We will be performing site maintenance on our learning platform at learn.aaos.org on Sunday, February 5th from 12 AM to 5 AM EST. We apologize for the inconvenience.

“No one study—no matter how good it is or how profound it seems—should change the course of clinical care. You must have the data corroborated and have the experiment repeated by someone else.” —James D. Heckman, MD

AAOS Now

Published 4/1/2009
|
Annie Hayashi

Massive fraud revelations stun orthopaedics

Fabricated data discredit prominent pain management researcher

Revelations about a well-known pain management researcher have hit orthopaedics, anesthesia, and other medical fields, resulting in more than 20 scientific articles being identified as containing fabricated data.

Scott S. Reuben, MD, was one of the most prolific investigators in the field of anesthesia and analgesia, particularly for orthopaedic perioperative and postoperative pain management. His work on multimodal analgesia appeared in numerous peer reviewed medical journals, and he was frequently invited to speak on the subject.

So when an investigation begun last year by Baystate Medical Center in Springfield, Mass., recently found that Dr. Reuben had fabricated part or all of the data used in 21 of his studies since 1996, the news surprised and shocked both the anesthesia and the orthopaedic communities.

Both The Journal of Bone and Joint Surgery (JBJS) and Anesthesia &Analgesia, which were among the journals that have published Dr. Reuben’s studies, have posted retractions on their Web sites. Anesthesiologists and orthopaedists who had used Dr. Reuben’s treatment protocols were left with many unanswered questions.

“It is mind-boggling,” said Santhanam Suresh, MD, professor of anesthesiology and pediatrics, Children’s Memorial Hospital, Northwestern University’s Feinberg School of Medicine, and section co-editor for the Academy’s Orthopaedic Knowledge Online management of pain in orthopaedics Web site.

“Those of us in the anesthesia community who deal with acute postoperative pain management have adopted some of Dr. Reuben’s techniques because of his solid, Level I evidence. It appeared to prove that the addition of a COX-2 inhibitor with pregabalin or gabapentin might actually reduce the amount of postoperative pain,” he said.

“It definitely sets our understanding of perioperative analgesic management back,” said Steven L. Shafer, MD, editor-in-chief of Anesthesia & Analgesia, which has retracted 10 of Dr. Reuben’s previously published studies.

“Dr. Reuben advanced certain concepts in multimodal analgesia throughout his career and some of that work has not been reproduced by other laboratories. That work must be considered impeached since we don’t know what is true,” he stated.

Two abstracts trigger investigation
The deception was uncovered last year when two abstracts submitted by Dr. Reuben for Baystate Medical Center’s annual “Research Week” were found to lack “necessary institutional review board oversight,” according to Hal B. Jenson, MD, MBA, Baystate’s chief academic officer.

“This prompted Baystate to conduct an investigation of Dr. Reuben’s past research, which eventually uncovered an extensive history of fabrication dating back to 1996,” said Baystate spokesperson, Jane Albert, director of public affairs.

According to Baystate Medical Center’s Policies and Procedures for Misconduct in Research and Scholarly Activities, fabrication is defined as “making up data or results and recording or reporting them.”

Dr. Reuben practiced at Baystate but was employed by Springfield Anesthesia Services, Inc. As a part of this group, he served as Baystate’s director of acute pain and conducted research at the facility supported, in part, with grants from Pfizer. He has been on medical leave from his positions since May 2008.

According to a statement released by Baystate, Dr. Reuben is “barred from research and educational activities at Baystate for at least 10 years.”

Dr. Reuben participated in the investigation and, according to Dr. Jenson, offered his full cooperation although his “recall of his earlier work was limited.”

When it was determined that some or all of the data had been fabricated in 21 of Dr. Reuben’s studies, the “stakeholders were all identified and contacted,” said Dr. Jenson.

“Although these circumstances may be viewed as a failure that occurred, there is also a positive outcome—the review process did work to identify the fabrication. That is important to remember,” Dr. Jenson said.

Red flags?
For years, no one in either the anesthesia or orthopaedic communities suspected that Dr. Reuben was fabricating data.

“I went back and looked at the reviews of Dr. Reuben’s papers and not once did the question of data fabrication come up,” said Dr. Shafer.

“I actually created one of the graphs in one of his retracted papers myself. I didn’t like the graph that he had created and I had a publication deadline. So I asked him to send me the spreadsheet and did the graph the way it needed to be done. There was no indication in the data that I was working with a fabricated spreadsheet,” he explained.

According to JBJS editor James D. Heckman, MD, “We rely upon the honest reporting of the authors. Fortunately, virtually everyone we deal with is honest in trying to do the right thing to advance science. Most people are scrupulous, meticulous in their disclosure of any problems or any potential problems.”

The fact that Dr. Reuben conducted prospective, randomized, controlled studies may have increased the medical community’s confidence in his work.

“Dr. Reuben was using the gold standard. You expect someone coming from a reputable university setting to go through the rigors of a proper scientific presentation,” Dr. Suresh said.

  • In retrospect, however, several editors cited the following possible “red flags” that could have called the data into question:
  • The results of Dr. Reuben’s studies on adding COX-2 inhibitors along with pregabalin or gabapentin were always positive.
  • Although some of Dr. Reuben’s findings were corroborated by other investigators, many were not.
  • Any new investigational drug trials must be registered at clinicaltrials.gov, a Web site sponsored by the National Institutes of Health. None of Dr. Reuben’s studies were registered.

According to Dr. Suresh, “When clincal trials are registered, data are generally under greater scrutiny. Important information is required, including the IRB approvals that initially led to Dr. Reuben’s investigation.”

Dr. Suresh also indicated Dr. Reuben’s research came at a time when the anesthesia community was looking for adjuvants to opioids for effective analgesia.

“I thought multimodal analgesia, particularly for orthopaedic surgery, was a revolutionary step forward in managing pain. That certainly may be questionable at this point,” he said.

Cleaning up the contamination
The journal editors acknowledge that the 21 studies identified by Baystate Medical Center as including fabricated data are just the tip of the iceberg. (See “
Fraudulent data”) According to Pub Med, Dr. Reuben has published 72 studies since 1991. The ISI Science Citation Index indicates that Dr. Reuben’s work has been cited in 763 other journal articles—adding yet another layer of complexity to an already very complicated situation.

“A question mark hangs over all of his work. I think we owe it to our readers and to the scientific and medical community to try and make some sort of determination about his other work,” said Dr. Shafer. “I am having discussions with other editors-in-chief about what we can do.”

Though Dr. Shafer concedes that journal editors do not have the resources or legal authority to con­duct a complete investigation into all of Dr. Reuben’s work, he would like to make some effort “to identify those papers where other authors might come forward” with data and source documents and “allow those papers to stay as part of the unimpeached medical literature.”

He is also making changes at Anesthesia & Analgesia to help prevent this from occurring again. “We will now be requiring more than one author to vouch for both the integrity of the raw data and the data analysis,” he said.

At JBJS, Dr. Heckman notified subscribers that one article that was identified by Baystate Medical Center as fraudulent was being retracted, as was a Current Concepts Review from 2007. The decision to retract the review article was based on the fact that there are “several places where statements of fact regarding perioperative pain management are only supported by Dr. Reuben’s work,” said Dr. Heckman.

In addition to “cleansing” his own journal, Dr. Heckman would like to look at the orthopaedic literature as a whole. “Perhaps we need a work group of orthopaedic editors to look at how pervasive Dr. Reuben’s work is and how much it may have compromised the orthopaedic literature.”

Both Drs. Shafer and Heckman agreed, however, that no number of safeguards will prevent someone who is intent on fabricating data from doing so.

Moving forward
The use of multimodal analgesia for orthopaedic surgery may be beneficial in certain cases, according to Dr. Suresh. “But specific use of some of the drugs used in Dr. Reuben’s studies lack scientific merit at the present time,” he said.

He believes further “scrutiny of the use of these drugs is necessary and further prospective, randomized, controlled trials are required to justify the use of these drugs.”

Dr. Heckman has a caution for his readers: caveat lector—let the reader beware. “No one study—no matter how good it is or how profound it seems—should change the course of clinical care. You must have the data corroborated and have the experiment repeated by someone else.

“Only when you have equivalent data from different, independent sources should you ever really make a substantial change in your practice,” he said.

Annie Hayashi is the senior science writer for AAOS Now. She can be reached at hayashi@aaos.org

Additional References:
Letter from James D. Heckman, editor-in-chief, The Journal of Bone and Joint Surgery
http://www.ejbjs.org/misc/retraction_apr09_ltr.pdf

Article from Anesthesiology News,Fraud Case Rocks Anesthesiology Community