Method for evaluating prediction models that apply the results of randomized trials to individual patients

Table 6 Net benefit for the Dutasteride data.

T	Treat all patients	Treat by prediction	Advantage of prediction
0.5%	0.026	0.027	0.001
1.0%	0.021	0.024	0.002
2.0%	0.011	0.017	0.006
3.0%	0.001	0.011	0.009
4.0%	-0.009	0.012	0.012
5.0%	-0.019	0.009	0.009
7.5%	-0.044	0.002	0.002
10.0%	-0.069	0.002	0.002
12.5%	-0.094	0.000	0.000
15.0%	-0.119	-0.001	-0.001

T is the treatment threshold corresponding to the minimum reduction in absolute risk required to consider opting for treatment. The units are the equivalent of the number of events per patient, where an event is either acute urinary retention or surgical intervention for BPH. The "advantage of prediction" column shows the increment in net benefit of prediction compared to the better of "treat all" and "treat none".

ISSN: 1745-6215