Rhode Island Supreme Court Decision In Case Involving Suicide Validates Uncertainty as a Limit in Present Tort Law

Editor’s note: Today’s blogpost was initially published in the November 2012 issue of Medical Liability Monitor, the industry’s premier source for consistent, reliable coverage and fresh perspectives on medical professional liability insurance and risk management issues. Click on the preceding hyperlink to visit the Monitor’s website, where subscription information is available.

by Paul Boylan
In a medical malpractice case decided earlier this year by the Rhode Island Supreme Court (Almonte v. Kurl, 46 A.3d 1), a jury returned a verdict in favor of an emergency room physician who treated an intoxicated and allegedly suicidal man. The patient earlier told police he wanted to kill himself, but was calm and coherent in the emergency room. He told the physician he was not suicidal and did not want to be referred to a psychiatrist. Following blood and urine tests, the doctor met with the patient, determined he was not at imminent risk of suicide and discharged him. About 36 hours later, however, the patient took his own life.

Almonte is non-exceptional in that it involves insufficient trial proof of liability, but it is instructive because the absence of proof is due to the inherent uncertainties surrounding suicide risk prediction. The reasoning in Almonte points to limitations of current common law regarding duty and causation and suggests that simpler rules would work better.
The reasoning central to Almonte is that suicide involves inherent difficulties in proving “the traditional required elements of causation,” according to the Court’s ruling. Those difficulties are “uniquely complex and challenging” because “even under accepted standards of care,” predictions of suicide are “notoriously difficult” and compounded by the fact that the patient “may be actively working at cross-purposes to the practitioner’s goals.”

The ER physician in Almonte had no clear-cut guidance. The patient history and observations gave no clear answers. The plaintiff (the patient’s executor) alleged the physician should have referred the patient for psychiatric evaluation. Almonte holds that without expert medical testimony at trial to establish failure to perform as required by a standard of care, the plaintiff could not prevail.

The issue of duty in Almonte required expert testimony. The concept of duty is troublesome if framed as a discussion of what in fact happened after treatment, i.e., the tragic suicide event. Whether or not duty is performed is something determined within a specific factual context with the information available at the time, and it is evaluated within those limits. If the standard of care is satisfied, subsequent events in principle are not relevant to criticize or characterize the prior performance of duty.

Where there is room for debate as to the specific content of the ER physician’s duty, any standard of care first described retrospectively in a courtroom long after the facts will be, or appear to be, at most a retrospective guess. Unless the standard of care can be articulated before the events and is knowable in advance, it is not clear that any subsequent finding of culpable negligence can be fair or rational. If a qualified expert had testified that the absence of a psychiatric referral was substandard and caused the suicide in Almonte, that testimony would raise new questions. By what standard is that retrospective testimony deemed reliable? If the treating physician faced uncertainty as to prediction of events, those same uncertainties must surely carry over into any later assessment by a trial expert.

If a standard of duty or care is identifiable only when later events seem to be predictable in hindsight, it is not clear that any duty can be described. Even if some consequences are subsequently deemed “foreseeable,” that provides nothing useful for real-life actors such as an ER physician. But defining duty by reference to later events is how common law presently works. The leading case concerning duty taught in most U.S. law schools says that “the orbit of the danger as disclosed to the eye of reasonable diligence would be the orbit of the duty” and “the risk reasonably to be perceived defines the duty to be obeyed.” Palsgraf v. L.I.R.R., 248 N.Y. 339 at 343, 344 (1928).

Identification of duty based on qualitatively more “foreseeable” later-in-time events raises questions as to fairness. If proximate cause is not predictable, foreseeability is arguably an illusion. What we may call foreseeable is what in retrospect seems foreseeable. What tells us that events seemed predictable, however, is solely our retrospective judgment or perception.  What is not foreseen as predictable provides zero guidance for the actor on the scene as to the content of duty. Therefore, what seems to be predictable only in retrospect is not a fair or reliable basis for imposition of a duty in retrospect. By pointing to problems in identifying causes, Almonte raises basic questions about how we identify duty. Where causes are multiple and difficult to assess, we are forced to be more humble about efforts to describe duty.

If the causes of suicide are unpredictable, there is reason to ask whether any medical provider acting in good faith can ever be justly charged with liability for making or not making a psychiatric referral. A more workable rule might be that no liability can be assessed if the medical judgment is undertaken with adequate medical basis and in good faith. Hospitals could establish a clear and simple protocol such as, if one or more informants (e.g., police or family) report that a patient is talking about suicide, those facts alone support a “bright-line” rule in favor of referral for psychiatric evaluation.

Almonte more broadly suggests that when uncertainty is identified and the risk of loss is great, the best available rule may be external to the actor and almost arbitrary. One analogy is to rules employed at four-way stop signs, which simplify the risks of traffic errors when an intersection has no signals. Each car presents a greater risk unless the actors follow arbitrary rules. Stop-sign rules illustrate that simple rules, even if somewhat arbitrary, can sometimes improve on existing concepts. That is a useful insight when, as in Almonte, the natural world presents innocent actors with disorder and lack of certainty as to how to proceed in the face of serious risks.

###

Paul Boylan is a Boston-based shareholder in national law firm LeClairRyan. He has more than 30 years of experience as trial counsel in complex or high-value civil litigation. His clients include insurers, professionals, corporate officers or directors and high net-worth individuals.