Do we need evidence for feeding?

Abstract

Asking the question “do we need evidence for feeding?” raises several concerns that are closely linked in clinical practice. What is the nature of the evidence in medicine? What are the specificities of the evidence in clinical nutrition, management of the evidence and of the uncertainty, as well as ethical implications? Searching for the evidence is an old issue that became more relevant in the eighties with the development of the evidence-based medicine (EBM) theory and the increasing importance of randomised controlled trials. Except some specific clinical situations, the level of proof is often weak in clinical nutrition. Two examples illustrate this situation: the glycemic control and the early enteral nutrition in critically ill patients. Some methodological issues may explain the discrepancies between the studies, but more deeply, these discrepancies demonstrate that, in opposition with the view of EBM promoters, the proof has always two components: a factual and a contextual one. Facts do not occur alone, a scientific fact cannot be isolated from its emerging context. Asking the question “do we need evidence for feeding?” refers to the specificities of clinical nutrition in comparison with other treatments. Nutrition is not only a “cure” but also a “care” with all its complexity. Stopping nutrition is associated with death. To initiate, continue or stop nutritional support should be associated with an ethical approach. In summary, in clinical practice, physicians are not facing an automatic application of the evidence but have to deal with uncertainty and manage ethical issues.