[Skip to Navigation]
Sign In
December 2018

Programming and Coding Errors in a Secondary Analysis of a Randomized Clinical Trial

JAMA Cardiol. 2018;3(12):1256. doi:10.1001/jamacardio.2018.3916

In the article titled “Identification of Patients With Stable Chest Pain Deriving Minimal Value From Noninvasive Testing: The PROMISE Minimal-Risk Tool, A Secondary Analysis of a Randomized Clinical Trial,”1 published online February 15, 2017, and in the April 2017 issue of JAMA Cardiology, statistical programming and coding errors changed the number of patients included in the minimal-risk and other cohorts, the derivation and validation subsets, and the model development set and affected the risk probabilities. All analyses were recalculated after correcting these errors; the direction of statistical significance changed in some comparisons of baseline clinical characteristics, as did 2 factors associated with minimal risk in the final derivation model.2 The overall conclusions remained unchanged. Additionally, the online risk calculator tool was taken down because the programming to support the tool needs correction; reference to this tool as well as eFigure 3 in the Supplement were removed. This article was corrected online.

Fordyce  CB, Douglas  PS, Roberts  RS,  et al; Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) Investigators.  Identification of patients with stable chest pain deriving minimal value from noninvasive testing: the PROMISE minimal-risk tool, a secondary analysis of a randomized clinical trial.  JAMA Cardiol. 2017;2(4):400-408. doi:10.1001/jamacardio.2016.5501PubMedGoogle ScholarCrossref
Fordyce  CB, Douglas  PS, Udelson  JE.  Errors in programming and coding affecting cohorts included in the study deriving and validating the PROMISE minimal-risk tool  [published online November 21, 2018].  JAMA Cardiol. doi:10.1001/jamacardio.2018.3897Google Scholar