EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access · CC BY 4.0

Inequality in preconception care utilization based on the opportunity inequality framework: evidence from explained machine learning

Xiaodong Wu, Hai Fang, Tingting Xu

Frontiers in Public Health · 07.05.2026 · 14

DOI: 10.3389/fpubh.2026.1732192 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Frontiers in Public Health, online erschienen am 07.05.2026. Verfasst von Xiaodong Wu und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Public Health, Capital Medical University, Beijing, China. Hauptthemen laut MEDLINE: Healthcare Disparities, Machine Learning, Patient Acceptance of Health Care, Preconception Care. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Despite the universal availability of free preconception care (PCC) in China, substantial disparities in utilization persist. It remains unclear how much of this reflects structural barriers versus individual shortfalls, and clarifying this balance is critical for designing equitable, accountable maternal health policy.

Methods

We conducted a multi-center cross-sectional survey across ten medical facilities in China from July 1, 2023, to July 1, 2024, including 8,866 pregnant women. Guided by the Equality of Opportunity framework, we applied logistic regression to assess PCC factors and three machine learning models to assess factors’ contributions of PCC utilization. Shapley Additive Explanations (SHAP) decomposition quantified the contribution of individual-level “effort” versus structural “circumstance” factors. Subgroup analyses explored heterogeneity by education level, urban–rural residents, whether or not the annual PCC promotion was accepted at local MCH institutions, and pregnancy intention status.

Results

Overall PCC uptake was 42.4%, with disparities strongly associated with individual-level factors as well as structural determinants. Effort-related variables, particularly pregnancy intention and knowledge of PCC, were the strongest predictors across all models, accounting for 74.1–93.6% of explained variation. However, circumstance factors, including annual PCC promotion at local maternal and child health institutions, remained influential, especially among vulnerable subgroups such as women with unplanned pregnancies, for whom structural factors explained nearly half of the disparity.

Conclusion

Our findings reveal that even in systems offering universal coverage, access to PCC remains stratified by structural opportunity. Although individual-level knowledge and intention are key levers, they are themselves shaped by broader social conditions. Addressing PCC inequality thus requires integrated strategies that promote both individual empowerment and systemic equity.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Healthcare DisparitiesMachine LearningPatient Acceptance of Health CarePreconception CareAdultFemaleHumansPregnancyChinaCross-Sectional Studies

Stichwörter der Autor:innen

SHAPfree health servicehealth equalitymachine learningpreconception care

Autor:innen und Institutionen

  • School of Public Health, Capital Medical University, Beijing, China.
  • School of Public Health, Peking University, Beijing, China.
  • Institute for Global Health and Development, Peking University, Beijing, China.

Zitieren

APA

Wu, X., Fang, H., & Xu, T. (2026). Inequality in preconception care utilization based on the opportunity inequality framework: evidence from explained machine learning. Frontiers in Public Health, 14. https://doi.org/10.3389/fpubh.2026.1732192

Vancouver

Wu X, Fang H, Xu T. Inequality in preconception care utilization based on the opportunity inequality framework: evidence from explained machine learning. Frontiers in Public Health. 2026;14. doi:10.3389/fpubh.2026.1732192

BibTeX
@article{Wu2026,
  author = {Wu, Xiaodong and Fang, Hai and Xu, Tingting},
  title = {Inequality in preconception care utilization based on the opportunity inequality framework: evidence from explained machine learning},
  journal = {Frontiers in Public Health},
  year = {2026},
  volume = {14},
  doi = {10.3389/fpubh.2026.1732192}
}
Kostenlos starten →