The death of an adolescent in a medical trial last year has federal health officials investigating whether financial rewards earned by some researchers may affect patient safety and public health.

In a conference at the National Institutes of Health, officials said Tuesday that a new era of biotechnology economics, in which researchers may earn millions through stock deals with drug companies, has raised serious questions about scientific integrity and threatens public confidence in medical research.

"If a crisis in confidence develops, if research subjects no longer feel safe, then medical research will grind to a halt," said Jane Henney, commissioner of the Food and Drug Administration.

She said the death last September of 18-year-old Jesse Gelsinger, a test subject in a gene therapy trial, caused the government and research institutions to take a close look at the highly profitable arrangements that some researchers have with companies that are financing their research.

"Questions were raised about financial interests of investigators and whether these interests clouded their judgment in decisions that were made," said Henney.

Gelsinger died in an experiment at the University of Pennsylvania's gene therapy program led by Dr. James Wilson. It later was learned that a company founded by Wilson funded part of the Penn research. The company, Genovo Inc., later was sold to a bigger company and Wilson got a reported $13.5 million in stock.

Penn's gene therapy research has been suspended by the FDA, in part because the agency found that researchers did not report two other serious reactions from it.

Wilson has not commented publicly on violations cited by FDA, but will have an opportunity to challenge the agency's findings. He did say of Gelsinger: "We may have activated his immune system very early."

Tests showed that there was a swift rise of immune system proteins, just as there were with other patients who had been involved in the therapy. But for Gelsinger, the proteins never receded.

Dr. Thomas Bodenheimer, a clinical professor at the University of California, San Francisco, said financial conflict of interest among researchers risks scientific misconduct that can "impact the practice of tens of thousands of physicians" who are guided by the research in treating millions of patients.

"Investigators who allow bias or error to infect their work are practicing scientific misconduct," he said.

Bodenheimer said such bias includes reporting only good results in a drug trial, making claims not justified by the evidence and "outright fraud with fabrication of evidence."

Most drug trials, he said, are paid for by drug companies. Other trials are conducted by researches with a financial interest in the companies.

For instance, said Bodenheimer, an analysis of 70 studies on the safety of a heart drug found that 96 percent of authors who had drug company ties found the drugs to be safe, while among experts with no drug company connections only 37 percent said the drugs were safe.

When drug companies pay for research into the effectiveness of their own pain relievers, said Bodenheimer, 100 percent find their drug superior to a competitor's.

An analysis of company-funded studies published in peer-reviewed journals found that results of the studies favored the company's products 98 percent of the time.

"The take-home message of these studies is that company-funded trials have a high likelihood of favoring the company's products," said Bodenheimer.

Among other examples of conflict of interest:

Private physicians may be paid $1,000 to $5,000 for each patient recruited for a drug trial. This leads to doctors sometimes enrolling patients who don't belong in the drug trials. Some doctors have even enrolled patients who don't even have the disease being studied.

Up to 75 percent of people enrolled in cancer studies are younger than 65 because older people respond more poorly to chemotherapy. Yet, 63 percent of cancer patients in the United States are over 65. Excluding older people can make it appear that a drug works better than it really does.

Drug companies own the data from drug trials. Bodenheimer said in some cases companies release only results that make their drug look good "and bury data on less favorable" results.

In at least two cases, said Bodenheimer, drug companies threatened researchers who attempted to publish company-paid studies that put their drug in a poor light.

Many research studies are written by "ghost" authors, usually hired by a drug company. The company then pays a prominent researcher to be named as the author "though they contributed nothing except their prestige," said Bodenheimer.

And, yet, said Bodenheimer, thousands of doctors rely upon the results of these studies to make treatment decisions for their patients.

He said the data show "that scientific misconduct does take place in clinical drug trials, that conflict of interest is a risk factor for scientific misconduct and that something must be done about it."