Skip to main content

Table 4 Features of flexible parametric survival models reported by the prognostic model studies

From: The current application of the Royston-Parmar model for prognostic modeling in health research: a scoping review

Features of flexible parametric survival models

n

(%)

Benefits

 Additional insights of prognostic factors versus other survival analysis models

1

(8.3)

 Compute additional parameter estimates versus other survival analysis models

1

(8.3)

 Extrapolation using the linear tail of the restricted cubic spline

2

(16.7)

 Flexibly fit (/model) the baseline function

4

(33.3)

 Improved model accuracy

5

(41.7)

 Model time-dependent effects

1

(8.3)

 Validation in other settings

2

(16.7)

 No reported benefits

3

(25.0)

Limitations

 Overfitting

1

(8.3)

 Difficult to interpret models with more than one time-dependent effect

1

(8.3)

 No reported limitations

10

(83.3)