Adaption and Application of the Four Phase Trials to Traditional Chinese Medicines

Table 2

Main objectives and PICOS (Patient, Intervention, Comparator, Outcome, and Study design) of trials in different stages of testing in TCM.

	Phase I	Phase II	Phase III	Phase IV
Objectives	(1) General safety evaluation; (2) PICO specification	Pilot testing on efficacy and on acute, common side effects and harms	Full-scale testing on efficacy and on acute, common side effects	(1) Surveillance on rare chronic harms; (2) Investigation on mechanisms and active substances
Study designs	(1) Case series, n of 1 trials; (2) Analysis of routine data; (3) Observational studies; (4) Non-/quasi-RCTs, cross-over trials sequential trials, and so forth	(1) Analysis of routine data, observational studies; (2) Non-/quasi-, open-labeled, cross-over trials, sequential trials, small-scale RCTs, and so forth	Rigorously-designed RCTs, large multicentre RCTs	(1) Analysis of routine data; (2) Observational studies (cohort studies and case control studies); (3) Laboratory studies on mechanisms and active substances
Patients	(1) People with or without the target WM disease; (2) Sample size: dozens	(1) Patients with the target WM disease and specific characteristics; (2) Sample size: dozens to hundreds	(1) More representative patients; (2) Sample size: hundreds to thousands	(1) Usually those who have received the treatment and those who have not; (2) Sample size: thousands and often more
Interventions	CHMs with varying herb components, doses, routes, therapeutic durations, and so forth	CHM of relatively fixed combination of herbs, dose, route, and so forth, which are determined in phase I	CHM similar to that in phase II or more fixed	CHM similar to that in phase II and III possibly with minor modifications in real practice
Controls	No control, or no treatment, standard proven effective treatments or placebo as control	No treatment, or different doses or delivery routes, standard proven effective treatments, or placebo as control	Standard proven effective treatment or placebo	Usually those patients who have not taken the treatment
Outcomes	Routine laboratory tests (e.g., blood counts, renal and liver functions, and ECG) for detecting side effects and harms, and surrogate outcomes defined in WM/TCM for detecting benefits	Surrogate outcomes for beneficial effects and appropriate outcomes for acute common harms	Endpoint outcomes for benefits and appropriate outcomes for acute common harms	Endpoint outcomes for rare chronic harms such as deaths