Description

Wasson et al listed methodological standards that can be used to evaluate a published report of a clinical prediction rule. These can help to identify those rules that have been adequately described. The authors are from Dartmouth Medical Center, River Junction Veterans Administration Medical Center, Stanford University, Palo Alto Veterans Administration Medical Center, University of California Berkeley, Brigham and Women's Hospital and Harvard University.


 

Standards for reporting a clinical prediction rule:

(1) The outcome(s) are defined.

(2) The predictive findings are defined.

(3) The demographics of the patients (age, gender, race, etc) are reported.

(4) The site and situation of the study are described.

(5) There is a test describing the misclassification rate.

(6) The results of its clinical use are studied prospectively.

(7) The mathematical methods are adequately described.

 

where:

• These outline the input and output behaviors.

• A modification of the score would be to rate how well each standard has been met.

 

In addition, blinded assessments should be used when appropriate:

(1) of outcome

(2) of predictive findings if a retrospective study is performed

 


To read more or access our algorithms and calculators, please log in or register.