Document kDLMvDvQ4oN2gOvGQ8Y2rQ5q0

Vista Chemical Company 900 Threadneedle Houston, Texas 77079 (713) 588-3000 P.O. Box 1 9029 Houston, Texas 77224 Fax {713) 588-3236 T&k /U ef: MW: AJ0: RF April 24, 1990 Mr. Bob Romano Chemical Manufacturers Association 2501 M Street, N.W. Washington, DC 20037 Mrs. Sara Schotland Cleary, Gottlieb, Steen & Hamilton 1752 North, N. W., Suite 200 Washington, DC 20036 = >Ai>^ VISTA Mr . Ron V an Mynen Mr. Norm Gaines Union Carbide Corporation 39 Old Ridgebury Rd. Danbury, CT 06817 Gentlemen and Mrs. Schotland: This attached article describes the conclusions of a study that further supports our recent arguments on the EO ATSDR Profile. Bad science has a way of hanging around and continuing to do damage. We should try to get this report. Assuming it's good science, we may want to refer to it in the future. Sincerely, Thomas G. Grumbles, C.I.H. Environmental Quality Manager dlj Attachment VVV 0000X2926 Report concludes bad research hard to erase once published REUTER NEWS SERVfCE CHICAGO -- Erroneous scien tific research, once published, is hard to kill and keeps being cited in subsequent studies even after its authors have retracted it, research ers said. "We found that articles citing in valid, retracted work are abundant and ubiquitous in the scientific lit erature," said the study from the University of Kentucky. The authors said they located 82 retracted articles published from 1973 to 1983 which were later cited 733 times as support for sub sequent research. As an example the authors listed 14 articles published by John Dar- see, who worked at Harvard and Emory universities and was ac cused in 1981 of inventing data ob tained during a heart drug experi ment in dogs. Despite widespread publicity about the case, the study said, "Darsee's 14 completely retracted articles were subsequently citied 123 times alter their retraction." The study said that ho reliable source exists to locate fradulent or erroneous work, and.that most sci entific journals have no index for their own retractions., The researchers said the prob lem involves both large, well- known journals and more obscure publications. ,, . The report was published in the Journal of the American Medical Association. V li . . - In a related article in the same issue, Paul Friedman of the Univer sity of California in San Diego said most scientific journals had no written policies for responding to allegations of research fraud. He said efforts by his university to retract dozens of articles written by a faculty member accused of re search fraud had met with mixed success at best. Wv 000012927