Statistical Analysis
For descriptive analysis, we compared the baseline characteristics of the patients who had 3 or more EF measurements versus fewer than 3 measurements. Normally-distributed continuous variables were compared with t-test. Categorical variables were compared with chi-square test.
We categorized the study population into two groups based on whether their follow-up EF measurements were >35% or ≤35%. Subgroup analyses were performed among patients with ischemic and non-ischemic cardiomyopathy and those with normalized EF. In additional analyses, we also categorized the study population into three groups based on their follow-up EF measurements. Patients who had a >10% increase in EF were categorized as improved, those with a >10% decrease in EF were categorized as reduced, and patients with <10% change in EF in either direction were categorized as unchanged. We performed sensitivity analyses using >5% cutoff instead of 10%. Unadjusted logistic regression was used to comparing ischemic and non-ischemic proportions at first follow up. All analyses were performed using SAS (SAS Institute) version 9.4. A p value < 0.05 was accepted as significant.