AbstractBackgroundWhile psychological treatments are effective, a substantial portion of patients do not benefit enough. Early identification of those may allow for adaptive treatment strategies and improved outcomes. We aimed to evaluate the clinical usefulness of machine-learning (ML) models predicting outcomes in Internet-based Cognitive Behavioural Therapy, to compare ML-related methodological choices, and guide future use of these.MethodsEighty main models were compared.

。早期识别这些可能会导致适应性治疗策略和改善结果。我们旨在评估机器学习（ML）模型预测基于互联网的认知行为治疗结果的临床实用性，比较ML相关的方法选择，并指导未来使用这些方法。方法对几种主要模型进行比较。

Baseline variables, weekly symptoms, and treatment activity were used to predict treatment outcomes in a dataset of 6695 patients from regular care.ResultsWe show that the best models use handpicked predictors and impute missing data. No ML algorithm shows clear superiority. They have a mean balanced accuracy of 78.1% at treatment week four, closely matched by regression (77.8%).ConclusionsML surpasses the benchmark for clinical usefulness (67%).

。结果我们表明，最佳模型使用精心挑选的预测因子并估算缺失数据。没有ML算法显示出明显的优越性。他们在治疗第四周的平均平衡准确率为78.1%，与回归（77.8%）密切相关。结论SML超过了临床实用性的基准（67%）。

Advanced and simple models perform equally, indicating a need for more data or smarter methodological designs to confirm advantages of ML.Plain language summary.

高级模型和简单模型的性能相同，这表明需要更多的数据或更智能的方法设计来确认ML的优势。普通语言摘要。

While there are many therapy treatments that are effective for mental health problems some patients don’t benefit enough. Predicting whom might need more help can guide therapists to adjust treatments for better results. Computer methods are increasingly used for predicting the outcome of treatment, but studies vary widely in accuracy and methodology.

虽然有许多治疗方法对心理健康问题有效，但一些患者并没有获得足够的益处。预测谁可能需要更多帮助可以指导治疗师调整治疗方法以取得更好的效果。计算机方法越来越多地用于预测治疗结果，但研究的准确性和方法差异很大。

We examined a variety of models to test performance. Those examined were based on a several factors: what data is chosen, how the data is managed, as well as type of mathematical equations and function used for prediction. When used on ~6500 patients, none of the computer methods tested stood out as the best.

我们检查了各种模型以测试性能。这些检查是基于几个因素：选择什么数据，如何管理数据，以及用于预测的数学方程和函数的类型。当在约6500名患者身上使用时，没有一种测试的计算机方法是最好的。

Simple models were as accurate as more advanced. Accuracy of prediction of treatment outcome was good enough to inform clinicians’ decisions, suggesting they may still be useful tools in mental health care..

简单的模型与更先进的模型一样准确。预测治疗结果的准确性足以告知临床医生的决定，这表明它们可能仍然是心理健康护理中有用的工具。。

IntroductionEvidence-based psychological treatments are beneficial for many conditions, and fewer treatments fail if therapists are given predictions for each patient’s outcome, based on continuous monitoring of symptoms1,2. This enables the therapist to adjust treatment for patients risking failure, and is referred to as an Adaptive Treatment Strategy2.

引言基于证据的心理治疗对许多情况都是有益的，如果治疗师根据对症状的持续监测对每位患者的结果进行预测，那么治疗失败的可能性就会减少1,2。这使治疗师能够为有失败风险的患者调整治疗，被称为适应性治疗策略2。

This has shown to improve outcomes in traditional psychological treatments1,3.Internet-delivered Cognitive Behavioural Therapy (ICBT), i.e. digital, diagnosis-specific self-help material with brief therapist support, could help increase access to psychological treatment and shows effects similar to traditional face-to-face CBT4.

这已经证明可以改善传统心理治疗的结果1,3。互联网提供的认知行为疗法（ICBT），即数字的，特定于诊断的自助材料，有短暂的治疗师支持，可以帮助增加心理治疗的机会，并显示出类似于传统面对面的CBT4的效果。

This guided ICBT treatment format is extensively used in regular care worldwide5 as well as thoroughly researched6. As in traditional CBT, about 30–60% of ICBT patients do not benefit sufficiently7, but the use of an Adaptive Treatment Strategy has been shown to improve outcome2.Various approaches to predicting patients’ treatment outcomes have been used, and the use of machine learning, supposedly superior to regression models when trained on large samples with multiple predictors8, is growing within mental health care9.

这种指导性ICBT治疗格式广泛用于全球常规护理5以及彻底研究6。与传统的CBT一样，约30-60%的ICBT患者不能充分受益7，但使用适应性治疗策略已被证明可以改善结果2。已经使用了各种方法来预测患者的治疗结果，并且机器学习的使用在心理健康护理中正在增长，当使用多个预测因子对大样本进行训练时，机器学习被认为优于回归模型8。

In both guided ICBT and traditional psychotherapy, using only baseline data as predictors results in weak predictions10,11. However, it is well established that early symptom change during treatment is associated with the symptomatic treatment outcome12,13 and including predictors from the first weeks of treatment, as in Routine Outcome Monitoring14 and Adaptive Treatment Strategies, increases the accuracy of predictions compared to baseline predictors only15,16.

在指导性ICBT和传统心理治疗中，仅使用基线数据作为预测因子会导致预测不佳10,11。然而，众所周知，治疗期间的早期症状变化与对症治疗结果有关[12,13]，包括治疗前几周的预测因子，如常规结果监测[14]和适应性治疗策略，与基线预测因子相比，预测的准确性仅提高了15,16。

While early symptom change is fundamental as predictor, adding other predictors in conjunction with machine learning.

虽然早期症状变化是预测因子的基础，但结合机器学习添加其他预测因子。

Data availability

数据可用性

The data for all results are in Supplementary Data 3, this data also includes the source data to reproduce the figures in this paper. Specifications of predictors and datasets are specified in Supplementary Data 1. Underlying patient data is not in the supplement due to scope of the approved ethical application and current health care data management policy, and are available according to the ethical approval upon reasonable request..

所有结果的数据都在补充数据3中，该数据还包括用于重现本文中数字的源数据。预测变量和数据集的规格在补充数据1中指定。由于批准的道德应用范围和当前的医疗保健数据管理政策，基础患者数据不在附录中，并且在合理要求下可根据道德批准获得。。

Code availability

代码可用性

Code for analysis of the prediction results and reproduction of the results in the paper can be found at: https://github.com/intraverbal/paper_ipsy_outcome_pred39, also detailed in the supplementary information, also in the supplementary software 1. Code for analyses, imputation, and data set creation is also available in the supplementary information and supplementary software 1.

本文中预测结果分析和结果再现的代码可在以下网址找到：https://github.com/intraverbal/paper_ipsy_outcome_pred39，也在补充信息中详细介绍，也在补充软件1中。补充信息和补充软件1中还提供了用于分析，插补和数据集创建的代码。

Code for underlying handling of patient data is limited due to the scope of the approved ethical application and current health care data management policy..

由于批准的道德应用程序的范围和当前的医疗保健数据管理政策，患者数据的底层处理代码受到限制。。

ReferencesLambert, M. J., Whipple, J. L. & Kleinstäuber, M. Collecting and delivering progress feedback: a meta-analysis of routine outcome monitoring. Psychotherapy 55, 520–537 (2018).Article

。文章

Google Scholar

谷歌学者

Forsell, E. et al. Proof of concept for an adaptive treatment strategy to prevent failures in internet-delivered CBT: a single-blind randomized clinical trial with insomnia patients. Am. J. Psychiatry 176, 315–323 (2019).Article

Forsell，E.等人，《预防互联网CBT失败的适应性治疗策略的概念验证：一项针对失眠患者的单盲随机临床试验》。《美国精神病学杂志》176315–323（2019）。文章

Google Scholar

谷歌学者

Lutz, W. et al. Prospective evaluation of a clinical decision support system in psychological therapy. J. Consult. Clin. Psychol. 90, 90–106 (2022).Article

Lutz，W.等人。心理治疗中临床决策支持系统的前瞻性评估。J、 咨询。。心理学。90，90–106（2022）。文章

Google Scholar

谷歌学者

Andrews, G. et al. Computer therapy for the anxiety and depression disorders is effective, acceptable and practical health care: an updated meta-analysis. J. Anxiety Disord. 55, 70–78 (2018).Article

焦虑和抑郁障碍的计算机治疗是有效的，可接受的和实用的医疗保健：最新的荟萃分析。J、 焦虑不安。55,70-78（2018）。文章

Google Scholar

谷歌学者

Titov, N. et al. ICBT in routine care: a descriptive analysis of successful clinics in five countries. Internet Interv. 13, 108–115 (2018).Article

。互联网互联网。13108-115（2018）。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Andersson, G., Carlbring, P., Titov, N. & Lindefors, N. Internet interventions for adults with anxiety and mood disorders: a narrative umbrella review of recent meta-analyses. Can. J. Psychiatry 64, 465–470 (2019).Article

Andersson，G.，Carlbring，P.，Titov，N。＆Lindefors，N。成人焦虑和情绪障碍的互联网干预：最近荟萃分析的叙述性综述。可以。J、 精神病学64465-470（2019）。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Rozental, A., Andersson, G. & Carlbring, P. In the absence of effects: an individual patient data meta-analysis of non-response and its predictors in internet-based cognitive behavior therapy. Front. Psychol. 10, 589 (2019).Article

Rozental，A.，Andersson，G。＆Carlbring，P。在没有影响的情况下：基于互联网的认知行为疗法中无反应及其预测因子的个体患者数据荟萃分析。正面。心理学。。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Christodoulou, E. et al. A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models. J. Clin. Epidemiol. 110, 12–22 (2019).Article

Christodoulou，E.等人的系统评价显示，对于临床预测模型，机器学习比逻辑回归没有性能优势。J、 临床。流行病。110,12-22（2019）。文章

Google Scholar

谷歌学者

Aafjes-van Doorn, K., Kamsteeg, C., Bate, J. & Aafjes, M. A scoping review of machine learning in psychotherapy research. Psychother. Res. 31, 92–116 (2021).Article

Aafjes van Doorn，K.，Kamsteeg，C.，Bate，J。＆Aafjes，M。心理治疗研究中机器学习的范围界定评论。精神病患者。第31、92-116（2021）号决议。文章

Google Scholar

谷歌学者

Bone, C. et al. Dynamic prediction of psychological treatment outcomes: development and validation of a prediction model using routinely collected symptom data. Lancet Digital Health 3, e231–e240 (2021).Article

Bone，C.等人，《心理治疗结果的动态预测：使用常规收集的症状数据开发和验证预测模型》。柳叶刀数字健康3，e231–e240（2021）。文章

Google Scholar

谷歌学者

Hilbert, K. et al. Predicting cognitive behavioral therapy outcome in the outpatient sector based on clinical routine data: a machine learning approach. Behav. Res. Ther. 124, 103530 (2020).Article

Hilbert，K.等人。基于临床常规数据预测门诊部门的认知行为治疗结果：机器学习方法。行为。Res.Ther。124103530（2020）。文章

Google Scholar

谷歌学者

Beard, J. I. L. & Delgadillo, J. Early response to psychological therapy as a predictor of depression and anxiety treatment outcomes: a systematic review and meta-analysis. Depress. Anxiety 36, 866–878 (2019).Article

。按下按钮。焦虑36866-878（2019）。文章

Google Scholar

谷歌学者

Szegedi, A. et al. Early improvement in the first 2 weeks as a predictor of treatment outcome in patients with major depressive disorder: a meta-analysis including 6562 patients. J. Clin. Psychiatry 70, 344 (2009).Article

。J、 临床。精神病学70344（2009）。文章

Google Scholar

谷歌学者

Barkham, M., De Jong, K., Delgadillo, J. & Lutz, W. Routine outcome monitoring (ROM) and feedback: research review and recommendations. Psychother. Res. 33, 841–855 (2023).Article

Barkham，M.，De Jong，K.，Delgadillo，J。＆Lutz，W。常规结果监测（ROM）和反馈：研究综述和建议。精神病患者。。文章

Google Scholar

谷歌学者

Forsell, E. et al. Predicting treatment failure in regular care internet-delivered cognitive behavior therapy for depression and anxiety using only weekly symptom measures. J. Consult. Clin. Psychol. 88, 311–321 (2020).Article

Forsell，E.等人预测常规护理中的治疗失败互联网提供的认知行为疗法仅使用每周症状测量来治疗抑郁症和焦虑症。J、 咨询。。心理学。88311-321（2020）。文章

Google Scholar

谷歌学者

Hoogendoorn, M., Berger, T., Schulz, A., Stolz, T. & Szolovits, P. Predicting social anxiety treatment outcome based on therapeutic email conversations. IEEE J. Biomed. Health Inform. 21, 1449–1459 (2017).Article

Hoogendoorn，M.，Berger，T.，Schulz，A.，Stolz，T。＆Szolovits，P。基于治疗性电子邮件对话预测社交焦虑治疗结果。IEEE J.生物医学。健康信息。211449-1459（2017）。文章

Google Scholar

谷歌学者

Bennemann, B., Schwartz, B., Giesemann, J. & Lutz, W. Predicting patients who will drop out of out-patient psychotherapy using machine learning algorithms. Br. J. Psychiatry 220, 192–201 (2022).Article

Bennemann，B.，Schwartz，B.，Giesemann，J。＆Lutz，W。使用机器学习算法预测将退出门诊心理治疗的患者。Br.J.精神病学220192-201（2022）。文章

Google Scholar

谷歌学者

Eisenberg, J. M. & Hershey, J. C. Derived thresholds: determining the diagnostic probabilities at which clinicians initiate testing and treatment. Med. Decis. Mak. 3, 155–168 (1983).Article

Eisenberg，J.M。＆Hershey，J.C。衍生阈值：确定临床医生开始测试和治疗的诊断概率。医学博士。马克。3155-168（1983）。文章

Google Scholar

谷歌学者

Forsell, E., Jernelöv, S., Blom, K. & Kaldo, V. Clinically sufficient classification accuracy and key predictors of treatment failure in a randomized controlled trial of internet-delivered cognitive behavior therapy for insomnia. Internet Interv. 100554 https://doi.org/10.1016/j.invent.2022.100554 (2022).DeMasi, O., Kording, K.

Forsell，E.，Jernelöv，S.，Blom，K。＆Kaldo，v。在一项针对失眠的互联网认知行为疗法的随机对照试验中，临床上足够的分类准确性和治疗失败的关键预测因素。互联网互联网。100554https://doi.org/10.1016/j.invent.2022.100554（2022年）。DeMasi，O.，Kording，K。

& Recht, B. Meaningless comparisons lead to false optimism in medical machine learning. PLOS ONE 12, e0184604 (2017).Article .

&Recht，B。无意义的比较会导致医学机器学习中的错误乐观。PLOS ONE 12，e0184604（2017）。文章。

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Flint, C. et al. Systematic misestimation of machine learning performance in neuroimaging studies of depression. Neuropsychopharmacol 46, 1510–1517 (2021).Article

Flint，C.等人。抑郁症神经影像学研究中机器学习表现的系统错误估计。。文章

Google Scholar

谷歌学者

Makridakis, S., Spiliotis, E. & Assimakopoulos, V. Statistical and machine learning forecasting methods: concerns and ways forward. PLOS ONE 13, e0194889 (2018).Article

Makridakis，S.，Spiliotis，E。和Assimakopoulos，V。统计和机器学习预测方法：关注点和前进方向。PLOS ONE 13，e0194889（2018）。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Hedman, E. et al. Effectiveness of Internet-based cognitive behaviour therapy for panic disorder in routine psychiatric care. Acta Psychiatr. Scandinavica 128, 457–467 (2013).Article

Hedman，E.等人。基于互联网的认知行为疗法对常规精神病护理中惊恐障碍的有效性。精神病学学报。斯堪的纳维亚128457–467（2013）。文章

Google Scholar

谷歌学者

Hedman, E. et al. Effectiveness of Internet-based cognitive behaviour therapy for depression in routine psychiatric care. J. Affect. Disord. 155, 49–58 (2014).Article

Hedman，E.等人。基于互联网的认知行为疗法在常规精神病护理中对抑郁症的有效性。J、 影响。混乱。155,49-58（2014）。文章

Google Scholar

谷歌学者

El Alaoui, S. et al. Effectiveness of Internet-based cognitive–behavior therapy for social anxiety disorder in clinical psychiatry. J. Consult. Clin. Psychol. 83, 902–914 (2015).Article

El Alaoui，S.等人。基于互联网的认知行为疗法在临床精神病学中对社交焦虑症的有效性。J、 咨询。。心理学。83902-914（2015）。文章

Google Scholar

谷歌学者

Montgomery, S. A. & Asberg, M. A new depression scale designed to be sensitive to change. Br. J. Psychiatry 134, 382–389 (1979).Article

蒙哥马利（Montgomery，S.A.）和阿斯伯格（Asberg，M.）。一种新的抑郁症量表，旨在对变化敏感。Br.J.精神病学134382-389（1979）。文章

Google Scholar

谷歌学者

Houck, P. R., Spiegel, D. A., Shear, M. K. & Rucci, P. Reliability of the self-report version of the panic disorder severity scale. Depress. Anxiety 15, 183–185 (2002).Article

Houck，P.R.，Spiegel，D.A.，Shear，M.K。＆Rucci，P。恐慌症严重程度量表自我报告版本的可靠性。按下按钮。。文章

Google Scholar

谷歌学者

Fresco, D. M. et al. The Liebowitz Social Anxiety Scale: a comparison of the psychometric properties of self-report and clinician-administered formats. Psychol. Med. 31, 1025–1035 (2001).Article

Fresco，D.M.等人，《Liebowitz社交焦虑量表：自我报告和临床医生管理格式的心理测量特性比较》。心理学。医学311025-1035（2001）。文章

Google Scholar

谷歌学者

Fantino, B. & Moore, N. The self-reported Montgomery-Åsberg depression rating scale is a useful evaluative tool in major depressive disorder. BMC Psychiatry 9, 26 (2009).Article

Fantino，B。＆Moore，N。自我报告的蒙哥马利-Åsberg抑郁评定量表是重性抑郁障碍的有用评估工具。BMC精神病学9,26（2009）。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Furukawa, T. A. et al. Evidence-based guidelines for interpretation of the panic disorder severity scale. Depress Anxiety. 26, 922–929 (2009).Glischinski, M. et al. Liebowitz Social Anxiety Scale (LSAS): optimal cut points for remission and response in a German sample. Clin. Psychol.

Furukawa，T.A.等人，《惊恐障碍严重程度量表解释的循证指南》。抑制焦虑。26922-929（2009）。Glishinski，M。等人。Liebowitz社交焦虑量表（LSAS）：德国样本中缓解和反应的最佳切点。。心理学。

Psychother. 25, 465–473 (2018).Article .

精神病患者。25465-473（2018）。文章。

Google Scholar

谷歌学者

Karin, E., Dear, B. F., Heller, G. Z., Gandy, M. & Titov, N. Measurement of symptom change following web-based psychotherapy: statistical characteristics and analytical methods for measuring and interpreting change. JMIR Ment. Health 5, e10200 (2018).Article

。JMIR公司。健康5，e10200（2018）。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Mikolov, T., Chen, K., Corrado, G. & Dean, J. Efficient estimation of word representations in vector space. arXiv:1301.3781 [cs] (2013).Stekhoven, D. J. & Buhlmann, P. MissForest–non-parametric missing value imputation for mixed-type data. Bioinformatics 28, 112–118 (2012).Article

Mikolov，T.，Chen，K.，Corrado，G。＆Dean，J。向量空间中单词表示的有效估计。arXiv:1301.3781【cs】（2013）。Stekhoven，D.J.&Buhlmann，P.MissForest–混合类型数据的非参数缺失值插补。生物信息学28112-118（2012）。文章

Google Scholar

谷歌学者

Pearson, R., Pisner, D., Meyer, B., Shumake, J. & Beevers, C. G. A machine learning ensemble to predict treatment outcomes following an Internet intervention for depression. Psychol. Med. 1–12 https://doi.org/10.1017/S003329171800315X (2018).Moons, K. G. M., Donders, R. A. R. T., Stijnen, T.

Pearson，R.，Pisner，D.，Meyer，B.，Shumake，J.＆Beevers，C.G.是一个机器学习集合，用于预测互联网干预抑郁症后的治疗结果。心理学。医学1-12https://doi.org/10.1017/S003329171800315X（2018年）。穆斯（Moons），K.G.M.，唐德斯（Donders），R.A.R.T.，施蒂宁（Stijnen）。

& Harrell, F. E. Using the outcome for imputation of missing predictor values was preferred. J. Clin. Epidemiol. 59, 1092–1101 (2006).Article .

&Harrell，F.E。首选使用结果来估算缺失的预测值。J、 临床。流行病。591092-1101（2006）。文章。

Google Scholar

谷歌学者

van Ginkel, J. R., Linting, M., Rippe, R. C. A. & van der Voort, A. Rebutting existing misconceptions about multiple imputation as a method for handling missing data. J. Personal. Assess. 102, 297–308 (2020).Article

van Ginkel，J.R.，Linting，M.，Rippe，R.C.A。和van der Voort，A。反驳了关于多重插补作为处理缺失数据的方法的现有误解。J、 个人。评估。102297-308（2020）。文章

Google Scholar

谷歌学者

Pedregosa, F. et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011).

Pedregosa，F.等人，《Scikit learn：Python中的机器学习》。J、 马赫。学习。第122825-2830号决议（2011年）。

Google Scholar

谷歌学者

Hentati Isacsson, N. Software for Methodological choices and clinical usefulness for machine learning predictions of outcome in Internet-based cognitive behavioral therapy. (2024).Pechenizkiy, M., Tsymbal, A. & Puuronen, S. PCA-based feature transformation for classification: issues in medical diagnostics.

Hentati Isacsson，N。基于互联网的认知行为疗法中机器学习预测结果的方法选择和临床有用性软件。（2024年）。Pechenizkiy，M.，Tsymbal，A。＆Puuronen，S。基于PCA的分类特征转换：医学诊断中的问题。

In Proc. 17th IEEE Symposium on Computer-Based Medical Systems 535–540 (IEEE Comput. Soc, Bethesda, MD, USA). https://doi.org/10.1109/CBMS.2004.1311770 (2004).Chien, I. et al. A machine learning approach to understanding patterns of engagement with internet-delivered mental health interventions. JAMA Netw.

在过程中。第17届IEEE基于计算机的医疗系统研讨会535-540（IEEE Comput.Soc，Bethesda，MD，USA）。https://doi.org/10.1109/CBMS.2004.1311770（2004年）。Chien，I.等人。一种机器学习方法，用于理解互联网提供的心理健康干预的参与模式。JAMA网络。

Open 3, e2010791 (2020).Article .

开放3，e2010791（2020）。文章。

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Boman, M. et al. Learning machines in Internet-delivered psychological treatment. Prog. Artif. Intell. 8, 475–485 (2019).Article

Boman，M.等人，《互联网中的学习机器提供了心理治疗》。程序。人工制品。因特尔。8475-485（2019）。文章

Google Scholar

谷歌学者

Torous, J. & Walker, R. Leveraging digital health and machine learning toward reducing suicide-from panacea to practical tool. JAMA Psychiatry 76, 999–1000 (2019).Article

Torous，J.＆Walker，R.利用数字健康和机器学习将自杀从灵丹妙药减少到实用工具。JAMA精神病学76999-1000（2019）。文章

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Schibbye, P. et al. Using early change to predict outcome in cognitive behaviour therapy: exploring timeframe, calculation method, and differences of disorder-specific versus general measures. PLoS ONE 9, e100614 (2014).Webb, C. A. et al. Personalized prognostic prediction of treatment outcome for depressed patients in a naturalistic psychiatric hospital setting: A comparison of machine learning approaches.

Schibbye，P。等人。使用早期变化预测认知行为疗法的结果：探索时间范围，计算方法以及特定疾病与一般措施的差异。PLoS ONE 9，e100614（2014）。Webb，C.A.等人，《自然精神病医院环境中抑郁症患者治疗结果的个性化预后预测：机器学习方法的比较》。

J. Consult. Clin. Psychol. 88, 25–38 (2020).Article .

J、 咨询。。心理学。88,25-38（2020）。文章。

PubMed Central

公共医学中心

Google Scholar

谷歌学者

Download referencesAcknowledgementsThis work was mainly supported by The Swedish Research Council (VR), The Erling Persson family foundation (EP-Stiftelsen), and The Swedish ALF agreement between the Swedish government and the county councils, with additional funding by the Swedish Foundation for Strategic Research (SSF), Psykiatrifonden, and Thuring’s Foundation.

下载参考文献致谢这项工作主要得到了瑞典研究委员会（VR），二灵佩尔松家庭基金会（EP Stiftelsen）以及瑞典政府与县议会之间的瑞典ALF协议的支持，并得到了瑞典战略研究基金会（SSF），Psykiatrifonden和Thuring基金会的额外资助。

The funding sources were not involved in any part of the study.FundingOpen access funding provided by Karolinska Institute.Author informationAuthors and AffiliationsCentre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, & Stockholm Health Care Services, Region Stockholm, SwedenNils Hentati Isacsson, Fehmi Ben Abdesslem, Erik Forsell & Viktor KaldoDepartment of Computer Science, RISE Research Institutes of Sweden, Stockholm, SwedenFehmi Ben AbdesslemDivision of Psychiatry, University College London, London, UKFehmi Ben Abdesslem & Magnus BomanDepartment of Medicine Solna, Clinical Epidemiology Division, Karolinska Institutet, Stockholm, SwedenMagnus BomanDepartment of Psychology, Faculty of Health and Life Sciences, Linnaeus University, Växjö, SwedenViktor KaldoAuthorsNils Hentati IsacssonView author publicationsYou can also search for this author in.

资金来源未参与研究的任何部分。资金由卡罗林斯卡研究所提供的开放获取资金。作者信息瑞典卡罗琳学院临床神经科学系精神病学研究的作者和附属机构，斯德哥尔摩地区，斯维登尼尔·亨塔蒂·艾萨森，费米·本·阿卜杜斯勒姆，埃里克·福塞尔和维克托·卡尔多瑞典RISE研究所计算机科学系，斯德哥尔摩，斯维登费米·本·阿卜杜斯勒姆，伦敦大学学院精神病学系，UKFehmi-Ben-Abdesslem＆Magnus-BomanDepartment of Medicine Solna，临床流行病学系，卡罗琳学院，斯德哥尔摩，斯维登马格努斯·鲍曼心理学系生命科学，林奈大学，Växjö，SwedenViktor KaldoAuthorsNils Hentati IsacssonView作者出版物您也可以在中搜索这位作者。

PubMed Google ScholarFehmi Ben AbdesslemView author publicationsYou can also search for this author in

PubMed Google ScholarFehmi Ben AbdesslemView作者出版物您也可以在

PubMed Google ScholarErik ForsellView author publicationsYou can also search for this author in

PubMed Google ScholarMagnus BomanView author publicationsYou can also search for this author in

PubMed Google ScholarMagnus BomanView作者出版物您也可以在

PubMed Google ScholarViktor KaldoView author publicationsYou can also search for this author in

PubMed Google ScholarContributionsAll authors contributed extensively to the work presented in this paper. N.H.I., F.B.A., E.F., M.B., and V.K. were involved in conceptualization and design, with N.H.I and V.K. leading. N.H.I. leads data management with support from F.B.A., E.F., M.B., V.K.

PubMed谷歌学术贡献所有作者对本文介绍的工作做出了广泛贡献。N、 H.I.，F.B.A.，E.F.，M.B.和V.K.参与了概念化和设计，N.H.I和V.K.领先。N、 H.I.在F.B.A.，E.F.，M.B.，V.K.的支持下领导数据管理。

N.H.I. lead the methodological development and F.B.A., E.F., M.B., V.K. supported in the research investigation. N.H.I. conducted all analyses. N.H.I., F.B.A., E.F., M.B., and V.K. contributed to writing and reviewing the paper with N.H.I leading.Corresponding authorCorrespondence to.

N、 H.I.领导方法学发展，F.B.A.，E.F.，M.B.，V.K.在研究调查中得到支持。N、 H.I.进行了所有分析。N、 H.I.，F.B.A.，E.F.，M.B。和V.K.在N.H.I的领导下为撰写和审阅论文做出了贡献。对应作者对应。

Nils Hentati Isacsson.Ethics declarations

尼尔斯·亨塔蒂·艾萨克森。道德宣言

Competing interests

相互竞争的利益

The authors declare no competing interests.

作者声明没有利益冲突。

Peer review

同行评审

Peer review information

同行评审信息

Communications Medicine thanks Brian Schwartz and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available

通信医学感谢Brian Schwartz和另一位匿名审稿人对这项工作的同行评审所做的贡献。同行评审文件可用

Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Supplementary informationPeer Review FileSupplementary InformationDescription of Additional Supplementary FilesSupplementary Data 1Supplementary Data 2Supplementary Data 3Supplementary software 1Reporting SummaryRights and permissions.

Additional informationPublisher的注释Springer Nature在已发布地图和机构隶属关系中的管辖权主张方面保持中立。补充信息同行评审文件补充信息其他补充文件的描述补充数据1补充数据2补充数据3补充软件1报告摘要权限。

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.

开放获取本文是根据知识共享署名4.0国际许可证授权的，该许可证允许以任何媒体或格式使用，共享，改编，分发和复制，只要您对原始作者和来源给予适当的信任，提供知识共享许可证的链接，并指出是否进行了更改。

The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

本文中的图像或其他第三方材料包含在文章的知识共享许可中，除非在材料的信用额度中另有说明。如果材料未包含在文章的知识共享许可证中，并且您的预期用途未被法律法规允许或超出允许的用途，则您需要直接获得版权所有者的许可。

To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/..

要查看此许可证的副本，请访问http://creativecommons.org/licenses/by/4.0/..

Reprints and permissionsAbout this articleCite this articleHentati Isacsson, N., Ben Abdesslem, F., Forsell, E. et al. Methodological choices and clinical usefulness for machine learning predictions of outcome in Internet-based cognitive behavioural therapy.

。

Commun Med 4, 196 (2024). https://doi.org/10.1038/s43856-024-00626-4Download citationReceived: 25 July 2023Accepted: 03 October 2024Published: 10 October 2024DOI: https://doi.org/10.1038/s43856-024-00626-4Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard.

Commun Med 4196（2024）。https://doi.org/10.1038/s43856-024-00626-4Download引文接收日期：2023年7月25日接收日期：2024年10月3日发布日期：2024年10月10日OI：https://doi.org/10.1038/s43856-024-00626-4Share本文与您共享以下链接的任何人都可以阅读此内容：获取可共享链接对不起，本文目前没有可共享的链接。复制到剪贴板。

Provided by the Springer Nature SharedIt content-sharing initiative

由Springer Nature SharedIt内容共享计划提供