{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:56Z","timestamp":1785545036035,"version":"3.56.0"},"reference-count":60,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR AI"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Anxiety disorders are highly prevalent among adults with autism, with 20%\u201065% experiencing at least one diagnosable anxiety disorder. While mindfulness-based interventions have demonstrated efficacy for anxiety reduction, treatment response varies considerably across individuals. Machine learning approaches offer potential for identifying who is most likely to benefit from smartphone-based mindfulness interventions, enabling personalized treatment recommendations.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and evaluate machine learning models to predict individual treatment response to a smartphone-based mindfulness intervention for adults with autism. We identified baseline characteristics that distinguish responders from nonresponders and explored few-shot learning with large language models (LLMs) as a complementary approach for low-data clinical prediction.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We conducted a secondary analysis of a randomized controlled trial comparing a 6-week smartphone-based mindfulness intervention with a waitlist control group in adults with autism. Among 73 participants who completed the intervention, we defined responders as those achieving a \u22657-point reduction in State-Trait Anxiety Inventory state anxiety scores. Baseline predictors included demographic variables; autism trait measures; and self-report questionnaires assessing anxiety symptoms, perceived stress, affect, and mindfulness. To determine which machine learning model was most predictive of response, we trained 6 different models (logistic regression, random forest, extreme gradient boosting [XGBoost], tabular data network [TabNet], TabICL, and Tabular Prior-Data Fitted Network [TabPFN]) using nested 10-fold cross-validation with inner 5-fold cross-validation for hyperparameter tuning and evaluated GPT-4o few-shot learning with tokenized features at 20 to 70 shots.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Random forest achieved the highest predictive performance for state anxiety response (area under the curve [AUC] 0.79, 95% CI 0.66\u20100.91), followed by TabPFN (AUC 0.78, 95% CI 0.64\u20100.94) and logistic regression (AUC 0.77, 95% CI 0.73\u20100.81). Higher baseline state anxiety (standardized \u03b2 coefficient=1.20,\n P<\/jats:italic>\n <.001) predicted better treatment response, while higher Autism Spectrum Quotient at baseline (standardized \u03b2 coefficient=\u22120.17,\n P<\/jats:italic>\n =.001), older age (standardized \u03b2 coefficient=\u22120.18,\n P<\/jats:italic>\n =.02), and lower childhood pretend play scores (standardized \u03b2 coefficient=\u22120.93,\n P<\/jats:italic>\n =.007) were associated with poorer response. Few-shot learning with 7-feature tokenization achieved an accuracy of 0.867 at 70 shots, compared to an accuracy of 0.733 for random forest. Prediction of trait anxiety changes was substantially weaker (AUCs 0.46\u20100.68), likely reflecting the inherent stability of this personality dimension.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Machine learning models successfully identified baseline characteristics predicting state anxiety response to a smartphone-based mindfulness intervention in adults with autism. Few-shot learning with LLMs demonstrated superior performance to traditional machine learning when provided with compact, high-signal feature representations, offering a promising approach for clinical prediction in small-sample settings. These findings demonstrate the feasibility of precision psychiatry in digital mental health interventions for adults with autism. As online mental health interventions become ubiquitous, patients and clinicians can know whether a particular intervention is more or less likely to benefit an individual patient.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/89054","type":"journal-article","created":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T11:10:07Z","timestamp":1781262607000},"page":"e89054-e89054","source":"Crossref","is-referenced-by-count":0,"title":["Large Language Model Few-Shot Learning for Predicting Individual Treatment Response to Smartphone-Based Mindfulness in Autistic Adults With Anxiety: Secondary Analysis of a Randomized Controlled Trial"],"prefix":"10.2196","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4185-6564","authenticated-orcid":false,"given":"Gun","family":"Ahn","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3912-3505","authenticated-orcid":false,"given":"Cindy E","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-0233-5822","authenticated-orcid":false,"given":"Aixin","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-5763-1935","authenticated-orcid":false,"given":"Wonchang","family":"Choi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-0951-4558","authenticated-orcid":false,"given":"Seoin","family":"Ahn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Clark","family":"Roberts","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1158-5692","authenticated-orcid":false,"given":"John D E","family":"Gabrieli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,7,23]]},"reference":[{"issue":"11","key":"R1","doi-asserted-by":"publisher","first-page":"2112","DOI":"10.1002\/aur.2808","article-title":"Associations between co-occurring conditions and age of autism diagnosis: implications for mental health training and adult autism research","volume":"15","author":"Jadav","journal-title":"Autism Res"},{"issue":"6","key":"R2","doi-asserted-by":"publisher","first-page":"519","DOI":"10.1016\/j.jpsychores.2015.03.009","article-title":"Mindfulness-based stress reduction for healthy individuals: a meta-analysis","volume":"78","author":"Khoury","journal-title":"J Psychosom Res"},{"issue":"6","key":"R3","doi-asserted-by":"publisher","first-page":"1504","DOI":"10.1007\/s12671-025-02558-z","article-title":"Smartphone mindfulness intervention reduces anxiety symptoms and perceived stress in autistic adults: a randomized controlled trial","volume":"16","author":"Li","journal-title":"Mindfulness (N Y)"},{"issue":"2","key":"R4","doi-asserted-by":"publisher","first-page":"15","DOI":"10.47895\/amp.vi0.8140","article-title":"The effectiveness of mindfulness-based interventions versus cognitive behavioral therapy on social anxiety of adolescents: a systematic review and meta-analysis","volume":"59","author":"Dones","journal-title":"Acta Med Philipp"},{"issue":"4","key":"R5","doi-asserted-by":"publisher","first-page":"269","DOI":"10.1159\/000511880","article-title":"A three-arm randomized clinical trial comparing the efficacy of a mindfulness-based intervention with an active comparison group and fluoxetine treatment for adults with generalized anxiety disorder","volume":"90","author":"Costa","journal-title":"Psychother Psychosom"},{"issue":"1","key":"R6","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1001\/jamapsychiatry.2025.3439","article-title":"Uptake, adherence, and attrition in clinical trials of depression and anxiety apps: a systematic review and meta-analysis","volume":"83","author":"Liu","journal-title":"JAMA Psychiatry"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.2196\/44220","article-title":"Experiences of using digital mindfulness-based interventions: rapid scoping review and thematic synthesis","volume":"25","author":"Osborne","journal-title":"J Med Internet Res"},{"issue":"16","key":"R8","doi-asserted-by":"publisher","first-page":"4843","DOI":"10.1017\/S0033291724003118","article-title":"Advancing the personalized advantage index (PAI): a systematic review and application in two large multi-site samples in anxiety disorders","volume":"54","author":"Meinke","journal-title":"Psychol Med"},{"issue":"7","key":"R9","doi-asserted-by":"publisher","first-page":"898","DOI":"10.1176\/ps.2009.60.7.898","article-title":"Dropout from outpatient mental health care in the United States","volume":"60","author":"Olfson","journal-title":"Psychiatr Serv"},{"issue":"1","key":"R10","doi-asserted-by":"publisher","DOI":"10.1186\/s12916-017-0849-x","article-title":"The new field of \u201cprecision psychiatry\u201d","volume":"15","author":"Fernandes","journal-title":"BMC Med"},{"issue":"11","key":"R11","doi-asserted-by":"publisher","first-page":"912","DOI":"10.1002\/da.22302","article-title":"Internet psychotherapy and the future of personalized treatment","volume":"31","author":"Andrews","journal-title":"Depress Anxiety"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpr.2025.102593","article-title":"Machine learning in the prediction of treatment response for emotional disorders: a systematic review and meta-analysis","volume":"120","author":"Curtiss","journal-title":"Clin Psychol Rev"},{"issue":"1","key":"R13","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01360-w","article-title":"Estimation of minimal data sets sizes for machine learning predictions in digital mental health interventions","volume":"7","author":"Zantvoort","journal-title":"NPJ Digit Med"},{"key":"R14","volume":"33","author":"Brown","journal-title":"Adv Neural Inf Process Syst"},{"key":"R15","unstructured":"Hegselmann S Buendia A Lang H Agrawal M Jiang X Sontag D . TabLLM: few-shot classification of tabular data with large language models. In: Proceedings of The 26th International Conference on Artificial Intelligence and Statistics. Vol 206. Proceedings of Machine Learning Research; 2023:5549-5581. URL: https:\/\/proceedings.mlr.press\/v206\/hegselmann23a\/hegselmann23a.pdf [Accessed 12-07-2026]"},{"issue":"3","key":"R16","doi-asserted-by":"publisher","first-page":"706","DOI":"10.1097\/00000539-200003000-00036","article-title":"The visual analog scale allows effective measurement of preoperative anxiety and detection of patients\u2019 anesthetic concerns","volume":"90","author":"Kindler","journal-title":"Anesth Analg"},{"issue":"2","key":"R17","doi-asserted-by":"publisher","first-page":"180","DOI":"10.1093\/bja\/74.2.180","article-title":"Assessment of preoperative anxiety: comparison of measures in patients awaiting surgery for breast cancer","volume":"74","author":"Millar","journal-title":"Br J Anaesth"},{"issue":"2","key":"R18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1159\/000478993","article-title":"State and trait anxiety scores of patients receiving intravitreal injections","volume":"2","author":"Kayikcioglu","journal-title":"Biomed Hub"},{"issue":"3","key":"R19","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1023\/A:1005592401947","volume":"30","author":"Lord","journal-title":"J Autism Dev Disord"},{"issue":"3","key":"R20","doi-asserted-by":"publisher","first-page":"488","DOI":"10.1016\/j.neuron.2018.01.015","article-title":"SPARK: a US cohort of 50,000 families to accelerate autism research","volume":"97","author":"SPARK Consortium","journal-title":"Neuron"},{"issue":"8","key":"R21","doi-asserted-by":"publisher","first-page":"3383","DOI":"10.1007\/s10803-021-05218-y","article-title":"Validation of autism diagnosis and clinical data in the SPARK cohort","volume":"52","author":"Fombonne","journal-title":"J Autism Dev Disord"},{"key":"R22","unstructured":"Constantino JN Gruber CP . Social Responsiveness Scale, Second Edition (SRS-2). Western Psychological Services; 2012."},{"issue":"1","key":"R23","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/a:1005653411471","article-title":"The autism-spectrum quotient (AQ): evidence from Asperger syndrome\/high-functioning autism, males and females, scientists and mathematicians","volume":"31","author":"Baron-Cohen","journal-title":"J Autism Dev Disord"},{"issue":"1","key":"R24","doi-asserted-by":"publisher","DOI":"10.1186\/s13229-021-00445-7","article-title":"The Comprehensive Autistic Trait Inventory (CATI): development and validation of a new measure of autistic traits in the general population","volume":"12","author":"English","journal-title":"Mol Autism"},{"key":"R25","doi-asserted-by":"crossref","unstructured":"Spielberger CD . State-Trait Anxiety Inventory for Adults. 1983. [doi: 10.1037\/t06496-000]","DOI":"10.1037\/t06496-000"},{"issue":"11","key":"R26","doi-asserted-by":"publisher","first-page":"1179","DOI":"10.1016\/j.jclinepi.2010.04.011","article-title":"The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008","volume":"63","author":"Cella","journal-title":"J Clin Epidemiol"},{"key":"R27","unstructured":"Cohen S . Perceived stress in a probability sample of the United States. In: Spacapan S Oskamp S , editors. The Social Psychology of Health. Sage Publications; 1988:31-67."},{"issue":"6","key":"R28","doi-asserted-by":"publisher","first-page":"1063","DOI":"10.1037\/0022-3514.54.6.1063","article-title":"Development and validation of brief measures of positive and negative affect: the PANAS scales","volume":"54","author":"Watson","journal-title":"J Pers Soc Psychol"},{"issue":"4","key":"R29","doi-asserted-by":"publisher","first-page":"822","DOI":"10.1037\/0022-3514.84.4.822","article-title":"The benefits of being present: mindfulness and its role in psychological well-being","volume":"84","author":"Brown","journal-title":"J Pers Soc Psychol"},{"issue":"1","key":"R30","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1177\/1073191105283504","article-title":"Using self-report assessment methods to explore facets of mindfulness","volume":"13","author":"Baer","journal-title":"Assessment"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.eclinm.2021.101042","article-title":"Interoceptive training to target anxiety in autistic adults (ADIE): a single-center, superiority randomized controlled trial","volume":"39","author":"Quadt","journal-title":"EClinicalMedicine"},{"issue":"6","key":"R32","doi-asserted-by":"publisher","first-page":"1425","DOI":"10.1017\/s0033291799001336","article-title":"Recovery rates in generalized anxiety disorder following psychological therapy: an analysis of clinically significant change in the STAI-T across outcome studies since 1990","volume":"29","author":"Fisher","journal-title":"Psychol Med"},{"issue":"1","key":"R33","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","journal-title":"Mach Learn"},{"key":"R34","doi-asserted-by":"crossref","unstructured":"Chen T Guestrin C . XGBoost: a scalable tree boosting system. In: KDD \u201916: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery; 2016:785-794. [doi: 10.1145\/2939672.2939785]","DOI":"10.1145\/2939672.2939785"},{"issue":"8","key":"R35","doi-asserted-by":"crossref","first-page":"6679","DOI":"10.1609\/aaai.v35i8.16826","volume":"35","author":"Arik","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"R36","unstructured":"Qu J Holzm\u00fcller D Varoquaux G Morvan ML . TabICL: a tabular foundation model for in-context learning on large data. In: Proceedings of the 42nd International Conference on Machine Learning. Proceedings of Machine Learning Research; 2025:50817-50847. URL: https:\/\/proceedings.mlr.press\/v267\/qu25d.html [Accessed 12-07-2026]"},{"issue":"8045","key":"R37","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1038\/s41586-024-08328-6","article-title":"Accurate predictions on small data with a tabular foundation model","volume":"637","author":"Hollmann","journal-title":"Nature"},{"issue":"5","key":"R38","doi-asserted-by":"publisher","first-page":"605","DOI":"10.3748\/wjg.v28.i5.605","article-title":"Machine learning models and over-fitting considerations","volume":"28","author":"Charilaou","journal-title":"World J Gastroenterol"},{"key":"R39","unstructured":"Lundberg S Lee SI . A unified approach to interpreting model predictions. In: Proceedings of the 31st International Conference on Neural Information Processing Systems. Vol 30. Curran Associates; 2017:4765-4774. ISBN: 9781510860964"},{"issue":"4","key":"R40","doi-asserted-by":"publisher","first-page":"e231","DOI":"10.1016\/S2589-7500(21)00018-2","article-title":"Dynamic prediction of psychological treatment outcomes: development and validation of a prediction model using routinely collected symptom data","volume":"3","author":"Bone","journal-title":"Lancet Digit Health"},{"issue":"6","key":"R41","doi-asserted-by":"publisher","DOI":"10.1192\/bjo.2024.789","article-title":"Externally validated clinical prediction models for estimating treatment outcomes for patients with a mood, anxiety or psychotic disorder: systematic review and meta-analysis","volume":"10","author":"Burghoorn","journal-title":"BJPsych Open"},{"issue":"3","key":"R42","doi-asserted-by":"publisher","first-page":"231","DOI":"10.1016\/s0887-6185(01)00060-3","article-title":"State and trait anxiety revisited","volume":"15","author":"Endler","journal-title":"J Anxiety Disord"},{"key":"R43","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpr.2020.101928","article-title":"Specificity of trait anxiety in anxiety and depression: meta-analysis of the State-Trait Anxiety Inventory","volume":"82","author":"Knowles","journal-title":"Clin Psychol Rev"},{"issue":"4","key":"R44","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1001\/jamapsychiatry.2020.4364","article-title":"Internet-based cognitive behavioral therapy for depression: a systematic review and individual patient data network meta-analysis","volume":"78","author":"Karyotaki","journal-title":"JAMA Psychiatry"},{"key":"R45","doi-asserted-by":"publisher","DOI":"10.1016\/j.brat.2021.103972","article-title":"The effect of depression and anxiety symptom severity on clinical outcomes and app use in digital mental health treatments: meta-regression of three trials","volume":"147","author":"Mohr","journal-title":"Behav Res Ther"},{"issue":"3","key":"R46","doi-asserted-by":"publisher","first-page":"166","DOI":"10.1089\/acm.2014.0285","article-title":"Decreased symptoms of depression after mindfulness-based stress reduction: potential moderating effects of religiosity, spirituality, trait mindfulness, sex, and age","volume":"21","author":"Greeson","journal-title":"J Altern Complement Med"},{"issue":"8","key":"R47","doi-asserted-by":"publisher","first-page":"434","DOI":"10.1016\/j.brat.2013.04.004","article-title":"Which treatment worked better for whom? Moderators of group cognitive behavioral therapy versus adapted mindfulness based stress reduction for anxiety disorders","volume":"51","author":"Arch","journal-title":"Behav Res Ther"},{"issue":"3","key":"R48","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1136\/ebmental-2022-300464","article-title":"Do mindfulness-based programmes improve the cognitive skills, behaviour and mental health of children and adolescents? An updated meta-analysis of randomised controlled trials","volume":"25","author":"Dunning","journal-title":"Evid Based Ment Health"},{"key":"R49","doi-asserted-by":"publisher","DOI":"10.1016\/j.paid.2015.09.040","article-title":"Mind full of ideas: a meta-analysis of the mindfulness\u2013creativity link","volume":"93","author":"Lebuda","journal-title":"Pers Individ Dif"},{"issue":"4","key":"R50","doi-asserted-by":"publisher","first-page":"1370","DOI":"10.1007\/s10803-024-06643-5","article-title":"Alexithymia, inner thinking patterns, and perceptions of mental health therapy strategies among autistic adults","volume":"56","author":"Mazurek","journal-title":"J Autism Dev Disord"},{"issue":"3","key":"R51","doi-asserted-by":"publisher","first-page":"189","DOI":"10.1002\/(sici)1098-240x(199806)21:3<189::aid-nur2>3.0.co;2-d","article-title":"Imaging ability and effective use of guided imagery","volume":"21","author":"Kwekkeboom","journal-title":"Res Nurs Health"},{"key":"R52","unstructured":"Sprague Z Yin F Rodriguez JD et al. To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning. arXiv. Preprint posted online on May 7, 2025. [doi: 10.48550\/arXiv.2409.12183]"},{"key":"R53","doi-asserted-by":"publisher","DOI":"10.2196\/75030","article-title":"Aiding large language models using clinical scoresheets for neurobehavioral diagnostic classification from text: algorithm development and validation","volume":"4","author":"Lin","journal-title":"JMIR AI"},{"issue":"2","key":"R54","doi-asserted-by":"publisher","first-page":"74","DOI":"10.1038\/nrneurol.2013.278","article-title":"Disentangling the heterogeneity of autism spectrum disorder through genetic findings","volume":"10","author":"Jeste","journal-title":"Nat Rev Neurol"},{"key":"R55","doi-asserted-by":"publisher","DOI":"10.1016\/j.janxdis.2024.102825","article-title":"Which client with generalized anxiety disorder benefits from a mindfulness ecological momentary intervention versus a self-monitoring app? Developing a multivariable machine learning predictive model","volume":"102","author":"Zainal","journal-title":"J Anxiety Disord"},{"key":"R56","doi-asserted-by":"publisher","DOI":"10.2196\/67210","article-title":"Prescriptive predictors of mindfulness ecological momentary intervention for social anxiety disorder: machine learning analysis of randomized controlled trial data","volume":"12","author":"Zainal","journal-title":"JMIR Ment Health"},{"key":"R57","doi-asserted-by":"publisher","DOI":"10.1177\/20552076211060659","article-title":"Predicting therapy outcome in a digital mental health intervention for depression and anxiety: a machine learning approach","volume":"7","author":"Hornstein","journal-title":"Digit Health"},{"issue":"1","key":"R58","doi-asserted-by":"publisher","DOI":"10.1038\/s41398-022-02133-3","article-title":"Predicting remission after internet-delivered psychotherapy in patients with depression using machine learning and multi-modal data","volume":"12","author":"Wallert","journal-title":"Transl Psychiatry"},{"issue":"11","key":"R59","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0272685","article-title":"Deep learning for the prediction of clinical outcomes in internet-delivered CBT for depression and anxiety","volume":"18","author":"Prasad","journal-title":"PLoS One"},{"key":"R60","unstructured":"Gunahn\/ml4mindfulness_autism. GitHub. URL: https:\/\/github.com\/gunahn\/ML4Mindfulness_Autism [Accessed 16-07-2026]"}],"container-title":["JMIR AI"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/ai.jmir.org\/2026\/1\/e89054","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:24:46Z","timestamp":1785536686000},"score":1,"resource":{"primary":{"URL":"https:\/\/ai.jmir.org\/2026\/1\/e89054"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,23]]},"references-count":60,"URL":"https:\/\/doi.org\/10.2196\/89054","relation":{},"ISSN":["2817-1705"],"issn-type":[{"value":"2817-1705","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,23]]},"article-number":"v5i11e89054"}}