• State the rationale for disclosing medical errors.
• Describe key principles in effective error disclosure.
• Appreciate that physicians are reluctant to criticize colleagues.
• Outline a process for disclosure of an error made by another institution.

A healthy 4-year-old boy presented to an emergency department (ED) with 3 days of vomiting associated with lethargy and fevers. He had been exposed to another child with streptococcal pharyngitis (strep throat) the previous week but otherwise had been well until the symptoms began. He received a full evaluation in the ED. He was found to have a low-grade fever and was a little sleepy with some redness in his throat. The laboratory tests were unremarkable and a head computed tomography (CT) was reported as normal by the radiologist. A rapid test for streptococcal pharyngitis was positive.

The child was admitted to the hospital for ongoing care and given intravenous hydration and antibiotics. Over the next 24 hours, the child became increasingly confused, disoriented, and lethargic. The following morning, his condition worsened and he had a respiratory arrest. He was placed on a ventilator and transferred to the intensive care unit (ICU).

In the ICU, he was noted to have fixed and dilated pupils on neurologic exam, a sign of serious neurologic injury. A repeat CT scan of the brain revealed severe cerebral edema (swelling of the brain) with evidence of herniation of the brain through the base of the skull.

He was transferred from this hospital to a tertiary care center for ongoing management. At the tertiary care center, the child was evaluated by neurology and neurosurgical teams. Further testing revealed a diagnosis of venous sinus thromboses (blood clots in the veins of the brain), which had led to edema and herniation. Unfortunately, the brain damage was too advanced and the child was determined to have no chance to survive.

As part of their routine evaluation, the neurology, neurosurgical teams, and the radiologists at the tertiary care center reviewed the CT scan that had been done in the original ED. Although the findings were subtle, they found that the scan was not normal (as had been reported) but demonstrated clear evidence of cerebral edema. The initial hospital had not recognized these findings and therefore had not pursued further work-up for the cause, which would have been indicated. The neurology and neurosurgical teams thought that if the brain swelling had been recognized at the time, the child could have been transferred earlier, received surgical management, and might have survived.

When it was clear the child could not survive, the pediatricians met with the mother and father to explain that their child was brain dead. Angry and upset, the parents asked repeatedly, "How could this happen? How could the CT scan have been normal and then be so bad in less than 48 hours?"

Due to concerns of legal liability, the hospital administration and the risk management department at the tertiary care hospital had instructed the physicians and other providers to not disclose the misinterpretation of the original CT scan. In fact, they were instructed not to comment on the care provided by the initial hospital in any way. Therefore the parents were never told that an error had been made that may have contributed to their child's death.

Few conversations strike as much fear in physicians' hearts as talking with patients about medical errors.(1) Health care workers strongly favor being open with patients, but find the process of disclosing errors extremely challenging—especially when they perceive that someone else may have been primarily responsible for the error. As a result, patients and families often learn little about tragic events such as the error in this case.

Over the last decade, considerable advances have arisen around error disclosure. First, important standards and guidelines have been developed by organizations such as the National Quality Forum and the Institute for Healthcare Improvement.(2,3) Second, research has explored patients' preferences for disclosure; results emphasize that patients want to learn about harmful errors that have occurred in their care.(4,5) Last, emerging evidence shows that a combination of disclosure programs and early offers of monetary compensation can help efficiently and productively resolve challenging cases involving error.(6,7)

Despite these critical advancements, studies also document a substantial gap between patients' desire for disclosure and our ability as clinicians to meet these expectations.(8) This gap reflects major unanswered questions regarding how best to communicate with patients about harmful errors.

The rationale favoring disclosure is strong.(9) Patients clearly want harmful errors disclosed. Furthermore, disclosure of medical errors to patients has several benefits. Disclosure allows patients to make informed health care decisions. Of course, the fact that the patient has died does not mean that the information disclosed is no longer relevant to decision-making around the event. This family may seek compensation, and the information that the child's death was potentially preventable would be important. Disclosure is also a significant component of truth telling and demonstrates respect for patient autonomy. While some have argued that on occasion limited disclosure can be in the patients' best interest by decreasing their anxiety about an error or preserving their trust in the competence of their clinicians, such "benevolent deception represents a significant breach in our professional obligation to be truthful with patients and their families."(10,11)

While the disclosure process is complex and needs to be customized to the needs of each patient, several key principles can serve as guideposts to help clinicians plan these challenging conversations (Table).

In this case, the health care workers at the tertiary care center remained silent when faced with the opportunity to disclose. But these parents would certainly want an explanation. This case highlights two important issues, namely how to talk with a patient or family when the error occurred elsewhere, and when there is uncertainty about what happened.

This case is complicated by the fact that the disclosing physicians are not the physicians who made the error. While research on disclosure has proliferated, little empirical information addresses health care workers' attitudes about talking with patients regarding other health care workers' errors. However, there is evidence that confirms the reluctance of clinicians to speak critically of other providers. In a recent survey of 1900 US physicians, 17% had direct personal knowledge of a physician colleague who is incompetent to practice medicine, and only 67% had reported this colleague to the authorities.(11) Reticence to impugn care delivered by another health care worker seems to have longstanding roots. The 1927 edition of the Mayo Clinic Desk Book notes that, "Great care must be used that the opinions of other men in the clinic or of home physicians are not belittled by word or look. A word dropped in conversation may upset entirely the confidence in the opinion given by another consultant."(12) Dr. William J. Mayo, in a 1930 memorandum, further emphasized the delicateness of these conversations: "If you want to call a man a liar and a thief, do not write him, go to see him and tell him in person, but be sure that you can run faster than he can."(13) The concept of medicine as self-regulating is central to our notions of professionalism. However, we are socialized from an early age not to "tattle," and we are reluctant to openly criticize or police our colleagues.

Considerations of power and self-interest also complicate these situations. Especially when the person considering disclosing an error is subordinate to the physician who made the error, fears of retaliation can ensue. A provider may also choose limited disclosure, hoping that our colleagues who might be involved in a future case where we made an error would show similar discretion. Lastly, disclosing another health care worker's error may precipitate a malpractice suit, which is especially worrisome given physicians' concerns about the fairness of our tort system.(14) In this case, fear of malpractice lawsuit may have motivated the risk managers at the tertiary care hospital to specifically recommend avoiding disclosure of potential errors at the other hospital.

The second challenging piece of this case is the various unknowns. As in many complex medical cases, this case has numerous dimensions of uncertainty, making it difficult to know whether an error occurred and whether that error harmed the patient. Did the initial radiologists miss cerebral edema that any competent radiologist would have noticed, or was this edema only visible in retrospect, or by the more specialized experts at the referral site? Would earlier diagnosis have improved this child's outcome? What information was available to the initial clinicians at the time specific decisions were made?

The presence of multiple unknowns may lead the disclosing physicians to speculate about the care provided at the referring hospital. When communicating with the parents, they may imply that an error was made and that they would not have made a similar mistake. Speculation of this nature can have significant downsides. If the parents conclude that the death might have been preventable, this doubt will be with them forever, regardless of what a formal review determines. Even phrasing that sounds innocuous, such as, "You might check with the first hospital about their final reading of the CT scan" could be interpreted by the parents as a definitive statement that care was substandard.

The challenges presented in this case reflect an outdated culture surrounding disclosure. Disclosure has traditionally been seen as a conversation between an individual doctor and their patient, discussing an error for which that same doctor was responsible.(15) However, we make errors as health care teams and therefore, we should consider how to disclose as a team. Furthermore, health care is often delivered by several institutions, so "team" should be envisioned even more broadly to include all the providers who have interacted with the patient.

Therefore, in cases where a potential error occurred under other providers, the critical first step is for the accepting physicians to speak directly with the referring care providers; collectively, they are all part of the same health care team for that patient. These discussions require careful planning with quality and risk management specialists at both hospitals. While conversations with other physicians about potential problems with the care provided can be very difficult, a thoughtful discussion with the physicians at the referring hospital is critical to starting a root cause analysis to determine whether errors took place.

Ideally, what would follow would be a formal process between the two involved organizations around event investigation and disclosure. Such a collaborative approach would enhance the root cause analysis process, as the addition of multiple perspectives often uncovers new and important information. In addition, if it was determined that an error occurred, a collaborative approach to disclosure, in which both the referring and accepting physicians talked together with the parents about what happened, would reduce miscommunications, especially those regarding whether the child's death was preventable, that can easily result from separate discussions.

Of course, there are significant barriers to this collaborative approach. First, there are the divergent financial self-interests of the malpractice insurers who would be involved. In addition, physicians at the referring hospital might worry that their presence in a disclosure conversation with this family could impugn the quality of care that they had delivered. Careful planning of these conversations can minimize these concerns. Several ongoing projects, funded by the Agency for Healthcare Research and Quality (AHRQ) Patient Safety and Medical Liability Reform grant portfolio, are testing such collaborative approaches.

In the most straightforward scenario, in which there was a consensus that the initial radiology reading was in error and that chances of improved patient outcome would have been much higher if the delay in diagnosis had not taken place, the physicians at the initial hospital would take the lead on talking with parents about what happened. Patients clearly want to hear about harmful errors from the involved provider.(5,16) Including the physicians at the initial hospital also allows those physicians to take responsibility and apologize, promoting psychological healing for both parties.

The situation becomes more complicated if all agreed that there was a clear-cut harmful error, but the doctors at the referring hospital chose not to disclose to the parents. In such a situation, should the accepting physicians share their beliefs directly with the family about the care problems at the referring hospital?

There is not yet consensus on what situations merit mandatory disclosure of another health care worker's error. One can imagine clinical situations in which the rationale for mandatory disclosure of another health care worker's error would be high, such as when the information would allow the patient to avoid significant future harm. However, for situations such as this case, no clear standard yet exists. One important practical realization emerging from ongoing work in disclosure is that leaving disclosure decisions completely to the discretion of the involved clinicians may be problematic. Strong psychological pressures accompanying these events (not wanting to admit a mistake) can bias the decision-making of the most thoughtful clinicians, leading to erroneous conclusions about whether and how such events should be disclosed.

Ideally, organizations would have a neutral third party (such as an ethics committee) consider cases involving disagreements between providers about whether and how disclosure should occur and make binding recommendations. Such recommendations recognize that disclosure is fundamentally an institutional responsibility, not the sole domain of the attending physician. Taking an institutional approach to complex cases involving communicating with patients about another health care worker's error can help ensure that these challenging but important conversations happen more frequently and in a way that meets patient and family needs.

In the case discussed, the two hospitals should have had an open dialogue about the case. If they determined that a clear error occurred, providers should have found a way to disclose the error openly and honestly to the parents. This outcome would have been ethical, collaborative, and patient centered.

Take-Home Points

• Although patients strongly favor hearing about medical errors in their care, providers often do not disclose.
• Clear disclosure of medical errors is ethical, respects patient autonomy, and may allow for better informed decision-making.
• If other providers have committed an error, optimal strategies involve full collaboration in error investigation and joint disclosure if an error did occur.
• Hospitals should consider instituting disclosure policies and utilizing a neutral third party such as an ethics committee to mediate the most challenging disclosure cases.

Thomas H. Gallagher, MD
Associate Professor, Departments of Medicine and Bioethics & Humanities

University of Washington

Faculty Disclosure: Dr. Gallagher is supported by grants from the Greenwall Foundation AHRQ (RO1HS16506, R18HS19531). Thanks to Lauren Lipira, MSW, for assistance with manuscript preparation.

1. Gallagher TH. A 62-year-old woman with skin cancer who experienced wrong-site surgery: review of medical error. JAMA. 2009;302:669-677. [go to PubMed]

2. National Quality Forum. Safe Practices for Better Healthcare-2009 Update. Washington, DC: National Quality Forum; 2009. [Available at]

3. Conway J, Federico F, Stewart K, Campbell MJ. Respectful Management of Serious Clinical Adverse Events. Cambridge, MA: Institute for Healthcare Improvement; 2010. [Available at]

4. Mazor KM, Simon SR, Gurwitz JH. Communicating with patients about medical errors: a review of the literature. Arch Intern Med. 2004;164:1690-1697. [go to PubMed]

5. Gallagher TH, Waterman AD, Ebers AG, Fraser VJ, Levinson W. Patients' and physicians' attitudes regarding the disclosure of medical errors. JAMA. 2003;289:1001-1007. [go to PubMed]

6. Mello MM, Gallagher TH. Malpractice reform—opportunities for leadership by health care institutions and liability insurers. N Engl J Med. 2010;362:1353-1356. [go to PubMed]

7. Kachalia A, Kaufman SR, Boothman R, et al. Liability claims and costs before and after implementation of a medical error disclosure program. Ann Intern Med. 2010;153:213-221. [go to PubMed]

8. Gallagher TH, Studdert D, Levinson W. Disclosing harmful medical errors to patients. N Engl J Med. 2007;356:2713-2719. [go to PubMed]

9. Lo B. Resolving Ethical Dilemmas: A Guide for Clinicians. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2005. ISBN: 9780781753579.

10. Gallagher TH, Bell SK, Smith KM, Mello MM, McDonald TB. Disclosing harmful medical errors to patients: tackling three tough cases. Chest. 2009;136:897-903. [go to PubMed]

11. DesRoches CM, Rao SR, Fromson JA, et al. Physicians' perceptions, preparedness for reporting, and experiences related to impaired and incompetent colleagues. JAMA. 2010;304:187-193. [go to PubMed]

12. Mayo Clinic Desk Book. Rochester, MN: Mayo Clinic; 1927.

13. Mayo WJ. People Files Collection. No. MHU-0675. Located at: Mayo Clinic Libraries, Historical Archives, Rochester, MN.

14. Carrier ER, Reschovsky JD, Mello MM, Mayrell RC, Katz D. Physicians' fears of malpractice lawsuits are not assuaged by tort reforms. Health Aff (Millwood). 2010;29:1585-1592. [go to PubMed]

15. Truog RD, Browning DM, Johnson JA, Gallagher TH. Talking with Patients and Families about Medical Error: A Guide for Education and Practice. Baltimore, MD: The Johns Hopkins University Press; 2010. ISBN: 0801898048.

16. Iedema R, Sorensen R, Manias E, et al. Patients' and family members' experiences of open disclosure following adverse events. Int J Qual Health Care. 2008;20:421-432. [go to PubMed]

Table. Principles in Preparing for Disclosure Conversations.