An n-of-1 Trial Service in Clinical Practice: Testing the Effectiveness of Liuwei Dihuang Decoction for Kidney-Yin Deficiency Syndrome

Abstract

Objective. To describe the clinical use of n-of-1 RCTs for kidney-Yin deficiency syndrome that is a traditional Chinese medicine syndrome in publicly clinical practice in China. Methods. Our study included patients with kidney-Yin deficiency syndrome, using a within-patient, randomized, double-blind, crossover comparison of Liuwei Dihuang decoction versus placebo. Outcome Measures. Primary outcome measures included number of individual completion rates, response rate, and post-n-of-1 RCTs decisions. Secondary measures were the whole group score of individual Likert scale, SF-36 questionnaire. Results. Fifty patients were recruited and 3 were not completed. Forty-seven patients completed 3 pairs of periods, 3 (6.38%) were responders, 28 (59.57%) were nonresponders, and 16 (34.05%) were possible responders. Doctors and patients used the trial results to making decision. Three responders stayed on the medication management, 28 nonresponders ceased the LDD, 7 patients of the 16 possible responders could not give clear decision, and the others kept the same medication station. Among the whole group, neither the individual Likert score nor the SF-36 showed any statistical differences between LDD and placebo. Discussion. More attention should be paid to choose experienced TCM doctor as investigator and keep the simulant same with test medication in n-of-1 RCTs of TCM and sufficiently biological half-life period of Chinese medicine compound.

Figure 1

Flowchart for the n-of-1 trial. In this example, “weeks 1–4” is a treatment period. Weeks 1–4 and weeks 5–8, combined, are the first pair of treatment periods. The whole n-of-1 trial consists of the 3 pairs of treatment periods.

Figure 2

Example of a responder to LDD. The responder's Likert score and SF-36 show clear differences between months 1, 3, and 5 (placebo) and months 2, 4, and 6 (LDD). Higher scores of Likert score indicate worse behavior. Higher scores of SF-36 indicate better behavior.

Figure 3

Meta-analysis of individual Likert score between LDD and placebo.

Figure 4

Meta-analysis of SF-36 total score between LDD and placebo.