- Email: [email protected]
Scientific fraud: action needed in China
Editorial
Scientific fraud: action needed in China
iStockphoto
The printed journal includes an image merely for illustration
For Acta Crystallographica Section E editorial see http://journals.iucr.org/e/ issues/2010/01/00/me0406/ See more on China’s research output see http://www.cas.cn/ xw/kjsm/gndt/200911/ t20091130_2677058.shtml
On Dec 19, 2009, editors at Acta Crystallographica Section E alerted the scientific community to a disgraceful pattern of fraud involving papers they had published in 2007. At least 70 false crystal structures were reported— mainly from two groups led by Hua Zhong and Tao Liu, both at Jinggangshan University, Jian, China. All authors have now agreed to retraction of 41 papers published by Zhong and 29 by Liu. It is rather surprising that wrongdoing on such a scale evaded detection during peer review and, considering that crystal structures are deposited in public databases upon publication, that the truth has been uncovered so slowly. In China, the government controls almost all funding for research. As in other countries, to gain funding researchers need to publish as many papers in high impact journals as possible. According to Science Citation Index and other resources, Chinese authors published 271 000 papers in 2008, roughly 11·5% of the world’s total. This incident is not the first time that scientific fraud has occurred in China. Regulations to monitor state-funded research projects were announced in 2006 by the Ministry of Science and Technology
in response to six high-profile cases of scientific misconduct. A new circular was issued on March 19, 2009, aimed at preventing misconduct in higher education institutions—punishment for breaching the new rules could involve warnings, dismissal, or legal action. Research programmes could be suspended or terminated, funding could be withdrawn, or awards and honours revoked. Such extensive fraud is disappointing—not only does it indicate a substantial waste of research time and money, but it is likely that, whatever punishments do result, damage to the reputations of the researchers, institutions, and journal concerned is likely to be disproportionately great. Clearly, China’s Government needs to take this episode as a cue to reinvigorate standards for teaching research ethics and for the conduct of research itself, as well as establishing robust and transparent procedures for handling allegations of scientific misconduct to prevent further instances of fraud. For Hu Jintao’s goal of China becoming a research superpower by 2020 to be credible, China must assume stronger leadership in scientific integrity. ■ The Lancet
General practitioner or computerised algorithm?
iStockphoto
The printed journal includes an image merely for illustration
94
In a recent interview in The Guardian, Mike Richards, the UK’s cancer czar, explained his plans to equip within 5 years every general practitioner with a computerised algorithm to predict cancer risk. The algorithm will be on the basis of risk factors such as age, weight, and symptoms—a tall order, because knowledge of risk factors varies for different cancers. The UK lags behind continental Europe in the diagnosis rate of cancer and the hope is that earlier diagnosis will lead to improved outcomes. The downside is that more referrals will use more resources, and could mean invasive tests and toxic therapies that have small overall effect on survival. If the proposed algorithm actually works, cancer incidence will increase too—so it will seem as if the UK has worsening cancer rates. No details of the predictive value of the algorithms were released, nor how they compare with the current capability of general practitioners to refer for cancer. The evidence base for these algorithms seems worryingly slim.
An individual’s understanding of risk is another problem: is a 2% lifetime risk good or bad news, for a 60-year-old, say? The glass-is-half-full individuals will go on their merry way. Those who see their glass as half empty will be worried. Not to mention further confusion with talk of statistical uncertainty. Are those who ponder the fullness or emptiness of the glass reassured or worried when their general practitioner turns to a computer? The answer could go either way. After all, specialisation in general practice is meant to equip the doctor with wide-ranging skills in diagnosis, and the need to refer versus the need to give reassurance. General practitioners do not come across many new cases of cancer each year, but are on high alert. The mantra is: if in doubt, refer. Having new software does not get around the real problem—individuals who do not see a general practitioner. Perhaps the algorithms should be available in pharmacies, nurses’ clinics, or simply on the web. But we are sceptical. ■ The Lancet www.thelancet.com Vol 375 January 9, 2010
Scientific fraud: action needed in China
iStockphoto
The printed journal includes an image merely for illustration
For Acta Crystallographica Section E editorial see http://journals.iucr.org/e/ issues/2010/01/00/me0406/ See more on China’s research output see http://www.cas.cn/ xw/kjsm/gndt/200911/ t20091130_2677058.shtml
On Dec 19, 2009, editors at Acta Crystallographica Section E alerted the scientific community to a disgraceful pattern of fraud involving papers they had published in 2007. At least 70 false crystal structures were reported— mainly from two groups led by Hua Zhong and Tao Liu, both at Jinggangshan University, Jian, China. All authors have now agreed to retraction of 41 papers published by Zhong and 29 by Liu. It is rather surprising that wrongdoing on such a scale evaded detection during peer review and, considering that crystal structures are deposited in public databases upon publication, that the truth has been uncovered so slowly. In China, the government controls almost all funding for research. As in other countries, to gain funding researchers need to publish as many papers in high impact journals as possible. According to Science Citation Index and other resources, Chinese authors published 271 000 papers in 2008, roughly 11·5% of the world’s total. This incident is not the first time that scientific fraud has occurred in China. Regulations to monitor state-funded research projects were announced in 2006 by the Ministry of Science and Technology
in response to six high-profile cases of scientific misconduct. A new circular was issued on March 19, 2009, aimed at preventing misconduct in higher education institutions—punishment for breaching the new rules could involve warnings, dismissal, or legal action. Research programmes could be suspended or terminated, funding could be withdrawn, or awards and honours revoked. Such extensive fraud is disappointing—not only does it indicate a substantial waste of research time and money, but it is likely that, whatever punishments do result, damage to the reputations of the researchers, institutions, and journal concerned is likely to be disproportionately great. Clearly, China’s Government needs to take this episode as a cue to reinvigorate standards for teaching research ethics and for the conduct of research itself, as well as establishing robust and transparent procedures for handling allegations of scientific misconduct to prevent further instances of fraud. For Hu Jintao’s goal of China becoming a research superpower by 2020 to be credible, China must assume stronger leadership in scientific integrity. ■ The Lancet
General practitioner or computerised algorithm?
iStockphoto
The printed journal includes an image merely for illustration
94
In a recent interview in The Guardian, Mike Richards, the UK’s cancer czar, explained his plans to equip within 5 years every general practitioner with a computerised algorithm to predict cancer risk. The algorithm will be on the basis of risk factors such as age, weight, and symptoms—a tall order, because knowledge of risk factors varies for different cancers. The UK lags behind continental Europe in the diagnosis rate of cancer and the hope is that earlier diagnosis will lead to improved outcomes. The downside is that more referrals will use more resources, and could mean invasive tests and toxic therapies that have small overall effect on survival. If the proposed algorithm actually works, cancer incidence will increase too—so it will seem as if the UK has worsening cancer rates. No details of the predictive value of the algorithms were released, nor how they compare with the current capability of general practitioners to refer for cancer. The evidence base for these algorithms seems worryingly slim.
An individual’s understanding of risk is another problem: is a 2% lifetime risk good or bad news, for a 60-year-old, say? The glass-is-half-full individuals will go on their merry way. Those who see their glass as half empty will be worried. Not to mention further confusion with talk of statistical uncertainty. Are those who ponder the fullness or emptiness of the glass reassured or worried when their general practitioner turns to a computer? The answer could go either way. After all, specialisation in general practice is meant to equip the doctor with wide-ranging skills in diagnosis, and the need to refer versus the need to give reassurance. General practitioners do not come across many new cases of cancer each year, but are on high alert. The mantra is: if in doubt, refer. Having new software does not get around the real problem—individuals who do not see a general practitioner. Perhaps the algorithms should be available in pharmacies, nurses’ clinics, or simply on the web. But we are sceptical. ■ The Lancet www.thelancet.com Vol 375 January 9, 2010