How Often Are Patient-Important Outcomes Represented in Neonatal Randomized Controlled Trials? An Analysis of Cochrane Neonatal Reviews

Abstract

<b><i>Background:</i></b> Research findings based on patient-important outcomes (PIOs) provide more useful conclusions than those that are based on surrogate outcomes. It is unclear to what extent PIOs are represented in neonatal randomized controlled trials (RCTs). <b><i>Objectives:</i></b> We determined the proportion of PIOs in neonatal RCTs included in Cochrane Neonatal reviews. <b><i>Methods:</i></b> We extracted up to 5 outcomes from each RCT included in Cochrane Neonatal reviews published until January 2018, with independent determination of PIOs among authors followed by a discussion leading to a consensus. We defined PIOs as outcomes that matter to patient care, such as clinical events or physiological or laboratory parameters that are widely used to guide management. <b><i>Results:</i></b> Among 6,832 outcomes extracted from 1,874 RCTs included in 276 reviews, 5,349 (78.3%) were considered PIOs; 461 studies (24.5%) included 5 or more PIOs, 1,278 (68.2%) included 1–4 PIOs, while 135 (7.2%) had no PIO included. PIOs were observed more often among dichotomous than among continuous outcomes (94.9 vs. 61.5%; RR: 1.54; 95% CI: 1.50–1.58), and more among subjective than among objective outcomes (95.9 vs. 76.8%; RR: 1.25; 95% CI: 1.22–1.28). Newer studies were more likely to have a greater number of PIOs (adjusted OR: 1.033 [95% CI: 1.025–1.041] with each publication year). <b><i>Conclusions:</i></b> The large and increasing representation of PIOs over the years suggests an improving awareness by neonatal trialists of the need to incorporate important outcomes in order to justify the utilization of resources. Further research should explore the reasons for non-inclusion or non-reporting of PIOs in a small proportion of RCTs.

The Choice of Population and Outcomes in Neonatal Trials on Hyperbilirubinemia: Are They Relevant? An Analysis of Cochrane Neonatal Reviews

“…The findings in this study serve as an example of clinical research in which a surrogate outcome that is widely used in practice is favored over the outcomes that actually matter to the patients. They echo the finding of a previous study that assessed data from all Cochrane Neonatal reviews, which showed that 7.2% of all the RCTs included in Cochrane Neonatal reviews reported only surrogate outcomes without any PIO [29]. Although surrogate outcomes, like serum bilirubin, are popular in clinical research due to their relative ease of measurement, overreliance on them may result in misleading interpretation of trial results and inappropriate practice recommendations [30-32].…”

Evaluation of the Quality of Perinatal Trials: Making the GRADE

“…To be valid and applicable, and to inform policy and practice, RCTs need to assess effects on the outcomes of greatest importance to infants and their families and health services [28]. A recent study of nearly 2,000 RCTs showed a consistent increase in the proportion of trials reporting infant- and family-important outcomes over the past few decades [29]. Some areas of neonatal research, however, assess “surrogate,” rather than infant- and family-important, outcomes more commonly.…”