Skip to main content
. Author manuscript; available in PMC: 2013 Feb 1.
Published in final edited form as: JACC Cardiovasc Imaging. 2012 Feb;5(2):131–140. doi: 10.1016/j.jcmg.2011.11.011

Figure 3. Net reclassification improvement by echocardiography and NT-proBNP according to baseline heart failure risk.

Figure 3

Net reclassification improvement (NRI) across the Health ABC HF Risk categories with addition of the echocardiographic score (A), NT-proBNP (B), or both (C) to the Health ABC HF model; in (D) echocardiographic score is added after refinement of risk with NT-proBNP. Five different scenarios are presented with varying weights (from 1:1 to 16:1) for correct reclassification of events vs. nonevents, representing increasing importance of events; 1:1 denotes that correct reclassification is equally important for events and nonevents whereas 16:1 denotes that correct reclassification of an event is 16 times more important than correct reclassification of a nonevent. Clinically relevant correct reclassification (indicated by >10% positive bars) was observed with echocardiography (A) and NT-proBNP (B) across assigned weights for participants with baseline 5%–10% (green bars) or 10%–20% (yellow bars) heart failure risk as clinically determined by the Health ABC HF Risk Score. Modest synergistic gains were observed when NT-proBNP and echocardiography were used simultaneously (C). When risk was first refined by NT-proBNP (D), echocardiography provided modest incremental benefit. wNRI: average weighted NRI, the weighted percentage of participants correctly reclassified.