Predictors of health and well-being among new French immigrants: Longitudinal, data-driven approach, and machine learning analysis
4 affiliations
- 1Department of Psychology, UR 7273 CRP-CPO, University of Picardie Jules Verne, Amiens, France
- 2Addiction Medicine, Department of Psychiatry, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland
- 3School of Law, University of Louvain, Louvain, Belgium
- 4Department of Human Sciences, Letters and Communication, TÉLUQ University of Quebec, Montreal, Canada
Abstract
While past research has identified income, employment, social capital, and living conditions as key predictors of EU immigrants’ health and well-being (HWB), few studies have used machine learning with longitudinal data to explore this in depth. This study aimed to (1) assess HWB in a random sample of new immigrants legally residing in France and (2) use a data-driven machine learning approach to identify key predictors of HWB three years after receiving a French residence permit. Data came from the ELIPA-2 longitudinal survey, including 4,053 new immigrants (57% male, mean age 34.25) from three waves (2019, 2020, 2022). Most participants (71%) were from Africa. The analysis involved 93 predictors using descriptive statistics, correlations, and machine learning regression models. Findings suggest that HWB levels in the studied population were high (mean score = 25 on a 4–33 scale). The most important predictors of HWB, in ranked order, were: HWB in the country of origin and during the early years of residence in France, the length and nature of residence permits, financial and economic status, housing conditions, life background in the country of origin, social relationships and working conditions in the host country, and emigration motives. The fact that the study population reports relatively high levels of health and well-being can be interpreted in light of the sample composition, with 71% of participants originating from African countries where average health indicators are comparatively lower. Consequently, it may reflect a “healthy immigrant effect” and/or the influence of protective conditions that help sustain favorable health outcomes.
Related
- Determine and compare limb specific predictors based on the foot pressure index of right and left foot among young adults in Saudi Arabia using machine learning models
- Beyond benchmark accuracy: machine-learning turnover-number predictors require system-level validation
- Explainable Machine Learning and Epigenomic Profiling Decipher the Topological Determinants of Lentiviral Integration and Longitudinal Persistence in SCID-X1 Gene Therapy
- Machine learning prediction of 1-year thrombosis risk in membranous nephropathy: integration of renal pathology and early longitudinal trajectories
- HQML-ICU: hybrid quantum-classical machine learning for ICU mortality prediction using MIMIC-IV electronic health records
BibTeX
@article{cruz2026predictors,
title = {{Predictors of health and well-being among new French immigrants: Longitudinal, data-driven approach, and machine learning analysis}},
author = {Germano Vera Cruz and Clarice Da Rosa and Yasser Khazaal and Noémie Tronche and Coraline Quenson and Lucas Cauwet and Lonzozou Kpanake},
journal = {Journal of Public Health Research},
year = {2026},
month = jul,
doi = {10.1177/22799036261462907},
url = {https://u-picardie.hal.science/hal-05757855v1},
note = {CC BY-NC},
}
RIS
TY - JOUR
TI - Predictors of health and well-being among new French immigrants: Longitudinal, data-driven approach, and machine learning analysis
AU - Germano Vera Cruz
AU - Clarice Da Rosa
AU - Yasser Khazaal
AU - Noémie Tronche
AU - Coraline Quenson
AU - Lucas Cauwet
AU - Lonzozou Kpanake
PY - 2026
DA - 2026/07/01
JO - Journal of Public Health Research
DO - 10.1177/22799036261462907
UR - https://u-picardie.hal.science/hal-05757855v1
AB - While past research has identified income, employment, social capital, and living conditions as key predictors of EU immigrants’ health and well-being (HWB), few studies have used machine learning with longitudinal data to explore this in depth. This study aimed to (1) assess HWB in a random sample of new immigrants legally residing in France and (2) use a data-driven machine learning approach to identify key predictors of HWB three years after receiving a French residence permit. Data came from the ELIPA-2 longitudinal survey, including 4,053 new immigrants (57% male, mean age 34.25) from three waves (2019, 2020, 2022). Most participants (71%) were from Africa. The analysis involved 93 predictors using descriptive statistics, correlations, and machine learning regression models. Findings suggest that HWB levels in the studied population were high (mean score = 25 on a 4–33 scale). The most important predictors of HWB, in ranked order, were: HWB in the country of origin and during the early years of residence in France, the length and nature of residence permits, financial and economic status, housing conditions, life background in the country of origin, social relationships and working conditions in the host country, and emigration motives. The fact that the study population reports relatively high levels of health and well-being can be interpreted in light of the sample composition, with 71% of participants originating from African countries where average health indicators are comparatively lower. Consequently, it may reflect a “healthy immigrant effect” and/or the influence of protective conditions that help sustain favorable health outcomes.
SN - 2279-9028
SN - 2279-9036
ER -