“Dr. Young, when was the last time you took care of a sick child?”

This is a question I am asked frequently during cross-examination while testifying in court. The attorney cross-examining me implies to the jury through this question that his experts — in this case practitioners of clinical medicine — are more qualified than I am to determine the cause of an injury or death because they see patients in a clinic. Frequently, I get to admit that I am not a board-certified pediatrician, ophthalmologist, neurologist, neurosurgeon, radiologist, or … name a specialist.

What attorneys and many, many others — even physicians — fail to realize is that patient care no more qualifies one to practice forensic medicine than running marathons qualifies one to compete in a swim meet: both activities require athletes but the athletic events are different. Patient care and forensic medicine are two different activities, two horses of a different color.

Do you know the difference between diagnostic medicine and forensic medicine? The distinction is more important than you might think.

A diagnosis is defined as “the identification of the nature and cause of something (of any nature).” As applied to medicine, it is “the identification of the nature and cause of an illness¹.” Patients visit physicians to find out the nature and cause of their unpleasant and worrisome symptoms. They do not know what is causing their troubles but they visit the doctor with the hope that he or she will identify what is wrong and then successfully treat it.

On the other hand, forensic medicine does not involve patient care. Forensic medicine benefits society at large and has the same purposes as the other forensic sciences: to determine what happened and who is responsible for what happened².

Note that the critical question for diagnostic medicine involves an issue of present concern, but the critical questions for forensic medicine involve a past event. Unlike a patient who does not understand the cause of his symptoms, the questions of primary importance to a court of law involve behaviors that someone could observe if that someone was at the right place at the right time to observe those behaviors. The questions involve behaviors that are potentially criminal or have other legal consequences.

For example, the assignment of the cause and manner of death on a death certificate — a legal document used in probate and frequently admitted as evidence in court — addresses the critical questions of forensic medicine. The cause of death answers the question, “What happened?” and the manner of death — a one-word designation placed on every death certificate — answers the question, “Who is responsible for what happened?” If the manner of death is “natural,” then no one — other than, perhaps, the victim through legal behaviors that may involve his “lifestyle” — is responsible for the death. If the manner is “suicide” — one of the “violent” and “unnatural” manners of death — then the victim is responsible for the death in a way that is not permitted legally. If the manner is “homicide,” then someone other than the victim is responsible for the death in a way that also may not be permitted legally. And if the manner is “accident,” then someone may or may not be responsible and that responsibility could be argued in a court of law. Finding the cause of a natural death may be important to several in society, particularly diagnostic physicians who are “curious” about disease, but this importance is secondary to the legal consequences resulting from a violent or unnatural cause — an injury or injuries caused by past behaviors.

Just as the critical questions differ, so do the goals of each practice of medicine.

One cannot perform appropriate treatment for an underlying cause unless that underlying cause is accurately identified. Also, one cannot properly provide a “forecast of the future course of a disease or disorder” (prognosis) without an accurate diagnosis³.

Also, a court or other agency cannot administer matters in society justly or protect the public without knowing the truth behind critical past events. These are two entirely different sets of goals.

What about the requirements to practice each? The requirements are the same.

This is why there is confusion between the two practices of medicine. Both require the same training in the same basic medical sciences, and those identical requirements allow the public to assume that the two are interchangeable.

What about the specialists in each area of medicine who devote their entire activity to that area? These also differ.

Note that the practitioners of diagnostic medicine are in adequate numbers — they even compete with one another to perform and get paid for a limited number of diagnostic procedures — but the practitioners devoted to forensic medicine are few and far between. One might argue that forensic medicine practitioners should be on the endangered species list! Why is this important? Because the supply of forensic medicine practitioners may be meager but the demand for their services is huge! We live in a litigious society, so the need for forensic expertise in both civil and criminal matters is almost endless. Consequently, diagnostic medicine practitioners — wittingly or unwittingly — find themselves responding to calls to be involved in the legal arena.

One would think that this would not be a problem, but as it turns out, it is a very big problem, so big that it has led to failure after failure in the courtroom. The failure of science in the courtroom in many cases is not disputed⁴. Could it be that a source of these problems is from a confusion of the analytical techniques of diagnostic medicine with those of forensic medicine?

As a child, I recall visiting the “funhouse” at a local amusement park. One portion of the funhouse had several distorting mirrors. The images reflected from these mirrors were amusing and grotesque, altering the size and shapes of our bodies in a surreal fashion. As I have demonstrated in previous articles, illogical analytical techniques for past events lead to incorrect conclusions²,⁵,⁶,⁷,⁸. Perhaps in a metaphorical sense, illogical analytical techniques for past events distort the truth like a funhouse mirror, allowing grotesque images to appear rather than accurate reflections.

What kind of funhouse mirror reflection would we receive if we analyzed forensic cases with the analytical techniques of diagnostic medicine?

Consider what happens when a patient visits a physician. The patient provides a “chief complaint” and a “history of present illness” where he or she describes “symptoms” (problems experienced by a patient, such as chest pain, shortness of breath, nausea, vomiting, diarrhea, etc.). The doctor notes this information, asks for a “past medical history” (previous illnesses, injuries, allergies, surgeries, family history, social history, etc.), performs a “review of systems” to see if there are other problems related to other organ systems in the patient, and performs a “physical examination” — looking for discoverable “signs” that would indicate the underlying problem. Much of this involves “forward reasoning” (inferring deductively from cause to effect). The past issues and events that the patient witnessed may help the physician understand his or her problems in the present, but often forward reasoning in diagnostic medicine is not sufficient. The patient often does not know what is wrong with him or her; if what was wrong was something the patient could see, then he or she would seek immediate treatment for an obvious problem rather than a diagnosis.

The physician then engages mostly in “backward reasoning” (inferring abductively from effect to cause) to guide his diagnostic approach in the selection of laboratory, imaging, and other diagnostic studies⁶. Often, physicians will use a differential diagnosis to guide the selection of appropriate diagnostic tests. A differential diagnosis is a list of potential causes for a patient’s problem. Like other forms of backward reasoning, the differential diagnosis technique is subject to the same biases and cognitive limitations as other forms of backward reasoning, such as availability (the inability to imagine more than a few items in a list) and anchoring (the inability to consider other new options that are completely different from what was previously considered)². Still, the trial-and-error approach to diagnostic testing will hopefully provide other leads to consider if the initial considerations do not pan out. Diagnostic physicians cannot use a differential diagnosis to uncover every possible condition: only conditions that have been discovered and characterized scientifically. One cannot order tests for something that doctors are not aware of.

This same diagnostic method has also influenced research in medicine. In order to diagnose and treat medical conditions and to provide prognoses, ailments have to be placed into various nosological categories. Such categorizations allow the study of medical problems in various populations. If certain diagnostic and treatment protocols work sufficiently well for members of a population, the thinking goes, the same protocol may work in a particular patient. Statistical methods are applied in these population studies to assess the probability of efficacy. Diagnostic physicians may alter the categories as more is learned about a specific malady. Such categorizations are useful only as they approximate the truth about the condition in specific patients.

Now back to the funhouse mirror. What happens when forensic medical practitioners adopt diagnostic analytical techniques?

Consider the child who is admitted to the hospital with a head injury. The child abuse pediatrician is notified about the suspicion that the child may have a “diagnosis” of “abusive head injury⁹,” so this consultant views the medical record; obtains a history from persons other than the suspected perpetrator; performs a physical examination; views the results of laboratory testing, imaging, and an eye examination for retinal hemorrhages; considers other conditions in a differential diagnosis; orders additional tests; then provides a “diagnosis” once conditions other than child abuse have been “ruled out.” The child abuse pediatrician fails to acknowledge that:

1. A diagnosis is for conditions that exist in the present, not behaviors that occurred in the past. As such, no observation made by a clinician nor any diagnostic study he or she may order can evaluate something that no longer is occurring.
2. A consultant cannot reliably determine “what happened” and “who is responsible for what happened” without learning from the person or persons who observed first-hand “what happened” and “who is responsible for what happened.” Relying on hearsay information obtained from a medical record or from others who were not present for the event is not sufficient, particularly when those providing the information were nowhere near the past event in place or time.
3. A consultant cannot reliably determine “what happened” and “who is responsible for what happened” from physical evidence — even after performing additional diagnostic tests. I have previously discussed fallacies of both deductive and inductive logic if one claims to determine past events from physical evidence⁸.
4. A consultant cannot reliably “rule out” all medical conditions that may mimic the consequences of “abusive head trauma” because there are conditions yet to be discovered and characterized scientifically. Also, the cognitive limitations of availability and anchoring may greatly limit the number of possibilities that are considered — even among those conditions that have been discovered and characterized scientifically. A consultant cannot consider conditions that never enter his or her conscious thought.
5. A consultant cannot make assessments of what is likely or not likely to have happened and apply them to a specific case because to do so is to commit the ecological fallacy: aggregate statistics that describe a population cannot apply with reasonable certainty to an individual case.
6. A consultant cannot rely on population studies of forensic issues that rely on diagnostic analytical techniques because such studies multiply the fallacies described above and utilize categories that oversimplify the complexities of human function and dysfunction.
7. The consultant cannot offer opinions on the case in court “to a reasonable degree of medical certainty” when such inferences are merely hypotheses (speculations) that cannot be tested¹,⁷.

What about forensic matters evaluated through forensic analytical techniques, such as the Forensic Scientific Method² and the Inferential Test⁵? In the case above, the child abuse consultant views all available primary witness information and all pertinent physical evidence. This would require the consultant to view primary witness accounts (hopefully not hearsay) and crime scene investigation findings from police investigative records, the ambulance trip sheets, the emergency department records, photographs of the child taken before and after the incident, crime scene photographs, medical records, autopsy photographs, and any other item that would fill in a continuous timeline of events from before the questioned event to the present. The doctor would then compare the witness accounts with the physical evidence from medical tests, an autopsy (if one is performed), and items from the crime scene, explaining which physical evidence items are consistent and which are inconsistent with witness accounts. The attempt would be to verify if what a witness claims makes sense scientifically. Opinions can then be expressed with certainty for items inferred deductively (forward from cause to effect), but the consultant would use terms of uncertainty for items inferred abductively (backward from effect to cause)⁵,⁷.

Consider the critical questions of diagnostic and forensic medicine once again:

If you are stumped by these critical questions in a case, where should you go for answers?

If you are a diagnostician and are asked to consult on forensic matters, I hope you will use the information above to your benefit. After all, who wants to make mistakes that send people to prison for crimes they did not commit?

¹ http://en.wiktionary.org/wiki/diagnosis. Accessed 21 Apr 2011.

² Young TW. Forensic Science and the Scientific Method.  February 13, 2008.

³ http://en.wiktionary.org/wiki/prognosis. Accessed 21 April 2011.

⁴ National Research Council. Summary. In: Strengthening Forensic Science in the United States: A Path Forward. Washington, DC: The National Academies Press; 2009. Pp. 1-33.

⁵ Young TW. An Inferential Test for Expert Testimony.  April 2, 2009.

⁶ Young TW. Is Sherlock Holmes’ “Reasoning Backwards” a Reliable Method for Discovering Truth? Analyses of Four Medicolegal Cases.  September 7, 2010.

⁷ Young TW. Attorneys and Judges, You Can Stop the Madness Now. September 18, 2010.

⁸ Young TW. Putting It All Together: The Logic Behind the Forensic Scientific Method and the Inferential Test.  January 11, 2011.

⁹ Christian C, Endom EE. Evaluation and diagnosis of abusive head trauma in infants and children. In: Lindberg DM, Drutz JE, Nordli DR. UpToDate, version 19.1; June 7, 2010.  Accessed 22 April 2011.