Statistical suggestions for long-term outcomes for non-ischemic dilated cardiomyopathy.

Grant support: Nil.

Conflicts of interest: Nil.

We thank Bursa N. for the statistical suggestions for our article entitled, ‘Long-term outcomes of non-ischemic dilated cardiomyopathy patients with left ventricular ejection fraction ≤19% on medical therapy’. We agree with Bursa’s observations regarding the use of Mann–Whitney U test for comparison of two groups in non-parametric distribution.

Bursa N. has highlighted the effect of ‘Multicollinearity’ in logistic regression model and suggested us to use forward and backward variable selection method. After putting these variables ‘baseline NYHA class, heart failure hospitalizations, use of beta-blockers, aldosterone antagonists, and angiotensin converting enzyme inhibitors/angiotensin receptor blockers’ in forward method, we found higher baseline NYHA class (p value = 0.02) and absence of use of aldosterone antagonists (p value = 0.009) as the predictor of mortality in binary logistic regression model. However, our primary outcome was to find the long-term transplant free survival of the patients, the raw data for long-term survival had been included in supplementary Tables 1 and 2.