabung","year":"2019","journal-title":"Psychother Psych Med"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"162","DOI":"10.13109\/zptm.2009.55.2.162","article-title":"Psychometrische \u00dcberpr\u00fcfung einer verk\u00fcrzten Version der\u202f\u00bbHamburger Module zur\nErfassung allgemeiner Aspekte psychosozialer Gesundheit f\u00fcr die therapeutische\nPraxis\u00ab\u202f(HEALTH-49)","volume":"55","author":"S Rabung","year":"2009","journal-title":"Z f\u00fcr Psychosom Med Psychother"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1093\/bioinformatics\/btr597","article-title":"MissForest\u2014non-parametric missing value imputation for mixed-type data","volume":"28","author":"D J Stekhoven","year":"2012","journal-title":"Bioinformatics"},{"key":"ref21","volume-title":"Statistical power analysis for the behavioral sciences. 2nd ed","author":"J Cohen","year":"1988"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"1478","DOI":"10.1177\/0956797617714580","article-title":"Is Romantic Desire Predictable? Machine Learning Applied to Initial Romantic\nAttraction","volume":"28","author":"S Joel","year":"2017","journal-title":"Psychol Sci"},{"key":"ref23","doi-asserted-by":"crossref","first-page":"2225","DOI":"10.1016\/j.patrec.2010.03.014","article-title":"Variable selection using Random Forests","volume":"31","author":"R Genuer","year":"2010","journal-title":"Pattern Recognit Lett"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18637\/jss.v028.i05","article-title":"Building Predictive Models in R Using the caret Package","volume":"28","author":"M Kuhn","year":"2008","journal-title":"J Stat Softw"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"182","DOI":"10.1055\/s-0042-105939","article-title":"Erfassung von negativen Antwortverzerrungen \u2013 Entwicklung und Validierung des\nBeschwerdenvalidierungstests BEVA","volume":"55","author":"F Walter","year":"2016","journal-title":"Rehabilitation"},{"key":"ref29","first-page":"293","article-title":"Erfolg, Effekt, Outcome, Nutzen oder Wirkung? Begriffe und Konzepte im Diskurs\ngesundheitsbezogener Handlungswissenschaften","author":"T Meyer","year":"2014","journal-title":"Pflege Ges"}],"container-title":["Die Rehabilitation"],"original-title":[],"language":"de","link":[{"URL":"http:\/\/www.thieme-connect.de\/products\/ejournals\/pdf\/10.1055\/a-2822-5523.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,20]],"date-time":"2026-08-20T22:39:51Z","timestamp":1787265591000},"score":1,"resource":{"primary":{"URL":"http:\/\/www.thieme-connect.de\/DOI\/DOI?10.1055\/a-2822-5523"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,20]]},"references-count":26,"journal-issue":{"issue":"04","published-online":{"date-parts":[[2026,8,20]]},"published-print":{"date-parts":[[2026,8]]}},"URL":"https:\/\/doi.org\/10.1055\/a-2822-5523","archive":["Portico","CLOCKSS"],"relation":{},"ISSN":["0034-3536","1439-1309"],"issn-type":[{"value":"0034-3536","type":"print"},{"value":"1439-1309","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,20]]}}}