CHAPTER 3.1 88 described by Breslow and Day.31 For first HF hospitalization, we did not estimate SMRs because existing prediction models for hospitalization are largely influenced by admission policies within individual health settings and hence have insufficient discriminative model performance.32 The largest RCTs (BEAUTIFUL and SHIFT) in this analysis only included patients who were in sinus rhythm (SR) and the BEAUTIFUL study included a population who had CAD; therefore, sensitivity analyses were conducted in subsets of registry patients who were (i) in SR or (ii) diagnosed with CAD. Missing data was multiply imputed by chained equations using the mice package in R. The number of imputations was set at 20.33 Statistical significance was set at level 0.05. Statistical analysis was performed using the R statistical software version 3.6.1 (R Core Team, 2019) and Stata SE Version 15 (StataCorp LP, College Station, TX).34,35 RESULTS Eligibility for potential trial enrollment Out of 46 914 patients from the registries, 14 584 were females (31.1%). After applying the harmonized set of eligibility criteria, 8294 out of 14 584 (56.9%) females and 17 818 out of 32 330 (55.1%) males in the registries were considered eligible for RCT for a final RCT-eligible group of 26 104 (31.8% females). Cancer was the most restricting criteria with 27.4% of females and 28.1% of males excluded. ACEI and ARB use at baseline differed between sexes with 15.7% of females and 11.6% of males who were ineligible for not taking ACEI or ARBs (Supplementary Table 2). In the RCT population, the observed number of females was significantly lower with 3 663 out of 16 917 (21.7%) patients (Figure 1). Baseline characteristics Baseline characteristics for the RCT population, RCT-eligible and RCT-ineligible patients stratified by sex are shown in Table 1. Overall, patients in the RCTs were younger compared to RCT-eligible and RCT-ineligible patients, with similar directions for both females (66.3 vs 73.9 vs 76.7 years) and males (62.8 vs 69.8 vs 73.8 years) in the three groups respectively. Compared to males, females were significantly older in all three groups (Table 1). Similarly in females and males, a minority of the RCT population had a LVEF below 30% (28.7% and 32.6%) as opposed to both the registry
RkJQdWJsaXNoZXIy MjY0ODMw