Thirty-five RCTs were included in meta-analyses (n=3,571 participants). Only two trials clearly reported a random component in the sequence generation process. None of the trials reported withdrawal and/or loss to follow-up rates. There was no blinding for participants, investigators or outcome assessors in any of these trials. Five trials had a high risk of bias relating to selective outcome reporting.
Chinese herbal medicine versus Western conventional medicine
Compared with cisapride alone, Chinese herbal medicine alone or with cisapride was associated with a significant decrease in the rate of failure to respond to treatment (RR 0.24, 95% CI 0.17 to 0.34; eight RCTs).
Compared with polyethylene glycol 4000 alone, Chinese herbal medicine alone or with polyethylene glycol 4000 was associated with a significant decrease in the rate of failure to respond to treatment (RR 0.14, 95% CI 0.06 to 0.34; six RCTs).
Compared with mosapride alone, Chinese herbal medicine alone or with mosapride was associated with a significant decrease in the rate of failure to respond to treatment (RR 0.33, 95% CI 0.23 to 0.46; six treatment arms).
Compared with phenolphthalein alone, Chinese herbal medicine in addition to phenolphthalein was associated with a significant decrease in the rate of failure to respond to treatment (RR 0.24, 95% CI 0.13 to 0.46; three RCTs).
Significant heterogeneity was only observed in the above outcome for the comparison of Chinese herbal medicine and/or mosapride versus mosapride alone (I2=70%).
One RCT reported that a traditional Chinese medicine combined with itopride led to a significant increase in the rate of overall effectiveness compared with itopride alone (p<0.05). One RCT reported that a traditional Chinese medicine combined with bifidobacteria led to a significant increase in the rate of overall effectiveness compared with bifidobacteria alone (p<0.01).
Chinese herbal medicine versus non-pharmaceutical interventions: One RCT reported that, compared with massage, Chinese herbal medicine led to a significant decrease in the rate of overall effectiveness (p<0.05).
Few studies reported adverse events of Chinese herbal medicines; these adverse events mainly included abdominal pain and diarrhoea.
Results of comparing a Chinese herbal medicine with another type of Chinese herbal medicine were also reported.