focalinterest.com

AI Summary of Peer-Reviewed Research

This page presents an AI-generated summary of a published research paper. The original authors did not write or review this article. [See full disclosure ↓]

Publishing process signals: STRONG — reflects the venue and review process. — venue and review process.

XGBoost predicted healthy aging in older adults

—

in
Multiple adults of varying ages participate in a hula hoop activity together in an indoor gymnasium, with one woman in a yellow-green shirt actively twirling a hula hoop while others hold hoops and watch in a supportive community setting.
Research area:GerontologyHealth disparities and outcomesCohort study

What the study found

The study found that an XGBoost model was the best-performing method for predicting healthy aging in adults aged 50 and older. The authors also report that health insurance type was the strongest predictive feature in their analysis.

Why the authors say this matters

The authors conclude that their findings underscore the significant role of health insurance in contributing to healthy aging. The study suggests that understanding social determinants of health, or social and economic factors linked to health, may help clarify healthy aging pathways.

What the researchers tested

The researchers used retrospective cohort data from the All of Us Research Program registered tier dataset v7. They analyzed participants aged 50 and older who answered at least one social determinants of health survey question and had electronic health record data, and they trained logistic regression, multi-layer perceptron, and XGBoost models to predict a composite healthy aging outcome.

What worked and what didn't

The best model was XGBoost with random oversampling, which achieved AUROC 0.793, F1 score 0.697, recall 0.739, precision 0.659, and accuracy 0.716. The study reports that this model outperformed logistic regression and multi-layer perceptron, and it showed predictive parity across race and sex groups with similar positive and negative predictive values.

What to keep in mind

The abstract does not describe detailed limitations beyond using one cohort and the available survey and electronic health record data. The healthy aging outcome was defined by a composite score based on comorbidities, cognitive conditions, and mobility function, so the findings apply to that specific definition.

Key points

  • The study analyzed 99,935 participants aged 50 and older.
  • Healthy aging was defined using comorbidities, cognitive conditions, and mobility function.
  • XGBoost with random oversampling was the best-performing prediction model.
  • Health insurance type was the most predictive feature in the SHAP analysis.
  • The XGBoost model showed predictive parity across race and sex groups.

Disclosure

Research title:
XGBoost predicted healthy aging in older adults
Authors:
Wei‐Han Chen, Yao-An Lee, Huilin Tang, Chenyu Li, You Lü, Yu Huang, Rui Yin, Melissa J. Armstrong, Yang Yang, Gregor Stiglic, Jiang Bian, Jingchuan Serena Guo
Institutions:
Indiana University – Purdue University Indianapolis, Indiana University – Purdue University Indianapolis, Indiana University Health, Regenstrief Institute, Regenstrief Institute, Regenstrief Institute, University of Edinburgh, University of Florida, University of Florida, University of Florida, University of Florida, University of Florida, University of Florida Health, University of Florida Health, University of Indianapolis, University of Maribor, University of Pittsburgh, University of Pittsburgh, Vibrant Data (United States)
Publication date:
2026-03-06
DOI:
10.1371/journal.pone.0342292
OpenAlex record:
View
AI provenance: This post was generated by gpt-5.4-mini (OpenAI). The original authors did not write or review this post.