Incorporating longitudinal variability in prediction models: A comparison of machine learning and logistic regression in a cohort study with long follow-up.

IF 3 3区 医学 Q1 PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH
L M de Groot, J W R Twisk, A A L Kok, M W Heymans
{"title":"Incorporating longitudinal variability in prediction models: A comparison of machine learning and logistic regression in a cohort study with long follow-up.","authors":"L M de Groot, J W R Twisk, A A L Kok, M W Heymans","doi":"10.1016/j.annepidem.2025.07.060","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Clinical prediction models benefit from longitudinal data. While the predictive value of a predictor's mean and change over time is well-established, the role of variability around this change is underexplored. Machine Learning methods can be effective in analyzing longitudinal data with long follow-up periods. This study evaluated the predictive value of mean, change, and variability, comparing Random Forest, Lasso regression, and logistic regression.</p><p><strong>Methods: </strong>We compared models including only mean and change to models also incorporating variability. Predictor selection, interpretability, and performance were compared across methods. Performance was assessed using AUC, sensitivity, specificity, PPV, NPV, and calibration. Data were drawn from the Longitudinal Aging Study Amsterdam to predict depression using 81 longitudinal parameters. Models were trained on 70 % and validated on 30 % of the data. To ensure robustness, analyses were repeated over 500 random splits, and aggregated results were reported.</p><p><strong>Results: </strong>Including variability improved AUCs for all methods. Predictor selection overlapped across models, and regression coefficients aligned with Random Forest partial dependence plots. Lasso showed the highest training AUC but poorer test performance, while logistic regression and Random Forest showed more stable results. Calibration was acceptable, though predicted risks remained below 0.6.</p><p><strong>Conclusion: </strong>Machine Learning methods did not outperform logistic regression. Nonetheless, incorporating variability in longitudinal predictors enhances prediction, especially with expected changes in predictors, e.g., ageing populations.</p>","PeriodicalId":50767,"journal":{"name":"Annals of Epidemiology","volume":" ","pages":"51-65"},"PeriodicalIF":3.0000,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annals of Epidemiology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1016/j.annepidem.2025.07.060","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0

Abstract

Purpose: Clinical prediction models benefit from longitudinal data. While the predictive value of a predictor's mean and change over time is well-established, the role of variability around this change is underexplored. Machine Learning methods can be effective in analyzing longitudinal data with long follow-up periods. This study evaluated the predictive value of mean, change, and variability, comparing Random Forest, Lasso regression, and logistic regression.

Methods: We compared models including only mean and change to models also incorporating variability. Predictor selection, interpretability, and performance were compared across methods. Performance was assessed using AUC, sensitivity, specificity, PPV, NPV, and calibration. Data were drawn from the Longitudinal Aging Study Amsterdam to predict depression using 81 longitudinal parameters. Models were trained on 70 % and validated on 30 % of the data. To ensure robustness, analyses were repeated over 500 random splits, and aggregated results were reported.

Results: Including variability improved AUCs for all methods. Predictor selection overlapped across models, and regression coefficients aligned with Random Forest partial dependence plots. Lasso showed the highest training AUC but poorer test performance, while logistic regression and Random Forest showed more stable results. Calibration was acceptable, though predicted risks remained below 0.6.

Conclusion: Machine Learning methods did not outperform logistic regression. Nonetheless, incorporating variability in longitudinal predictors enhances prediction, especially with expected changes in predictors, e.g., ageing populations.

在预测模型中纳入纵向可变性:机器学习和逻辑回归在长期随访队列研究中的比较。
目的:临床预测模型受益于纵向数据。虽然预测器的平均值和随时间变化的预测值已经确立,但围绕这种变化的变异性的作用尚未得到充分探讨。机器学习方法可以有效地分析长时间随访的纵向数据。本研究通过比较随机森林、Lasso回归和逻辑回归,评估均值、变化和变异的预测价值。方法:我们将仅包含平均值和变化的模型与包含变异的模型进行比较。对不同方法的预测器选择、可解释性和性能进行比较。使用AUC、灵敏度、特异性、PPV、NPV和校准来评估性能。数据来自阿姆斯特丹纵向衰老研究,使用81个纵向参数来预测抑郁症。模型在70%的数据上进行训练,在30%的数据上进行验证。为了确保稳健性,对500多个随机分裂进行了重复分析,并报告了汇总结果。结果:纳入变异性可改善所有方法的auc。预测器选择在模型之间重叠,回归系数与随机森林部分相关图对齐。Lasso的训练AUC最高,但测试性能较差,而logistic回归和Random Forest的结果更稳定。校正是可接受的,但预测风险仍低于0.6。结论:机器学习方法并不优于逻辑回归方法。尽管如此,将可变性纳入纵向预测因子可以加强预测,特别是考虑到预测因子的预期变化,例如人口老龄化。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Annals of Epidemiology
Annals of Epidemiology 医学-公共卫生、环境卫生与职业卫生
CiteScore
7.40
自引率
1.80%
发文量
207
审稿时长
59 days
期刊介绍: The journal emphasizes the application of epidemiologic methods to issues that affect the distribution and determinants of human illness in diverse contexts. Its primary focus is on chronic and acute conditions of diverse etiologies and of major importance to clinical medicine, public health, and health care delivery.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信