Clinical prediction models in children that use repeated measurements with time-varying covariates: a scoping review

Alastair Fung; Miranda Loutet; Daniel E Roth; Elliott Wong; Peter J Gill; Shaun K Morris; Joseph Beyene

doi:10.1016/j.acap.2024.03.016

Clinical prediction models in children that use repeated measurements with time-varying covariates: a scoping review

Acad Pediatr. 2024 Mar 30:S1876-2859(24)00115-3. doi: 10.1016/j.acap.2024.03.016. Online ahead of print.

Authors

Alastair Fung¹, Miranda Loutet², Daniel E Roth³, Elliott Wong⁴, Peter J Gill⁵, Shaun K Morris⁶, Joseph Beyene⁷

Affiliations

¹ Division of Paediatric Medicine (A Fung, DE Roth, and PJ Gill), Hospital for Sick Children, Toronto, Ontario, Canada; Dalla Lana School of Public Health (A Fung, M Loutet, DE Roth, PJ Gill, SK Morris, and J Beyene), University of Toronto, Toronto, Ontario, Canada; Centre for Global Child Health (A Fung, M Loutet, DE Roth, and SK Morris), Hospital for Sick Children, Toronto, Ontario, Canada. Electronic address: alastair.fung@sickkids.ca.
² Dalla Lana School of Public Health (A Fung, M Loutet, DE Roth, PJ Gill, SK Morris, and J Beyene), University of Toronto, Toronto, Ontario, Canada; Centre for Global Child Health (A Fung, M Loutet, DE Roth, and SK Morris), Hospital for Sick Children, Toronto, Ontario, Canada.
³ Division of Paediatric Medicine (A Fung, DE Roth, and PJ Gill), Hospital for Sick Children, Toronto, Ontario, Canada; Dalla Lana School of Public Health (A Fung, M Loutet, DE Roth, PJ Gill, SK Morris, and J Beyene), University of Toronto, Toronto, Ontario, Canada; Centre for Global Child Health (A Fung, M Loutet, DE Roth, and SK Morris), Hospital for Sick Children, Toronto, Ontario, Canada; Temerty Faculty of Medicine (DE Roth, E Wong, PJ Gill, and SK Morris), University of Toronto, Toronto, Ontario, Canada; Child Health Evaluative Sciences (DE Roth, PJ Gill, and SK Morris), Hospital for Sick Children Research Institute, Toronto, Ontario, Canada.
⁴ Temerty Faculty of Medicine (DE Roth, E Wong, PJ Gill, and SK Morris), University of Toronto, Toronto, Ontario, Canada.
⁵ Division of Paediatric Medicine (A Fung, DE Roth, and PJ Gill), Hospital for Sick Children, Toronto, Ontario, Canada; Dalla Lana School of Public Health (A Fung, M Loutet, DE Roth, PJ Gill, SK Morris, and J Beyene), University of Toronto, Toronto, Ontario, Canada; Temerty Faculty of Medicine (DE Roth, E Wong, PJ Gill, and SK Morris), University of Toronto, Toronto, Ontario, Canada; Child Health Evaluative Sciences (DE Roth, PJ Gill, and SK Morris), Hospital for Sick Children Research Institute, Toronto, Ontario, Canada.
⁶ Dalla Lana School of Public Health (A Fung, M Loutet, DE Roth, PJ Gill, SK Morris, and J Beyene), University of Toronto, Toronto, Ontario, Canada; Centre for Global Child Health (A Fung, M Loutet, DE Roth, and SK Morris), Hospital for Sick Children, Toronto, Ontario, Canada; Temerty Faculty of Medicine (DE Roth, E Wong, PJ Gill, and SK Morris), University of Toronto, Toronto, Ontario, Canada; Child Health Evaluative Sciences (DE Roth, PJ Gill, and SK Morris), Hospital for Sick Children Research Institute, Toronto, Ontario, Canada; Division of Infectious Diseases (SK Morris), Hospital for Sick Children, Toronto, Ontario, Canada.
⁷ Dalla Lana School of Public Health (A Fung, M Loutet, DE Roth, PJ Gill, SK Morris, and J Beyene), University of Toronto, Toronto, Ontario, Canada; Department of Health Research Methods, Evidence and Impact (J Beyene), Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada.

PMID: 38561061
DOI: 10.1016/j.acap.2024.03.016

Abstract

Background: Emerging evidence suggests that clinical prediction models that use repeated (time-varying) measurements within each patient may have higher predictive accuracy than models that use patient information from a single measurement.

Objective: To determine the breadth of the published literature reporting the development of clinical prediction models in children that use time-varying predictors.

Data sources: MEDLINE, EMBASE and Cochrane databases.

Eligibility criteria: We included studies reporting the development of a multivariable clinical prediction model in children, with or without validation, to predict a repeatedly measured binary or time-to-event outcome and utilizing at least one repeatedly measured predictor.

Synthesis methods: We categorized included studies by the method used to model time-varying predictors.

Results: Of 99 clinical prediction model studies that had a repeated measurements data structure, only 27 (27%) used methods that incorporated the repeated measurements as time-varying predictors in a single model. Among these 27 time-varying prediction model studies, we grouped model types into nine categories: time-dependent Cox regression, generalized estimating equations, random effects model, landmark model, joint model, neural network, K-nearest neighbor, support vector machine and tree-based algorithms. Where there was comparison of time-varying models to single measurement models, using time-varying predictors improved predictive accuracy.

Conclusions: Various methods have been used to develop time-varying prediction models in children, but there is a paucity of pediatric time-varying models in the literature. Incorporating time-varying covariates in pediatric prediction models may improve predictive accuracy. Future research in pediatric prediction model development should further investigate whether incorporation of time-varying covariates improves predictive accuracy.

Keywords: clinical decision rules; longitudinal studies; pediatrics; risk factors; time-dependent.

Publication types

Review