{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T03:01:23Z","timestamp":1790218883440,"version":"4.1.0"},"reference-count":31,"publisher":"SAGE Publications","license":[{"start":{"date-parts":[[2026,9,20]],"date-time":"2026-09-20T00:00:00Z","timestamp":1789862400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Population Health Management"],"abstract":"<jats:p>Effective care management programs depend on accurately identifying patients who are likely to account for a disproportionate share of future health care expenditures. Unlike traditional risk-adjustment models, which emphasize population-level predictive accuracy, care management targeting is fundamentally a top-k decision problem focused on maximizing expenditure capture. This study used 2000\u20132023 MEPS data to compare regression-, classification-, and learning-to-rank machine learning approaches for predicting next-year health care expenditures. Primary outcomes included precision-at-k, cost capture-at-k, weighted cost capture-at-k, and AUC. Models were benchmarked against a persistence-based floor baseline using prior-year expenditures and an oracle upper bound representing perfect future information. Classification-based models consistently achieved the strongest operational targeting performance. Neural network and XGBoost classifiers outperformed regression and ranking approaches in identifying future high-cost individuals and maximizing expenditure capture. Differences in cost capture were statistically significant across yearly evaluations. More complex modeling extensions provided little additional benefit. Prior-year utilization and expenditure variables contributed substantially to predictive accuracy; however, models retained approximately 80%\u201390% of full-model performance after removal of these features, indicating meaningful predictive signal from sociodemographic characteristics, chronic disease burden, functional limitations, and self-reported health status. For care management programs operating under fixed enrollment constraints, classification-based prediction approaches appear better aligned with operational objectives than regression or ranking. Additional performance gains are more likely to arise from richer clinical and longitudinal data than from increased model complexity alone. Future AI-enabled care management systems may further improve resource allocation by integrating cost-risk prediction with estimates of patient engagement and intervention impactability.<\/jats:p>","DOI":"10.1177\/19427891261488894","type":"journal-article","created":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T05:16:15Z","timestamp":1789967775000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["Optimizing Care Management Targeting: A Machine Learning Comparison of Regression, Classification, and Ranking Approaches for Identifying Future High-Cost Patients"],"prefix":"10.1177","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-8560-6552","authenticated-orcid":false,"given":"Chunhui","family":"Mo","sequence":"first","affiliation":[{"name":"Independent Researcher, Powell, Ohio, USA."}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2026,9,20]]},"reference":[{"key":"e_1_3_2_2_1","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2014.18171"},{"key":"e_1_3_2_3_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11606-021-07333-z"},{"key":"e_1_3_2_4_1","first-page":"1312","article-title":"Supervised learning methods for predicting healthcare costs: Systematic literature review and empirical evaluation","volume":"2017","author":"Morid MA","year":"2017","unstructured":"3.Morid MA, , Kawamoto K, , Ault T, , Dorius J, , Abdelrahman S. Supervised learning methods for predicting healthcare costs: Systematic literature review and empirical evaluation. AMIA Annu Symp Proc. 2018 2017;2017:1312\u20131321.","journal-title":"AMIA Annu Symp Proc. 2018"},{"key":"e_1_3_2_5_1","doi-asserted-by":"publisher","DOI":"10.1089\/pop.2017.0122"},{"key":"e_1_3_2_6_1","first-page":"1912","article-title":"Using massive health insurance claims data to predict very high-cost claimants: A machine learning approach","author":"Maisog JM","year":"2019","unstructured":"5.Maisog JM, , Li W, , Xu Y, et al. Using massive health insurance claims data to predict very high-cost claimants: A machine learning approach. ArXiv 2019:1912; doi: 10.48550\/arXiv.1912.13032","journal-title":"ArXiv"},{"key":"e_1_3_2_7_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10742-022-00275-x"},{"key":"e_1_3_2_8_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-021-01743-z"},{"key":"e_1_3_2_9_1","first-page":"2311","article-title":"Machine learning for an explainable cost prediction of medical insurance","author":"Orji U","year":"2023","unstructured":"8.Orji U, , Ukwandu E. Machine learning for an explainable cost prediction of medical insurance. ArXiv 2023:2311; doi: 10.48550\/arXiv.2311.14139","journal-title":"ArXiv"},{"key":"e_1_3_2_10_1","doi-asserted-by":"publisher","DOI":"10.1080\/13696998.2026.2630598"},{"key":"e_1_3_2_11_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12938-018-0568-3"},{"key":"e_1_3_2_12_1","doi-asserted-by":"publisher","unstructured":"11.Burges C Shaked T Renshaw E et al. Learning to rank using gradient descent. In: Proceedings of the 22nd International Conference on Machine Learning - ICML \u201805. ACM Press; 2005:89\u201396. doi:10.1145\/1102351.1102363","DOI":"10.1145\/1102351.1102363"},{"key":"e_1_3_2_13_1","unstructured":"12.Burges CJC. From RankNet to LambdaRank to LambdaMART: An Overview. Microsoft Res Tech Rep MSR-TR-2010-82."},{"key":"e_1_3_2_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/s0167-6296(01)00086-8"},{"key":"e_1_3_2_15_1","volume-title":"Pattern Recognition and Machine Learning","author":"Bishop CM","year":"2006","unstructured":"14.Bishop CM. Pattern Recognition and Machine Learning. Springer: New York, NY; 2006."},{"key":"e_1_3_2_16_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-84858-7"},{"key":"e_1_3_2_17_1","unstructured":"16.Moeller J Cohen S. Methodology Report #13: Projecting National Medical Expenditure Survey Data: A Framework for MEPS Projections. Agency for Healthcare Research and Quality: Rockville MD; 2026. Available from: https:\/\/meps.ahrq.gov\/data_files\/publications\/mr13\/mr13.shtml [Last Accessed: June 3 2026]."},{"key":"e_1_3_2_18_1","doi-asserted-by":"publisher","DOI":"10.5600\/mmrr.002.04.a09"},{"key":"e_1_3_2_19_1","doi-asserted-by":"publisher","DOI":"10.1093\/intqhc\/mzz135"},{"key":"e_1_3_2_20_1","unstructured":"19.AHRQ. 2018 Office-Based Medical Provider Visits. AHRQ: Rockville MD. Available from: https:\/\/meps.ahrq.gov\/data_stats\/download_data\/pufs\/h206g\/h206gdoc.shtml [Last Accessed: June 5 2026]."},{"issue":"19","key":"e_1_3_2_21_1","first-page":"52372","article-title":"Care management of patients with complex health care needs","author":"Berry-Millett R","year":"2009","unstructured":"20.Berry-Millett R, , Bodenheimer TS. Care management of patients with complex health care needs. Synth Proj Res Synth Rep 2009(19):52372.","journal-title":"Synth Proj Res Synth Rep"},{"key":"e_1_3_2_22_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12913-015-1063-1"},{"key":"e_1_3_2_23_1","doi-asserted-by":"publisher","DOI":"10.1177\/00469580211064118"},{"issue":"12","key":"e_1_3_2_24_1","first-page":"e472","article-title":"Self-reported health and functional status information improves prediction of inpatient admissions and costs","volume":"17","author":"Perrin NA","year":"2011","unstructured":"23.Perrin NA, , Stiefel M, , Mosen DM, , Bauck A, , Shuster E, , Dirks EM. Self-reported health and functional status information improves prediction of inpatient admissions and costs. Am J Manag Care 2011;17(12):e472\u2013e478.","journal-title":"Am J Manag Care"},{"key":"e_1_3_2_25_1","doi-asserted-by":"publisher","DOI":"10.7326\/M16-0961"},{"key":"e_1_3_2_26_1","doi-asserted-by":"publisher","DOI":"10.3233\/JEM-2000-0173"},{"key":"e_1_3_2_27_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1475-6773.2009.01080.x"},{"key":"e_1_3_2_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jclinepi.2019.02.004"},{"key":"e_1_3_2_29_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10198-021-01363-4"},{"key":"e_1_3_2_30_1","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMp1413945"},{"key":"e_1_3_2_31_1","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocaa143"},{"key":"e_1_3_2_32_1","doi-asserted-by":"publisher","DOI":"10.2196\/18084"}],"container-title":["Population Health Management"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/19427891261488894","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/19427891261488894","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/19427891261488894","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T02:46:41Z","timestamp":1790218001000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/19427891261488894"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,20]]},"references-count":31,"alternative-id":["10.1177\/19427891261488894"],"URL":"https:\/\/doi.org\/10.1177\/19427891261488894","relation":{},"ISSN":["1942-7891","1942-7905"],"issn-type":[{"value":"1942-7891","type":"print"},{"value":"1942-7905","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,20]]},"article-number":"19427891261488894"}}