Comparison of methodologies for constructing non-time-varying outcome prediction models based on longitudinal data_Chinese Journal of Evidence-Based Medicine

Authors：

WEI Wanqiang ^1,2,3 , REN Yan ^1,2,3 , SUN Xin ^1,2,3 , LIU Chunrong ^1,2,3 ,  JIA Yulong ^1,2,3 ,  TAN Jing ^1,2,3

1. Chinese Evidence-based Medicine Center, West China Hospital, Sichuan University, Chengdu 610041, P. R. China;
2. NMPA Key Laboratory for Real World Data Research and Evaluation in Hainan, Chengdu 610041, P. R. China;
3. Sichuan Center of Technology Innovation for Real World Data, Chengdu 610041, P. R. China;

Corresponding author：

JIA Yulong, Email: jiayulong@wchscu.cn; TAN Jing, Email: tanjing84@outlook.com

Keywords：

Longitudinal data; Non-time-varying outcome; Predictive modeling; Severe postpartum hemorrhage

DOI：

10.7507/1672-2531.202501108

Video：

Export PDF Favorites Scan Get Citation

Abstract Full text Figures/Tables Video References Cited by

Objective To explore the utilization of longitudinal data in constructing non-time-varying outcome prediction models and to compare the impact of different modeling approaches on prediction performance. Methods Clinical predictors were selected using univariate analysis and Lasso regression. Non-time-varying outcome prediction models were developed based on latent class trajectory analysis, the two-stage model, and logistic regression. Internal validation was performed using Bootstrapping resampling, and model performance was evaluated using ROC curves, PR curves, sensitivity, specificity and other relevant metrics. Results A total of 49 629 pregnant women were included in the study, with mean age of 31.42±4.13 years and pre-pregnancy BMI of 20.91±2.62kg/m². Fourteen predictors were incorporated into the final model. Prediction models utilizing longitudinal data demonstrated high accuracy, with AUROC values exceeding 0.90 and PR-AUC values greater than 0.47. The two-stage model based on late-pregnancy hemoglobin data showed the best performance, achieving AUROC of 0.93 (95%CI 0.92 to 0.94) and PR-AUC of 0.60 (95%CI 0.56 to 0.64). Internal validation confirmed robust model performance, and calibration curves indicated a good agreement between predicted and observed outcomes. Conclusion For the longitudinal data, the two-stage model can well capture the dynamic change trajectory of the longitudinal data. For different clinical outcomes, the predictive value of repeated measurement data is different.

1.	郑卫军, 何凡. 重复测量资料的统计分析方法. 预防医学, 2020, 32(8): 863-864.
2.	Chen YH, Ferguson KK, Meeker JD, et al. Statistical methods for modeling repeated measures of maternal environmental exposure biomarkers during pregnancy in association with preterm birth. Environ Health, 2015, 14: 9.
3.	于玥琳, 胥洋, 王俊峰, 等. 临床预测模型中处理时依性变量的策略及进展. 中华流行病学杂志, 2023, (8): 1316-1320.
4.	WHO Guidelines Approved by the Guidelines Review Committee. WHO recommendations for the prevention and treatment of postpartum haemorrhage. Geneva: World Health Organization, 2012.
5.	中华医学会妇产科学分会产科学组. 前置胎盘的诊断与处理指南(2020). 中华妇产科杂志, 2020, 55(1): 3-8.
6.	陈代娟, 徐金凤, 田园, 等. 凶险性前置胎盘术前评估及手术诊治研究现状. 中华妇幼临床医学杂志(电子版), 2020, (6): 634-640.
7.	中华医学会妇产科学分会产科学组, 中国医师协会妇产科分会母胎医学专委会. 胎盘植入性疾病诊断和处理指南(2023). 中华围产医学杂志, 2023, (8): 617-627.
8.	Alba AC, Agoritsas T, Walsh M, et al. Discrimination and calibration of clinical prediction models: users' guides to the medical literature. JAMA, 2017, 318(14): 1377-1384.
9.	Abdullahi AS, Suliman A, Khan MA, et al. Temporal trends of hemoglobin among pregnant women: The Mutaba'ah study. PLoS One, 2023, 18(12): e0295549.
10.	Lennon H, Kelly S, Sperrin M, et al. Framework to construct and interpret latent class trajectory modelling. BMJ Open, 2018, 8(7): e020683.
11.	Randall DA, Patterson JA, Gallimore F, et al. Haemoglobin trajectories during pregnancy and associated outcomes using pooled maternity and hospitalization data from two tertiary hospitals. Vox Sang, 2019, 114(8): 842-852.
12.	Young MF, Nguyen P, Tran LM, et al. Maternal hemoglobin concentrations across pregnancy and child health and development from birth through 6-7 years. Front Nutr, 2023, 10: 1114101.
13.	O'Connor S, Blais C, Mésidor M, et al. Great diversity in the utilization and reporting of latent growth modeling approaches in type 2 diabetes: a literature review. Heliyon, 2022, 8(9): e10493.
14.	Elsensohn MH, Klich A, Ecochard R, et al. A graphical method to assess distribution assumption in group-based trajectory models. Stat Methods Med Res, 2016, 25(2): 968-982.
15.	贾玉龙, 任燕, 魏万强, 等. 基于重复测量数据预测非时变结局发生风险的二阶段建模方法. 中国循证医学杂志, 2024, 24(11): 1360-1364.
16.	Galwey NW. Estimation of random effects in mixed models: best linear unbiased predictors (BLUPs). Hoboken: John Wiley & Sons, Ltd, 2014: 165-191.
17.	Liu C, Xu Y, Li J, et al. Development and validation of a predictive model for severe postpartum hemorrhage in women undergoing vaginal delivery: a retrospective cohort study. Int J Gynaecol Obstet, 2022, 157(2): 353-358.
18.	Mehrnoush V, Ranjbar A, Farashah MV, et al. Prediction of postpartum hemorrhage using traditional statistical analysis and a machine learning approach. AJOG Glob Rep, 2023, 3(2): 100185.
19.	Song Z, Wang X, Zhou Y, et al. Development and validation of prognostic nomogram for postpartum hemorrhage after vaginal delivery: a retrospective cohort study in China. Front Med (Lausanne), 2022, 9: 804769.
20.	Bull LM, Lunt M, Martin GP, et al. Harnessing repeated measurements of predictor variables for clinical risk prediction: a review of existing methods. Diagn Progn Res, 2020, 4: 9.
21.	苏璇, 王远军. 面向高维不平衡医学数据的特征选择算法. 小型微型计算机系统, 2024, 45(2): 309-318.
22.	闫慈, 田翔华, 阿拉依·阿汗, 等. 基于AdaBoost法在代谢综合征不平衡数据分类中的应用. 现代预防医学, 2017, 44(21): 3850-3852,3862.
23.	Zhao J, Feng Q, Wu P, et al. Learning from longitudinal data in electronic health record and genetic data to improve cardiovascular event prediction. Sci Rep, 2019, 9(1): 717.
24.	Robinson MR, Patxot M, Stojanov M, et al. Postpartum hemorrhage risk is driven by changes in blood composition through pregnancy. Sci Rep, 2021, 11(1): 19238.
25.	Kraft M, Richardson M, Hallmark B, et al. The role of small airway dysfunction in asthma control and exacerbations: a longitudinal, observational analysis using data from the ATLANTIS study. Lancet Respir Med, 2022, 10(7): 661-668.
26.	Madlala HP, Bengtson AM, Hannan L, et al. Maternal weight trajectories and associations with infant growth in South African women. BMC Public Health, 2023, 23(1): 2055.
27.	Park B, Chang Y, Ryu S, et al. Trajectories of breast density change over time and subsequent breast cancer risk: longitudinal study. BMJ, 2024, 387: e079575.
28.	Rod NH, Bengtsson J, Budtz-Jørgensen E, et al. Trajectories of childhood adversity and mortality in early adulthood: a population-based cohort study. Lancet, 2020, 396(10249): 489-497.
29.	Griffin MM, Avtushka V, Venkatesh P, et al. Reticulocyte hemoglobin trend in pregnancy. Am J Obstet Gynecol, 2023, 229(4): 471-472.
30.	Peng Z, Si S, Cheng H, et al. The Associations of maternal hemoglobin concentration in different time points and its changes during pregnancy with birth weight outcomes. Nutrients, 2022, 14(12): 2542.
31.	Burden CA, Smith GC, Sovio U, et al. Maternal hemoglobin levels and adverse pregnancy outcomes: individual patient data analysis from 2 prospective UK pregnancy cohorts. Am J Clin Nutr, 2023, 117(3): 616-624.

1. 郑卫军, 何凡. 重复测量资料的统计分析方法. 预防医学, 2020, 32(8): 863-864.
2. Chen YH, Ferguson KK, Meeker JD, et al. Statistical methods for modeling repeated measures of maternal environmental exposure biomarkers during pregnancy in association with preterm birth. Environ Health, 2015, 14: 9.
3. 于玥琳, 胥洋, 王俊峰, 等. 临床预测模型中处理时依性变量的策略及进展. 中华流行病学杂志, 2023, (8): 1316-1320.
4. WHO Guidelines Approved by the Guidelines Review Committee. WHO recommendations for the prevention and treatment of postpartum haemorrhage. Geneva: World Health Organization, 2012.
5. 中华医学会妇产科学分会产科学组. 前置胎盘的诊断与处理指南(2020). 中华妇产科杂志, 2020, 55(1): 3-8.
6. 陈代娟, 徐金凤, 田园, 等. 凶险性前置胎盘术前评估及手术诊治研究现状. 中华妇幼临床医学杂志(电子版), 2020, (6): 634-640.
7. 中华医学会妇产科学分会产科学组, 中国医师协会妇产科分会母胎医学专委会. 胎盘植入性疾病诊断和处理指南(2023). 中华围产医学杂志, 2023, (8): 617-627.
8. Alba AC, Agoritsas T, Walsh M, et al. Discrimination and calibration of clinical prediction models: users' guides to the medical literature. JAMA, 2017, 318(14): 1377-1384.
9. Abdullahi AS, Suliman A, Khan MA, et al. Temporal trends of hemoglobin among pregnant women: The Mutaba'ah study. PLoS One, 2023, 18(12): e0295549.
10. Lennon H, Kelly S, Sperrin M, et al. Framework to construct and interpret latent class trajectory modelling. BMJ Open, 2018, 8(7): e020683.
11. Randall DA, Patterson JA, Gallimore F, et al. Haemoglobin trajectories during pregnancy and associated outcomes using pooled maternity and hospitalization data from two tertiary hospitals. Vox Sang, 2019, 114(8): 842-852.
12. Young MF, Nguyen P, Tran LM, et al. Maternal hemoglobin concentrations across pregnancy and child health and development from birth through 6-7 years. Front Nutr, 2023, 10: 1114101.
13. O'Connor S, Blais C, Mésidor M, et al. Great diversity in the utilization and reporting of latent growth modeling approaches in type 2 diabetes: a literature review. Heliyon, 2022, 8(9): e10493.
14. Elsensohn MH, Klich A, Ecochard R, et al. A graphical method to assess distribution assumption in group-based trajectory models. Stat Methods Med Res, 2016, 25(2): 968-982.
15. 贾玉龙, 任燕, 魏万强, 等. 基于重复测量数据预测非时变结局发生风险的二阶段建模方法. 中国循证医学杂志, 2024, 24(11): 1360-1364.
16. Galwey NW. Estimation of random effects in mixed models: best linear unbiased predictors (BLUPs). Hoboken: John Wiley & Sons, Ltd, 2014: 165-191.
17. Liu C, Xu Y, Li J, et al. Development and validation of a predictive model for severe postpartum hemorrhage in women undergoing vaginal delivery: a retrospective cohort study. Int J Gynaecol Obstet, 2022, 157(2): 353-358.
18. Mehrnoush V, Ranjbar A, Farashah MV, et al. Prediction of postpartum hemorrhage using traditional statistical analysis and a machine learning approach. AJOG Glob Rep, 2023, 3(2): 100185.
19. Song Z, Wang X, Zhou Y, et al. Development and validation of prognostic nomogram for postpartum hemorrhage after vaginal delivery: a retrospective cohort study in China. Front Med (Lausanne), 2022, 9: 804769.
20. Bull LM, Lunt M, Martin GP, et al. Harnessing repeated measurements of predictor variables for clinical risk prediction: a review of existing methods. Diagn Progn Res, 2020, 4: 9.
21. 苏璇, 王远军. 面向高维不平衡医学数据的特征选择算法. 小型微型计算机系统, 2024, 45(2): 309-318.
22. 闫慈, 田翔华, 阿拉依·阿汗, 等. 基于AdaBoost法在代谢综合征不平衡数据分类中的应用. 现代预防医学, 2017, 44(21): 3850-3852,3862.
23. Zhao J, Feng Q, Wu P, et al. Learning from longitudinal data in electronic health record and genetic data to improve cardiovascular event prediction. Sci Rep, 2019, 9(1): 717.
24. Robinson MR, Patxot M, Stojanov M, et al. Postpartum hemorrhage risk is driven by changes in blood composition through pregnancy. Sci Rep, 2021, 11(1): 19238.
25. Kraft M, Richardson M, Hallmark B, et al. The role of small airway dysfunction in asthma control and exacerbations: a longitudinal, observational analysis using data from the ATLANTIS study. Lancet Respir Med, 2022, 10(7): 661-668.
26. Madlala HP, Bengtson AM, Hannan L, et al. Maternal weight trajectories and associations with infant growth in South African women. BMC Public Health, 2023, 23(1): 2055.
27. Park B, Chang Y, Ryu S, et al. Trajectories of breast density change over time and subsequent breast cancer risk: longitudinal study. BMJ, 2024, 387: e079575.
28. Rod NH, Bengtsson J, Budtz-Jørgensen E, et al. Trajectories of childhood adversity and mortality in early adulthood: a population-based cohort study. Lancet, 2020, 396(10249): 489-497.
29. Griffin MM, Avtushka V, Venkatesh P, et al. Reticulocyte hemoglobin trend in pregnancy. Am J Obstet Gynecol, 2023, 229(4): 471-472.
30. Peng Z, Si S, Cheng H, et al. The Associations of maternal hemoglobin concentration in different time points and its changes during pregnancy with birth weight outcomes. Nutrients, 2022, 14(12): 2542.
31. Burden CA, Smith GC, Sovio U, et al. Maternal hemoglobin levels and adverse pregnancy outcomes: individual patient data analysis from 2 prospective UK pregnancy cohorts. Am J Clin Nutr, 2023, 117(3): 616-624.

Chinese Journal of Evidence-Based Medicine

Latest ArticlesComparison of methodologies for constructing non-time-varying outcome prediction models based on longitudinal data

Abstract Full text Figures/Tables Video References Cited by

Format

Content