Placebo effects emerging from the expectations of relatives, also known as placebo by proxy, have seldom been explored. The aim of this study was to investigate whether in a randomized controlled trial (RCT) there is a clinically relevant difference in long-term outcome between very preterm infants whose parents assume that verum (PAV) had been administered and very preterm infants whose parents assume that placebo (PAP) had been administered. The difference between the PAV and PAP infants with respect to the primary outcome-IQ at 5 years of age-was considered clinically irrelevant if the confidence interval (CI) for the mean difference resided within our pre-specified ±5-point equivalence margins. When adjusted for the effects of verum/placebo, socioeconomic status (SES), head circumference and sepsis, the CI was [-3.04, 5.67] points in favor of the PAV group. Consequently, our study did not show equivalence between the PAV and PAP groups, with respect to the pre-specified margins of equivalence. Therefore, our findings suggest that there is a small, but clinically irrelevant degree to which a preterm infant's response to therapy is affected by its parents' expectations, however, additional large-scale studies are needed to confirm this conjecture.
Keywords: long-term outcome; placebo by proxy; preterm infants; randomized controlled trial.