Skip to Main Content
Follow us

2.1.3 Disclosure of Adverse Events

Maggie Constantine, MD, FRCPC; and Irene Sadek, MD, FRCPC

Educational Objectives

  1. To understand the ethical issues related to disclosure of medical information
  2. To identify the potential benefits and harms of disclosure of medical information
  3. To examine potential methods of disclosure of medical information


As the resident member of your hospital's ethics committee, you have been asked to advise the Director of Blood Transfusion, who has requested advice on the following situation. Mrs. S., a 45-year-old woman, is currently an in-patient receiving chemotherapy for the treatment of a malignancy. She has consented to transfusion support with blood products as part of her treatment and has thus far received several units of blood products.

The implementation of West Nile Virus (WNV) nucleic acid testing of donated blood units commenced in July 2003 and was undertaken by your local blood manufacturer. The purpose of this testing is to prevent the transmission of WNV infection through the blood supply. Once a donor tests positive for WNV, he or she is notified of the positive test result, and the corresponding products are quarantined and removed from supply or, if already distributed to a hospital, are recalled. As this testing was being implemented, in areas with high levels of activity of WNV, the blood manufacturer quarantined and recalled plasma products from your hospital to prevent the potential transmission of WNV through these untested products.

Upon being notified of this recall of blood products, you initiate the process to locate these units. Unfortunately, only three of the eight blood products could be returned to your blood manufacturer for destruction. The other five units were transfused to Mrs. S. Although Mrs. S. remains immunosuppressed from chemotherapy, she is currently clinically stable and well.

The majority of persons infected with WNV are asymptomatic, with symptoms occurring in approximately 20% of those infected. The most common manifestation is an acute febrile illness with symptoms generally lasting three to 10 days. A much less common manifestation, occurring in less than 1% of the cases, is neurologically invasive disease. This is characterized by encephalitis, muscle weakness, flaccid paralysis and other neurologic abnormalities. The rates of this complication are higher in immunosuppressed patients. Currently, treatment remains largely supportive.


  1. Is there an ethical obligation to notify Mrs. S. of the potential, although unconfirmed, risk of WNV infection in the blood products she has received?
  2. What potential benefits and harms may arise out of the disclosure of this information to Mrs. S.?
  3. Is there an ethical justification for withholding this information from Mrs. S.?
  4. If this information is to be disclosed to Mrs. S., what should be discussed with her?
  5. If this information is not to be disclosed to Mrs. S., what alternatives are available?
  6. What is the legal concept of material risk, and how does it apply to this case?


Q1. Is there an ethical obligation to notify Mrs. S. of the potential, although unconfirmed, risk of WNV infection in the blood products she has received?

During the later part of the 20th century, a paradigm shift began with respect to the patient-physician relationship, from one of paternalism to one of increasing respect for the autonomy of the patient. Along with this increasing respect for the freedom of patients has been the increasing acknowledgement of the expanding role of patients in the decision-making process regarding their health care. This cooperative partnership, rather than a paternalistic relationship, between patient and physician has increased the need of physicians to fully discuss the aspects of diagnosis, treatment options and prognosis required for the patient to make informed decisions.

This element of full disclosure of information is not entrenched in the traditional codes of ethics governing physician behaviour. The Hippocratic Oath does not mention veracity, nor does the Declaration of Geneva of the World Medical Association or the American Medical Association in its "Principles of Medical Ethics."1 In the Canadian Medical Association's "CMA Code of Ethics," article 21 states: "Provide your patients with the information they need to make informed decisions about their medical care, and answer their questions to the best of your ability."2 As with other codes of medical ethics, veracity is hinted at, but not explicitly stated, nor is it stated elsewhere in this code of ethics.

What is full disclosure, and is there an obligation for full disclosure between physician and patient? Disclosure within the patient-physician relationship encompasses the communication of information to patients that is comprehensive and accurate. This communication should enhance the understanding of the patient with respect to his or her diagnosis, prognosis and therapeutic options. According to Beauchamp and Childress, three arguments contribute to the justification of an obligation to veracity:3

  1. Respect is owed to others.
  2. Obligation of veracity is closely connected to the obligations of fidelity and promise keeping.
  3. Relationships between physicians and patients are ultimately dependent on trust, and adherence to veracity is essential to foster trust.

Therefore, this need for disclosure of information is not only important to the process of informed consent for the patient, but also fundamental to the ongoing relationship between physician and patient. As stated above, the relationship between patient and physician is partly based upon a commitment of trust and openness on the part of the physician. With regard to medical error or adverse events, where the non-disclosure of such events can lead to a breakdown of the patient-physician relationship, this need for disclosure is particularly important. Suspicions of cover-up and of being lied to may result, which has been shown to increase the rates of malpractice cases.

Overall, there is an obligation to disclose all relevant information to patients to ensure their ability to make informed decisions with regard to their health care and also to foster an open patient-physician relationship.

Q2. What potential benefits and harms may arise out of the disclosure of this information to Mrs. S.?

As stated above, veracity is an obligation that physicians have to their patients, but it is not absolute. Although this case clearly does not involve medical error or an adverse event, it does involve the ethical decision of whether or not to disclose information to Mrs. S. about her possible contact with WNV. Analogous to this situation may be one in which a patient has had testing to investigate a possible malignancy. Suppose that the test results are indeterminate and the physician has already organized additional testing to clarify the situation. Should the patient be told of the indeterminate results of the initial testing, or can this information be withheld from the patient until the final testing results are available?

Potential benefits from discussing the test results with the patient may include a better understanding for the patient in terms of testing procedures and alleviation of the anxiety related to waiting for the initial test results. The ongoing communication may further foster a sense of trust in the physician. Keeping the patient informed of the test results may emphasize to the patient the necessity to undergo further testing. As for this case, one might also argue that, although the risk of Mrs. S.'s exposure to WNV is unknown, she should nevertheless be informed of this risk. If she is not informed and should she develop symptoms of WNV, particularly of the more serious neuroinvasive type, she may perceive the withholding of this information as a cover-up, and this may result in the breakdown of the patient-physician relationship.

On the other hand, one might argue that an important potential harm arising from disclosing an indeterminate test result is that this information would only serve to cause undue anxiety for the patient. One major difference between the analogy of the indeterminate test result and this case of possible WNV exposure is that further testing is not possible in this case. The information about the risk of WNV is as ambiguous, if not even more so, than an indeterminate test result. In this case, the risk of WNV exposure cannot be determined with any certainty, and one might argue that disclosing such vague information to the patient would unburden the physician and would only cause the patient undue anxiety.

Q3. Is there an ethical justification for withholding this information from Mrs. S.?

The word autonomy, from the Greek autos ("self") and nomos("rule"), was initially used in reference to state governance and has since been adapted and extended to the individual. This personal autonomy refers to self-rule free from the interference of others and to the freedom to make meaningful choices. Therefore, the respect needed for an autonomous patient involves respecting a patient's right to make choices and to act on his or her personal values and beliefs. This respect for the autonomy of patients ensures that, as physicians, we do not treat patients as a means to our own ends.

As with disclosure (see above), autonomy is not absolute, and there may be instances in which one or both may not be upheld. Paternalism, as a model for the patient-physician relationship, is most often cited as an example as a model in which conflicts arise from the physician's desire for beneficence and the patient's need for autonomy. The paternalistic model is often compared to the relationship between incompetent children and their need for beneficent parental guidance. An example of this is the incompetent psychiatric patient who needs involuntary medical care due to ongoing self-harm. There are grades of paternalism, varying from the weak form, in which interventions are sought to protect patients from their non-autonomous actions, to the strong form, in which action is taken for patients who remain autonomous. Arguments for limited disclosure or non-disclosure of information may come under a paternalistic model and, depending on the state of the patient, may be viewed as either weak or strong paternalism. Although these lapses in veracity are generally seen as wrong, they may be viewed as justified "benevolent deception."

Therefore, one may argue that telling Mrs. S. about her possible exposure to WNV may cause her undue worry and anxiety. This is a reason often cited by family and physicians when withholding information from patients. Within the cancer care literature, the withholding of diagnostic information from patients based upon the wishes of family members or from a paternalistic role of the physician has largely been refuted. Survey work has shown that patients do in fact want to be informed, and although there is increased anxiety after disclosure, the anxiety is thought to be generally short-lived and the benefits of an open physician-patient relationship are apparent. Of course, the disclosure of information must respect the wishes of the patient, and if the wish has been expressed beforehand that certain types of information not be disclosed to the patient, then this wish must be respected.

But, unlike a diagnosis of cancer, the certainty of Mrs. S.'s exposure to WNV is unclear, and so one might also argue that, given the large degree of uncertainty, the disclosure of this unknown risk may be an "unburdening" of information upon the patient with little perceived benefit. The lack of perceived benefit may be further argued on the basis that, even if Mrs. S. should show symptoms of WNV infection � even of the more serious neuroinvasive type � treatment remains largely supportive and there is no available prophylactic treatment.

Q4. If this information is to be disclosed to Mrs. S., what should be discussed with her?

If one chooses to disclose the possible exposure to WNV to Mrs. S., a frank discussion about the lack of certainty of her exposure must be included along with a discussion of the possible symptoms of WNV infection. Information should be provided in terms that she can understand. Mrs. S. must be reassured that she would be monitored for possible symptoms on an ongoing basis. "Truth dumping," that is, the unburdening of information on the patient, must be avoided.

Q5. If this information is not to be disclosed to Mrs. S., what alternatives are available?

If one chooses not to disclose the possible exposure to WNV to Mrs. S., a reasonable alternative may be to monitor Mrs. S. closely for symptoms of WNV infection and, should they occur, to treat her supportively through the infection.

Q6. What is the legal concept of material risk, and how does it apply to this case?

The following is a brief discussion, and a more extensive discussion can be found in the appropriate legal cases and texts.4

The legal implication of disclosure of information has been extensively explored in the area of consent. Key to this discussion is how a physician determines which risks need to be disclosed to a patient prior to obtaining consent for a procedure or treatment. A risk is determined to be "material" to the patient's health if it is a serious risk such as paralysis or death, even if the chance of the risk is so low that it is seen as a mere possibility. This definition of risk was further explored in White v. Turner (1981), 120 D.L.R. (3d) 269, at 284-85, in which the judge stated: "Material risks are "significant risks that pose a real threat to the patients' life, health or comfort. . . . Unusual or special risks are somewhat extraordinary, uncommon and not encountered every day, but they are known to occur occasionally... must be disclosed to a reasonable patient, whether or not they are material." Although physicians play a key role in the determinant of risk as material or not, this responsibility ultimately falls to the trier of fact if a case is taken to court. As to Mrs. S., one could argue that the risk of WNV exposure for her is a material risk to her health because it carries with it a potentially serious complication such as paralysis. But the assessment of risk as material may not necessitate its disclosure to a patient as part of the duty of disclosure because an emotional assessment of the patient by the physician is also central to this discussion. How much information is it appropriate to disclose to this particular patient given his or her unique emotional state? Depending on the emotional state of the patient, a physician may be justified in the withholding or generalization of risk discussion. Known as the concept of therapeutic privilege, this is a legally contentious concept and is similar to the ethical concept of "benevolent deception" (see above). Ultimately, each case must be treated individually with a full assessment of the risk and the emotional status of the patient.


In conclusion, there is an ethical obligation on the part of the physician to maintain an open relationship with the patient, one that is based on honesty and trust. The respect for patient autonomy is important to ensure that patients are able to make informed decisions about their health care. The ethical obligation to disclose information that may have an unclear and uncertain impact on the patient is not always clear, and disclosure of such information should not be done in a way to unburden the physician. Potential benefits and harms of disclosure should be carefully considered.


  1. American Medical Association
  2. Canadian Medical Association. CMA code of ethics (update 2004). Ottawa: Canadian Medical Association; 2004. Available from:
  3. Beauchamp TL, Childress JF. Principles of biomedical ethics. 5th edition. New York: Oxford University Press; 2001.
  4. Picard EI. Legal liability of doctors and hospitals in Canada. 2nd edition. Toronto: Carswell Legal Publications; 1984.


  • Cantor MD. Telling patients the truth: a systems approach to disclosing adverse events. Quality and Safety in Health Care 2002; 11(1): 7–8.
  • Disclosure and performance: expecting more, and better. CMAJ: Canadian Medical Association Journal 2005; 173(5): 457.
  • Gallagher TH, Waterman AD, Ebers AG, Fraser VJ, Levinson W. Patients' and physicians' attitudes regarding the disclosure of medical errors . JAMA: Journal of the American Medical Association 2003; 289(8): 1001–7.
  • Gold M. Is honesty always the best policy? Ethical aspects of truth telling. Internal Medicine Journal 2004; 34: 578–80.
  • Hebert PC, Levin AV, Robertson G. Bioethics for clinicians: 23. Disclosure of medical error. CMAJ: Canadian Medical Association Journal 2001; 164(4): 509–13.
  • Hingorani M, Wong T, Vafidis G. Patients' and doctors' attitudes to amount of information given after unintended injury during treatment: cross sectional, questionnaire survey. British Medical Journal 1999; 318: 640–1.
  • Hobgood C, Peck CR, Gilbert B, Chappell K. Medical error — what and when: what do patients want to know? Academic Emergency Medicine 2002; 9: 1156–61.
  • Kagan AR. Statistics is not a surrogate for judgement. American Journal of Clinical Oncology 2005; 28: 327–8.
  • Kraman SS, Hamm G. Risk management: extreme honesty may be the best policy. Annals of Internal Medicine 1999; 131(12): 963–7.
  • Meykens FJ, Hietanen P, Tannock IF. Talking to a patient. Journal of Clinical Orthodontics: JCO 2005; 23(19): 4463–4.