On the night of his 12th wedding anniversary, Andrew Friedman was terrified.
This brilliant surgeon and researcher at Brigham and Women's Hospital and Harvard Medical School feared that he was about to lose everything -- his career, his family, the life he'd built -- because his boss was coming closer and closer to the truth: For the past three years, Friedman had been faking -- actually making up -- data in some of the respected, peer-reviewed studies he had published in top medical journals.
``It is difficult for me to describe the degree of panic and irrational thought that I was going through,'' he would later tell an inquiry panel at Harvard.
On this night, March 13, 1995, he had been ordered in writing by his department chair to clear up what appeared to be suspicious data.
But Friedman didn't clear things up.
``I did something which was the worst possible thing I could have done,'' he testified.
He went to the medical record room, and for the next three or four hours he pulled out the permanent medical files of a handful of patients. Then he covered up his lies, scribbling in the information he needed to support his study.
``I created data. I made it up. I also made up patients that were fictitious,'' he testified.
Friedman's wife met him at the door when he came home that night. He wept uncontrollably. The next morning he had an emergency appointment with his psychiatrist.
But he didn't tell the therapist the truth, and his lies continued for 10 more days, during which time he delivered a letter, and copies of the doctored files, to his boss. Eventually he broke down, admitting first to his wife and psychiatrist, and later to his colleagues and managers, what he had been doing.
Friedman formally confessed, retracted his articles, apologized to colleagues and was punished. Today he has resurrected his career, as senior director of clinical research at Ortho-McNeil Pharmaceutical Inc, a Johnson & Johnson company.
His case, recorded in a 2.1m-high stack of documents at the Massachusetts Board of Registration in Medicine, tells a story of one man's struggle with power, lies and the crushing pressure of academia.
The story itself is more common than most people might realize.
Allegations of research misconduct reached record highs in the United States last year -- the Department of Health and Human Services received 274 complaints, which was 50 percent higher than 2003 and the most since 1989, when the federal government established the program to deal with scientific misconduct.
Chris Pascal, director of the federal Office of Research Integrity, said its 28 staffers and US$7 million annual budget haven't kept pace with the allegations. The result: Only 23 cases were closed last year. Of those, eight individuals were found guilty of research misconduct. In the past 15 years, the office has confirmed about 185 cases of scientific misconduct.
Research suggests this is but a small fraction of all the incidents of fabrication, falsification and plagiarism. In a survey published June 9 in the journal Nature, about 1.5 percent of 3,247 researchers who responded admitted to falsification or plagiarism. One in three admitted to some type of professional misbehavior.
Some cases have made headlines:
Eric Poehlman was convicted this year of fabricating research data to obtain a US$542,000 federal grant while working as a professor at the University of Vermont College of Medicine.