AW: For a study that was published in the Journal of the American Medical Association in 1991, we asked interns and residents in three of the top internal medicine programs in the country to report to us the worst mistake that they had made in caring of patients, what happened to the patient, what happened to them, what they did, how they were treated by the institution and what happened subsequently. Virtually everyone who responded reported a significant error, most of which harmed patients, yet only half told their attending physician and less than a quarter told the patient or the family. In 1997 we published an analysis in the Journal of Internal Medicine on ethical and practical considerations in disclosing mistakes. We came down very strongly with the position that physicians are obligated to disclose errors that harm patients to patients and/or their families, largely because the patient stands to benefit from it. While there are potentially some risks or downsides for the physician, the doctor-patient relationship can also, paradoxically, improve sometimes with these discussions. We are currently conducting work on disclosure of medical errors. We are interested in the question of what is the best way for a physician, hospital or other health care worker to tell patients or families about incidents that harm them. We have developed a series of vignettes which describe real cases where patients were harmed and in which a doctor is disclosing to a patient what happened. One involves an overlooked mammogram, another a chemotherapy overdose, a third is of a physician who is slow to answer pages about a hospitalized patient and as a result the patientcondition deteriorates. We are showing these vignettes to people who volunteer to view them and we’re asking them if they trust the physician, what their impressions are of the physician, and also if they would sue. We are just now getting the results of that study, but our hypothesis is that a full apology and acknowledgement of responsibility will be received better by patients and families.

AW: At this point research has been fairly limited, mostly to either case studies, natural experiments or surveys. An important study published in the Lancet reported on closed claims: patients were asked what they would have wanted to hear at the time of an incident. What they described was wanting to learn what happened, that someone would accept responsibility, that steps were being taken to prevent future similar incidents, and they wanted an apology. A proactive policy was put into place ten years ago at the Lexington Kentucky VA Medical Center of disclosing incidents to patients and offering them compensation. Perhaps the most interesting result is that, while the number of claims increased over the years, their claims experience has actually been more favorable. They are in the top quartile for the number claims filed but in the bottom quartile for the total payout per year, compared to a reference group of other VA hospitals. They’ve been sued very rarely and the settlements are generally small and prompt, with relatively little involvement of the legal profession. Other studies, and they are relatively few, have surveyed patients about what their preferences would be about disclosure and, in general, patients want to know when these events occur. The preponderance of results from those studies is that a full disclosure is likely to result in the patient trusting the physician more, but would not necessarily change their proclivity to sue.

AW: There are a number of approaches. The first is to sweep them under the rug and hope no-one notices – that is, to not disclose. This may still be the most common strategy for incidents that are not suspected, and the benefit is that perhaps it will never be discovered. The downside is that, if it is discovered, it is very likely that patients will be angry, lawsuits will include damages and coverage in the press is likely to mention the word "cover-up" rather prominently. When it’s evident that an error has occurred one can simply describe what happened. The question is whether one should apologize, and apologies come in three varieties. The first is a full, personal, sincere apology which acknowledges responsibility. The second is something more of an impersonal or non-specific apology: "I’m sorry that you are having a bad day." The third is no apology at all, which could include accepting responsibility personally or on behalf of the institution, or not. At the moment there is a diverse opinion about apology, but I think the tide is swinging toward the understanding that apology is almost always beneficial. There’s a worry that apology is tantamount to an admission of guilt, but it appears that patients do not seem to see it that way. In our own institution, I don’t think there has ever been a case where the fact that the physician apologized wound up figuring into a lawsuit, and if it did it was only to the positive: patients observing that the physician was honest, sincere, seemed to understand that they had suffered, and thought better of the physician as a result.

In the initial disclosure discussion, it makes sense to simply tell the patient everything you know about what happened – no more and no less. One certainly shouldn’t speculate about what might have happened and one shouldn’t take responsibility for something for which it is not evident that you are responsible. If it’s only apparent that something bad happened, you certainly can say, "It looks like something bad has happened. We are not sure exactly what has occurred. We are looking into it and we will let you know as soon as we find something out." If you were the patient, while it would be upsetting to get that news, you would feel like you are being appropriately informed about what is going on and it could help you or your family cope better with medical problems, which are certainly upsetting all by themselves.

It is best to have the initial discussion as soon as you are aware that an adverse event occurred and as soon as you can say something intelligent about it. The more that discussion occurs in real time, the more it really is part of the normal discourse between physician and patient informing them about their condition, and the less it seems like something that ought to merit a malpractice suit. The longer one delays and withholds information, the more one is uncomfortable and it becomes more evident to the patient that something is amiss, not just with their condition but with how they’re being handled. People become enraged, frankly, when things are covered up from them. They believe that it’s their right to know, and we believe the same thing.

AW: Historically, physicians have gotten advice in many cases from both insurance companies and their defense attorneys not to say anything. Patients describe the experience of previously friendly and forthcoming physicians and nurses suddenly clamming up after there’s been a bad outcome for a patient, and find this very chilling and distressing. In the past, physicians got advice which was as extreme as, "You may not discuss anything with a patient, certainly not admit any responsibility and certainly not apologize to the patient. In fact, you may not say anything to the patient without first clearing it with me, your lawyer. If you do, your malpractice coverage may be voided." Not surprisingly, that’s had a chilling effect, which I believe is pernicious. Paradoxically, when health care workers do clam up, patients become angry and if anything, it appears that the likelihood of a lawsuit increases. Generally, the most common reason for lawsuits, it appears, is patients trying to get some kind of information about a bad outcome when that information is not forthcoming through official channels. This is beginning to change. Our own chief counsel, Department of Risk Management and our malpractice insurer believe that full disclosure with apology is the right thing to do, and probably the best thing to do for the institution. We have a disclosure policy in place which dictates that it’s the right of the patient to learn about mistakes made in their care, and it’s the obligation of physicians at the hospital to disclose these events to patients and families. I think that, when our physicians become aware of this, in many ways they are relieved: it appears that this has happened to other people before. There is a policy. There are guidelines for how to do this. And the institution is behind them.

AW: Certainly, disclosure of an event that was unsuspected could lead to a settlement or even a lawsuit. Despite this, disclosure is the right thing to do and, if there was an injury, it is very likely that the information will come out through some different channel. The physicians and institutions really have one chance to talk to patients about the incident that happened in real time, before it appears that the information is only coming out because the cover-up had failed. I think that patients are remarkably charitable with physicians who are simply describing these incidents occurring as part of the dialogue that’s hopefully been going on with them throughout their care, and I think that’s the way it should be treated. The Lexington, Kentucky VA experience suggests that proactive investigation, full disclosure, and even offering to settle early and for a reasonable amount when it seems appropriate is likely to have good outcomes for the patients. Those who need compensation are likely to get it promptly and the total costs that are incurred by the institution, overall, are likely to be the same or perhaps even decreased. The surveys that have been done in terms of liability risk suggest that disclosure of an unsuspected incident is likely to have a neutral impact. On the one hand, people may give you credit for being honest and be less likely to sue. On the other hand, after finding out they had been injured by an incident which they had not known about, some of those people are likely to seek compensation. For incidents where it is obvious that that the patient was injured by medical care, I think there is no downside to full disclosure, apology and trying to come to some agreement with the patient that is reasonable.

AW: The only data we have currently is anecdotal, although some of those anecdotes are quite compelling. I’ve heard patients and plaintiffs attorneys on their behalf say that an apology takes the emotion out of the incident or takes away some of the patient’s anger at the physician. They may still be angry that it occurred, but focus more appropriately on trying to pursue health rather than striving to beat the doctors and hospitals in order to get what they feel is the truth that they deserve. Part of what they do at the Lexington, Kentucky VA is express their regret that these things have occurred, and they describe patients and their lawyers as being surprised, initially, but then being able to put those issues behind them and get down to the business of figuring out if a settlement is required, what would be a reasonable settlement amount and how to do it in a reasonable time. They describe everyone as becoming much more sensible about these incidents, sensible not being the word that’s usually associated with malpractice suits.

My own belief is that apologies are part of treating the patient with the respect and compassion that they deserve. I think you lose nothing by expressing your own regrets and apologizing. The apology should be appropriate to the situation. If it’s something you have done you should say, "I am sorry that I did this." If it’s something that happened through systemic causes in the institution, you could certainly say that you are sorry that it happened. If there’s no way to determine exactly what happened, it may also make sense to take responsibility for the institution: "I’m sorry that this happened. This happened on my watch. It shouldn’t have happened. I take responsibility for it."

AW: Error and apology are part of a cultural change that is really necessary in institutions. There needs to be greater acknowledgement all around that we are fallible, that the mistakes are inevitable, and they will happen even in the best institutions and even involving the best and most well-meaning clinicians. I think patients need to understand, too, that the people who are taking care of them are human, and things can and do go wrong. I think that increased awareness of the inevitability of errors allows institutions to place appropriate emphasis on managing their institutions with that reality. Errors will always happen; the trick is to manage those risks, manage that uncertainty, manage those errors, and to still come out in the end with the best outcomes for patients. It is difficult for leaders to make decisions, to shift priorities, to improve safety without having any data. While disclosure to patients is important, there currently is inadequate disclosure of incidents to colleagues and institutions. Without those data, it is impossible for institutions to know where they should be putting their efforts.