To help clinicians and patients understand prediction models, model presentation can play an important role. For logistic models the standard method for visualization is currently the nomogram (1). In a nomogram each predictor is represented by a line segment, where this segment is longer for predictors with larger predictive value. We developed an alternative method based on color bars (2). This method represents predictors by color bars of equal length. The relative strength of predictors is accounted for by the colors used in the bars: in essence, more predictive predictors have a color bar that changes color more quickly.

References

Balachandran VP, Gonen M, Smith JJ, DeMatteo RP. Nomograms in oncology: more than meets the eye. Lancet Oncol. 2015;16(4):e173-e80.

Van Belle V, Van Calster B. Visualizing risk prediction models. PLoS ONE 2015;10(7):e0132614. doi:10.1371/journal.pone.0132614  Manual  R package