SCIENTIFIC FRAUD

As mounting scandals shake the public trust, researchers struggle to reconstruct
science's shattered reputation.

On an early spring afternoon in March 1987, two scientists hunched over a small
desk in a hotel room in Frankfurt, Germany-the pair pressured by their
governments to end more than two years of bitter battles over who first isolated
the AIDS virus. Robert Gallo, one of the National Institutes of Health's most
powerful and protected superstars, and Luc Montagnier, a virologist with Paris'
Pasteur Institute, hammered out a "definitive scientific history" of their
stunning achievements.

The account that emerged from the hotel-room meeting spawned an agreement
between the French and American governments. Montagnier and Gallo would share
the credit for the discovery and their governments would jointly patent the
rights to the test for detecting the AIDS virus. Officially, the occasion marked
the end of a controversy that erupted in 1985 when evidence surfaced that
Gallo's AIDS-causing HIV virus was virtually the genetic twin of a virus sent to
him in 1983 by Montagnier. The matter rested, however, until 1989 when the
Chicago Tribune published a 50,000-word expose on Robert Gallo that finally
prodded federal investigators to probe more deeply into the events in his lab in
1983.

Now one of the more ignominious chapters in the annals of science appears to be
grinding to a conclusion. A blue-ribbon scientific panel, set up to monitor an
internal NIH investigation of the affair, accused Gallo of "intellectual
recklessness of a high degree." The panel members, nominated by the National
Academy of Sciences, criticized Gallo for his failure to acknowledge having
grown and studied an AIDS virus sent to him by the French. Gallo is also the
target of a federal inquiry investigating charges of perjury and patent fraud
related to his patent application for the AIDS test.

As Omni went to press, the NIH was reportedly ready to clear Gallo of charges of
misconduct. The unpublished report addresses-and ultimately rejects-most of the
charges against Gallo. The resolution, however, is tainted by years of acrimony
and the wrenching downfall of Gallo, one of the giants of science and the only
scientist ever to win two Albert Lasker awards, perhaps our nation's most
prestigious biomedical honor. The battle for scientific supremacy, however,
destroyed more than egos. While Gallo and Montagnier were wasting precious time
taking potshots at each other across the Atlantic-instead of using their
considerable talents to find a cure-this plague was claiming thousands Of lives.
And on this score, the French are hardly saints. In 1985, the health officials
in France stalled licensing the American test for the AIDS virus to give French
scientists a chance to devise their own. In the intervening five months,
virtually half the hemophiliacs in France became infected with HIV through the
untested blood supply. The unconscionable act eventually erupted into a
national scandal that threatened to topple the Mitterand government.

In the United States, the Robert Gallo debacle is just the latest in a rash of
recent scandals that have rocked the scientific community and tarnished its once
pristine image. Incidents of simply sloppy science, misconduct, plagiarism,
manipulation or faking of data, and outright criminal behavior have made
front-page news with alarming regularity. Even Nobel Laureate David Baltimore's
career suffered when he stubbornly refused to admit the possibility that a
colleague committed fraud in a scientific paper he co-authored. But the Gallo
affair underscores, perhaps more than others, that when scientists go astray, it
represents no mere breach of protocol. People's lives are put at stake.

Indeed, these disturbing revelations of wrongdoing by reputable researchers have
eroded public confidence, prompted Congressional oversight committees to launch
costly investigations, and forced universities and federal agencies to develop
more rigorous policing mechanisms to flush out the charlatans. Some worry about
the fate of the next generation of scientists. And oth-- fate of the next
generation of scientists. And others question what this means for the future of
science, a collaborative enterprise that must be conducted in an atmosphere of
trust to thrive. "Honor is integral to the scientific method," says Bernadine
Healy, director of the National Institutes of Health. "Without it, science
would crumble."

Whether misconduct is more prevalent, however, cannot be documented. The
agencies that dispense many of the federal research grants-the National
Institutes of Health (NIH), which has a budget of $9 billion and supports about
100,000 scientists, and the National Science Foundation (NSF), which runs on
$2.6 billion a year-do not keep statistics on fraud. But out of about 18,000
grants the NSF made in 1991, only 52 cases of misconduct were reported to the
agency, according to Donald Buzzelli, a staff associate for oversight in the
NSF's Office of Inspector General.

But scientists themselves report the actual incidence of miscounduct is much
higher. In a November 1991 survey of 1,500 scientists conducted by the American
Association for the Advancement of Science (AAAS), more than a quarter of
respondents said they had witnessed faking, falsifying, or outright theft of
research in the past decade. "There's no hard evidence, but my gut feeling is
the problem has gotten much worse in the past five years," says Walter Stewart,
an NIH staff scientist who's participated in several fraud investigations.
"There's been a collapse of the professional consensus that you have to behave
correctly. But the consequences of failing to face up to ethical necessities
are devastating and profound. Science is about discovering the truth-if
dishonesty is condoned in one sphere, it spreads like a disease and infects the
whole process."

Of course, charges of cheating, corner cutting, and deception are not new. Many
great scientists-Galileo, Mendel, Newton, and Dalton, among them-fudged or
concocted data to make their theories more compelling or to demolish the
arguments of their rivals. But on the whole, science used to resemble the
priesthood; individuals were called to the profession, which operated on the
honor system. Scientists were, says one Congressional aide, "the white-coated
guardians of truth searching after the grail of knowledge."

Researchers chafed at outside intrusions and insisted their self-correcting
system of internal checks was enough to catch any miscreants. Peer review
panels scrutinize grant applications before any money is awarded. Editors of
scientific journals send articles submitted for publication to "referees," other
experts in the field, to determine the significance and validity of the
findings. And once articles are published, researchers attempt to replicate the
results to double-check their accuracy.

In isolated instances, such as the cold fusion controversy, the system still
appears to work. In March of 1989, the University of Utah announced that two
chemists, Martin Fleischmann and Stanley Pons, had harnessed the source of the
sun's energy and had produced cold fusion in a test tube of heavy water.
Excited chemists and physicists formed teams to confirm these astonishing
findings. As the weeks passed, though, the scientists could find no evidence of
fusion and concluded the Utah experiment was flawed.

"My guess is when they made their claim, they genuinely believed it. Therefore
they were guilty of sloppy science-not fraud," says Robert Park, a professor of
physics at the University of Maryland and director of the Washington, DC, office
of the American Physical Society. But the chief reason scientists tried to
replicate cold fusion is because these experiments could be done on a tabletop
with their lunch money.

In today's financially constrained climate, though, researchers rarely repeat
each other's work: "Nobody gets funding to do replications, so science is not
the self-cleansing apparatus it once was," says Jules Hallum, director of the
NIH's Off ice of Scientific Integrity (OSI). Flawed or even flagrantly spurious
science has slipped through these supposedly fail-safe mechanisms.

Deception, by its nature, must be clandestine. So the job of keeping science
clean has fallen to whistle blowers. Yet, far from welcoming or supporting
whistle blowers, some research institutions treat them shabbily. Sometimes they
even drive them out of the profession-raising serious doubts as to whether
science can be trusted to govern itself. Something has gone awry. But what?

"This is all about money," argues Robert Bell, an economics professor at
Brooklyn College and author of Impure Science (John Wiley), a troubling-and
exhaustively researched-indictment of shoddy scientific practices. "Like the
Pentagon's defense contractors, the science community has evolved into another
patronage system which enriches those at the top. Universities have a vested
interest in not finding anyone guilty of fraud. Because if they do, they may
have to return the delinquent researcher's grants. When someone blows the
whistle, universities set up investigatory panels, which are almost inevitably
kangaroo courts that cover up abuses."

Indeed, science is big business, and all the awesome advances of the past two
decades cost plenty. The electron microscopes, ultracentrifuges, and gene
sequencers that have emerged in the past decade as standard laboratory equipment
cost well into five figures. Sophisticated imaging devices, like PET scanners,
cost upwards of $1 million. Laboratories like Baltimore's, where as many as 60
people-graduate students, postdoctoral fellows, technicians, and junior
researchers-struggle for space, equipment, attention, and glory, require
hundreds of thousands of dollars a year to run.

Unfortunately, this dramatic escalation in costs for equipment and research
comes at a time when funding sources are dwindling. Nobel laureate physicist
Leon Lederman, the former president of the AAAS, conducted a recent survey of
researchers at 50 universities. He calculated that 1990 federal funding for
scientific research at these institutions was only 20 percent higher than in
1968, while the number of scientists has doubled. Small wonder that the
jockeying for grants often looks more like Roller Derby than rivalry among
highly educated professionals.

Almost by default, top scientists have been cast in the role of empire builders.
They jet to scientific meetings around the world to network with equally
eminent colleagues or woo big university donors, while their underlings grind
out massive volumes of research and grant proposals to keep the laboratory
engines lubricated with money. Rarely do science's superstars conduct the
hands-on research that made them famous or mentor the new crop of young
scientists or even closely monitor the results of every experiment conducted in
their laboratories.

Added to this is the fierce competition to make a key discovery. And progress
up the ranks depends more on the number of papers a scientist has published than
on competence. "The game of acquiring a long list of publications is a
relatively new development as evidenced by the fact that just two decades ago,
the current problems with paper inflation were unthinkable," note William Broad
and Nicholas Wade in Betrayers of the Truth (Simon and Schuster). "In 1958,
James Watson"-who won the 1962 Nobel Prize for deciphering the structure of DNA
in 1953--had on his curriculum vitae eighteen papers ... Today, the bibliography
of a candidate facing a similar climb often lists fifty or even one hundred
papers."

Yet nature refuses to behave the way researchers would like. Often experiments
don't turn out the expected results, so scientists are forced to try a fresh
approach-and pray they get more money to keep going.

But the pressures to hit a home run each time tempt scientists to cut corners,
to round off numbers so that results appear more impressive,' to overlook
anomalous findings that would put the data in a less favorable light-or to just
cheat. While "most scientists are scrupulously honest and deplore scurrilous
behavior," says the NIH's Healy, they are nevertheless reluctant to inform on
others. "They don't rally to alter corrupt situations because they see whistle
blowers get burned," says Bell. "They're seen as spoilers who've upset the
patronage gravy train."

Like Robert Sprague. Just before Christmas in 1983, the psychology professor
and director of the Institute for Research on Human Development at the
University of Illinois, made the most agonizing decision of his life. He sent a
lengthy letter to the National Institute of Mental Health (NIMH), accusing
Stephen Breuning, a former protege and close friend, of falsifying data.

Breuning, while a psychologist at the University of Pittsburgh, had carried out
drug studies on severely retarded children who often mutilated themselves by
banging their heads against the wall. These children were calmed with powerful
tranquilizers known as neuroleptics. But neuroleptics had an unfortunate side
effect: They seemed to cause tardive dyskinesia-involuntary muscle spasms such
as flapping arms, or wagging tongues.

Breuning questioned whether these drugs were beneficial. Experiments he
conducted between 1980 and 1983 indicated they did more harm than good and that
controversial stimulants, including Ritalin, far more effectively controlled the
wild, self-destructive behavior with fewer adverse side effects. Colleagues
praised his work, which helped shape public health policy.

But Sprague felt there was something fishy about Breuning's work. There were
glaring inconsistencies and his results were just a little too perfect. In the
real world, nobody produces 100-percent correlations. Poring over Breuning's
data, Sprague realized that his friend had cheated. He was heartsick-and
outraged.

After he sent his detailed dispatch outlining three separate incidents of fraud
to the NIMH, Sprague expected swift action. In fact, the NIMH ordered the
University of Pittsburgh to convene a panel to look into one of Sprague's
allegations. Breuning consequently confessed to filing research reports
containing false statements when he was confronted with the charges.

But Sprague didn't expect that his life would turn into a Franz Kafka novel:
that the NIMH would make him the target of a probe or sharply cut his grants.
NIMH officials stoutly deny these incidents had any relation to the Breuning
case even though the institute had fully funded Sprague the preceding 17 years.
Nor did the scientist imagine this gut-wrenching affair would drag on for more
than three years before a glacially slow NIMH panel rendered its final verdict:
"Virtually all of Breuning's work was fabricated and that Sprague's work and
accusations were beyond reproach." The University of Pittsburgh was forced to
return $163,000 in federal grants and Breuning was subsequently convicted on
criminal charges for falsifying medical research.

What happened to Margot O'Toole is even more egregious. In 1986, she worked as
a postdoctoral fellow in the MIT laboratory of Thereza Imanishi-Kari. That year,
Imanishi-Kari co-authored- with MIT's Whitehead Institute's director, David
Baltimore, a scientific paper published in the journal Cell. The article
purported to show that antibodies expressed by one mouse could be made to mimic
those of another mouse. O'Toole accidentally came across Imanishi-Kari's
laboratory notes and noticed serious discrepancies. "It was obvious from these
records that the experiment had not yielded the published results," she says.
O'Toole took her concerns to Imanishi-Kari and later reported her suspicions to
Baltimore and MIT officials. Baltimore dismissed her claims, calling O'Toole a
"discontented" postdoctoral fellow. Suddenly, O'Toole became a pariah: She lost
her position and could not find another one. Finally, she took a job answering
phones for her brother's moving company.

O'Toole's ordeal spotlights how science's allegedly self-correcting mechanisms
broke down at every checkpoint. It never occurred to the referees assigned to
review the paper that there could be anything even slightly shady in an article
co-authored by David Baltimore, one of the world's most distinguished
scientists. Fellow scientists likely wasted countless hours and money trying to
build on the bogus findings. Investigators at Tufts University, MIT, and,
initially, at the NIH performed little more than a perfunctory probe of
O'Toole's charges. As the New York Times wrote in a scathing editorial, "the
initial investigations of Dr. O'Toole's complaints smacked of an oldboy network
drawing up the wagons to protect scientific reputations."

The truth came to light only when the case fell under the noses of NIH
scientists Walter Stewart and Ned Feder and the House Subcommittee on Oversight
and Investigations, headed by Representative John Dingell. But Baltimore can
only blame himself for the scale of the scandal. His defiant insistence that
his critics didn't know what they were talking about escalated this relatively
minor transgression into what is dubbed a "scientific Watergate." He even
orchestrated a letter-writing campaign among his colleagues to stop Dingell's
inquiry. And Baltimore retracted the article and halfheartedly apologized to
O'Toole only after the Secret Service forensically analyzed Imanishi-Kari's
laboratory notes, proving her data entries were falsified at a later date.

After five years of vilification before she was finally vindicated, would
O'Toole do it again? "Absolutely. Because at the root of this were the most
fundamental principles of the entire profession," says O'Toole, who now works
for a biotech company in Cambridge.

But these jolting scandals may ultimately prove salutary. A number of hopeful
signs demonstrate the obviously shaken scientific community-acting out of
enlightened self-interest-intends to clean house and stop the university
cover-ups, the federal foot dragging, and the witch hunts of whistle blowers,
all of which allow abuses to flourish.

The OSI's staff may grow from 19 investigators to 28, including three lawyers,
and, to ensure objectivity, the office may be moved away from the NIH. The NSF
has issued stricter guidelines regarding misconduct and is keeping a tighter
rein on universities. And a 22-member panel of the National Academy of Sciences
(NAS) in Washington, DC, has worked for more than two years to hammer out the
principles of good scientific conduct.

The NIH also now requires ethics programs in graduate students' curriculum. "We
operated under the false belief that young scientists would learn ethics in the
laboratory by osmosis or by being exposed to good role models," says Stephanie
J. Bird, a special assistant to the MIT's associate provost, recruited to
develop ethics programs there. "But now there's a recognition that ethics need
to be explicitly articulated." Other schools, including Harvard and Dartmouth,
now have similar ethics programs.

In 1989, the American Association for the Advancement of Science gave its
Scientific Freedom and Responsibility Award to Robert L. Sprague. In accepting
one of science's highest honors, Sprague observed that it wasn't surprising some
scientists cheat. After all, they're only human. "What is surprising," charged
the soft-spoken, bespectacled psychologist, who hardly resembles a
rabble-rousing renegade, "is that the system of science actually works against a
speedy, appropriate adjudication of suspected misconduct. This is intolerable
in a civilized society."

If the science community can eliminate the harmful politicking that has
corrupted the research process and restore the high ethical standards-the
boldness of vision and the sense of mission that historically has made American
science great-it will be because of people like Robert Sprague. "Science better
learn to police itself," warns the NIH's Bernadine Healy. "The future of
research is intimately linked with the seriousness to which we address the issue
of scientific conduct. Because if we lose the trust on which research is built,
we lose everything."



GRAND ILLUSIONS

THE TOP TEN KNOWN OR SUSPECTED SCIENCE FRAUDS Even history's greats misconducted
science.

*
1. Galileo Gall-liar? Galileo Galilei (1564-1642) is considered the founder of
the modern scientific method. But he wrote about experiments that were so
difficult to reproduce that many doubt he actually conducted them.

*
2. Star-crossed Science. Compelling evidence indicates Johannes Kepler
(1571-1630), the father of modern astronomy, doctored his calculations to
bolster his theory that the planets move in elliptical orbits, not in circles,
around the sun.

*
3. A Matter of Some Gravity. Isaac Newton (1642-1727) crunched numbers to make
the predictive power of his universal gravitational theory carry more weight.
Scientists have since noted he "adjusted" his calculations on the velocity of
sound and on the processions of the equinoxes so they would support his theory.

*
4. Bad Chemistry. John Dalton, the great nineteenth-century chemist, reported
numerous findings from experiments conducted in 1804-1805 that no chemist since
has been able to reproduce. Scientists now believe he fudged his data.

*
5. Mendelian Misconduct. Abbe Gregor Mendel's (1865) experimental results were
so perfect that later researchers were convinced he falsified his data, which
formed the basis of modern genetics.

*

6. Fishy Physics. Robert A. Millikan won the Nobel prize in 1923 for measuring
the electrical charge of an electron. But scientists later discovered he failed
to report the unfavorable results of related research conducted between 1910 and
1913.

*
7. Skullduggery. The Piltdown Man is generally considered the greatest
scientific hoax of all time. In 1908, a part of a skull hailed as proof of the
missing link between apes and humans was unearthed in an English gravel pit on
Piltdown Common, in Sussex. In the 1950s, however, researchers using modern
dating techniques revealed the skull was actually an ape jaw with part of a
human skull attached that had been stained to appear old.

*
8. Spurious Superiority. Sir Cyril Burt, a pioneering British psychologist,
deliberately made up more than three decades of data, from the mid 1940s until
1966, to back up his bogus theory on the relationship between heredity and
intelligence. He claimed human intelligence was 75 percent inherited, thereby
reinforcing the British class system.

*
9. Of Mice and Mendacity. William T. Summerlin, a researcher at the
Sloan-Kettering Institute for Cancer Research, colored white skin grafts black
with a felt-tip pen to fake the results of skin-transplant experiments in mice
in the mid Seventies. He was trying to prove that human skin, if maintained in
an organ culture for several weeks, becomes universally transplantable without
risk of rejection.

*
10. Heartbreaker. John Darsee, a heart researcher at Emory University in
Atlanta and at Harvard in the early 1980s, falsified data that formed the basis
for about 100 scientific publications on heart disease. The Darsee case was
especially troubling because 47 other researchers co-authored his papers and
never caught on to the fraud.