Table 4.

Data analysis method and modes of model presentation of the studies

Study	Missing data described	Missingness handling described	Statistical/data analysis methods used for model development	Mode of model presentation
Bengtson 2022	Yes	Yes	Lasso regression	NR
Man 2021	No	No	Multivariable Cox proportional hazards regression	Risk prediction equation
Bartáková 2021	No	No	Univariate and multivariate logistic regression with backward stepwise prediction algorithm	Risk score
Joglekar 2020	Yes	Yes	Univariate and multivariate logistic models were constructed to determine an eventual statistically significant effect of any relevant variable and ROC analysis was applied to test the final models. Machine learning and traditional analysis	NR
Muche 2020	No	No	Multivariable logistic regression	Risk prediction equation
Khan 2019	No	No	Stepwise multiple (both ways) logistic analysis and machine learning approach	Decision tree
Kondo 2018	No	No	Multivariable logistic regression analysis, decision-curve analysis	NR
Allalou 2016	No	No	Machine learning	Decision tree
Ignell 2016	Yes	No	Multivariable regression analysis	Model equation
Köhler 2016	No	No	Lasso method for Cox regression	Risk score
Bartáková 2015	No	No	Logistic regression analysis and ROC analysis	NR
Lappas 2015	Yes	No	Student’s t test, multivariate logistic regression analysis	NR
Cormier 2015	No	No	General linear model procedure and using the type-III sum of squares and logistic regression and ROC analysis	NR
Kwak 2012	Yes	Yes	Multiple logistic regression analysis	NR
Kjos SL1995	No	No	Multivariate regression analysis	NR

NR not reported, ROC receiver operating characteristic