United States District Court, M.D. Georgia, Albany Division
W. LOUIS SANDS, District Judge.
Defendants Stewart Parnell, Michael Parnell, and Mary Wilkerson move to exclude the expert testimony of Dr. Ian Williams. The Court held a Daubert hearing on September 2, 2014. For the reasons that follow, Defendants' motion to exclude Dr. Ian Williams' expert testimony is DENIED.
I. Procedural background
Defendants Stewart Parnell, Michael Parnell, and Mary Wilkerson are charged in a seventy-six-count, fifty-two-page indictment arising from the sale of salmonella-contaminated peanuts at the Peanut Corporation of America ("PCA"). The case is before the Court on Defendants' motion to exclude the expert testimony of Dr. Ian Williams. In its April 23, 2013 scheduling order, the Court ordered the Government to provide all discovery and expert witness notices by June 17, 2013. On June 11, 2014, however, the Government disclosed Dr. Williams as an expert witness. Following a hearing on various motions, the Court found that exclusion of Dr. Williams as an expert was not warranted on the basis of the late disclosure. Subsequently, Defendants requested a Daubert hearing on eight of the Government's expert witnesses. Because that request was untimely, the Court denied the request for all witnesses save Dr. Williams. The request was granted as to Dr. Williams because his disclosure was late. The Government was ordered to give two days' notice before calling Dr. Williams to testify. The Government gave that notice on Friday, August 29, 2014, and the Court held a Daubert hearing on Tuesday, September 2, 2014.
II. Dr. Williams' expected testimony
At the Daubert hearing, Dr. Williams' testified that he intends to testify that PCA was the source of the 2008-2009 salmonella outbreak. He bases that opinion on the following information. Dr. Williams is the Chief of the Outbreak Response Branch of the Centers for Disease Control and Prevention ("CDC") in Atlanta, Georgia. Dr. Williams' group coordinates the public health response to multistate bacterial outbreaks in the United States.
Pulsed-field gel electrophoresis ("PFGE") is a laboratory DNA test that identifies unique characteristics of different strains of bacteria. PFGE is a well-established technique that is standardized and easily conducted. The PFGE process is conducted by running a sample through a testing device that produces an image of multiple lines. The lines represent the "fingerprint" of the tested strain of bacteria. The PFGE image can be compared to other test results by trained analysts to determine whether two samples contain the same strain of bacterium. The image is saved as a TIFF file, and that file is uploaded onto Pulse-Net. PulseNet is an electronic network that contains images of PFGE samples obtained from patients at hospitals and doctors' offices, as well as images uploaded by governmental agencies. PulseNet has been in operation approximately twenty years and contains about 500, 000 human samples.
Dr. Williams' Outbreak Response Branch is responsible for investigating bacterial outbreaks. Most of that group's work involves salmonella. Around November 2008, Dr. Williams' team learned that 166 people had been hospitalized with salmonella-related illnesses. For salmonella-related illnesses, the CDC assumes that thirty unreported cases exist for each reported case. For that reason, Dr. Williams considered the referenced hospitalizations an indication of a salmonella outbreak. Dr. Williams' team found that two similar types of Salmonella Typhimurium had infected salmonella patients and therefore concluded that the salmonella was likely from the same or related source. At that time, Dr. Williams' team was tasked with identifying the source of the salmonella outbreak.
Dr. Williams began an investigation and sent to infected individuals questionnaires that involved a series of questions. The investigation focused on three major pillars of evidence. The first pillar is epidemiological evidence. That pillar of evidence involves the identification of the people involved, similarities among their past actions and behaviors, and their demographics, geographic locations, and general health conditions. The second pillar is trace-back evidence. That pillar involves trying to identify a common source of infection among the group of infected people. The third pillar involves testing of foods from the patients' homes.
Based on the evidence collected, Dr. Williams' team concluded that peanut butter and peanut products were the likely source of the outbreak. In each of the facilities affected by salmonella, King Nut peanut butter was found. King Nut peanut butter was traced back to the PCA facility in Blakely, Georgia. Samples taken by FDA inspectors from PCA facilities were compared to samples taken from suspected food items and infected individuals. Dr. Williams' team found that the PFGE patterns from those samples matched. For those reasons, Dr. Williams concluded that PCA was the source of the outbreak. After the recall of PCA products, over the next few months, the number of salmonella-related illnesses returned nearly to zero. That decline in salmonella-related illnesses confirmed to Dr. Williams that PCA was the source of the outbreak.
Dr. Williams and his team published the results of their investigation in the Journal of Internal Medicine. Their findings were based on test results uploaded to PulseNet which were obtained by doctors, FDA employees, and other scientists. Reliance on test results obtained by other people is customary in Dr. Williams' field. Because the CDC responds to multistate outbreaks and ongoing emergencies, it would be impracticable for Dr. Williams' team to personally take samples from all affected individuals.
Defendants stipulate that Dr. Williams is an expert and that PFGE is an accepted scientific method in his field. However, Defendants object to testimony that (1) illnesses were caused by salmonella because such testimony would be prejudicial and violate the Confrontation Clause, and (2) for each one reported illness, there was likely an additional thirty unreported cases ...