Do sick newborn infants benefit from participation in a randomized clinical trial?

BACKGROUND: Adult participants in randomized controlled trials often have better outcomes than patients who are eligible but not enrolled. OBJECTIVE: To examine whether newborn infants who were allocated to placebo in an investigational drug trial had better outcomes than infants who were eligible but not randomized (eligible NR). STUDY DESIGN: During a randomized controlled trial of antithrombin therapy in premature infants with respiratory distress syndrome, data were collected prospectively on all 76 infants in the eligible NR group. Study outcomes were compared with those of all 61 infants who were randomized to placebo. The same exogenous surfactant was used in all patients. RESULTS: In the placebo group the mean (SD) birth weight was 1201 (314) g, mean (SD) gestational age was 28.8 (2.3) weeks, and 51% were male. In infants in the eligible NR group, mean (SD) birth weight was 1141 (262) g, mean (SD) gestational age was 28.3 (2. 3) weeks, and 58% were male; 57% of infants in both groups had been exposed to steroids before birth. The median duration of mechanical ventilation was reduced from 6.2 days in the eligible NR group to 4. 8 days in the placebo group (P =.008). There was also a trend toward less frequent and less severe intraventricular hemorrhage in trial participants. CONCLUSIONS: These data are consistent with the hypothesis that sick newborn infants may benefit from participation in a randomized controlled trial.