Positive Health Online

Abstract

VICKERS and colleagues, Research Council for Complementary Medicine, London UK investigated potential research bias by analysing the results of clinical trials originating in various countries .

Background

Methodology

Sources were abstracts from Medline, January 1966-June 1995. Two separate studies were conducted. The first comparing clinical outcome of subjects receiving acupuncture compared to groups receiving placebo, no treatment to a nonacupuncture intervention. The second study compared the results of randomised or controlled trials of interventions apart from acupuncture published in China, Japan, Russia/USSR or Taiwan with those published in England. Determination of inclusion, outcome and classification of trial by country of origin were performed by blinded reviewers.

Results

252 of 1085 abstracts of acupuncture trials met the inclusion criteria. All trials which originated in China, Japan, Hong Kong and Taiwan were positive as were 10 of 11 studies published in Russia/USSR . In the nonacupuncture intervention studies, 405 of 1100 abstracts met the inclusion criteria. Compared to China (99%), Japan (89%), Russia/USSR (97%) and Taiwan (95%), where the results of the test treatment were superior to controls, only 75% of trials published in England gave the test treatment as superior to control. No trial published in China or Russia/USSR found a test treatment to be ineffective .

Conclusion

Certain countries publish an unusually high proportion of positive results, which could be the result of publication bias. Researchers undertaking systematic reviews need to consider how to manage research data from these countries.

References

Vickers A et al. Do certain countries produce only positive results? A systematic review of controlled trials.Control Clin Trials 19(2): 159-66 April 1998.