Case Discussion: Truthfulness in the Physician-Patient Relationship - Physician Fallibility, Medical Error and Disclosure

A 9-month-old boy, Andre, comes to the clinic with his mother for his well-child visit. His mother specifically wants to address his immunization status and make sure he is up to date. She hands you his state immunization record. You see that he hasn't received his third Haemophilus influenzaetype B (Hib) or inactivated poliovirus (IPV) immunization yet.

Because his mother is in a hurry to get to her own doctor's appointment, you help expedite Andre's catch-up shots by asking your medical assistant (MA) to give them as you leave the patient's room.

Ten minutes later, as you begin to write a note in Andre's chart, you notice a note written three months earlier, when Andre saw one of your partners. The note states that Andre received Hib#3 and IPV#3 as part of that visit. It occurs to you that the immunization card was probably not completed after Andre's last visit. You run to see if you can stop the MA from administering the shots, but it's too late: you meet her coming out of Andre's room just having given them.

Your patient has received an immunization that he shouldn't have. Do you think that having given him the shot twice, instead of once, should be considered a medical error?

Yes, this should be considered a medical error. There are several definitions of what constitutes medical error, but one commonly cited is from the Institute of Medicine (IOM): failure of a planned action to be completed as intended or the use of a wrong plan to achieve an aim. Relevant examples of medical errors using this definition include wrong-site surgery, chemotherapy dosing error and, as in our case, the mistaken administration of a vaccine.

Despite the IOM definition, medical error can be difficult to define. For instance, is it considered medical error if a planned action is successfully completed, but just not on time? Consider the case of an inpatient with sepsis who is scheduled to receive an IV antibiotic every six hours but receives a dose three hours late because his nurse was busy with another patient. There is no consensus on whether this is considered a medical error. Because the definition of medical error can be problematic, empirical data for the prevalence of medical error can vary quite dramatically, depending on how inclusive or exclusive one makes the definition.

The patient likely will not suffer any harm from this second dose of vaccine. Does that make you think differently as to whether this was an error?

A medical error does not necessarily require that a patient has been harmed. Although the definition of medical error can vary, it is widely recognized that medical error is different from an adverse event. An adverse event is an injury caused by medical management rather than by the underlying disease or condition of the patient (IOM). Therefore, while an adverse event involves an injury, there does not have to be harm experienced by the patient for a medical error to have occurred.

Imagine that you are not only the physician in this case but that you are early in your training, such as an intern or resident. How easy or difficult is it to explain your mistake to your attending?

Most people have difficulty talking about an error they have made. It may be even more difficult for physicians. It has been long recognized that physicians are trained within a culture of perfectionism (Leape). This culture teaches that mistakes are unacceptable and are due to a failure of character, leading most trainees to believe that they must pursue perfection in order to be a good physician. Similarly, when an error occurs, most physicians feel that it occurred because of their negligence. The expectation that a doctor is infallible is often held by patients as well (Hilfiker).

This culture makes the disclosure of mistakes difficult. Acknowledging fallibility in the face of a culture that expects, and can make career judgments upon, infallibility, is very hard to do. Fortunately, there is a growing medical literature that calls for the realization that humans make mistakes and that we must design systems in the workplace that prevent human error (Leape). However, lingering expectations for physicians, and especially physicians-in-training, remain, and it is important to recognize these as potential barriers to one's ability to disclose mistakes.

Should you disclose your mistake to the patient's mother? Why or why not?

Yes, this mistake should be disclosed to the patient's mother. Several principles support the disclosure of medical error.

  • Honesty and fidelity
    • The American Medical Association's (AMA) Principles of Medical Ethics states that a physician "shall…be honest in all professional interactions." Many other specialty academies have similar stances. Disclosure is an important part of many professional physician codes because it serves the purpose of developing trust and cooperation essential for a successful therapeutic relationship. Nondisclosure can undermine public trust in medicine by being construed as deceitful as well as disrespectful of the well-being of patients (Matlow).
  • Nonmaleficence
    • Patients may be caused avoidable harm if they are injured further by the failure to disclose. An example of this would be failing to disclose that a surgical sponge was left in a patient after abdominal surgery. This most certainly leads to further harm if it is not immediately removed (Matlow, Hebert).
  • Justice
    • When a patient is harmed, they should be able to seek appropriate restitution or recompense (Hebert).
  • Respect for persons
    • This principle supports the notion that patients have a claim to information that might be of importance to them and relevant to making future decisions. When not told the truth, the patient's autonomy may be restricted in the sense that they are deprived of information that might be useful in making decisions (e.g., whether to change physicians, be more attentive to record keeping in the future, etc.). One might also argue that nondisclosure is disrespectful - one is making a decision for another person that he or she does not want to know or doesn't need to know this information (Matlow).

Most patients want even harmless errors disclosed. One recent study (Gallagher) revealed that nearly all patients want disclosure of all harmful errors along with information about what happened, why the error occurred, how the error's consequences will be mitigated and how recurrences will be prevented. They also want emotional support from their physician and an apology.

Honest disclosure of medical error should be the rule, and violating that principle would require justification. What are some of the arguments people might have against disclosing this probably harmless error?

Some may argue that there is no need to disclose in this case because the error was minor and very unlikely to lead to any harm. The parents may not realize an error has occurred, and telling them about it will just make things worse by undermining their trust in the medical system. But this argument rests on several assumptions that cannot be guaranteed. First, errors frequently are discovered after the fact, at which time it becomes clear that the error was not disclosed. Second, there is always the potential that this child will suffer a side effect in the next couple days. Third, a mistake has been made, so "protecting" the family from this knowledge in order to maintain their trust in the medical system is flawed reasoning. Failing to disclose in order to maintain the fiction that mistakes do not occur does not really protect patients and their families. It seeks to protect physicians and healthcare providers. Finally, it is worth considering the cost to the provider. Nondisclosure is a form of dishonesty, and being dishonest has a corrosive effect on a physician's character.

Physicians commonly worry about legal liability in admitting a mistake and disclosing an error. There are several issues worth mentioning in this regard. First, in this case, in all likelihood no harm has occurred. Medical malpractice requires a demonstration of harm. Even if harm had occurred, however, there is evidence in the literature that patients and families might be less likely to seek legal action when a physician has dealt with them honestly and forthrightly. In contrast, among the most common reasons families give for filing medical malpractice suits are the following:

  • They do not feel they have been told everything about their care or why there was a bad outcome.
  • They feel they have been lied to.
  • They feel there has been a cover-up.

Physicians also worry that the patient or family may lose trust in them and perhaps find another physician. This is not sufficient reason to justify nondisclosure. Families are owed an honest explanation of their care, regardless of the consequences to the physician. Even if trust is lost, and the family chooses to find a new physician, that is their choice to make. In reality, trust in the physician is more likely to be maintained when the physician has demonstrated a willingness to be honest even under difficult circumstances, rather than when a family discovers that the physician has withheld information from them.

Do physicians, by virtue of the physician-patient relationship, have a duty to disclose all medical errors?

No. Just as in routine patient care, we typically don't tell patients, for example, the percentage of eosinophils that are in their CBC differential, as this is largely technical and insignificant.

So, what errors should we disclose?

There are a couple of approaches here:

  • A good rule of thumb can be found in the informed-consent process, where the accepted standard is the concept of the "reasonable person." This concept maintains that a provider should disclose what a reasonable person in the patient's position would want to know in order to make an intelligent and informed treatment decision. In the context of medical error, then, a physician should share with their patients the information a reasonable person in the patient's position would want to know about his or her medical condition, including the diagnosis, prognosis, course and outcomes of treatment, complications and errors in their care (Moskop).
  • Critics of the above approach say that it doesn't seem helpful (what is reasonable, anyway?), and instead propose that errors that don't reach the patient need not be disclosed. The argument supporting this is that in these cases, the "system" in place was robust enough such that the error was recognized and corrected before it reached the patient. In other words, the system worked and did what it was designed to do. If we acknowledge that human error is inevitable, and the systems humans work in should be constructed to minimize or altogether prevent those human errors from reaching the patient, then errors in the healthcare setting that don't reach the patient are exactly what is hoped for.

Beware of the position of "therapeutic privilege," which allows the physician to judge that the disclosure of certain information will in itself be highly detrimental to the patient. Before embarking on this route, one should have well-thought-out reasons that far outweigh the benefits of disclosure.

Beyond disclosing an error to the patient, is there any reason for reporting a medical error to other individuals?

If errors are not brought to the attention of a patient-safety committee or quality-assurance office, efforts to improve the safety of medical practice may be undermined. Quality-improvement techniques designed to prevent the recurrence of errors are thwarted when it is unknown if an error even occurred. Singer states that "the lesson to all of us is that we should learn to love mistakes because they carry in them the kernel of their own elimination."

How should we approach medical error in practice?

Growing evidence shows that patients prefer complete candor and want to be told about all errors that occur in their medical care (adverse events or not). Specifically in the pediatric setting, 99% of parents wanted to be told of the error in the care of their child, regardless of its severity (Hobgood). Other results from this study show that a parent's desire for disclosure does not differ by race, gender, age or insurance status. Also, as mentioned previously, parents are more likely to seek legal action if an error is not disclosed and they learn of its occurrence through other means.

Despite this evidence, and more showing similar results, physicians favor nondisclosure more often than do patients. As stated above, physicians cite a variety of misguided reasons for justifying nondisclosure, including the wish not to upset the patient, fear of disciplinary or legal action, or aversion to implicating colleagues. Some physicians state that they don't disclose errors because they simply don't know how (Hebert, Matlow, Hobgood, Sheldon).

Might it ever be appropriate to deceive a patient?

In the context of a medical error that reaches the patient, no. If the medical error does not reach the patient, as discussed earlier, it is acceptable not to divulge the fact that an error occurred. However, if asked about the error directly by the patient, disclosure, not deception, is encouraged.

In medicine in general, however, there are rare situations in which deception may be appropriate. One must always keep in mind the perspective of the person from whom information is withheld (Bok). Sheldon discusses the case of a physician who finds herself caring for a woman in very critical condition after a car accident that has killed one of the patient's four children and severely injured another. If the woman, in the course of the physician's attempts to stabilize her tenuous medical condition, asks about her children, it may be appropriate to evade a direct response and "never actually indicate that one child is dead" (Sheldon). This seems to be a case in which harm to the mother might be significant enough to warrant temporary deception. In this case, however, it is important to note that the physician fully intends to disclose the truth eventually but has chosen not to do so at this time. In other words, the physician has made a decision about the appropriate timing of the disclosure, rather than a decision not to disclose at all.

Is disclosure of the error enough? Do you have any other obligation to the family?

Disclosure is only the beginning. There are three parts to an ethical response to a mistake (Crigger):

  • Disclosure: Saying honestly and accurately what has happened
  • Apology: Taking responsibility for what happened and expressing regret (one does not have to admit fault to take responsibility and apologize)
  • Amends: Taking appropriate steps to minimize any harm done by the mistake

How might we disclose errors to patients?

Here are some guidelines (Matlow, Hebert):

  • Disclose promptly what you know about the event. Concentrate on what happened and the possible consequences.
  • Take the lead in disclosure; don't wait for the patient to ask.
  • Be prepared for strong emotions.
  • Accept responsibility for outcomes. Don't act defensively or evasively.
  • Apologize with a simple "I'm sorry." Patients often appreciate this form of acknowledgement and empathy, which can sometimes even strengthen your relationship with the family.

How do you think you will feel about your mistake?

It is normal to feel inadequacy, uncertainty, anger, humiliation and guilt after making a mistake. One also may be concerned about the many times in the future where there will be repeated opportunities to make the same mistake. These can all take a toll on one's emotional health and be exacerbated by the fact that the culture of medicine is still struggling with how to address these effects. In the 1980s, Hilfiker wrote that "there is no permission given to talk about errors, no way of venting emotional responses." In the 1990s, Leape echoed that concern: "Seldom is there a process to evaluate the circumstances of a mistake and to provide support and emotional healing for the fallible physician." In 2000, Wu described the physician who makes a mistake as the "second victim" because of the emotional trauma that results. Although the patient has long been recognized as the victim, this perspective helps us realize that most physicians also suffer - though in a different way - after making a mistake.

Errors should be seen as opportunities for growth and learning. It is important to recognize the need for support during these times, and to find trusted colleagues with whom you can discuss your response to what has happened.

Conclusion with Suggestions

  • There is an ethical imperative to disclose medical errors based on the principles of justice, nonmaleficence and respect for persons. In addition, many professional codes in medicine require disclosure. There is also empirical evidence that patients want to be told when an error has occurred in their care. All of these reasons create a firm obligation on the part of the physician for disclosure.
  • Despite the codes and principles above, there is little guidance as to how and when to report medical errors. Furthermore, there may even be medical errors that are acceptable not to disclose. For example, a good argument can be made that errors that do not reach the patient (e.g., an order for 10 times the dose is caught by the nurse or pharmacist) do not need to be disclosed to patients or their families. In those cases, the system worked, even if one part of that system failed. Because the system worked, the error did not reach the patient: the order was corrected before the wrong dose was given. Humans make mistakes, and physicians are humans; the system in which we work should have processes designed that make errors less likely, and when the system works to catch and correct errors, there is no need to disclose these "close calls" to families.
  • Making a mistake, particularly one that harms a patient, is often accompanied by emotional turmoil and feelings of guilt, inadequacy and shame. It is important to recognize the need for support during these times and to use errors as an opportunity for personal growth and learning.
  • The culture of medicine is slowly changing to incorporate systems thinking and trust in order to promote a blame- and shame-free environment for physicians.
  • When disclosing an error, be timely, tell the patient and family what happened, take responsibility and apologize.

Truthfulness in the Physician-Patient Relationship

1. Instructor's Guide 2. Student's Guide 3. Case Discussion

This instructor's guide was developed by Douglas J. Opel, MD, senior fellow, Clinical Bioethics, and Douglas S. Diekema, MD, MPH, director of education, Treuman Katz Center for Pediatric Bioethics, Seattle Children's Hospital.

In addition to the copyright notice set forth in the link below, permission to display, cache and print unlimited copies of the Case-Based Teaching Guides referred to on this page is hereby granted, solely for educational purposes, without charge (other than charges solely to cover the costs of copying), and without alteration of the Materials in any way.