American Association for Clinical Chemistry
Better health through laboratory medicine
Patient Safety Focus: Current Concepts in the Disclosure of Serious Medical Errors to Patients

Patient Safety Logo

Current Concepts in the Disclosure of Serious Medical Errors to Patients
An Interview with Thomas Gallagher, MD

Thomas Gallagher, MD

Thomas Gallagher, MD, is an Associate Professor of Medicine and Medical History & Ethics in the University of Washington School of Medicine. An internist who cares for both inpatients and outpatients, Gallgher is also an internationally recognized expert in the field of error disclosure, having recently served on a consensus group that developed a national quality standard on the subject.

Michael Astion, MD, PhD, conducted this interview.

Q: What is your view on errors in medicine and the way they relate to disclosure?
Errors and adverse events are unavoidable because healthcare is a human enterprise. The patient safety movement supports transparency between patients and care providers regarding the disclosure of errors. Proper handling of error disclosures has the potential to enhance patient satisfaction, to help patients and caregivers develop mutual trust, and to decrease the chances that patients will sue their care providers.

Q: What is the attitude of patients regarding error disclosure?
Patients define errors broadly. They include many aspects of low quality in their definition of error: unnecessary waiting, poor bedside manner, and unpreventable complications of care. Patients want caregivers to disclose errors because they believe caregivers have the ethical obligation to be truthful. They want truthful error disclosure, but they also believe it is human nature for healthcare workers to hide or minimize errors.

Q: What do patients want when a harmful laboratory error occurs?
Patients want an explicit, comprehensible statement that an error occurred; a description of what happened, including the implications for their health; a description of why it happened and how future recurrences will be prevented; and for caregivers to say that they are sorry about the error.

Q: How do healthcare workers differ from patients regarding attitudes and experiences about disclosure?
Most of the healthcare worker data on error disclosure have been collected from physicians, nurses, and risk managers. These healthcare workers define errors more narrowly, often using medical definitions of error that emphasize undesirable patient outcome. Therefore, healthcare workers tend to emphasize the need to disclose errors to patients in those circumstances when errors harm patients or force patients to make unexpected decisions about their care. However, by focusing primarily on errors impacting patients, caregivers often feel it is justified to not inform patients if the error does not harm patients. Like patients, healthcare workers endorse the concept of error disclosure, and they want to be truthful. However, caregivers experience a variety of barriers that block them from making a full error disclosure and vary significantly regarding what they think should be revealed during a disclosure.

Q: Are most harmful errors disclosed to patients?
Currently, most harmful errors are not disclosed to patients. Estimates vary regarding the frequency of disclosure, but a reasonable estimate is that about 30% of harmful errors are disclosed. This appears to be the case internationally and across medical disciplines.

Q: What are some barriers to disclosure that create the gap between the desire to disclose and actual disclosure?
One group of barriers is ethical considerations. A question that frequently arises among healthcare workers is whether the disclosure will do more harm than good to the patient.

Q: Can you give some examples of when caregivers feel that disclosure might do more harm than good?
Consider a case where a patient is hopelessly ill and will die soon. Caregivers sometimes do not disclose a serious or fatal error in such cases because they feel it will only compound the agony of family and friends of the patient. Or consider a case when an error is minor and the patient experiences minimal or no harm. Some providers will not disclose in this type of case, because they feel the error disclosure will demoralize the patient and perhaps make the patient feel worse.

Q: What are some of the other barriers to disclosure mentioned by care providers?
Caregivers are worried that the disclosure could precipitate a lawsuit and financial damages (See Box below). They are concerned that error disclosure may cause them to be emotionally harmed due to loss of reputation and acknowledgement of personal failure. They report being uncomfortable reporting errors made by other care providers, some of whom are close colleagues or superiors. Last, care providers find these conversations awkward and feel they do not have the training or communication skills to provide error disclosure correctly.

Barriers that Hinder Error Disclosure by Healthcare Workers

  • Fear of harming the patient or the patient’s friends or family
  • Fear of litigation
  • Fear of financial or emotional damage (loss of reputation)
  • Awkwardness of the disclosure conversation
  • Lack of confidence regarding communication skills

Q: Is the fear of litigation and accompanying financial damage justified?
This question is not decided and will not be decided for a long time. Early studies from a Veteran’s Affairs hospital, the University of Michigan, and a malpractice insurer in Colorado suggest that, overall, there will be less litigation as error disclosure practices spread. These results are limited, however, and it is too early to draw a strong conclusion.

Q: What are the implications of the barriers to disclosure?
There are two major implications. The first is that the barriers decrease the frequency of disclosure. The second is that when caregivers choose to disclose, they choose their words too carefully.

Q: Can you explain what you mean by “choosing their words too carefully”?
One scenario that we have used in research is to ask physicians what they would do if their bad handwriting on an insulin order caused an insulin overdose that seriously harmed the patient. In this example, the patient is found unresponsive with critically low glucose. The patient is transferred to intensive care and then fully recovers. More than 65% of the physicians said they would disclose this error. If they were disclosing this error, more than 70% of the physicians would describe what happened as an error, saying something like: “Your blood sugar went too low because an error happened and you received too much insulin.” The remaining physicians would use a more generic description, saying, “Your blood sugar went too low because you got more insulin than you needed.”

Examples of Disclosure of a Laboratory Error by a Laboratory Technologist to a Nurse

The error was a data entry on a troponin, which caused an incorrect diagnosis of myocardial infarction. The request for read back has been omitted from the disclosure.

Choosing words too carefully: “I am calling to correct a troponin result. The troponin result on patient John Doe from March 14 at 14:52, which was reported as 57 ng/mL, has been changed to 0.02 ng/mL.”

Reasonable disclosure: “I am calling to inform you about a laboratory error. The troponin result on patient John Doe from March 14 at 14:52, which was reported as 57 ng/mL, has been changed to 0.02 ng/mL. This was due to a manual, data-entry error. In the laboratory, there was a specimen for troponin from another patient at the same time as John Doe’s specimen. That patient had a troponin of 57 ng/mL, and we incorrectly entered that patient’s results into John Doe’s record. We are sorry that we made this error. We are doing a further analysis on this error to look for ways to prevent its recurrence. We will contact you about this further analysis in the next 48 hours. Please feel free to call the laboratory supervisor if you have questions.”

Disclosing too much: “I am calling to inform you about a laboratory error. The troponin result on patient John Doe from March 14 at 14:52, which was reported as 57 ng/mL, has been changed to 0.02 ng/mL. This was due to a manual, data-entry error by a technologist named Joey. There was a specimen for troponin from another patient in the laboratory at the same time as John Doe’s specimen. That patient had a troponin of 258 ng/mL, and Joey got all confused as usual and incorrectly entered the patient’s results into John Doe’s record. We are sorry that Joey made this error. Joey has been having personal problems. He is in the middle of a messy relationship and taking care of his elderly parents. He has been making lots of these errors, and everybody has been talking about it. But management around here is too timid to do anything, and they are never around anyway, as they are usually taking some kind of fancy retreat or driving around in their boats. Nobody listens to us, and that is why these things happen. It is a good thing we didn’t kill the patient. This place stinks. If I had more money, I would retire. We are doing a further analysis, even though we all know what is going on. Please feel free to call the laboratory supervisor in a few days if you have additional questions.”

Q: Would they mention the handwriting? 
This is the interesting part. When given choices regarding what they would specifically say about the error, only one-third of physicians would specifically mention the cause of the error—in this case, their bad handwriting. Rather, physicians favored saying, “This occurred because of a miscommunication in your insulin order.” They would only volunteer more information if the patient asked for clarifying information.

Q: Are you saying there is a gap between what patients want and what physicians are giving them regarding the description of the error?
Yes. Patients want to know what specifically happened and how it will be prevented in the future. If you do not come out and say that a specific error occurred, you cannot meet the patient’s desire to know why it occurred and how recurrences will be prevented. In our studies, the approach favored by most physicians was to provide a nugget of information but not necessarily to share the information that patients have said they would like to hear, unless the patient asks clarifying questions. The patients want more than that nugget.

Q: Are patients receiving the apologies they desire?
Physicians are split regarding their approach to apology. For the bad handwriting scenario, we gave physicians three choices regarding how they would apologize. The first choice was no apology, the second was an expression of regret stating “I am sorry that this happened,” and the third was an apology stating, “I am so sorry you were harmed by this error.” Patients want the apology, but physicians split 50-50 between the expression of regret and the apology.

Q: Besides choosing their words too carefully, what are the other errors made in error disclosure?
Occasionally, we see too much disclosure, which is the opposite behavior. The caregivers, in their desire to be truthful and reduce some of the emotional burden of the error, hastily give the patient more information than the patients desire. Many times, this information is not accurate, since an accurate description of an error requires the time to collect and analyze data.

Q: What are some of the important developments occurring in the area of error disclosure?
Perhaps the most important development is the National Quality Forum’s (NQF) addition of standards for disclosure to its list of safe practices. I was part of the committee that developed the standard.

Q: Could you describe NQF’s standards for safe practices and why they are influential?
The NQF endorses a set of safe practices that are considered fundamental to quality care. Currently there are 32 such practices. These practices are based on evidence as well as expert opinion. The standards were developed by individual stakeholders working in collaboration with representatives from organizations interested in healthcare quality such as the Joint Commission, CMS, and the Agency for Healthcare Research and Quality.

Q: What are some of the key points of the NQF’s standard?
The standard contains the basic characteristics of an effective system for medical error disclosure. There are two key areas addressed by the safe practice list. The first is what information should be given to the patient. The second set of issues addressed by the safe practice list is the institutional support of disclosure so that care providers can be supported while meeting the needs of patients.

Elements of an Effective Disclosure Support System

  • Education to care providers and other healthcare workers
  • 24-hour availability of coaching /mentoring
  • Emotional support available to healthcare workers, patients, families of patients

Q: What are the patient issues?
The standard supports the list of patient desires discussed previously. It supports giving the patient facts about the error, especially those facts that support decision making. In addition, the standard states that formal apologies, rather than expressions of regret, should be given when a clear-cut error or system failure occurs.

Q: What are the institutional requirements to support disclosure?
One of the main institutional requirements is to give disclosure education to all healthcare workers who might participate in an error disclosure. Another requirement is to make sure that help, in the form of mentoring or coaching, is available to healthcare workers. The last recommendation is to provide emotional support to everybody involved in the error and the disclosure. This means healthcare workers, patients, and the families of patients.

Q: Who acts as the coach?
Coaches should be experienced people, and they are often in positions of authority. One frequently used coach is the medical director of the practice or the hospital. The chief nursing officer and risk managers can also be helpful coaches. No matter who is doing the coaching, the communication between the care provider, who is knowledgeable about the error and who may have played a role in the error, and the coach needs to be protected.

Q: What do you mean by protected communication?
It means that the communication cannot be legally construed as an admission of guilt regarding an error.

Q: By emotional support, do you mean employee counseling by a therapist?
That could be part of it in some unusual cases. More frequently, the basis of the emotional support is a little less formal. Specifically, I am referring to support by the medical director’s office, by departmental leadership, by supervisors, and by peers.

Q: Why does the NQF safe practice standard matter?
Hospital scores on NQF patient safety practices are part of a number of pay-for-performance initiatives and show up on publicly accessible Web sites that grade hospital performance. Although it will not be possible to grade hospitals on individual disclosures, it will be possible to grade them on whether they have the elements of an effective disclosure system in place, and that is a great start.

Q: Are you saying that organizations tend to behave better when they are being monitored?
Yes. The hope for error disclosure is that the spread of the NQF patient safety practice and posting of hospital scores regarding carrying out the practice will lead to strong error disclosure programs in every hospital.

Q: What are some other key advances in disclosure?
The other important developments are legal and are mostly taking place at the state level. Seven states have now mandated that serious errors be disclosed to patients, and many other states are developing similar legislation. To address legitimate concerns about legal actions caused by error disclosure, the laws usually include language stating that these disclosures cannot be used as evidence of guilt regarding the event described in the disclosure.

Q: Why are the legal developments important?
The most important aspect of the legal developments is that they represent a public policy endorsement of error disclosure. Along with the NQF safe practice standards, this endorsement provides a strong and probably irreversible movement toward effective error disclosure.

Q: Overall, you seem optimistic about error disclosure?
I told you that currently about 30% of errors that harm patients are disclosed. I am optimistic that transparency regarding errors including effective error disclosure, which meets the needs of both patients and care providers, will become the norm over the next several years.

Q: In most cases of laboratory errors, laboratorians do not disclose errors to the patient. Typically, the laboratorian discloses the error to a nurse or physician, who then makes a choice regarding disclosing the error to the patient. What are your recommendations regarding disclosure of laboratory errors by laboratorians to direct care providers?
The NQF safe practice standards describe an effective system for error disclosure for all healthcare workers. Therefore, the safe practice should be extrapolated to the situation you describe. To meet the patient’s needs, the laboratorian must give the caregiver the information needed to provide an optimal disclosure to the patient. This means speaking in comprehensible language and avoiding the most common problem, which is choosing words too carefully. Laboratory staff should also avoid the less common problem of hastily, emotionally, and inaccurately disclosing too much.

Q: When talking with physicians and nurses about laboratory errors, can laboratory staff talk about QC intervals, Westgard rule violations, out-of-range high instrument flags, universal aliquotting, Levy-Jennings charts, instrument linearity, and delta-checking?
Laboratory workers should not assume that physicians and nurses understand laboratory jargon.

Q: What about institutional support to laboratory workers and other workers in ancillary services?
The basic elements of a disclosure system, which are education, support by coaches, and emotional support, need to be available to laboratory staff, supervisors, and medical directors. When disclosing laboratory errors to nurses and physicians, laboratory staff faces some of the barriers to disclosure experienced by direct care providers when they disclose errors to patients. Laboratory staff may fear litigation, job loss, or emotional damage through loss of reputation, and they may find the disclosure conversation awkward. Institutional support is critical for overcoming the barriers that block error disclosures between healthcare professionals.

Q: Thanks for an informative interview on a topic not frequently addressed in clinical laboratory publications. You provided some great information here and during your presentation at AACC’s 2007 Annual Meeting.
It is my pleasure to work with the AACC on issues related to error disclosure, since communication about errors is significant for clinical laboratorians.

More on This Topic

A recent lecture by Dr. Gallagher, "Disclosing harmful medical errors to patients: what are the data telling us?" can be viewed at the University of Washington School of Medicine University of Washington School of Medicine Web site. Lecture date: November 7, 2007.

Gallagher TH, Studdert D, Levinson W. Disclosing harmful medical errors to patients. N Engl J Med 2007;356:2713–9.

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.

Gallagher TH, Waterman AD, Garbutt JM et al. US and Canadian physicians’ attitudes and experiences regarding disclosing errors to patients. Arch Intern Med 2006; 166:1605–1611.

Safe practices for better healthcare. Washington, DC: National Quality Forum, 2007. NQF Web site.

Patient Safety Focus Editorial Board

Michael Astion, MD, PhD
Department of Laboratory Medicine
University of Washington, Seattle

Peggy A. Ahlin, BS, MT(ASCP)
ARUP Laboratories
Salt Lake City, Utah 
James S. Hernandez, MD, MS
Mayo Clinic College of Medicine
Rochester, Minn.
Devery Howerton, PhD
Centers for Disease Control and Prevention
Atlanta, Ga.

Sponsored by ARUP Laboratories, Inc.