A step-by-step approach is usually followed in the development of new herbal medicines, but may ordinarily be less necessary for a study to validate the safety and efficacy of a traditional herbal medicine.

The point of entry to the trial phases will be determined by the nature and history of the herbal medicines being studied.

Clinical trials are generally designated in terms of a “phase”, although study designs appropriate for the clinical evaluation of a herbal medicine may, strictly speaking, fall on the borderline between two of the following classical definitions of the usual phases.

Phase I: First trials for a new compound or a new formulation that are generally carried out with a small number of healthy volunteers or patients suffering from the disease for which the herbal medicine is intended. The main purpose of a phase I trial is to observe tolerance to the herbal medicine and therefore to get an indication of the dose that might be used safely in subsequent studies.

Phase II: Studies on a limited number of patients to determine clinical efficacy and to further confirm safety. Such trials are preferably designed as randomized, double-blind, controlled studies, using for control groups either an existing alternative treatment or a placebo. The dosage schedules established in such studies are then used for a more extensive clinical study.

Phase III: A larger patient group is usually studied at several centres using a randomized double-blind design to validate preliminary evidence of efficacy obtained in earlier studies. Ordinarily, such trials are conducted under conditions which are as close as possible to the anticipated conditions of normal use.

Phase IV: Studies performed after the dosage form is available for general use. The main purpose of such studies is to detect toxic events that may occur so rarely that they are not detected earlier.

Individual countries may design clinical trials that follow the general principles embodied in the four phases mentioned above, namely, first to ensure general safety, then to determine efficacy and finally to use post-marketing surveillance to be certain that rare but serious adverse reactions are not occurring and to confirm the long-term efficacy.