572

Abstract

Introduction: Commercial funding of randomized controlled trials (RCTs) is associated with positive results in other clinical areas. Methods: We searched the Evidence in Pediatric Intensive Care Database (epicc.mcmaster.ca). This database searches MEDLINE, EMBASE, LILACS and CENTRAL for pediatric critical care RCTs using comprehensive search strategies. We included RCTs published in English that compared two interventions and reported a primary outcome. We excluded cross-over and non-inferiority trials. Pairs of reviewers screened studies for eligibility and abstracted data. Discrepancies were resolved by consensus. Results: We included 151 trials, randomizing a total of 21 222 children, published between 1986 and July 4, 2013. 26 (17%) RCTs reported funding from a commercial source. The primary outcome results were positive in 58 (38%) trials, negative in 5 (3%), and not statistically significant in 64 (42%) of trials. Of 26 trials reporting commercial funding, 9 (35%) had positive results, which was not different from those that those that did not report commercial funding (49 of 125 (39%); p=0.66). Commercially funded trials were more often multi-centered (46% vs. 21%; p=0.01) but were not more likely to study medications (58% in both groups), be stopped early (19% vs. 15%; p=0.61), or report blinding (73% vs. 54%; p=0.07) than those that did not report commercial funding. The median (IQR) number of participants enrolled was also not different (48 (35 to 181) vs. 60 (37 to 102) p=0.82). Using logistic regression, early stopping was independently associated with positive results (p=0.01) but commercial funding, year of publication, and use of a pharmaceutical intervention or blinding were not. Conclusions: Commercial funding is not common among pediatric critical care trials and does not appear to be associated with positive results in these trials.