“First, Do No Harm”. This is the first principle attributed to Hippocrates as part of the ancient oath of the medical profession. It expresses a fundamental ethic that should guide the practice of every physician. Today, medical science and technology have evolved to permit medical practitioners to stop or slow the progress of many life threatening diseases. The advance of modern medicine, and its power to heal, is unlike anything that existed when our tort law was developed. Chief among the arenas of progress in treatment and healing is treatment of cancer, a leading cause of premature, non-traumatic
death today. The Minnesota Supreme Court recently heard an appeal in the case of Dickhoff v. Green that may determine whether professional accountability will be possible for the vast number of people who need cancer treatment.
In this case, the Court is asked to consider whether traditional notions of proof of proximate cause should apply in medical malpractice cases in which there is a negligent delay of diagnosis and treatment of cancer.
Since medical malpractice claims, by their very nature, always involve plaintiffs who suffered preexisting illness or injury before encountering the defendant, how the court resolves this question could have a huge impact upon whether physicians can be held accountable for failing to do no harm, or whether the medical profession will be able to shield itself from accountability by relying upon the misfortune of the very patients who seek their help in relieving that misfortune.
One of the problems frequently encountered when there is a delay in diagnosis of cancer is “what is the harm?” If the cancer diagnosis is eventually made in time for some treatment, then unless the patient is on death’s door, there will be much to litigate in the question of what is the harm. Anyone who has had cancer or known someone with it, knows that prompt diagnosis and treatment is the key to success in battling the disease. Yet in spite of that general truth, can we really say that a few more days, weeks or months, will make much difference in the long run? Medical knowledge offers only indirect evidence of the truth in such circumstances. Mainly they offer statistics about 5 year survival rates, and treatment norms tied to “staging” of cancer. When cancer is said to be in “remission”, what it really means is not that it is necessarily gone, but that the disease has been so beaten back, that our imperfect diagnostic tools cannot find it. Whether the cancer is “gone” or merely “in hiding”, usually only time can tell. But regardless of whether the statistics offer much direction, there is an underlying truth in each case that delay has certainly not helped matters and that, quite possibly, any such delay has squandered the only opportunity for a cure.
If we remember back to the first year of law school, we can recall that one of life’s mysteries was the doctrine of “proximate cause”. As our professors brought us through the development of notions of tort causation, slowly one day we settled upon some core notions that seem to now have almost universal acceptance in American common law. First, a cause need not be the only cause to still be considered a legal cause of a harm for which there may be liability. Second, a legal cause need not be an indispensable cause, that is to say that a legal cause need not pass the “but for” test. Third, a legal cause need not be the primary or most significant cause. Fourth, a minor cause may not be a legal cause unless it is at least a “substantial” cause. The burden of proof is on the plaintiff to offer enough evidence to persuade a reasonable jury that the alleged negligence was a substantial factor in bringing about harm to the plaintiff.
In the Dickhoff case now before the court, the plaintiff’s expert opines that the delay of diagnosis and treatment permitted progression of the cancer to such an extent that the survival rate for treatment had gone from one in which survival was probable (60%) to one in which death had now become probable (dropping to a 40% survival rate). The defense contends that since the likelihood of death was 40% before seeking treatment, and the delay of treatment only meant it crept to 60%, then most of the risk of death (two-thirds of it to be exact) pre-existed the patient’s attempt to seek treatment. The defense submits that for the delay to be considered a legal cause of harm, it must exceed the natural disease process itself to be sufficient. In effect, they urge a “predominant cause” approach.
Plaintiff argues that since Minnesota previously rejected damage claims for a mere loss of a chance at recovery and that the cases holding this reaffirm the notion that plaintiff mush show damage to be a probability, merely the shifting of probability as occurred here is all that is required for the case to go to the jury. Of course this approach alone would also mean that if the probability of death was nudged from 49% to 51%, that a mere 2% bump in a risk that was still overwhelmingly due to the natural disease process, would be also compensable as a cause. This scenario does not exactly sound like the delay would have been a “substantial factor”. This formulation is at least troublesome under such a circumstance.
During oral argument on this appeal, Justice Stras asked a key question that put the appellant’s theory of predominant cause to the test of the settled common law construct of legal cause. Justice Stras noted that precedent held that when two causes combined to produce a single harm, we have recognized either cause as sufficient to be a legal cause, so long as it was found to be a “substantial factor”. Justice Stras asked why should not that conception of the law apply here? The defense response to this was that this formulation of the “substantial factor” test is typically applied in the context of two separate tortious causes and, that since here we have an underlying natural disease process that was not caused by a tort, this approach to causation should not apply in medical malpractice cases. At first blush, this response seems a plausible, and perhaps clever, distinction. Upon closer examination, however, the distinction fades in significance. It turns out that the distinction between tortious and innocent causes was not one considered in the development of the law. Nor is it a distinction that logic suggests ought to have much bearing on the question of whether the tortious cause has been proven to be significant enough to allow a jury to consider it.
If we remember back to the first year of law school, we may recall that the origin of the “substantial factor test”, perhaps first articulated in a Harvard Law Review article from 1911 was, according to Professor William Prosser, “picked up by the supreme court of Minnesota in a case of merging fires presenting an issue of causation in fact, and was used by the court as a substitute for the obviously inapplicable “but for” rule of causation.” W. Prosser, Law of Torts, 4th Ed., p. 248. The case cited by Professor Prosser was Anderson v. Minneapolis, St. P. & S.S.M.R. Co, 179 N.W. 45 (Minn. 1920).
Ninety two years later, the same Minnesota Supreme court is now revisiting this long settled case law in the pending Dickhoff appeal. In its 1920 Anderson ruling, the Court applied the “substantial factor” test where two distinct fires combined to burn and destroy the plaintiff’s property. One of those fires was allegedly of tortious origin caused by the defendant railroad. The other fire, however, was of unknown and possibly natural origin. The Court, assuming either fire may have been sufficient alone to cause the harm, upheld the use of the substantial factor even in this situation. The Court noted that if both fires had been started by different railroads, there would clearly be joint liability. The Court saw this application of law as persuasive in holding that the substantial factor should also apply when the concurrent cause was of unknown origin.
Whether or not the plaintiff’s formulation of the “tipping the scale” from probably survivable to probably not survivable is always going to be sufficient to make a prima facie case for the causal link to a physician’s negligence, the facts of this case do not present a particularly close question. There is little doubt that a reasonable jury may conclude that physician negligence causing a patient with a 40% risk of death to now have a 60% risk of death, presents enough of a shift in the odds of survival, so as to be a substantial factor in causing harm to the patient. Of course, a reasonable jury might also conclude it is not “substantial” either. But that, after all, is what we have juries for.