{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:58Z","timestamp":1785545038248,"version":"3.56.0"},"reference-count":32,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Form Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Depression in older adults is often underdiagnosed due to atypical symptom presentation and generational stigma, leading to delayed intervention. Early identification of individuals at risk of developing elevated depressive symptoms is therefore critical, but traditional approaches show limited predictive accuracy. To date, no study has applied machine learning (ML) models to predict clinically significant depressive symptoms at 2-year follow-up in older adults in the United Kingdom using data from the English Longitudinal Study of Ageing (ELSA). Moreover, the impact of encoding strategies for categorical health care variables has not been examined.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop and evaluate ML models to predict the clinically significant depressive symptoms at 2-year follow-up in older adults using ELSA data. We further compared ordinal and one-hot encoding strategies across different ML architectures and identified key predictors of depressive symptoms at follow-up.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n Data were drawn from 4 consecutive waves of ELSA, including participants aged \u226550 years without significant depressive symptoms at the baseline wave (waves 6\u20109). Clinically significant depressive symptoms were defined as 8-item Center for Epidemiologic Studies Depression Scale (CES-D 8) scores of \u22654 at the subsequent wave (waves 7\u201010). Over 120 features spanning sociodemographic, psychological, and health-related domains were analyzed. Eight ML models were applied, including tree-based ensembles, deep learning architectures for tabular data, distance-based methods, probabilistic methods, and linear methods. Model performance was assessed using the area under the receiver operating characteristic curve (AUROC) and\n F<\/jats:italic>\n 1<\/jats:sub>\n -score. Model interpretability was examined using Shapley additive explanations (SHAP). Sensitivity analyses assessed the robustness of results across alternative CES-D 8 thresholds (\u22653, \u22654, and \u22655) and encoding strategies.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Across waves, the best-performing models achieved mean AUROC scores of 0.72\u20100.73, with a peak of 0.75 in the highest-performing wave. Ordinal encoding consistently outperformed one-hot encoding across all ML models, yielding improvements in AUROCs and\n F<\/jats:italic>\n 1<\/jats:sub>\n -scores, with the greatest increase in tree-based methods. SHAP consistently identified loneliness, sleep disturbances, and low social engagement as strong predictors of elevated depressive symptoms at follow-up. Sensitivity analyses across CES-D 8 thresholds demonstrated robust feature importance, with AUROCs ranging from 0.67 to 0.82. Traditional ML models (random forest, extreme gradient boosting, and support vector machines) generally achieved higher performance than the deep learning models for this task.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Our findings demonstrate the feasibility of predicting clinically significant depressive symptoms at 2-year follow-up in UK older adults, with moderate accuracy. Ordinal encoding demonstrates superior performance for health care datasets with inherently ordered categorical features. The identification of consistent risk factors highlights opportunities for developing targeted clinical screening tools and preventive interventions. This study provides new evidence on depressive symptom prediction in the UK context, leveraging longitudinal data from ELSA, and contributes to advancing digital mental health research for aging populations.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/84744","type":"journal-article","created":{"date-parts":[[2026,7,2]],"date-time":"2026-07-02T15:12:03Z","timestamp":1783005123000},"page":"e84744-e84744","source":"Crossref","is-referenced-by-count":0,"title":["Prediction of Clinically Significant Depressive Symptoms at 2-Year Follow-Up in Older Adults: Machine Learning Study Using the English Longitudinal Study of Ageing"],"prefix":"10.2196","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-8716-5749","authenticated-orcid":false,"given":"Bahar","family":"Khorram","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9801-183X","authenticated-orcid":false,"given":"Ramin","family":"Nilforooshan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8591-9638","authenticated-orcid":false,"given":"Payam","family":"Barnaghi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5870-4030","authenticated-orcid":false,"given":"Samaneh","family":"Kouchaki","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,7,2]]},"reference":[{"key":"R1","unstructured":"Depression. World Health Organization. 2023. URL: https:\/\/www.who.int\/news-room\/fact-sheets\/detail\/depression [Accessed 17-07-2024]"},{"key":"R2","unstructured":"McManus S Bebbington P Jenkins R Brugha T . Mental health and wellbeing in England: adult psychiatric morbidity survey 2014. NHS Digital; 2016. URL: https:\/\/digital.nhs.uk\/data-and-information\/publications\/statistical\/adult-psychiatric-morbidity-survey\/adult-psychiatric-morbidity-survey-survey-of-mental-health-and-wellbeing-england-2014 [Accessed 15-06-2026]"},{"issue":"1-2","key":"R3","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1016\/j.jad.2007.06.005","article-title":"Risk factors for anxiety and depression in the elderly: a review","volume":"106","author":"Vink","journal-title":"J Affect Disord"},{"issue":"Suppl_3","key":"R4","doi-asserted-by":"publisher","first-page":"466","DOI":"10.1093\/geront\/gnw162.1870","article-title":"Suicidal thoughts and attempts among Chinese and Indian older adults: findings from WHO-SAGE","volume":"56","author":"Nguyen","journal-title":"Gerontologist"},{"key":"R5","doi-asserted-by":"publisher","DOI":"10.1590\/S1980-220X2019026603694","article-title":"Suicide in the elderly: an epidemiologic study","volume":"55","author":"Santos","journal-title":"Rev Esc Enferm USP"},{"key":"R6","unstructured":"The state of ageing 2023-24. Centre for Ageing Better. 2023. URL: https:\/\/ageing-better.org.uk\/our-ageing-population-state-ageing-2023-4 [Accessed 24-09-2024]"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2023.06.031","article-title":"Prediction of depression onset risk among middle-aged and elderly adults using machine learning and Canadian longitudinal study on aging cohort","volume":"339","author":"Song","journal-title":"J Affect Disord"},{"issue":"1","key":"R8","doi-asserted-by":"publisher","first-page":"363","DOI":"10.1146\/annurev.clinpsy.032408.153621","article-title":"Depression in older adults","volume":"5","author":"Fiske","journal-title":"Annu Rev Clin Psychol"},{"key":"R9","doi-asserted-by":"publisher","DOI":"10.1016\/j.psychres.2022.114511","article-title":"Prevalence of depression in older adults: a systematic review and meta-analysis","volume":"311","author":"Hu","journal-title":"Psychiatry Res"},{"issue":"3","key":"R10","doi-asserted-by":"publisher","first-page":"205","DOI":"10.11606\/issn.2176-7262.v53i3p205-214","article-title":"Prevalence and factors associated with depression in the elderly: a cross-sectional study","volume":"53","author":"Leite","journal-title":"Medicina (Ribeir\u00e3o Preto)"},{"issue":"1","key":"R11","doi-asserted-by":"publisher","first-page":"281","DOI":"10.1007\/s11280-021-00992-2","article-title":"Explainable depression detection with multi-aspect features using a hybrid deep learning model on social media","volume":"25","author":"Zogan","journal-title":"World Wide Web"},{"key":"R12","doi-asserted-by":"crossref","unstructured":"Orabi AH Buddhitha P Orabi MH Inkpen D . Deep learning for depression detection of Twitter users. Presented at: Proceedings of the Fifth Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic; Jun 5, 2018:88-97; New Orleans, LA. URL: http:\/\/aclweb.org\/anthology\/W18-06 [Accessed 12-06-2026] [doi: 10.18653\/v1\/W18-0609]","DOI":"10.18653\/v1\/W18-0609"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2021.104499","article-title":"A textual-based featuring approach for depression detection using machine learning classifiers and social media texts","volume":"135","author":"Chiong","journal-title":"Comput Biol Med"},{"issue":"3","key":"R14","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1176\/appi.prcp.20200031","article-title":"A machine learning approach to predicting new-onset depression in a military population","volume":"3","author":"Sampson","journal-title":"Psychiatr Res Clin Pract"},{"issue":"11-12","key":"R15","doi-asserted-by":"publisher","first-page":"1441","DOI":"10.1093\/milmed\/usab364","article-title":"Resilience and depression in military service: evidence from the national longitudinal study of adolescent to adult health (add health)","volume":"187","author":"Orak","journal-title":"Mil Med"},{"issue":"11","key":"R16","doi-asserted-by":"publisher","first-page":"519","DOI":"10.3390\/info11110519","article-title":"Random forest with sampling techniques for handling imbalanced prediction of university student depression","volume":"11","author":"Sawangarreerak","journal-title":"Information"},{"issue":"6","key":"R17","doi-asserted-by":"publisher","first-page":"120","DOI":"10.3390\/asi5060120","article-title":"Prediction of depression for undergraduate students based on imbalanced data by using data mining techniques","volume":"5","author":"Narkbunnum","journal-title":"ASI"},{"issue":"16","key":"R18","doi-asserted-by":"publisher","first-page":"2895","DOI":"10.1017\/S0033291720001579","article-title":"Prediction of depression cases, incidence, and chronicity in a large occupational cohort using machine learning techniques: an analysis of the ELSA-Brasil study","volume":"51","author":"Librenza-Garcia","journal-title":"Psychol Med"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.1016\/j.jad.2020.12.160","article-title":"Use of machine learning approach to predict depression in the elderly in China: a longitudinal study","volume":"282","author":"Su","journal-title":"J Affect Disord"},{"issue":"1","key":"R20","doi-asserted-by":"publisher","DOI":"10.1186\/s12888-024-06299-6","article-title":"Using machine learning to predict the probability of incident 2-year depression in older adults with chronic diseases: a retrospective cohort study","volume":"24","author":"Zheng","journal-title":"BMC Psychiatry"},{"issue":"12","key":"R21","doi-asserted-by":"publisher","first-page":"e141","DOI":"10.1002\/gps.4673","article-title":"Frailty and incident depression in community-dwelling older people: results from the ELSA study","volume":"32","author":"Veronese","journal-title":"Int J Geriatr Psychiatry"},{"key":"R22","doi-asserted-by":"publisher","DOI":"10.1016\/j.psyneuen.2015.11.005","article-title":"Higher serum dehydroepiandrosterone sulfate protects against the onset of depression in the elderly: Findings from the English Longitudinal Study of Aging (ELSA)","volume":"64","author":"Souza-Teodoro","journal-title":"Psychoneuroendocrinology"},{"issue":"2","key":"R23","doi-asserted-by":"publisher","first-page":"353","DOI":"10.1007\/s00127-021-02155-0","article-title":"The dynamic relationship between hearing loss, quality of life, socioeconomic position and depression and the impact of hearing aids: answers from the English Longitudinal Study of Ageing (ELSA)","volume":"57","author":"Tsimpida","journal-title":"Soc Psychiatry Psychiatr Epidemiol"},{"issue":"6","key":"R24","doi-asserted-by":"publisher","first-page":"1640","DOI":"10.1093\/ije\/dys168","article-title":"Cohort profile: the English longitudinal study of ageing","volume":"42","author":"Steptoe","journal-title":"Int J Epidemiol"},{"issue":"7","key":"R25","doi-asserted-by":"publisher","first-page":"2146","DOI":"10.1177\/10731911221138930","article-title":"The eight-item center for epidemiological studies depression scale in the English longitudinal study of aging: longitudinal and gender invariance, sum score models, and external associations","volume":"30","author":"Schlechter","journal-title":"Assessment"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpsychores.2018.07.009","article-title":"Depressive symptoms predict incident chronic disease burden 10\u202fyears later: findings from the English longitudinal study of ageing (ELSA)","volume":"113","author":"Poole","journal-title":"J Psychosom Res"},{"issue":"1","key":"R27","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1207\/s15327752jpa4901_13","article-title":"The satisfaction with life scale","volume":"49","author":"Diener","journal-title":"J Pers Assess"},{"issue":"6","key":"R28","doi-asserted-by":"publisher","first-page":"655","DOI":"10.1177\/0164027504268574","article-title":"A short scale for measuring loneliness in large surveys: results from two population-based studies","volume":"26","author":"Hughes","journal-title":"Res Aging"},{"issue":"8","key":"R29","doi-asserted-by":"publisher","first-page":"6679","DOI":"10.1609\/aaai.v35i8.16826","article-title":"TabNet: attentive interpretable tabular learning","volume":"35","author":"Arik","journal-title":"Proc AAAI Conf Artif Intell"},{"key":"R30","unstructured":"Huang X Khetan A Cvitkovic M Karnin Z . TabTransformer: tabular data modeling using contextual embeddings. arXiv. Preprint posted online on Dec 11, 2020. [doi: 10.48550\/arXiv.2012.06678]"},{"issue":"4","key":"R31","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbae344","article-title":"A novel deep machine learning algorithm with dimensionality and size reduction approaches for feature elimination: thyroid cancer diagnoses with randomly missing data","volume":"25","author":"Tutsoy","journal-title":"Brief Bioinform"},{"key":"R32","unstructured":"English longitudinal study of ageing. UK Data Service. 1998. URL: https:\/\/beta.ukdataservice.ac.uk\/datacatalogue\/series\/series?id=200011 [Accessed 19-06-2026]"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e84744","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:25:57Z","timestamp":1785536757000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e84744"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,2]]},"references-count":32,"URL":"https:\/\/doi.org\/10.2196\/84744","relation":{},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,2]]},"article-number":"v10i1-2e84744"}}