SAT-041 A META-ANALYSIS OF THE REPRESENTATIVENESS OF RANDOMIZED CONTROLLED TRIAL COHORTS IN END-STAGE KIDNEY DISEASE

Abstract

Systematic differences between patients included in randomized controlled trials (RCTs) and the general patient population may influence the generalizability of RCT findings. Comprehensive national registries of patients with end-stage kidney disease (ESKD) on dialysis provide a unique opportunity to compare trial and real-world patient cohorts. We aimed to determine if participants in large, multicentre dialysis trials were similar to the general dialysis population in terms of age, co-morbidities and mortality rate. Medline, PubMed and the Cochrane Central Register of Controlled Trials were systematically searched from 2007 to 2016. Data sources were published manuscripts, supplementary material and trial registration information. General dialysis population data was derived from US Renal Data System (USRDS). RCTs enrolling only participants on dialysis for ESKD; with ≥100 adult participants from ≥2 sites were included. Abstract screening and data extraction were performed independently by two researchers. Data were pooled using a random-effects model. The primary outcome was difference in mean age between the RCT and USRDS populations. Secondary outcomes included differences in mortality rate and co-morbidities. We identified 189 RCTs, enrolling 80,104 participants. Compared to the 2011 USRDS population, RCT participants were younger (58.9 [95% confidence interval 58.3-59.5] vs 61.2 years; P<0.001), more likely to be male (58.8% [57.5-60.0] vs 55.7%; P<0.001) and to have coronary artery disease (26.7% [22.1-31.4] vs 17.7%; P<0.001) and less likely to have diabetes (40.4% [36.9-43.8] vs 44.2%; P=0.035) or heart failure (19.9 [15.6-24.3] vs 29.8; P<0.001). The mortality rate during trial participation was less than half that of the USRDS population (8.92 [7.85-10.00] vs 18.59 per 100 patient-years; P<0.001). The differences in age, mortality and coronary artery disease remained when studies recruiting only from the USA were considered. Diabetes was more common in RCT participants from the USA than in the registry population. Participants in large, multicentre dialysis RCTs are younger, have a different pattern of co-morbidities and a lower mortality rate than the general dialysis population. This finding has implications for the generalization of trial results to the broader patient population and for future trial design.