We examined the relationships between the continuous predictor variables and our two outcome variables at both Phases 1 and 2 (see Table 1) using Pearson's correlation coefficients.
We propose a categorisation methodology for clinical-prediction models, using Generalised Additive Models (GAMs) with P-spline smoothers to determine the relationship between the continuous predictor and the outcome.
To further investigate the effects of the predictors identified from the DistLM analysis, the continuous predictors were binned.
The remaining 11 articles either categorized or dichotomized the continuous predictors of interest.
The odds ratios for continuous predictors estimate the change in odds for a unit increase in the continuous predictor.
The continuous predictor was age, and the categorical predictors were sex, education and race/ethnicity.
The number of categories into which the continuous predictor should be categorised depends partly on the relationship -graphical and numerical- between the predictor and the outcome: hence, the need to create more than two categories should be borne in mind.
We did not find any non-linear association between the possible continuous predictor variables and the outcome variables.
The nature of the association between continuous predictors and risk of microalbuminuria was studied with restricted cubic splines with three knots (two df).
Most models use logistic regression, with a cardinal assumption being linearity in the relationship between continuous predictors (for example, blood pressure) and the log odds of outcome (death).
