Feature Research

Data rising from the East

BMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d7480 (Published 05 December 2011) Cite this as: BMJ 2011;343:d7480

This article has a correction. Please see:

  1. Geoff Watts, freelance journalist
  1. 1London, UK
  1. geoff@scileg.freeserve.co.uk

The number of scientific publications coming out of China is increasing rapidly, but concerns remain about the quality. Geoff Watts reports

The rise of science in China has been spectacular. Flick through the pages of 20 year old issues of Nature, Science, and other international journals and you’ll find only a scattering of research reports from the country. At that time, as data from the Science Citation Index show, China’s scientists published only about 6000 papers a year. Compare this with the 2008 figure of well over 270 000, or 11.5% of the year’s global output.

No wonder a Royal Society report earlier this year suggested that by 2020 China could be the world’s most prolific producer of research.1 And what is true of Chinese science in general finds a strong echo in biomedicine. Chinese work is now commonplace in basic research journals such as Cell, Genetics, or Biochemistry. And, to a lesser extent, the same is becoming true of clinically oriented publications such as the Lancet and the New England Journal of Medicine.

The only cause for disquiet about this impressive and generally laudable embrace of science is a persistent undertone of claims (or proof) that not all the published Chinese work, even in peer reviewed international journals, is of high quality and entirely to be trusted. When it comes to Chinese language journals, the index of suspicion is higher still.

A recent experience of Greg Irving from the department of health services research at Liverpool University makes the point. At the September meeting in Warsaw of WONCA, the family doctors’ organisation, he spoke about a Cochrane review on the efficacy of hepatitis A vaccination. The data scrutinised for the study were overwhelmingly Chinese. “Out of the 11 studies that we identified,” he …

View Full Text

Log in

Log in through your institution


* For online subscription