Sunday, June 22, 2008

ORI Research misconduct survey reports the obvious - again

The Office for research integrity has just published a survey in Nature.

Sandra L. Titus, James A. Wells Lawrence J. Rhoades (2008-06-19). Repairing research integrity. Nature. 453, 980-982.

The survey suggests that most research misconduct in the United States is unreported to the ORI. Furthermore, fraud is not rare.
The US National Academy of Sciences has asserted that scientists share an 'obligation to act' when suspected research misconduct is observed. However the reported number of investigations submitted to ORI has remained low: on average 24 institutional investigation reports per year.

"The 2,212 researchers we surveyed observed 201 instances of likely misconduct over a three-year period. That's 3 incidents per 100 researchers per year. A conservative extrapolation from our findings to all DHHS-funded researchers predicts that more than 2,300 observations of potential misconduct are made every year. Not all are being reported to universities and few of these are being reported to the ORI."

"These numbers indicate a sizeable disconnect between what universities are seeing and the 24 investigations evaluated by the ORI annually."
There is a leadership problem, whitewashing, bullying, and sham investigation.
The leaders of institutions may also have concerns about handling research misconduct. Because public image is important to institutions, some may try to minimize reporting and keep unfavourable information from reaching the ORI and the press. An institution may choose to ignore conducting an investigation and instead they may simply dismiss an accused person or even a whistleblower in the hope that the problem will go away without needing further examination. Additionally, institutional leaders may wish to ignore or minimize allegations of possible research misconduct to protect the revenue that the researcher generates; some may avoid investigations because they are costly in terms of time and money.
The paper makes six recommendations
  1. Adopt zero tolerance. Social responsibility to the academic community and to the public who fund the research will be strengthened when it is apparent that an institution has a real commitment to integrity.
  2. Protect whistleblowers: 43% reported that institutions encouraged them to drop the allegation.
  3. Clarify how to report
  4. Train the mentors
  5. Use alternative mechanisms
  6. Model ethical behaviour: People imitate the behaviour of powerful role models. Institutions successfully stop cheating, for example, when they have leaders who communicate what is acceptable behaviour, encourage faculty members and staff to follow the policies, develop fair and appropriate procedures for handling misconduct cases, focus on ways to develop and promote ethical behaviour, and provide clear deterrents that are communicated(McCabe, D. L., Trevino, L. K. & Butterfield, K. D. 2001 Ethics Behav. 11, 219–232).
The conclusion is:
"Our study calls into question the effectiveness of self-regulation. We hope it will lead individuals and institutions to evaluate their commitment to research integrity."
There are some problems here. Much serious fraud takes place at the commercial-academic interface with the acquiescence of government. The ORI has no role here. Furthermore, carefully evolved re-definitions of fraud have made it difficult to deal with the most serious instances of fraud. The ORI fails to comment on most cases of fraud which arise.

So much for leadership, role models and the rest.

Despite the appropriate conclusions, the study adds nothing beyond the long trail of other ignored studies which have reported the same. This is a never-ending loop of many roads leading from nowhere to nothing. It's all a heady mix of righteous indignation, hand sitting and some occasional wrist slapping.

So let's say it again:
  1. Research fraud is not rare.
  2. Investigation of fraud is a sham
  3. Fraudsters are usually shielded from criticism except for the occasional random punishment that maintains the illusion.
  4. Those who state the obvious are abused.
  5. Official integrity bodies and regulators are entirely useless and are often complicit.
  6. It is not necessary to conduct research to demonstrate the obvious. In the pharmaceutical arena the list is endless (Vioxx, Avandia, Zetia, Actonel).
  7. In each and every instance there are attempts at cover-up, serious damage to the public and damage to science.
  8. Most importantly, there is failure of the very organizations who claim to act as guardians of integrity to open their mouths.

It reminds me of the very old story of the Officer and the Thug


A CHIEF OF POLICE who had seen an Officer beating a Thug was very indignant, and said he must not do so any more on pain of dismissal.
"Don't be too hard on me," said the Officer, smiling; "I was beating him with a stuffed club."
"Nevertheless," persisted the Chief of Police, "it was a liberty that must have been very disagreeable, though it may not have hurt. Please do not repeat it."
"But," said the Officer, still smiling, "it was a stuffed Thug."
In attempting to express his gratification the Chief of Police thrust out his right hand with such violence that his skin was ruptured at the armpit and a stream of sawdust poured from the wound.

He was a stuffed Chief of Police.

Earlier|Later|Main Page


Anonymous said...

The ORI's teeth were extracted around 1985. Since then they have been relegated to blubbings of the type reported in Nature this week.

Anonymous said...

The ORI has even refused to act on a complaint that was forwarded to it by the editors of a MAJOR publishing group, who told the whistleblower involved that "the institution's handling of the investigation was disgusting." The matter was nevertheless closed by ORI and the whistleblower was ultimately silenced (eg, bought off) by the institution through a legal settlement. Aside from the usual non-disparagement clauses, the settlement agreement also required the complainant to forfeit all documents related to the research misconduct allegations, as well as the right to put in future public records act requests for the respondent's research data, grant applications, or other incriminating documents. This is how scientific misconduct investigations are handled in the REAL world.