J Altern Complement Med. 2001 ; 7 Suppl 1(): S45-52.

Research on complementary and alternative medicine using randomized controlled trials.

Harlan WR.

Public Health Service, National Institutes of Health (NIH), Bethesda, MD 20892-9629, USA. [email protected]

In 1998, the National Institutes of Health (NIH) formed the National Center for Complementary and Alternative Medicine (NCCAM) from what had formerly been the Office of Alternative Medicine. This presentation opens with a brief discussion on the history of the NIH and the development of CAM at the NIH before moving on to the work of the NCCAM. The NCCAM is moving toward an integration of CAM therapies into conventional medicine, when there is evidence for the value of CAM. One of twenty-five institutes or centers at the NIH, the NCCAM looks at evidence-based medicine and public health. In this context, "public health" means educating the public about its health. The NCCAM supports training to conduct research and plays an important role in disseminating information to the public and to health providers about what works and what is safe. This evolves into the concept of evidence-based medical and public-health practices, that is, making decisions on the basis of evidence from scientifically rigorous studies that are sufficiently large to provide a confident estimate of biologically and medically important benefits and risks. In the hierarchy of generating scientific evidence, randomized controlled trials are considered the "gold standard." The NCCAM entertains proposals for studies that come spontaneously from investigators, or, upon identifying an existing need that is not being met by the investigative community, the NCCAM can initiate a request for proposals. Every proposal is subjected to a rigorous application and review process. Another possible step in the assessment of the evidence from clinical trials is to do a systematic analysis of several studies to bring together all the information that is available. Systematic reviews of smaller studies that individually might have an insufficient sample size can assist in making treatment decisions, but, importantly, they can lead the NCCAM in the development of future, definitive studies. Training to conduct research is especially important to CAM. This presentation outlines several approaches the NCCAM has to training (see http://nccam.nih.gov).