Three Harvard University researchers have been accused of breaking conflict-of-interest rules after failing to properly declare millions of dollars in fees from drugs firms.
An investigation by Senator Charles Grassley of Iowa found the psychiatrists underreported external earnings over a period of seven years.
One of those involved is Joseph Biederman, a renowned child psychiatrist whose research is linked to an increased use of antipsychotic medicines in children.
Dr Biederman was found to have earned at least $1.6 million (?810,000) - much of which was not declared - from pharmaceutical companies.
In evidence cited by The New York Times, the congressional investigation found that while he initially reported no income from pharmaceutical company Johnson & Johnson for 2001, for example, the firm's records show that it paid him $58,169 that year.
Two of his colleagues, Timothy E. Wilens and Thomas Spencer, are also accused of failing to disclose income from the drugs industry of $1.6 million and $1 million respectively.
All three deny deliberately concealing their earnings, and Harvard has now launched a formal investigation. In a statement, Dr Biederman said his interests were "solely in the advancement of medical treatment through rigorous and objective study" and added that he took conflict-of-interest policies "very seriously".
Dr Wilens said: "I have always believed that I was acting within the applicable, relevant guidelines and rules."
Dr Spencer also defended his paid-for appearances at professional conferences. "It was my sincere belief that I was at all times complying with the relevant policies and procedures as to outside income."
A spokeswoman for Harvard said: "The information released by Senator Grassley suggests that, in certain instances, each doctor may have failed to disclose outside income from pharmaceutical companies ... that should have been disclosed." She said the matter had been referred to a university conflict committee for review.
Dr Biederman is one of the most influential researchers in child psychiatry. However, he is also seen as a major contributor to the controversial 40-fold increase in the diagnosis of paediatric bipolar disorder between 1994 and 2003.
Many of his studies have been financed by drugs companies, a model that is not unusual in the US.
To protect research integrity, the National Institutes of Health require researchers to report earnings of $10,000 or more a year to their universities, which includes consulting fees from manufacturers of drugs being studied in federally financed trials.
