THE NEED FOR RAW DATA REPROCESSING
Laboratory fraud may not be as rare as you would like to believe. Bart Simmons (Preventing and Detecting Fraud, Hydrovision Online, Spring 1997) discusses the Eureka Laboratories, Inc. case which occurred in 1995. He states, "Eureka Laboratories, Inc. was fined $1.8 million and two chemists who worked at Eureka were convicted of federal fraud charges related to the manipulation of lab results for federal contracts. The convictions came after the federal government suspended Eureka from contracting because of allegations of fraudulent practices. The Eureka case, although one of the most prominent, is only one of many cases of laboratory fraud which have been uncovered nationwide."
Donald Budd, former owner of Texas Environmental Services, Beaumont, Texas, a laboratory that tested water for cities, refineries and other companies was sentenced to six years in federal prison after misrepresenting the lab's methods. (Reported in Lubbock Online News).
On June 3, in Portland, Ore., Robert Cyphers , former president and owner of UST Environmental Services, pleaded guilty to four felony counts involving the submission of written false statements to the government.
A criminal investigation is under way by the U.S. Department of Justice into the doctoring of test results from the U.S. EPA's Region V laboratories, a federal official told the Chicago Sun-Times on Feb. 14, 2000.
On January 7, 1998, Intertek Testing Services Environmental Services laboratory (ITS) made a voluntary disclosure that fraudulent practices were being conducted in their laboratory located in Richardson, Texas. ITS disclosed that employees had been improperly manipulating QC data during the time period later determined to be 1991-1997. The manipulations included peak shaving, peak juicing, and time traveling in order to make data appear to meet QC requirements. These practices affected hundreds of projects, thousands of clients, and hundreds of thousands of samples.
The above are just a few examples of cases of laboratory fraud. One of the most effective means of uncovering the most common types of laboratory fraud is raw data re-processing. Raw, unmanipulated data taken directly from the GC or GC/MS data system is re-processed by a third party. Re-processing consists of integrating all chromatographic peaks, assigning all target analyte identifications, and quantitating all results. The re-processed results are then compared to the original results to determine if there are any significant differences. If significant differences are found, the original results are examined to see if consistent (and appropriate) procedures were used or if improper procedures were used and if they were used fraudulently to make QC checks appear to pass. The resultant effect on actual sample results is then determined.