Interaction between the Kansas City Cardiomyopathy Questionnaire and the Pocock’s clinical score in predicting heart failure outcomes

PURPOSE: Heart failure (HF) is a complex syndrome. Its appropriate management should combine several health measurements. We assessed the relationship between the Kansas City Cardiomyopathy Questionnaire (KCCQ) and the Pocock’s clinical score. METHODS: We conducted a prospective registry of HF outpatients. The main outcome was occurrence of death or hospitalization during a 6-month follow-up. A multivariate logistic regression was performed, including the KCCQ overall summary score, the Pocock’s clinical score and their interaction in the model. RESULTS: From January 2008 to December 2010, 143 patients were involved. Mean age of patients was 68 years, and 74 % were men. KCCQ’s overall summary score and Pocock’s clinical score were inversely correlated (r = −0.24, p = 0.026). A total of 61 (42.7 %) events occurred. There was a high proportion of events (77.8 %) in patients with a Pocock’s clinical score >50 %, whatever the KCCQ score value. When the KCCQ score was ≤50 %, there was a low increase in risk as the Pocock’s clinical score increased (OR 2.0 [0.6; 6.6]). However, when the KCCQ score was between 50 and 75 or ≥75 %, there was a high increase in risk as the Pocock’s clinical score increased (OR 6.9 [1.2; 38.9] and OR 7.4 [0.8; 69.7], respectively). CONCLUSIONS: Patients with a high Pocock’s clinical score are at a high risk of death or hospitalization. For patients with a low Pocock’s clinical score, the KCCQ score can identify those at risk of these events. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s11136-015-1154-9) contains supplementary material, which is available to authorized users.