Recognizing and addressing the problems presented by differential diagnosis testimony

Introduction to differential diagnosis

In product liability and toxic tort litigation, plaintiff bears the burden of proving that the defendant, or an act of the defendant, caused his or her injury. To meet this burden, expert witnesses are retained to offer opinions on the cause of the plaintiff’s injuries and often employ a tool known as differential diagnosis.

Differential diagnosis is a patient-specific technique used by medical practitioners and physicians to determine from which of two or more diseases a patient is suffering, by comparing the patient’s symptoms to symptoms associated with known diseases.[1] In injury litigation, an expert witness offers a differential diagnosis opinion (more accurately, “differential etiology”) by listing all potential causes of the plaintiff’s injury, including the cause attributable to the defendant. Then the expert explains how she eliminated all other possible causes of injury until the only remaining cause is the one attributable to the defendant.[2]

Although plaintiffs commonly rely on differential diagnosis as proof of specific causation, courts should be wary of admitting expert testimony based on differential diagnosis due to problems applying a medical technique in litigation. This article explores some of the problems, including: (1) The perversion of the medical tool of differential diagnosis as used in the legal realm; and (2) the results-driven nature of the methodology supporting the differential diagnosis.

Using expert testimony to establish causation

To succeed on a claim of injuries, plaintiffs must prove causation. Typically in the context of product and toxic tort litigation, plaintiff bears the burden of proving both general causation – that the defendant’s product or alleged toxin is capable of causing the alleged injury – and specific causation – that the plaintiff’s specific injury was caused by defendant’s product or toxin.^[3] To prove causation, plaintiffs rely on expert witness testimony.

In determining the admissibility of expert testimony, courts apply either the majority Daubert standard[4] or the previously prevailing Frye standard[5]. The “general acceptance” standard mandated in Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923), is a more deferential standard, allowing courts to admit expert scientific testimony so long as it is generally accepted in the relevant scientific community.^[6] Seventy years after Frye was decided, the United States Supreme Court heard argument in Daubert v Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). In Daubert, the Court held that as trial court “gatekeepers,” judges must “ensure that any and all scientific testimony or evidence is not only relevant but reliable.”^[7]

The Daubert court listed several factors judges should consider as relevant, but not dispositive, when assessing the scientific validity of expert testimony, including: (1) Whether the scientific theory or technique can be and has been tested; (2) whether the theory or technique has been subjected to peer review publication; (3) the known or potential rate of error; (4) the existence and maintenance of standards controlling the technique’s operation; and (5) whether the theory or technique is generally accepted within the pertinent scientific community.[8] Thus, while Daubert retained the “general acceptance” aspect of Frye, it also heightened the standard, by requiring judges to look beyond whether the expert’s method is generally accepted.

In the nearly 25 years since Daubert was decided, a majority of courts have held that a reliable differential diagnosis provides a valid foundation for expert medical opinion using the Daubert standards.^[9] As a result, differential diagnosis has grown in popularity and familiarity in the courtroom. Nevertheless, courts need to remain vigilant when assessing the admissibility of differential diagnosis testimony, paying particular attention to the issues discussed below.

Early courtroom uses of differential diagnosis

Differential diagnosis testimony has been used in courts for decades by parties offering medical opinions. As early as 1940, a doctor testified he used differential diagnosis to identify a patient’s disease in support of the patient’s dispute that he was entitled to benefits under an insurance policy.^[10] By the 1980’s , courts no longer restricted differential diagnosis opinions to identifying a plaintiff’s illness or disease, rather the medical technique was stretched beyond “diagnosis” and encompassed opinions of the “cause” of the plaintiff’s illness.

For example, in In re Swine Flu Immunization Products Liability Litigation, 533 F. Supp. 703, 713 (D. Utah 1982), the District Court for the District of Utah concluded that the plaintiff proved by substantial evidence that the swine flu vaccine was the proximate cause of her neurological illness, transverse myelitis. The court relied on four of the five expert-witness neurologists, who testified they performed a differential diagnosis to rule out other possible causes of the plaintiff’s illness. Rather than using differential diagnosis to identify and diagnose the plaintiff’s condition, the experts were allowed to reach beyond differential diagnosis as it is used in the medical setting and offer a broad conclusion that the vaccine was the cause of plaintiff’s condition.^[11]

In the 1990’s the pendulum began to swing in the other direction with a significant increase in admissibility challenges to differential diagnosis. Differential diagnosis opinions became more closely scrutinized due to the increasing use of scientific experts in court, the rise of toxic tort actions and renewed interest in the criteria used to judge the admissibility of expert testimony.^[12] Given this history one might suggest that the use of differential diagnosis in support of opinions of the cause of a plaintiff’s injury, has been fully vetted. Familiarity with the phrase differential diagnosis and a general understanding of the technique, is not an excuse for complacency.

Deficiencies of differential diagnosis

Perversion of the medical technique as used in litigation

Physicians and practitioners who offer differential diagnosis testimony to support opinions of the specific cause of a plaintiff’s disease oftentimes distort the purpose of the medical technique. As discussed above, differential diagnosis is the process whereby medical practitioners determine the nature of a patient’s illness for purposes of diagnosis and possible treatment. In contrast, differential etiology is the process used by experts designated for trial to determine the cause of the patient’s illness for purposes of imposing liability. Both methods require process of elimination reasoning, but they lead to two fundamentally different determinations: Differential diagnosis is used to diagnose the patient’s condition, while differential etiology takes the diagnosis of the patient’s condition as a given and then uses the basic technique of elimination to determine the cause of the patient’s diagnosed condition.[13]

Although the distinction may seem slight, the implications are far-reaching. From a practical standpoint, doctors regularly engage differential diagnosis to identify the disease or illness responsible for the patient’s symptoms. Once the illness is identified the physician can determine the most appropriate treatment, i.e., additional testing, medication, or surgery.[14] Because physicians are primarily concerned with diagnosis and treatment, they rarely pursue the question of what caused the illness or disease, because in the medical context, the cause of the disease is irrelevant.[15]

Doctors receive extensive training in deducing disease or illness based on a patient’s symptoms, by performing a differential diagnosis of a patient. But doctors have far less training in deducing the cause of those diseases.[16] Accordingly, differential diagnosis performed by a physician in the clinical setting is aimed at diagnosing the patient’s disease and typically is not used to determine the cause of the patient’s disease.

A hypothetical example illustrates the point. A patient visits her doctor with complaints of respiratory symptoms and cough, fatigue and shortness of breath. The patient’s history included cigarette smoking, a family history of lung cancer and the patient had recently attended a large conference at a hotel in Philadelphia. The doctor conducts a general physical examination and orders a chest x-ray. The doctor uses differential diagnosis and considers a number of conditions that could present with the same general symptoms, including chronic obstructive pulmonary disease (“COPD”), lung cancer and lung infection. After reviewing the x-ray, the doctor rules out COPD and lung cancer. The doctor diagnoses the patient with a lung infection commonly referred to as pneumonia. A treatment plan of antibiotics is prescribed.

Every year more than three million Americans will be diagnosed with pneumonia, however not all pneumonias can be attributed to the same cause. In fact, there are a myriad of pathogens and a variety of methods of exposure that cause pneumonia. Some patients are exposed by breathing a bacteria lurking in the air, others may be exposed to different bacteria when handling infected birds or animals and still others may aspirate contents from the stomach or through the pharynx, leading to aspiration pneumonia. These patients all develop pneumonia, but the organism and method of exposure that caused their pneumonias are very different.

In the example of the hypothetical patient, the doctor diagnosed pneumonia and prescribed a broad-spectrum antibiotic known to be a successful treatment for common lung infections. Although the doctor concluded that the patient’s symptoms were due to pneumonia, the doctor’s diagnosis tells us nothing of what caused the pneumonia in this particular patient. In fact it is common in the clinical setting that determining the specific cause of a patient’s illness is not necessary for a doctor to make a diagnosis and treatment plan. It is here that the disconnect between the medical and legal application of differential diagnosis occurs. While diagnosing a patient’s condition, a treating doctor may not even consider the cause of a patient’s illness. Yet, when the patient becomes a plaintiff in a lawsuit, she is required to prove both the nature of her illness – pneumonia – and the cause of her pneumonia.

In Turner v. Iowa Fire Equipment Company, 229 F.3d 1202, 1205–07 (8th Cir. 2000), the plaintiff’s treating physician, Dr. Hof testified in deposition that the plaintiff was suffering from a variety of respiratory problems and further opined that such problems were caused by exposure to substances released when a fire extinguisher discharged. The Turner court grasped the fine distinction between diagnosis and determining the cause of the illness behind the diagnosis. The court admitted the diagnosis testimony but excluded the causation testimony.[17] The court explained:

Unlike his diagnosis of condition, Dr. Hof’s causation opinion was not based upon a methodology that had been tested, subjected to peer review, and generally accepted in the medical community. Significantly, Dr. Hof did not systematically rule out all other possible causes. He was clearly more concerned with identifying and treating [the plaintiff]’s condition than he was with identifying the specific substance that caused her condition.[18]

The court held that although “a causation opinion based upon a proper differential diagnosis (one that systematically rules out other possible causes) satisfies Daubert,” the exclusion of Dr. Hof’s causation testimony in this case was proper.[19]

Similarly, in Wynacht v. Beckman Instruments, Inc., 113 F. Supp. 2d 1205, 1208-09, 1211 (E.D. Tenn. 2000), the court admitted testimony of the plaintiff’s treating physician and allowed testimony identifying plaintiff’s condition, but excluded the physician’s opinion on medical causation because it was based on unreliable methodology.

If the identification of plaintiff’s illness or disease is in dispute, differential diagnosis can be a useful and reliable tool; however, rarely is the diagnosis seriously contested. Alternatively, when a plaintiff offers differential diagnosis as support for an opinion of the cause of plaintiff’s condition, the court must keep as a focal point of its analysis the distinction between a clinical diagnosis of a disease for treatment and a legal “diagnosis of the cause” of a disease, to ensure that such testimony is reliable and admissible.

Results-driven methodology

There is also a danger that an expert providing causation testimony based on differential diagnosis produced his or her results using subjective, results-driven methodologies, such as by concluding first the cause of the plaintiff’s injury and then working backwards to establish support for that conclusion. In considering expert testimony that a defendant caused a plaintiff’s injury, courts are in agreement that an expert does not need to rule out every conceivable cause of the plaintiff’s injury in order for the expert’s testimony to be admissible.[20]

However, a problem arises when the expert fails to provide any valid explanation connecting available scientific evidence with the cause to which he or she is testifying. Indeed, while a plaintiff’s expert is generally called for the primary purpose of testifying in favor of the plaintiff’s case, courts must ensure that the methodology used to support that testimony is reliable. If a plaintiff’s symptoms could have multiple causes, the expert cannot, without supporting his opinion, “simply pick[] the cause that is most advantageous to [the plaintiff]’s claim.”[21] As one court explained, “Mentioning some risk factors and moving on to a conclusion, without a specific explanation, is not a proper differential diagnosis. . . . [It] is simply a recitation of facts—this does not help the jury.”[22]

The district court in Castellow v. Chevron USA, 97 F. Supp.2d 780, 786 (S.D. Tex. 2000), held that an expert who formulates a conclusion first and then works backward to medical and scientific support for his conclusion does not withstand the reliability and relevancy requirements under Daubert.

Likewise, relying on Castellow, the court in Henricksen v. Conoco Phillips Company, 605 F. Supp. 2d 1142 (E.D. Wash. 2009) excluded an expert’s causation testimony based on an unreliable differential diagnosis. In Henricksen, the plaintiff’s experts failed to consider and exclude the likelihood that plaintiff’s acute myelogenous leukemia (“AML”), was caused by an unknown factor or was idiopathic, even though 80-90% of AML cases had no known cause.[23] The experts failed to address the possibility that the plaintiff’s AML was idiopathic. Additionally, the court recognized that comparing plaintiff’s AML with chemically induced or “secondary” AML “would not have served plaintiff’s purposes,” as the features characteristic in secondary AML were not present in plaintiff’s case.[24] The court ruled that such results-driven analysis or lack thereof, was unreliable and therefore inadmissible.[25]

Courts must ensure that the methodology used to formulate a causation opinion based on differential diagnosis may not be scientifically or medically supported and therefore inadmissible.

Conclusion

In the case of the hypothetical patient, now plaintiff, the treating doctor is called to testify at trial. The doctor tells the jury it is his opinion the plaintiff suffered from pneumonia. He explains in reaching his diagnosis, he used the well-accepted technique of differential diagnosis to conclude that the plaintiff’s symptoms were attributable to pneumonia, ruling out COPD or a lung tumor. In addition, his diagnosis was confirmed when the plaintiff responded successfully to the antibiotics prescribed and her symptoms resolved.

At this point the jury has heard the plaintiff was diagnosed and treated for pneumonia. The treating doctor offered nothing for a jury to conclude the cause of plaintiff’s pneumonia. Yet, all too often, a treating doctor’s invocation of the words “differential diagnosis” is taken out of the context of diagnosis and instead deemed sufficient proof of the specific cause of plaintiff’s injury.

The problems posed by the use of differential diagnosis in the legal setting, such as the perversion of this medical tool when used in the legal context and the results-driven nature of the methodology supporting differential diagnosis, will not diminish any time soon. However, recognizing a physician’s differential diagnosis of a plaintiff’s symptoms is not always the most reliable method for establishing specific causation; defense counsel should more readily object and courts should more readily exclude testimony that falls outside the standards governing the admissibility of expert testimony.

[1] Whiting v. Boston Edison Co., 891 F. Supp. 12, 21 n.41 (D. Mass. 1995); A. McGehee Harvey & James Bordley III, Differential Diagnosis 7 (2d ed. 1970); John B. Wong et al., Reference Guide on Medical Testimony 687, 690, in Federal Judicial Center, Reference Manual on Scientific Evidence (3d ed. 2011), available at http://www.fjc.gov/public/pdf.nsf/lookup/SciMan3D01.pdf/$file/SciMan3D01.pdf; Michael D. Green et al., Reference Guide on Epidemiology 549, 617–18, in Federal Judicial Center, Reference Manual on Scientific Evidence (3d ed. 2011), available at http://www.fjc.gov/public/pdf.nsf/lookup/SciMan3D01.pdf/$file/SciMan3D01.pdf.

[2] Baker v. Dalkon Shield Claimants Trust, 156 F.3d 248, 252-253 (1st Cir. 1998).

[3] In re Zicam Cold Remedy Mktg., Sales Practices, and Prods. Liab. Litig., 797 F. Supp. 2d 940, 941-946 (D. Ariz. 2011).

[4] Federal courts and a majority of state courts apply the Daubert standard. Alaska, Arkansas, Colorado, Connecticut, Delaware, Idaho, Indiana, Iowa, Kentucky, Louisiana, Maine, Michigan, Mississippi, Montana, Nebraska, New Mexico, North Carolina, Ohio, Oklahoma, Oregon, Rhode Island, South Carolina, South Dakota, Tennessee, Texas, Vermont, West Virginia and Wyoming have adopted Daubert or a similar test for determining the admissibility of scientific and other expert evidence. 90 A.L.R. 5th 453, §§ 3-27.

[5] Arizona, California, District of Columbia, Florida, Illinois, Kansas, Maryland, Michigan, Minnesota, Missouri, New Jersey, New York, North Dakota, Pennsylvania and Washington continue to apply the Frye test for determining the admissibility of scientific and other expert evidence. 90 A.L.R. 5th 453, §§ 28-43.

[6] Frye, 293 F. at 1014.

[7] Daubert,509 U.S. at 589.

[8] Id. at 593-94.

[9] Kennedy v. Collagen Corp., 161 F.3d 1226, 1228 (9th Cir. 1998); Baker, 156 F.3d at 252–53; Zuchowicz v. United States, 140 F.3d 381, 385–87 (2d Cir. 1998).

[10] Phx. Mut. Life Ins. Co. v. Harmegnies, 110 F.2d 20, 26 (8th Cir. 1940).

[11] In re Swine Flu, 533 F. Supp. at 714.

[12] Joseph Sanders & Julie Machal-Fulks, The Admissibility of Differential Diagnosis Testimony to Prove Causation in Toxic Tort Cases: The Interplay of Adjective and Substantive Law, 64 LAW & CONTEMP. PROBS. 107, 108 (2001) (citing Raynor v. Merrell Pharms., Inc., 104 F.3d 1371, 1376 (D.C. Cir. 1997) (Bendectin and fetal birth defects); Ambrosini v. Labarraque, 101 F.3d. 129 (D.C. Cir. 1996) (Depo-Provera and birth defects); Hall v. Babcock & Wilcox, 69 F. Supp. 2d 716 (W.D. Pa. 1999) (exposure to radiation and cancer); Pick v. Am. Med. Sys., 958 F. Supp. 1151 (E.D. La. 1997) (silicone penile implant and autoimmune disorder); Golod v. Hoffman La Roche, 964 F. Supp. 841 (S.D.N.Y. 1997) (Tegison and eye injury); Berry v. CSX Transp., 709 So. 2d 552 (Fla. App. 1998) (organic solvent exposure and toxic encephalopathy); Norfolk S. Ry. Co. v Baker, 514 S.E.2d 448 (Ga. App. 1999) (diesel exhaust exposure and nasopharyngeal cancer)).

[13] Michael D. Green et al., Reference Guide on Epidemiology 549, 617 n.211; see also Turner v. Iowa Fire Equip. Co., 229 F.3d 1202, 1208 (8th Cir. 2000) (distinguishing between differential diagnosis conducted for the purpose of identifying the disease from which the patient suffers and one attempting to determine the cause of the disease).

[14] John B. Wong et al., Reference Guide on Medical Testimony 687, 691, in Federal Judicial Center, Reference Manual on Scientific Evidence (3d ed. 2011), available at http://www.fjc.gov/public/pdf.nsf/lookup/SciMan3D01.pdf/$file/SciMan3D01.pdf.

[15] Michael D. Green et al., Reference Guide on Epidemiology 549, 617 n.211.

[16] 3 Mod. Sci. Evidence § 21:1 (2015-2016 Edition).

[17] Turner v. Iowa Fire, 229 F.3d at 1210.

[18] Id. at 1208.

[19] Id.

[20] Smith & Nephew, 259 F.3d at 202 (“A medical expert’s opinion based upon differential diagnosis normally should not be excluded because the expert has failed to rule out every possible alternative cause of a plaintiff’s illness.”); Johnson v. Mead Johnson & Co., LLC, 754 F.3d 557, 563 (8th Cir. 2014); Kannankeril v. Terminix Int’l, Inc., 128 F.3d 807 (3d Cir. 1997); Takeda Pharm, 239 Cal. App. 4th 555, 578 (2015).

[21] Viterbo v. Dow Chem. Co., 826 F.2d 420, 424 (5th Cir. 1987).

[22] In re Prempro Products Liab. Litig, No. 3:05CV00078-WRW, 2010 WL 8357351, at *2 (E.D. Ark. Sept. 26, 2010).

[23] Henricksen, 605 F. Supp. 2d at 1162.

[24] Id. at 1162–63.

[25] Id. at 1163.