{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T06:04:53Z","timestamp":1789538693546,"version":"build-2803163510"},"reference-count":31,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T00:00:00Z","timestamp":1789516800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Health Serv."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>The infant mortality rate is widely used in comparative health research, but less is known about how panel regression and machine-learning models perform in small macro-level health datasets when they are evaluated using comparable information and the same out-of-period test period.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Using a balanced panel of 25 European Union countries observed annually from 2000 to 2022 (575 country-year observations, no missing data) assembled from OECD and World Bank sources, we separated an explanatory analysis from a predictive analysis. The explanatory layer used a two-way fixed effects regression with country-clustered standard errors. The predictive layer trained pooled linear regression, tuned Random Forest, Gradient Boosting, and a Multilayer Perceptron, with and without explicit country information, on 2000\u20132017 and evaluated them on a held-out 2018\u20132022 window, alongside two naive benchmarks (persistence and historical country mean) and two prediction-compatible fixed effects specifications. Hyperparameters were tuned by expanding-window temporal cross-validation within the training period, leaving the test window untouched.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Under common out-of-period evaluation, a simple persistence benchmark achieved the strongest predictive performance (R\n                      <jats:sup>2<\/jats:sup>\n                      \u2009=\u20090.74), closely followed by tuned Gradient Boosting and Random Forest with country information (R\n                      <jats:sup>2<\/jats:sup>\n                      \u2009=\u20090.71 and 0.69). Explicit country encoding improved the tree-based models, but adding country information was not enough to improve the linear specifications substantially. The prediction-compatible fixed effects and pooled linear models both produced negative out-of-period R\n                      <jats:sup>2<\/jats:sup>\n                      values. Model rankings varied across evaluation windows: in an alternative pre-pandemic split, tuned nonlinear models outperformed persistence. In the explanatory fixed effects model, GDP per capita and the total fertility rate were significant under conventional country-clustered inference, although neither remained significant under a small-cluster wild-bootstrap sensitivity analysis; the substantive covariate block had a partial R\n                      <jats:sup>2<\/jats:sup>\n                      of 0.366 net of country and year effects.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>In this EU infant-mortality panel, much of the predictive signal came from persistent country-specific outcome history. Country information improved the tree-based models, while model flexibility also remained important. The strong within-panel fit of the fixed-effects model did not translate into strong out-of-period prediction.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frhs.2026.1893109","type":"journal-article","created":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T05:32:24Z","timestamp":1789536744000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Infant mortality across EU health systems: heterogeneity encoding, fixed effects regression, and machine learning in a small macro panel"],"prefix":"10.3389","volume":"6","author":[{"given":"Ecmel","family":"Kisa","sequence":"first","affiliation":[{"name":"Department of Technology, Faculty of Health Sciences and Technology, Kristiania University of Applied Sciences","place":["Oslo, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adnan","family":"Kisa","sequence":"additional","affiliation":[{"name":"Department of Health and Exercise, Faculty of Health Sciences and Technology, Kristiania University of Applied Sciences","place":["Oslo, Norway"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,16]]},"reference":[{"key":"B1","article-title":"Infant mortality rate (probability of dying between birth and age 1 per 1,000 live births)","year":"2024","journal-title":"Global Health Observatory"},{"key":"B2","year":"2025"},{"key":"B3","doi-asserted-by":"publisher","first-page":"e45652","DOI":"10.7759\/cureus.45652","article-title":"Trends and factors associated with mortality rates of leading causes of infant death: a CDC wide-ranging online data for epidemiologic research (CDC WONDER) database analysis","volume":"15","author":"Okobi","year":"2023","journal-title":"Cureus"},{"key":"B4","doi-asserted-by":"publisher","first-page":"288","DOI":"10.1080\/14034940600979171","article-title":"Socioeconomic determinants of infant mortality: a worldwide study of 152 low-, middle-, and high-income countries","volume":"35","author":"Schell","year":"2007","journal-title":"Scand J Public Health"},{"key":"B5","doi-asserted-by":"publisher","first-page":"10725","DOI":"10.3390\/ijerph182010725","article-title":"Government health expenditure and public health outcomes: a comparative study among EU developing countries","volume":"18","author":"Onofrei","year":"2021","journal-title":"Int J Environ Res Public Health"},{"key":"B6","doi-asserted-by":"publisher","first-page":"20220643","DOI":"10.1515\/biol-2022-0643","article-title":"Studying infant mortality: a demographic analysis based on data mining models","volume":"18","author":"Satti","year":"2023","journal-title":"Open Life Sci"},{"key":"B7","doi-asserted-by":"publisher","first-page":"24278","DOI":"10.1038\/s41598-025-04066-5","article-title":"Development and evaluation of machine learning training strategies for neonatal mortality prediction using multicountry data","volume":"15","author":"Silva","year":"2025","journal-title":"Sci Rep"},{"key":"B8","article-title":"Organisation for Economic Co-operation and Development and European Commission.","volume-title":"Health at a Glance: Europe 2024: State of Health in the EU Cycle","year":"2024"},{"key":"B9","first-page":"244","volume-title":"Health at a Glance 2025: OECD Indicators","year":"2025"},{"key":"B10","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1186\/s13561-020-00262-3","article-title":"The effects of health expenditure on infant mortality in sub-saharan Africa: evidence from panel data analysis","volume":"10","author":"Kiross","year":"2020","journal-title":"Health Econ Rev"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1588","DOI":"10.1080\/13504851.2017.1340565","article-title":"Infant mortality in Europe, socio-economic determinants based on aggregate data","volume":"24","author":"Tavares","year":"2017","journal-title":"Appl Econ Lett"},{"key":"B12","doi-asserted-by":"publisher","first-page":"107423","DOI":"10.1016\/j.compbiomed.2023.107423","article-title":"Infant death prediction using machine learning: a population-based retrospective study","volume":"165","author":"Zhang","year":"2023","journal-title":"Comput Biol Med"},{"key":"B13","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1186\/s13040-022-00308-8","article-title":"Machine learning algorithms for understanding the determinants of under-five mortality","volume":"15","author":"Saroj","year":"2022","journal-title":"BioData Min"},{"key":"B14","first-page":"1033","article-title":"Predictive modeling of neonatal mortality trends using machine learning: a cross-sectional data science approach","volume":"14","author":"Pal","year":"2025","journal-title":"J Neonatal Surg"},{"key":"B15","doi-asserted-by":"publisher","first-page":"e2443","DOI":"10.1002\/bdr2.2443","article-title":"Exploring ensemble learning techniques for infant mortality prediction: a technical analysis of XGBoost stacking AdaBoost and bagging models","volume":"117","author":"Verma","year":"2025","journal-title":"Birth Defects Res"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1785","DOI":"10.1007\/s10618-020-00728-2","article-title":"Predictive modeling of infant mortality","volume":"35","author":"Saravanou","year":"2021","journal-title":"Data Min Knowl Discov"},{"key":"B17","doi-asserted-by":"publisher","first-page":"388","DOI":"10.1186\/s12884-022-04699-8","article-title":"Application of machine learning methods for predicting infant mortality in Rwanda: analysis of Rwanda demographic health survey 2014\u201315 dataset","volume":"22","author":"Mfateneza","year":"2022","journal-title":"BMC Pregnancy Childbirth"},{"key":"B18","doi-asserted-by":"publisher","first-page":"101058","DOI":"10.1016\/j.ehb.2021.101058","article-title":"Inequality of opportunity in infant mortality in south Asia: a decomposition analysis of survival data","volume":"43","author":"Aizawa","year":"2021","journal-title":"Econ Hum Biol"},{"key":"B19","doi-asserted-by":"publisher","first-page":"394","DOI":"10.1159\/000516891","article-title":"Machine learning models for predicting neonatal mortality: a systematic review","volume":"118","author":"Mangold","year":"2021","journal-title":"Neonatology"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e60231","DOI":"10.2196\/60231","article-title":"Sample size requirements for popular classification algorithms in tabular clinical data: empirical study","volume":"26","author":"Silvey","year":"2024","journal-title":"J Med Internet Res"},{"key":"B21","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","year":"2025","journal-title":"Nature"},{"key":"B22","doi-asserted-by":"publisher","first-page":"754","DOI":"10.1016\/j.socscimed.2012.04.017","article-title":"Correlation or causation? Income inequality and infant mortality in fixed effects models in the period 1960\u20132008 in 34 OECD countries","volume":"75","author":"Avendano","year":"2012","journal-title":"Soc Sci Med"},{"key":"B23","doi-asserted-by":"publisher","first-page":"289","DOI":"10.1214\/10-STS330","article-title":"To explain or to predict?","volume":"25","author":"Shmueli","year":"2010","journal-title":"Stat Sci"},{"key":"B24","year":"2025"},{"key":"B25","doi-asserted-by":"publisher","first-page":"673","DOI":"10.1007\/s11135-006-9018-6","article-title":"A caution regarding rules of thumb for variance inflation factors","volume":"41","author":"O'Brien","year":"2007","journal-title":"Qual Quant"},{"key":"B26","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-38747-0","volume-title":"An Introduction to Statistical Learning: With Applications in R","author":"James","year":"2023"},{"key":"B27","doi-asserted-by":"publisher","first-page":"223","DOI":"10.1086\/259880","article-title":"On the concept of health capital and the demand for health","volume":"80","author":"Grossman","year":"1972","journal-title":"J Polit Econ"},{"key":"B28","doi-asserted-by":"publisher","first-page":"e0144908","DOI":"10.1371\/journal.pone.0144908","article-title":"Factors contributing to maternal and child mortality reductions in 146 low- and middle-income countries between 1990 and 2010","volume":"11","author":"Bishai","year":"2016","journal-title":"PLoS One"},{"key":"B29","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach Learn"},{"key":"B30","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy function approximation: a gradient boosting machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann Stat"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1123759","DOI":"10.3389\/fpubh.2023.1123759","article-title":"Government health expenditures and health outcome nexus: a study on OECD countries","volume":"11","author":"Anwar","year":"2023","journal-title":"Front Public Health"}],"container-title":["Frontiers in Health Services"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frhs.2026.1893109\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T05:32:26Z","timestamp":1789536746000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frhs.2026.1893109\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,16]]},"references-count":31,"alternative-id":["10.3389\/frhs.2026.1893109"],"URL":"https:\/\/doi.org\/10.3389\/frhs.2026.1893109","relation":{},"ISSN":["2813-0146"],"issn-type":[{"value":"2813-0146","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,16]]},"article-number":"1893109"}}