Lead Researcher Probed on Charges of Skewing Data
A prominent researcher whose work has been instrumental in persuading health officials to adopt a stricter standard for acceptable levels of childhood exposure to lead is being investigated on charges that he manipulated data in his seminal study on lead poisoning.
Since February, a five-member faculty team at the University of Pittsburgh has been investigating charges that Herbert L. Needleman, a professor of psychiatry and pediatrics at the school, skewed the results of a 1979 study that found that children are adversely affected even by very low levels of lead.
In response to the investigation, Dr. Needleman last week filed a lawsuit against the university and the National Institutes of Health, which requested the review. The suit alleges that the inquiry violates the researcher's constitutional right to due process and is flawed on other legal and procedural grounds as well.
Meanwhile, other researchers, notably one who examined earlier charges against Dr. Needleman for the U.S. Environmental Protection Agency, have come to his defense.
At issue is the reputation of an academic who authored what is considered to be a pioneering effort in the field. The study, published in the New England Journal of Medicine, found that children who were exposed to lead but had no symptoms of lead poisoning were nonetheless affected by the metal. The report found an inverse relationship between a child's exposure to lead and his I.Q. score.
The results reported by Dr. Needleman, who has become a leading advocate for efforts to reduce childhood lead exposure, and other scientists in subsequent studies, have led several federal agencies to lower acceptable lead levels.
Last year, new standards that were drafted by Dr. Needleman and other experts for the U.S. Centers for Disease Control recommended that all children be tested for the presence of lead in their blood by their first birthday. (See Education Week, Oct. 16, 1991.)
Under the new guidelines, children are considered to be at risk of becoming lead poisoned if they have lead readings that exceed 10 micrograms of lead per liter of blood, a level 60 percent lower than the old standard of 25 micrograms, which was a trigger for medical treatment.
Federal officials estimate that as many as 4 million preschool-age children may have blood-lead levels that exceed this new standard.
'Things Are Funny'
Although most health experts have praised the new standard, some have questioned the efficacy of spending so much money on testing all children for lead. Others believe that there is not enough scientific evidence to support the conclusion that children are adversely affected by very low levels of the metal.
One vocal critic of the new lead policy, Claire B. Ernhart, has also long questioned the veracity of Dr. Needleman's 1979 study. Ms. Ernhart, a professor of psychiatry and reproductive biology at Case Western Reserve University, said that she and another researcher familiar with the 1979 study's data base last year sent a letter outlining their concerns to the N.I.H., which has financed some of Dr. Needleman's work.
According to Ms. Ernhart, who ackowledged that she received about $60,000 annually in research money for about six years from an industry-linked group called the International Lead Zinc Research Association, Dr. Needleman's study is flawed because he excluded from his final results data collected from about 100 children. He also did not adequately account for factors besides lead that might affect a child's intelligence level, she said.
"You look at the study and things are funny,'' she said in an interview last week. She said that in response to her letter, the N.I.H. asked Pittsburgh to review Dr. Needleman's work.
Spokesmen for both the N.I.H. and the university said last week that such a review was being done at the N.I.H.'s request. But the N.I.H. spokesman could not confirm that it was Ms. Ernhart's letter that prompted the investigation.
James Rosenberg, the university's research-integrity officer, said that an internal three-member panel decided in December that the charges merited further review.
He said the new panel, which includes faculty members from various departments, will report on "the technical aspects'' of Dr. Needleman's study in May.
He said that in response to Dr. Needleman's lawsuit, the school will break with policy and hold open hearings during the investigation.
If Dr. Needleman is found guilty of scientific misconduct, he could face a range of penalties, from receiving a letter of reprimand to being required to publish a statement retracting his original findings, Mr. Rosenberg said.
A 'Continuing Vendetta?'
For his part, Dr. Needleman last week denied Ms. Ernhart's charges. "It's absolutely untrue,'' he said.
On the other hand, Dr. Needleman, who said he is being attacked by the lead industry, said he would "welcome a hard look. Maybe we can get this behind us.''
Dr. Needleman and his supporters noted that in the mid-1980's, a scientific panel under the aegis of the E.P.A. reviewed, and then dismissed, similiar charges.
"It is indeed a shame and a waste of time, talent, and taxpayer expense that Dr. Needleman is the subject of a continuing vendetta by persons and parties who seem consumed with a need to punish an early messenger of unpleasant news about the role of lead on the health of children,'' wrote Morton Lippmann, the head of the E.P.A. panel, in a letter to the N.I.H. in February.
Joel Schwartz, a senior E.P.A. scientist who received a MacArthur Foundation "genius'' award for his lead-related research, re-analyzed Dr. Needleman's data. The analysis showed that with the 100 excluded children and other variables included, the researcher's conclusions remain valid, he said.
Even if the Needleman study is found to be in error, he said, there
would be little effect on federal policy, because "there are literally
dozens of studies out there that show the adverse effects of lead'' at
even lower levels than outlined in the standards.
Vol. 11, Issue 29, Page 9Published in Print: April 8, 1992, as Lead Researcher Probed on Charges of Skewing Data