Reference Manual on Scientific Evidence Third Edition (2011) / Chapter Skim
Currently Skimming:
Reference Guide on Medical Testimony--John B. Wong, Lawrence O. Gostin, and Oscar A. Cabrera
Reference Guide on Medical Testimony--John B. Wong, Lawrence O. Gostin, and Oscar A. Cabrera
Pages 687-746
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 687... ...
Licensure and credentialing, 698 4. Continuing medical education, 700 B. Organization of Medical Care, 700 C. Patient Care, 702 1.
|
From page 688... ...
Risk communication, 737 3. Shared decisionmaking, 739 V. Summary and Future Directions, 740 Glossary of Terms, 742 References on Medical Testimony, 745 688
|
From page 689... ...
A survey of federal judges published in 2002 indicated that medical and mental health experts constituted more than 40% of the total number of testifying experts.1 Medical evidence is a common element in product liability suits,2 workers' compensation disputes,3 medical malpractice suits,4 and personal injury cases.5 Medical testimony may also be critical in certain kinds of criminal cases.6 The goal of this reference guide is to introduce the basic concepts of diagnostic reasoning and clinical decisionmaking, as well as the types of evidence that physicians use to make judgments as treating physicians or as experts retained by one of the parties in a case. Following this introduction (Section I)
|
From page 690... ...
(ruling that district court not in error for excluding expert medical testimony that relied on an improperly performed differential diagnosis) (citations omitted)
|
From page 691... ...
2005) (distinguishing differential diagnosis from differential etiology, with the former closer to the medical definition and the latter employed as a technique to determine external causation)
|
From page 692... ...
2000) (a medical malpractice case reversing a lower court's interpretation of the statutory phrase "clinical practice" because it did not comport with the legislature's intent that the statutory meaning reflect the medical definition)
|
From page 693... ...
The Supreme Court adopted an abuse of discretion standard in Joiner23 and affirmed it in Kumho.24 Although in most product liability cases the courts reached the same conclusion, inconsistent determinations regarding the admissibility of similar evidence may not constitute an abuse of discretion under the federal standard of review or in states with a similar standard.25 C elationship of Medical Reasoning to Legal Reasoning R As Section II.A suggested, the goal that guides the physician -- recommending the best therapeutic options for the patient -- means that diagnostic reasoning and the process of ongoing patient care and treatment involve probabilistic judgments concerning several working hypotheses, often simultaneously.
|
From page 694... ...
As Section IV will demonstrate, when analyzing the patient's symptoms and making a judgment based on the available medical evidence, a physician will not expressly identify a "proximate cause" or "substantial factor." For example, in order to recommend treatment, a physician does not necessarily need to determine whether a patient's lung ailment was more likely the result of a long history of tobacco use or prolonged exposure to asbestos if the optimal treatment is the same. In contrast, when testifying as an expert in a case in which an employee with a long history of tobacco use is suing his employer for possible injuries as a result of asbestos exposure in the workplace, physicians may need to make judgments regarding the likelihood that either tobacco or asbestos -- or both -- could have contributed to the injury.28 Physicians often will be asked to testify about patients from whom they have never taken a medical history or examined and make estimates about proximate cause, increased risk of injury, or likely future injuries.29 The doctor may even need to make medical judgments about a deceased litigant.30 Testifying in all such cases requires making judgments that physicians do not ordinarily make in their profession, making these judgments outside of physicians' customary patient encounters, and adapting the opinion in a way that fits the legal standard.
|
From page 695... ...
The goal of medical school is to prepare students in the art and science of medicine for graduate medical education.32 Of the 4 years of medical school, the first 2 years are typically spent studying preclinical basic sciences involving the study of the normal structure and function of human systems (e.g., through anatomy, biochemistry, physiology, behavioral science, and neuroscience) , followed by the study of abnormalities and therapeutic principles (e.g., through microbiology, immunology, pharmacology, and pathology)
|
From page 696... ...
citizens and foreign nationals.35 Because educational standards and curricula outside the United States and Canada vary, the Education Commission for Foreign Medical Graduates has developed a certification exam to assess whether these graduates may enter Accreditation Council for Graduate Medical Education (ACGME) accredited residency and fellowship programs.36 self-education on the subject, and that she had critically reviewed medical literature for the FDA.
|
From page 697... ...
. 41. Accreditation Council for Graduate Medical Education, Specialty Programs with Dependent and Independent Subspecialties, available at http://www.acgme.org/acWebsite/RRC_sharedDocs/ sh_progs_depIndSubs.asp (last visited Feb.
|
From page 698... ...
Besides awarding medical licenses, state medical boards also investigate complaints, discipline physicians who violate the law, and evaluate and rehabilitate physicians. The Federation of State Medical Boards represents the 70 medical boards of the United States and its territories, and its mission is "promoting excellence in medical practice, licensure, and regulation as the national resource and voice on behalf of state medical boards in their protection of the public."45 Credentialing typically involves verifying medical education, postgraduate training, board certification, professional experience, state licensure, prior credentialing outcomes, medical board actions, malpractice, and adverse clinical events.
|
From page 699... ...
.49 The American Osteopathic Association (AOA) certifies osteopathic physicians in 18 osteopathic specialty boards (e.g., emergency medicine, internal medicine, obstetrics and gynecology, family medicine, pediatrics, surgery, and others)
|
From page 700... ...
Although hospitals increasingly belong to a network or system, most community physicians belong to practices involving 10 or fewer physicians.56 Concerns about the safety of the organization of medical care first arose from the Harvard Medical Practice Study which found that adverse events occurred in 52. American Osteopathic Association, Continuing Medical Education, available at http:// www.osteopathic.org/inside-aoa/development/continuing-medical-education/Pages/default.aspx (last visited Feb.
|
From page 701... ...
Leape et al., The Nature of Adverse Events in Hospitalized Patients: Results of the Harvard Medical Practice Study II, 324 New Eng.
|
From page 702... ...
. In Crossing the Quality Chasm, the IOM emphasized care delivery that should accommodate individual patient choices and preferences and be customized on the basis of patients needs and values.64 62. 2001 CQHCA Report, supra note 54, at 44, 5-6.
|
From page 703... ...
Social history usually includes education, employment, and social relationships and provides a socioeconomic context for developing or coping with illness and an employment context for exposure to environmental or toxin risks. Finally, the review of systems is a comprehensive checklist of symptoms that might or might not arise from the various organ systems and is an ancillary means to capture symp 65. Medical Professionalism Project: ABIM Foundation, Medical Professionalism in the New Millennium: A Physician Charter, 136 Annals Internal Med.
|
From page 704... ...
many diseases 69. 2001 CQHCA Report, supra note 54, at 27.
|
From page 705... ...
Goodman, Toward Evidence-Based Medical Statistics, 1: The p Value Fallacy, 130 Annals Internal Med.
|
From page 706... ...
Physicians finding those three symptoms in a patient with hypertension may overestimate the patient's likelihood of having pheochromocytoma based on representativeness bias, overestimating the likelihood of a less common disease just because case findings resemble those found in that disease.78 Other cognitive errors include availability (overestimating the 75. Elstein, supra note 73; George A Miller, The Magical Number Seven Plus or Minus Two: Some Limits on Our Capacity for Processing Information, 63 Psychol.
|
From page 707... ...
83. Elstein, supra note 73. 84. Arthur S
|
From page 708... ...
That is, not everyone with disease has a positive test; these are called false-negative test results. Similarly, some individuals who are healthy may mistakenly have positive tests; these are called false-positive test results.
|
From page 709... ...
Screening 1000 women for breast cancer. 1000 Women Prevalence = 0.8% 8 with Breast Cancer 992 without Breast Cancer Sensitivity Specificity = 90% = 93% 7 with 1 with 69 with 923 with Positive Negative Positive Negative Test Test Test Test Probability of Breast Cancer 7 with +Mammogram = 9% (Predictive Value Positive)
|
From page 710... ...
The probability of breast cancer among those with a positive mammogram is termed the "predictive value positive." Similarly, if the test were negative, the likelihood of breast cancer in those with a negative mammogram ("false reassurance rate") would be 1 divided by 924 (1 woman with breast cancer and a negative test and 923 women without breast cancer who have negative tests in Figure 1)
|
From page 711... ...
Terms such as "sensitivity," "specificity," and "predictive value negative or positive" are called conditional probabilities because they express the likelihood of a particular result based on a particular condition (e.g., a positive test result among those with disease) or the likelihood of a particular condition among those with a particular result (e.g., disease among those with a positive test)
|
From page 712... ...
Screening 1000 individuals with a positive family 91. Gigerenzer, supra note 87, at 42.
|
From page 713... ...
Even with a specificity of 99.99%, if a test screens for 10,000 genes simultaneously, then 63% of individuals will have at least one false-positive test result. Based simply on the genetic test results alone, neither individuals nor physicians would be able to distinguish those with true-positive results from those with false-positive results, thereby potentially leading to inappropriate monitoring or treatment for all with positive test results.
|
From page 714... ...
Similarly, a very positive test has a different interpretation than a weakly positive test, and Bayes' rule can quantify the difference. Results from multiple tests can be combined with Bayes' rule by applying Bayes' rule to the first test result and then reapplying Bayes' rule to subsequent test results.
|
From page 715... ...
can suddenly rupture and cause heart attacks. Not surprisingly, revascularization trials involving either bypass surgery or percutaneous interventions such as stenting or angioplasty do not diminish the risk of having a heart attack or improve survival for most patients.107 Although treating physicians108 may testify with regard to both general and specific causation, as with use of evidence for causation, their standards for evi 104. Kassirer et al., supra note 73, at 63–66.
|
From page 716... ...
However, in a randomized trial limited to patients with documented weakened heart, one particular beta blocker was found to not confer a survival benefit, and as a result the heart failure guidelines limited their beta blocker recommendation to just those three drugs with documented mortality benefit in trials.110 Although treating physicians may be aware of patient-specific risk factors such as smoking or family history, they may not routinely review specialized aspects of such data, for example, toxicology, industrial hygiene, environment, and some aspects of epidemiology. Additional experts may assist in distinguishing general from specific causation by using their specialized knowledge to weigh the relative contribution of each putative causative factor to determine "reasonable medical certainty" or "reasonable medical probability." The determination of general causation involves medical and scientific literature review and the evaluation of epidemiological data, toxicological data, and dose–response relationships.
|
From page 717... ...
Screening for disease in symptomatic, a otherwise healthy patients has become widely accepted and promulgated. 114 Screening differs from diagnostic testing used to elucidate the cause of symptoms or loss of function because screening involves apparently healthy individuals.115 Although screening may prevent the development of disease-related morbidity and mortality, positive test results (both false positive and true positive)
|
From page 718... ...
119. Black, supra note 115, at 1280. 120. Charles K
|
From page 719... ...
Normative prescriptive decision models for reasoning in the presence of uncertainty suggest that whether and which tests get ordered should depend on the sensitivity and specificity of the test as discussed in Section IV.A.2, supra, but also the risk of mortality or morbidity from the test, and the benefit and risk of treatment.125 In general, for sufficiently low probabilities of disease, no tests should be ordered and no treatment given. For sufficiently high probabilities of disease, 121. Black, supra note 115.
|
From page 720... ...
.127 Just as some therapies may eventually be found to be harmful or not beneficial, tests initially felt to be useful may be found to be less valuable.128 Among other potential biases,129 this may occur because of the choice of study population used to determine the test's sensitivity and specificity. For example, an FDAapproved rapid test for HIV infection has a reported specificity of 100%, implying that any positive tests must indicate truly infected individuals, yet one of the populations in which testing is recommended is women who have had prior children and are in labor but have not yet had an HIV test during the pregnancy.130 In 15 multiparous women, this rapid HIV test resulted in one false-positive test result in the 15 women tested, yielding a specificity of 93%,131 and so not all pregnant women with positive tests can be assumed to be truly infected.
|
From page 721... ...
136. John E Wennberg, Unwarranted Variations in Healthcare Delivery: Implications for Academic Medical Centres, 325 BMJ 961–64 (2002)
|
From page 722... ...
; 2000 CQHCA Report, supra note 58. 139. Michael J
|
From page 723... ...
is that the strength of medical evidence supporting a therapy or strategy is hierarchical. When ordered from strongest to weakest, systematic review of randomized trials (meta-analysis)
|
From page 724... ...
Guyatt et al., Users' Guides to the Medical Literature: A Manual for EvidenceBased Clinical Practice (2d ed.
|
From page 725... ...
In some cases, treatments that have no effect on mortality or are potentially harmful continued to be recommended by several clinical experts.159 155. Goodman, supra note 74, at 998.
|
From page 726... ...
."165 In their evaluation of mammography, the USPSTF "recommends against routine screening mammography in women aged 40 to 49 years" (see 160. Committee to Advise the Public Health Service on Clinical Practice Guidelines, Institute of Medicine, Clinical Practice Guidelines: Directions for a New Program 8 (Marilyn J Field & Kathleen N
|
From page 727... ...
Preventive Services Task Force Recommendation Statement, 147 Annals Internal Med.
|
From page 728... ...
Rosoff, Evidence-Based Medicine and the Law: The Courts Confront Clinical Practice Guidelines, 26 J Health Pol., Pol'y & L
|
From page 729... ...
Donald Schön argued that regardless of the professional field, "An artful practice of the unique case appears anomalous when professional competence is modeled in terms of application of established techniques to recurrent events" and that specialization "fosters selective inattention to practical competence and professional artistry."179 In the case of a patient with peanut allergies and heart disease, allergy guidelines recommend avoiding beta blockers, but heart disease guidelines recommend beta blockers because they have been shown to prolong life in patients with heart disease. An allergist would recommend against taking a beta blocker, yet a cardiologist would recommend taking it.180 Well-performed randomized trials provide the least biased estimates of treatment benefit and harm by creating groups with equivalent prognoses.
|
From page 730... ...
In early randomized controlled trials, oral antiarrhythmic drugs such as encainide and flecainide were approved by FDA on the basis of their ability to suppress these extra heartbeats in patients who had had a myocardial infarction. Years after 183. Dennis T
|
From page 731... ...
190. Clarke, supra note 152, at 815.
|
From page 732... ...
. Some patients with very low risk of dying from their heart attack who received tPA likely were harmed because their risk of intracranial hemorrhage exceeded the benefit.194 In practice then, even in a randomized controlled trial demonstrating survival benefit, on average, those benefits may not accrue to every patient in that trial that received treatment.
|
From page 733... ...
. In one study, lumpectomy and radiation appeared to have a lower risk of breast cancer recurrence with 5 women reported to have had breast cancer recurrences following lumpectomy and radiation versus 195. Guyatt, supra note 150, at 8; see also supra Section IV.C.3.
|
From page 734... ...
201. Jacobson, supra note 199, at 999. 202. Karen Sepucha et al., Policy Support for Patient-Centered Care: The Need For Measurable Improvements In Decision Quality, Health Affairs Supp.
|
From page 735... ...
test based on recommendations from the U.S. Preventive Services Task Force, the American College of Physicians–American Society of Internal Medicine, the American Medical Association, the American Urological Association, the American Cancer Association, and the American Academy of Family Physicians.
|
From page 736... ...
adults who had made a medical decision found that Internet use 211. King & Moulton, supra note 206, at 432–34; Daniel Merenstein, A Piece of My Mind: Winners and Losers, 291 JAMA 15–16 (2004)
|
From page 737... ...
. 216. 1983 Kassirer, supra note 203, at 889.
|
From page 738... ...
Nelson et al., Screening for Breast Cancer: An Update for the U.S. Preventive Services Task Force, 151 Annals Internal Med.
|
From page 739... ...
Preventive Services Task Force Recommendation Statement, 151 Annals Internal Med.
|
From page 740... ...
For example, consider lymphoma, a blood cancer that used to be classified simply by appearance under the microscope as 231. Michael J Barry, Health Decision Aids to Facilitate Shared Decision Making in Office Practice, 136 Annals Internal Med.
|
From page 741... ...
Communication is critical because most decisions involve tradeoffs, in which case individual patient preferences for the outcomes that may be unique to patients and that may affect decisionmaking should be considered. In summary, medical terms shared in common by the legal and medical professions have differing meanings, for example, differential diagnosis, differential etiology, and general and specific causation.
|
From page 742... ...
The last stage of narrowing the differential diagnosis to a final diagnosis by testing the validity of the diagnosis for its coherency, adequacy, and parsimony. differential diagnosis.
|
From page 743... ...
A change in the likelihood of the potential diagnostic hypotheses (differential diagnosis) in response to additional information, e.g., symptoms, signs, and lab test results.
|
From page 744... ...
The predictive value positive or positive predictive value is the probability of disease in those known to have a positive test result. The predictive value negative or negative predictive value is the probability of disease in those known to have a negative test result.
|
From page 745... ...
. Gordon Guyatt et al., Users' Guides to the Medical Literature: Essentials of Evidence-Based Clinical Practice (2d ed.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.