The Criminalization of Environmental Quality Assurance - United States v. Jeffus, et al.

Edwin J. Tomko and Peter K. Wahl
Akin, Gump, Strauss, Hauer & Feld, L.L.P.
etomko@akingump.com pwahl@akingump.com

Thanksgiving 2001 was a poignant holiday for eight former employees of Intertek Testing Services Environmental Laboratories. As the holiday approached, the employees were acquitted of federal charges of mail fraud, wire fraud, defrauding the federal government, and conspiracy to commit these offenses. The eight defendants were charged with a total of 77 felony counts and all faced the very real possibility of spending several years in a federal prison. The acquittals came after nearly two months in trial. These defendants would not have made for exciting viewing on any of this country's television programs portraying criminal trials. One defendant was accompanied to the courthouse regularly by her husband and three young boys. Another defendant missed the first part of the trial because she had just given birth to her first child. Our client, a Chinese immigrant, stated after the trial was over that the experience had posed more trauma to him than his experience in the Chinese Cultural Revolution in which he was ordered to serve at a State-run work farm in intolerable conditions for several years.

How these individuals found themselves in the maelstrom of a federal felony trial requires a short and oversimplified primer on laboratory analysis technique. The laboratory analysis of environmental samples is most commonly done by gas chromatography. A gas chromatograph processes environmental samples and produces a chart known as a chromatogram. In contaminated samples, the chromatogram portrays an irregular line with peaks of differing sizes. Laboratory analysts measure the peaks to identify chemicals in the sample and their concentration. The measurement process is called "integration" and is either done by a computer program or manually.

Gas chromatography methods commonly contain quality assurance requirements in a purported attempt to establish the reliability of analysis. These quality assurance steps most commonly focus upon the analysis of specially prepared "spike samples" of known composition and concentration. The measurement of values in the spike sample is compared against the known values for that sample to provide quality assurance of future analysis. If the quality assurance measurement "passes," the analyst may proceed with analysis of actual environmental samples. The prosecution of the eight defendants in Jeffus revolved around allegations that the defendants were using "improper" manual integration in their quality assurance work in order to "pass" specified criteria and that they then proceeded with analysis of actual environmental samples.

How the government's case was put together is also instructive into how the case proceeded. A routine internal audit of analytical data at Intertek in 1998 revealed chromatographic peaks that did not look like "proper" manual integrations in the opinion of auditors. This data was reported along the internal chain-of-command, clients were notified, and further investigation was conducted by Intertek, clients, counsel for the company, and government investigators. Intertek and government investigators worked so closely together that Intertek offered untrained government investigators the assistance of its own employees in operating laboratory computer systems, understanding the chromatographic data contained therein, and also reviewing a voluminous amount of hardcopy data. Intertek submitted data and other information regarding the manual integrations at issue to the federal government as part of applications to EPA and Department of Defense voluntary disclosure programs. Intertek's applications for voluntary disclosure were ultimately denied and 13 employees of the company were indicted in the Northern District of Texas shortly thereafter. The helpfulness of the Intertek employees appears to have been rewarded by government investigators and prosecutors by the use of voluntarily produced information as incriminating evidence at trial. Ironically, government investigators would have been incapable of retrieving a significant portion of this evidence without the assistance of the defendants against whom it was offered.

The finer details of Jeffus may be most interesting to someone who has previously worked in an environmental laboratory. Nevertheless, a number of very valuable lessons were learned from the defense perspective in Jeffus.

• The government in a criminal case premised upon violations of quality assurance may overestimate the importance that a jury will place upon quality assurance. In the instant case, the defense focused the jury on proof that the manual integrations at issue in the quality assurance procedures had marginal, if any, effect upon the accuracy of actual sample results and no impacts upon human health or the environment.
• In an environmental criminal case, all means taken to prevent or minimize references to environmental harm will be worthwhile. The less personally-affected a jury feels during an environmental criminal trial, the more likely the jury will believe that the case would have been more suitably brought in a civil forum, or not brought at all. In the instant case, a Motion to Exclude Government References of Environmental Harm was filed by defense counsel in limine and also argued at trial. This motion was denied. However, it is believed to have been effective in informing the judge of the outer limits of relevance and the unfair prejudice that would be caused by inflaming the jury in a case involving environmental issues. The motion also conveyed to the prosecution that all statements and witness questions related to environmental harm would be closely monitored by defense counsel.
• A prosecution needs to be based on clearly articulated standards and an obvious breach of those standards. The basis of the government's fraud allegations in the instant case were that the defendants falsely represented to clients that the laboratory analysis performed by them complied with several different EPA methods. The government attempted to demonstrate that the quality assurance portions of test methods were circumvented by the defendants in repeatedly portraying to the jury what had been determined to be "improper" manual integrations. However, the government could not point to any relevant published source that defined the difference between a "proper" versus an "improper" manual integration. Additionally, the Indictment explained that fraud was committed by the defendants' false representation that they had complied with EPA-approved methods of quality assurance. However, the government's principal expert deviated from the Indictment's theory in opining that an "improper" manual integration standing alone was criminal conduct even absent proof of intent to deceive. Further, several of the EPA-approved methods focused upon by the government were shown at trial to not even contain the quality assurance requirements central to the government's case in its broad allegations of fraud.
• In a multi-defendant case in which conspiracy is alleged, the government may not be sufficiently concerned with associating individual criminal acts to individual criminal defendants. Laboratory analysts at Intertek trusted each other. As such, these analysts conducted laboratory analysis using previously logged-in computer terminals, used each others passwords when theirs would not work, and did not log-off computers when completed with analysis. This situation, which the government attempted to downplay, created facts very amenable to the defense, such as defendants being logged-in to a computer and being indicted for allegedly fraudulent manual integrations associated with this log-in at the same time they were on vacation in the Caribbean, out of work sick, and on maternity leave.
• The government may improperly consider the acceptance of plea agreements to be an outright success in a multi-defendant case. In Jeffus, 13 individuals were indicted on felony counts. Five defendants ultimately plead guilty, four to misdemeanors, and pursuant to the plea agreements, no pleading defendant was sentenced to jail. After trial, the government attorneys and investigators cited the plea agreements as a demonstration of a successful case.
• The investigation of the instant case was primarily conducted by the EPA's Criminal Investigation Division, Quality Assurance staff, and Office of Inspector General. However, no environmental crimes were alleged; all charges were brought under Title 18 of the United States Code. What occurred in Jeffus, and is continuing in ongoing investigations involving the Environmental Crimes Section, is a blurring of the lines between economic fraud cases and environmental impact cases. The prosecution of the fraud statutes by the Environmental Crimes Section, without formal input or review by the Criminal Division, may be an area that is worthy of study at the Justice Department.