Commentary: Systems, Sensitivity, and “Sorry”

Fix the problem, not the blame.—Sean Connery giving Wesley Snipes his take on Japanese business strategy, in the film Rising Sun (20th Century Fox, 1993)

I thank the editors of the Journal for the opportunity to discuss this useful and timely article, in which Dr. Steiner¹ describes, in part, taking a systemic corrective approach to errors, rather than attempting to assign individual responsibility and blame. I will discuss the liability risk aspects of the approach described and will make some general observations.

Looking at adverse outcomes from the viewpoint of the system requires a certain moral atmosphere in the institution as a whole. It is similar to the moral atmosphere that makes free acknowledgment of countertransference in psychotherapy a natural and discussable aspect of treatment rather than a shameful character flaw. Like all moral atmospheres, it must come from the institution’s leadership. Such an atmosphere requires transcending the natural human tendency, when under stress, to adopt the paranoid position of projection and externalization of blame, as if to say: “We did not fail as a system or institution; only one particular bad apple was negligent in this case.”

We all accept the inevitability of accidents, but medicine (compared with, say, the airline industry) comes late to the analysis. In an excellent book, Normal Accidents, Charles Perrow² presents the theory that accidents occur when two factors mesh: complexity and “close coupling.” The latter means a system in which an error or oversight has immediate consequences down the line. An example from the cases in Dr. Steiner’s article might be (1) the complexity of dealing with many inpatients who have constantly changing mental states, and (2) the situation when those in the chain of command are unclear about the rules for sharps and the consequences—self-injury—are immediate.

Some accidents result from ambiguity. In one hospital, an order was written, “4 mg for four days.” The intent was to prescribe 1 mg each day for four days, a total of 4 mg. The order was read as 4 mg each day for four days, resulting in a total of 16 mg, a marked overdose. Though an understandable error, this event was described in one local newspaper as “The Three Stooges have taken over the hospital.” This latter caricature captures some public perceptions of errors in medicine.

What do patients want when an adverse outcome occurs? They want acknowledgment of the harm, an expression of regret and remorse, and a remedy. The last desire is very often expressed as the wish to prevent anyone else from being harmed.³ Historically, the patients’ wishes and even the doctors’ inclinations toward candor and regret have been in severe tension with the threat of liability. Let us look at each element.

Acknowledgment has powerful risk management effects. The failure of the clinical staff to acknowledge a wrong, sometimes combined with premature referral to the hospital’s lawyer, frustrates the family’s wish to know what happened. Both responses may drive the family into the attorney’s office. In fact, my caseload includes malpractice cases triggered by the treater’s refusal to return phone calls. Some people sue just to get the facts.

Regret and remorse are both best conveyed by an apology. Efforts to introduce apology into the culture of medical institutions are beginning to grow. Many years ago, I suggested⁴ that apology had a specific value in restoring the doctor-patient alliance and preventing suit. Others have reached the same conclusion.⁵ In Massachusetts, this approach receives statutory support from an “apology law” which states: Statements, writings or benevolent gestures expressing sympathy or a general sense of benevolence relating to the pain, suffering or death of a person involved in an accident and made to such person or the family of such person shall be inadmissible as admission of liability in a civil action (Ref. 6, ch 233, § 26d).

In translation, this law states that to say you are sorry is not to say you are culpable or liable. Such statutory support obviously clears the way for freer use of apology.

Remedy addresses the expressed concern that similar things might happen to others. Applying this approach might mean describing what actually happened or sharing the remedial actions or plans for same with the aggrieved patient or family.

All the approaches discussed thus far are usually implemented by the institution’s risk managers. Formerly, risk managers of the old school were said to urge treaters to back away from the family, have no contact, and refer all calls to the hospital’s attorney. This approach often led to the “I’ll see your attorney and raise you my attorney” response. The modern risk manager may suggest meeting with the family for both explanatory reasons and to aid in the process of grief when appropriate.

Dr. Steiner’s portrayal of a systems approach¹ brings to mind an incident in which a woman was hospitalized for trying to drown herself in a bathtub and then succeeding in doing just that in a hospital bathtub while on suicide precautions. Certainly, this scenario seems to reveal negligence. A wider view showed that the bathroom was directly across the hall from a seclusion room, where a large, strong, out-of-control manic patient was being taken down and secluded, requiring the involvement of almost all the staff on the ward. In the melee, with attention focused on that patient, the woman sneaked into the bathroom and drowned herself.

Can plaintiffs’ attorneys deal with a systems view rather than naming individual or corporate defendants? When a systems view is presented, such attorneys usually sue the institution as a whole, motivated at least in large part by the “deep pockets” theory. The unanswered question so far is: what do juries make of this systems view? Can they grasp a systemic problem without a specifically identified scapegoat? My own view is that the legal outcome would depend on whether explaining the systems view sounds like a feeble rationalization for culpability, based on the attempt to “spread the blame around,” rather than an understanding of something fairly complicated. Unfortunately, complicated scenarios do not play well in court.

The pool of data is growing as to the actual success rates of apology, increased acknowledgment of error, and a systems approach to remedy. Such statistical support must overcome the fearful resistance caused by the risk of liability.

• American Academy of Psychiatry and the Law