When has enough evidence accumulated to change neonatal practice?

Randomized clinical trials are the best method to assess the safety and efficacy of therapeutic interventions. However, it is not always clear how much evidence from randomized trials is required to change clinical practice. Throughout the history of neonatal medicine, some therapies were subject to excessive and unnecessary testing through replication of clinical trials. Other therapies were adopted into clinical practice with insufficient evidence. In only a few cases was the right amount of evidence accumulated to drive a change in practice. Here we present a case history for each of these three scenarios. Arising from these, we suggest principles to identify when enough evidence exists for a therapy to become standard practice.