The physician who knew too much: a comment on Watters v. White.
With the sequencing of the human genome in 2003, the role of medical genetics in health care continues to expand. While allowing for new opportunities, especially in personalised medicine, this developing field also creates uncertainty about the duties physicians owe their patients. An area of particular controversy is the scope of the physician's duty to inform relatives of a patient of potentially important genetic information. The 2010 Quebec Superior Court decision Liss v Watters (1) dealt with such an issue. The central allegation in this case was that two physicians failed to inform the mother and relatives of a patient affected by a genetic condition of their carrier status. The court was presided over by Judge Riordan, who attempted to outline the scope and limitations of a physician's duty to inform in the realm of medical genetics. Essentially, Judge Riordan expanded the scope of the duty to inform to include close relatives of patients with heritable genetic conditions. In doing so, he found two physicians liable for the wrongful birth of a child with genetic defects. One physician appealed in Watters v White (2), the subject of this commentary. In this case, the Quebec Court of Appeal unanimously overturned the trial judgment, concluding that the duty to inform could not be expanded without undermining the physician's duty to uphold professional secrecy. Leave to appeal to the Supreme Court was denied.
While the alleged faults in this case took place decades ago, Watters v White poses key questions about physician duties to communicate genetic information. First and foremost, does a physician have a duty to inform patients' relatives of serious genetic risks? If so, what is the scope of this duty? What sorts of genetic risks must be communicated? Which relatives should be informed? Does physician-patient confidentiality bar such communications? If so, must physicians seek alternative means to ensure that at-risk individuals are informed? Important issues of causation also arise in Watters v White. Here, a pernicious genetic risk afflicts a large extended family across several generations. The lacuna between fault (a failure to inform) and injury (the wrongful birth of an affected child) spans nearly 30 years. This case commentary begins by outlining the facts of this case (Part I). Next, the reasoning and conclusions of both the trial and Court of Appeal judgments are discussed (Part II and III). The commentary concludes with a discussion of the implications of this case for physicians in today's rapidly evolving genetic milieu (Part IV).
The facts of this case are as tragic as they are complex. They are summarized in the timeline in Figure 1, and the pedigree in Figure 2. The genetic disease in question is Pelizaeus-Merzbacher Disease ("PMD"), a serious, sex-linked genetic disorder. In PMD, the myelin sheath that protects cells of the central nervous system does not form properly. (3) This leads to a host of symptoms including impaired sight, respiratory and digestive problems, spastic quadriplegia, and premature death. (4) The inheritance pattern of this condition is of central importance to these two judgments. Females are carriers of the condition. All the daughters of a carrier female have a 50% risk of becoming carriers--meaning they will not display any symptoms of the disease. A carrier's male children, however, have a 50% risk of being affected (5) (see Figure 2).
The plaintiffs in the court of first instance were Audrey Liss, along with her cousin's daughter Marcy White--both carriers of PMD. In 1970, Audrey Liss and her husband Frank Carin gave birth to a son affected by PMD, Corey Carin. Corey was referred for consultation to a paediatric neurologist, Dr. Gordon Watters. Dr. Watters examined Corey, asked his parents about their immediate families, and raised the possibility that their son could be suffering from Plott's Disease, a rare sex-linked disorder that shares some symptoms with PMD, and has a similar inheritance pattern. (6)
A year later, Corey was hospitalized and tests were carried out to confirm Dr. Watters' hypothesis of Plott's Disease. During his stay, only Corey's father, Frank Carin, accompanied him. Frank met with Dr. Watters and Dr. Naomi Fitch, a researcher in genetics at Montreal's Jewish General Hospital. Dr. Fitch informed Frank and Dr. Watters that Audrey's aunt, Edythe Sacks, had previously given birth to two boys who suffered from similar symptoms and died a few months after being born. Both physicians explained to Frank that if Corey was indeed suffering from Plott's Disease, this would have serious implications for Audrey, who would be at risk of having another child with the same disease. The physicians added that there were also implications for the women in Audrey's extended family. (7)
After receiving the test results, Dr. Watters eliminated PMD and concluded that Corey had Plott's Disease. He called Frank to inform him of the results and explained to him the implications of this diagnosis on Audrey. He offered to make himself available to discuss the matter with her. Frank, however, informed Dr. Watters that Audrey was not able to handle such information, and that he would personally undertake to inform Audrey of the diagnosis at the appropriate time. (8)
In 1973, Dr. Watters and Dr. Fitch published an article in Clinical Genetics on Corey's diagnosis. The article suggested--without specifying names--that Corey and Edythe Sacks' two sons were all affected by Plott's Disease. (9) The article included a family tree constructed by Dr. Fitch, who was able to construct it after securing Edythe's boys' medical records following her authorization. Dr. Watters never knew the identities of all the individuals in the family tree (though he knew of Edythe's sons). The article was never sent to Audrey Liss, her husband Frank Carin, or her aunt Edythe Sacks. (10)
In 1974, Dr. Watters met with Frank again, but did not ask him whether he had informed Audrey of the implications of Corey's diagnosis. He saw no reason why Frank would not have done so. Dr. Watters never contacted Audrey or Edythe Sacks, (11) and was unaware that Frank had never informed Audrey of the diagnosis. The couple had, however, separated by 1974. (12)
After moving to Israel in the late 1970s, Audrey married Claudio Liss, with whom she had several children, including Yochai Liss in 1988. After his birth, Yochai showed very similar symptoms to those of Corey. At age two, Yochai was diagnosed with leukodystrophy, "a category of diseases, almost exclusively hereditary, one of which is PMD." (13) He died in 2006; the leukodystrophy was identified as PMD. (14) Audrey noticed the resemblance between Yochai's symptoms and those of her first child, Corey. Her extended family did not make the connection.
Four years before Yochai's death, Marcy White--Edythe Sacks' granddaughter --gave birth to Jacob, who was born with similar symptoms to those of Yochai, Corey and the sons of Edythe Sacks. (15) Andrew Trossman, his father, made the connection between Jacob and Edythe's sons and decided to investigate the matter further. He discovered, after obtaining the medical chart of one of Edythe's sons, that Jacob had similar symptoms. Audrey was informed of the situation and she confirmed that Jacob's symptoms were similar to Corey's. A year later, Jacob was diagnosed with PMD. Beverly Sacks--Marcy's mother--and Edythe Sacks were both tested and confirmed as carriers of a specific mutation associated with PMD. (16) Audrey was also tested and confirmed as a carrier of the same mutation. It became clear that, back in the 70s, Corey had suffered from PMD, not Plott's disease. (17)
Both Audrey Liss and Marcy White (along with her husband Andrew Trossman) sued Dr. Watters for the wrongful birth of their affected sons.
The central allegation against Dr. Watters pertained to his failure to inform anyone in the family, especially, Audrey, Beverly and Edythe, of his diagnosis of Plott's disease in 1971. They claimed that had he informed them, Audrey and Marcy would have chosen not to give birth to Yochai and Jacob respectively. They also sued Dr. Fitch for failing to inform Beverly or Edythe of the genetic implications of Corey's disease. The Jewish General Elospital was sued as the employer of Dr. Fitch. In addition, Dr. Watters counter-sued Audrey in warranty, claiming that she failed to inform her family of her son's diagnosis. (18)
II. Liss v Watters: the Quebec Superior Court's Judgment
The court of first instance found Dr. Watters and Dr. Fitch liable. (19) The court awarded Marcy White and Andrew Trossman a total of $6,000,276, holding Dr. Watters and Dr. Fitch responsible for a share of 75% and 25% respectively. (20) It should be noted that Dr. Fitch died before the trial and never testified. (21) The trial judge also found that Dr. Watters failed to ensure that Audrey Liss was effectively warned and counselled on her carrier status, but her claim was dismissed for failure to prove causation. (22)
According to Judge Riordan, Dr. Watters committed three faults: 1) failure to communicate directly with Edythe and Beverly Sacks; 2) failure to request permission from Frank Carin to contact Edythe and Beverly Sacks; and 3) failure to persist and ensure that Audrey Liss was effectively warned and counselled.
The judge relied on the reasonable person standard and a two-fold test to establish an obligation to inform in his analysis of the first fault, that is, the failure to communicate directly with Edythe and Beverly Sacks. (23) The two-fold test was satisfied when 1) there was a serious illness, and 2) the third parties (Edythe and Beverly Sacks in this case) needed to be within the physician's "radius of contact." In other words, "the stronger, i.e., the more direct, the relationship, the more likely that the radius of contact is one that requires the physician to act." (24) According to the judge, both of these criteria were satisfied, even in the absence of a patient-physician relationship with Dr. Watters.
Judge Riordan played down the tension between the obligation to inform (arising after the two-fold test was fulfilled) and the obligation of confidentiality. He maintained that "[t]he physician is the one who best understands the consequences of the case for other family members" and that he "must be a leader, not a follower, in the information process. This applies no matter what obligation of confidentiality might bind him." (25) Experts who presented the duty of confidentiality as it existed in 1970 saw their opinions sidelined by the court, which relied on the 1991 Quebec case Roberge v Bolduc (26) to conclude that the practice of one's peers is not determinative of reasonable and diligent conduct. (27) The failure to warn Edythe and Beverly was therefore considered a fault under the general rules of civil liability.
Second, if confidentiality was an impediment, Judge Riordan concluded that Dr. Watters was at fault for not seeking Frank's permission to contact Edythe, knowing from the family tree depicted in the 1973 article that she had an adult daughter (Beverly). Moreover, the court deemed that the act of simply waiting for Audrey to come to terms with her son's diagnosis, and then informing her extended family, also constituted a fault under the general rules of civil liability. (28)
Third, the court decided that Dr. Watters should not have relied on an open-ended promise from Frank to contact Audrey. The failure to be persistent and ensure that Audrey was effectively warned, given the seriousness of the situation, was negligent. According to Judge Riordan, Dr. Watters should also have followed through with Dr. Fitch. (29) She was privy to the diagnosis, and in contact with Edythe Sacks (from whom she obtained consent to retrieve her sons' medical charts for the 1973 article); Dr. Watters could have reached Edythe through Dr. Fitch.
Proving causation was considered difficult due to the lapse of time since the 1970s. Hence, the Court relied on the presumption of fact to evaluate and determine causality. Judge Riordan concluded--given the available evidence --that had Dr. Watters or Dr. Fitch warned Edythe or Beverly, one of them would have told Marcy about the sex-linked condition. (30) According to Judge Riordan, there was also enough evidence to establish that had Marcy been informed of those genetic implications, she would have terminated her pregnancy and Jacob would not have been born. (31)
As mentioned earlier, Audrey's case was dismissed at the causation stage. The Court found that her actions and behaviour reflected a need to distance herself from Corey's condition and its implications. Indeed, she subsequently gave birth to four children without requesting specific testing. (32) Therefore, even if Dr. Watters had informed her of potential reproductive risks, she may not have taken heed. This, according to the judge, was "fatal to the causal link between Dr. Watters' fault towards her" (33) and the damages she alleged.
III. Watters v White--the Court of Appeal's Unanimous Decision
The Quebec Court of Appeal unanimously ruled to dismiss Marcy and Andrew's action against Dr. Watters. The reasons were delivered by Judge Kasirer, who found that the trial judge erred in both his finding of Dr. Watter's fault, and his finding of its causal link with the birth of Marcy White's affected child.
a. Fault: "The Duty to Inform" and the Trouble with Disclosure
As to the issue of fault, Judge Kasirer noted that the Superior Court judge erred in law by not identifying the appropriate standard of conduct. (34) Indeed, when establishing the fault committed by Dr. Watters, Judge Riordan analyzed the doctor's actions under the reasonable person standard rather than the reasonable physician standard. He also decided to develop his own test based on the gravity of the illness and the radius of contact "without regard to prevailing practices in the profession". (35) According to Judge Kasirer, this two-step test had the effect of throwing
the burden on Dr. Watters of disproving the existence of a duty to inform family members who were not his patients of the implication to them of his diagnosis. [...] Instead, the judge should have placed the burden of proving Dr. Watters' fault--the failure to act as the reasonable neurologist would have in the circumstances--squarely on the shoulders of the plaintiffs. (36)
In his reasons, Judge Kasirer emphasized that in Quebec civil law, a physician has an obligation of means and not of result. He also disapproved of this particular use of Roberge, especially if it discharged the court from considering the prevailing practices in a medical profession. He writes: "where courts do not have the necessary expertise to assess technical matters relating to medical treatment, due regard must be given to the opinion of experts." (37) Furthermore, he added that Judge Riordan's two-fold test (gravity of the illness and radius of contact) establishing a duty to inform ran afoul not only of the reasonable physician standard but also relied on "unsound" analogies from American precedents. (38) Judge Kasirer concluded that the radius of contact test would create a vague and onerous obligation for physicians, as it requires them to go beyond the patient-physician relationship and disclose information to third parties they had never met before. (39)
As for the duty of confidentiality--which was sidelined by Judge Riordan--Judge Kasirer characterized it as the "cornerstone of the doctor-patient because Corey's picture appears in the text. (44) Judge Kasirer concluded that "informing Edythe that she was a carrier brought with it a violation of the duty of confidentiality to Corey." (45) He also found that exceptions to the duty of confidentiality, such as the common law "duty to warn" (discussed below), did not apply.
Applying the standard of the reasonable physician, Judge Kasirer found that Dr. Watters did not commit a fault by not communicating with Beverly and Edythe. (46) He also found that the failure to encourage Frank to communicate with Audrey's relatives was also not a fault, and that encouraging Dr. Fitch to follow-up with Edythe and Beverly would have breached his patient's confidentiality. (47) In short, Dr. Watters fulfilled his duty to inform by telling his patient's representative, Frank, of the familial implications of the diagnosis. (48) This also implied that Dr. Watters' failure to get consent from Frank to contact Edythe and Beverly was not a fault, since he had no direct duty to inform them. (49)
b. Causation: The Wrong Disease and the "Egregious Omission"
The causation analysis in this case is complicated by the fact that Dr. Watters misdiagnosed Corey Carin's condition. His diagnosis was Plott's disease; it was confirmed years later that the affected male children actually had PMD. Importantly, however, both conditions share similar symptoms and pattern of inheritance. (50) Indeed, both conditions are X-linked recessive; females carry the mutation, which in turn affects 50% of their male offspring. For the trial judge, the mistaken diagnosis did not alter causation: even if Marcy White believed she was a carrier for Plott's, she would have chosen to give birth to a son. The Court of Appeal, however, found that the means of diagnosis was a major relevant difference between the two conditions. (51) Importantly, carrier testing was available for Marcy White for PMD, but not for Plott's disease. (52) In addition, when Marcy White was pregnant with Jacob Trossman, in utero testing was available for PMD, but not for Plott's disease. So if Corey's Plott's diagnosis had been brought to Marcy's attention, she would have had the compounded uncertainty of not knowing whether or not she was a carrier, or whether or not her fetus was affected.
According to Judge Kasirer, three links had to be established in the causal chain: 1) that Dr. Watters did not take adequate steps to inform Edythe and Beverly of their risk of being carriers; 2) that if Edythe had known of the diagnosis, she would have shared it with her daughter Beverly, who in turn, would have shared it with her daughter Marcy; and, 3) that if Marcy had been so informed, she would have terminated her pregnancy and Jacob would not have been born. Judge Kasirer found the first two steps were established, but found the third link problematic. (53) Indeed, Dr. Watters' diagnosed Corey with Plott's Disease while Marcy was later tested and confirmed as a carrier for PMD. In other words, the trial judge made an error by ignoring the misdiagnosis:
With respect for the judge, the appellant is right that the error is a palpable one. It is manifest from the account the judge himself gave of the case that he did not take into account the difference in the manner that Plott's Disease and PMD are diagnosed in his analysis of causation [...] His [the trial Judge] allusions to 'PMD' in paragraph  and 'the disease' in paragraph  are suggestive of the same manifest error of confounding Plott's and PMD. (54)
The trial judge had relied on a presumption of fact to establish causation: "that Marcy would have aborted had she tested for PMD and learned she was a carrier and her fetus was afflicted with PMD." (55) The Court of Appeal found that the misdiagnosis undermined this presumption:
that presumption of fact cannot properly be established from the diagnosis for Plott's for which there was no test. It is not a deduction of an unknown fact (i.e. the choice to terminate the pregnancy) from a known fact (i.e. that Corey or the Sacks' boys had PMD) because that latter fact was not known at the time. Article 2849 C.C.Q. has no possible application to establish the third link in the causal chain. (56)
Judge Kasirer maintained that nothing in the evidence allowed for the establishment of a presumption of fact that was "serious, precise and concordant", as required by article 2849 CCQ. (57) The trial judge presumed that, even without a direct test, if Marcy White simply had knowledge of the high statistical probability she was a carrier of Plott's, she would have aborted a male fetus. Again, the Court of Appeal disagreed. Because of the compounded uncertainty, Judge Kasirer calculated that Marcy White would have known only that she had a 12.5% chance of having an affected male child. The only evidence of Marcy White's "risk tolerance" was that she aborted her next child, but this could not be said to reflect her previous state of mind at the time she was pregnant with Jacob Trossman. (58)
Finally, Judge Kasirer found that Frank's inaction broke the chain of causation. While Marcy and Andrew insisted that absent Dr. Watters' fault, Marcy would have been informed of his diagnosis and ended her pregnancy, Judge Kasirer disagreed. He concluded that Frank's "egregious" (59) failure to inform his wife of the diagnosis made it quite unlikely that encouragement from Dr. Watters would have made a difference. While complex, this case reflects the core challenges of assessing causation in genetics: predicting familial communication patterns and decision making relating to genetic risk information over long periods of time.
a. The Duty to Inform
In Quebec, the duty to inform, also known as "l'obligation de renseigner" (60) or "l'obligation d'information du patient" (61) contains four different elements: 1) informing the patient of his diagnosis/medical condition; 2) explaining the nature, objectives of, and identifying the individual who will be executing, the proposed intervention; 3) disclosing the known risks; and 4) identifying the therapeutic options available to the patient. What is at stake here is the medical condition component. (62) This duty has a similar form in common law jurisdictions, where it is sometimes referred to as the "duty to disclose." (63) In short, physicians have an obligation to inform patients if they are affected by a medical condition. (64) This information can facilitate understanding of, and consent to, subsequent medical interventions. (65) In addition, communicating a patient diagnosis has important implications for lifestyle and reproductive decisions. Confidentiality is traditionally not an issue in the context of a duty to inform, because it remains bounded by the patient-physician relationship. (66) The central question in Watters v White asks whether the duty to inform in Quebec can extend beyond the patient.
The trial judge in Liss v Watters found a duty to inform extended to both the patient's mother (Audrey Liss), and two of her extended female relatives (Edythe and Beverly Sacks). Judge Kasirer overturned this decision for three reasons. First, he questioned the relevance of American precedents the trial judge relied upon to decide the case in Quebec (discussed below). (67) Second, he refused to acknowledge that the duty to inform extended beyond the patient. As Philips-Nootens and colleagues explain, the duty to inform must include an individual responsible for the transmission of the information, and a recipient of the information. The physician is responsible for transmitting the information to the patient. (68) As for the recipient, "[...] the duty to inform is owed to the patient and not to their relatives, except when they act as representatives of the patient or are authorized to decide on his/her behalf". Although consultation with the family is advised, it cannot constitute an obligation. (69) Hence, Judge Kasirer was right to decide that "[o]nce Frank was given the information, including a sense of the risk that other family members might face as a result of the sex-linked character of the disorder with which Corey has been diagnosed, Dr. Watters had fulfilled his obligation to inform his patient." (70) Third, as explained in detail in Section III(a), the trial judge overlooked confidentiality. According to Judge Kasirer, whether there is a duty to inform a patient's relatives is a matter of expert evidence, one that needs to be considered in relation to the physician's duty of confidentiality. Judge Kasirer reinforced the status of this duty by insisting the burden of proof lay with the plaintiff to prove that confidentiality did not pose a barrier to communication.
Would the outcome be the same elsewhere in Canada? While there are no Canadian cases on the duty to inform in the genetics context, the American jurisprudence Judge Kasirer rejected may be more readily welcomed outside Quebec. One such case is Pate v Threlkel, where a physician did not inform a patient with cancer that her condition was genetically inheritable. (71) Years later, the patient's daughter developed the same cancer. The court found that a physician's duty to prevent harm extends to family members where the standard of care "creates a duty that is obviously for the benefit of certain identified third parties." (72) However, the physician was only required to directly inform the patient of the heritable nature of the condition, not to seek out family members. Such a duty would be too arduous, and would potentially conflict with patient confidentiality. (73) In a related case, Safer v Pack, the court explicitly left open the question of whether a physician may have to take more extensive steps than simply informing the patient. (74) Molloy v Meier (75) is a recent case more representative of the issues in Liss v Watters, yet not addressed by either Judge Riordan or Judge Kasirer in the ruling. In that case, the judge ruled a duty to inform was owed to a biological mother of an adopted child who was affected by a heritable genetic condition. The court found that "the physician's duty regarding genetic testing and diagnosis extends beyond the patient to the biological parents who foreseeably may be harmed ..." (76)
There is, however, good reason to doubt that these cases represent a statement of positive law. (77) First, all three cases were appeals of summary judgments, meaning that both the facts and the prevailing standards of care alleged by the plaintiffs were assumed to be true. Second, the duty of confidentiality was largely overlooked in these cases, in part because no experts were called upon to testify. Third, each case involved physicians who failed to inform even the patient, thus any conclusion that they owed a duty to go beyond this level of disclosure could be considered obiter. (78) Nevertheless, the aforementioned cases do reflect a judicial willingness to extend the duty to inform in the genetics context.
That being said, a duty to inform patients' relatives of genetic risks is not recognized in Quebec. As such, simply communicating the familial implications to the patient will satisfy any duty that physicians may have towards a patient's family members in the genetic context, all but shuts the door on establishing a more extensive duty. It will require plaintiffs to present expert evidence of the standard of care that a reasonable physician in similar circumstances would have taken additional steps. In addition, plaintiffs will have the burden of proving that the duty to inform was not limited by considerations of confidentiality. As we discuss in the next section, it is highly unlikely that physicians will be obliged to inform at-risk relatives when confidentiality limits the duty to warn.
b. The Duty to Warn
The language of both judgments tends to conflate two distinct notions: the duty to inform and the duty to warn. Both the trial and appeal judges use the two terms interchangeably. While discussing the scope of the duty to inform, both ask whether Dr. Watters should "warn" Beverly and Edythe Sacks of their condition. (79) The duty to inform, discussed earlier, does not necessarily implicate confidentiality. The duty to warn, however, is a common law notion that describes an exception to confidentiality. Judge Kasirer explicitly defines the duty to warn:
There is a narrow category of exception whereby non-consensual disclosure is justified by considerations of public health, urgency or imminent danger. It may be noted that it was not until 1999 that the Supreme Court of Canada, in a different context, recognized a legal basis for a professional's duty to warn third parties in breach of a duty of confidentiality and only then when the danger faced by the third party was imminent. (80)
A "duty to warn" may exist between physicians and certain third parties to the doctor-patient relationship. (81) This duty outlines an exception to confidentiality where a physician must divulge patient information to prevent a serious and imminent injury to a third party. (82) The landmark case establishing this duty in the US is Tarasofff. (83) A psychologist was found to have a duty to warn a third party upon receiving information from a psychotic patient that threatened the safety of the interested third party. Three conditions were required to justify the confidentiality breach: 1) the professional must have a special relationship with the potential aggressor or victim; 2) the potential victim must be identifiable; and 3) the injury must be serious and foreseeable. The intentional character of the injury, however, was not thought to be a pertinent element to the analysis in Tarasoff. (84) In Smith v Jones, the Supreme Court of Canada recognized a similar duty where a psychiatrist retained by legal counsel assessed a sexual assault defendant. The defendant confided that he had fantasies of raping and killing prostitutes. The Court laid out a three-part test to determine if the duty to warn relevant authorities warranted a breach of solicitor-client privileges:
Three factors should be taken into consideration in determining whether public safety outweighs solicitor-client privilege: (1) Is there a clear risk to an identifiable person or group of persons? (2) Is there a risk of serious bodily harm or death? (3) Is the danger imminent? These factors must be defined in the context of each situation and different weights will be given to each, and to the various aspects of each, in any particular case. (85)
Canadian courts have not recognized a duty to warn in situations "when both patients and third parties are joint victims of a predisposition to suffer harm, such as from common genetic inheritance." (86) The conditions outlined in Smith v Jones suggest why. First, the imprecision and interpretation of genetic risk often cannot translate into "clear" risk, meaning disclosure may not be of appreciable benefit to a third party. (87) Second, genetic risk information is difficult to qualify as "imminent" or urgent. (88) While the risks of harm to the threatened individuals in Smith v Jones and Liss v Watters were both serious, the urgency in the latter case was not apparent. Indeed, it took decades for the risk to Marcy White to materialize. The duty to warn is not an evident solution for ensuring genetic risk information reaches a patient's relatives. It is only available under very strict conditions that seem ill suited to genetic information. Judge Kasirer was right to express caution before imposing such strict duty in the highly uncertain realm of genetics.
c. The Quebec Code of Ethics and the Discretion to Breach Professional Secrecy
The Quebec Code of Ethics may offer a preferable solution when confidentiality is at issue. The Code of Ethics of Physicians (89) does not codify the common law duty to warn. Rather than imposing a duty to warn on physicians, the Code grants them a discretion to breach professional secrecy under certain conditions. Articles 20 and 21 of the Code are worded permissively so as to grant physicians the discretion to breach professional secrecy "when there are compelling and just grounds related to the health or safety of the patient or of others." (90) Article 21 lists factors that must be considered when exercising this exception:
2) the identity of the person exposed to danger or of the group of persons exposed to danger;
3) the identity of the person to whom the communication was made, specifying, according to the case, whether it was the person or persons exposed to danger, their representative or the persons likely to come to
4) the act of violence he aimed to prevent;
5) the danger he had identified;
6) the imminence of the danger he had identified;
7) the information communicated.
The list of factors reflects the common law duty to warn, but the Code does not clearly insist that every factor be met to justify a breach of confidentiality. Nor does it create a duty; the use of the word "may" implies a discretion. As outlined in Article 21, however, the physician must at least consider each condition before deciding to breach professional secrecy. Guidelines have already been outlined in the genetics research context to guide physicians in exercising this discretion. Quebec's Network of Applied Genetic Medicine (91) permits researchers to disclose genetic information to the biological relatives of a participant in spite of the participant's refusal, under the following conditions:
1. There must be elevated risk for the family members in developing a serious illness;
2. The biological family members are identifiable; and
3. The illness can be avoided by preventative measures or controlled by approved treatments. (92)
Serious genetic risks are compelling grounds to justify breaching professional secrecy in an effort to warn a patient's relatives. Granting physicians this discretion offers a workable solution to the conflicting interests at play. Because the common law duty to warn is not just an exception to confidentiality, but also a legal duty owed to a third party, its application will remain strictly limited. The discretion granted to physicians in Quebec may allow a proportional approach that more adequately balances the patient's interest in confidentiality with the interest of family members, allowing for the communication of serious genetic risk information, ft should be noted, however, that the authors have not encountered evidence that this discretion is currently exercised in the genetic context.
In addition to proving fault, Marcy White was also required to prove that Dr. Watter's failure to communicate Corey's diagnosis caused her to give birth to an affected child. Judicial determination of causation in the genetic context is complicated by two factors arising in Watters v White. First is the question of inter-familial communication. If informed of her carrier status, would Edythe Sacks have passed on this information to her daughter and granddaughter? This question is particularly difficult when the causal chain stretches over several decades, and requires communications between several individuals. The second challenge for the judge is evaluating decisions made in the face of risk: would Marcy White have terminated her pregnancy had she been informed of her carrier risk? Before discussing these challenges, we outline the legal test for causation in duty to inform cases.
In common law provinces, the plaintiff must not only prove that she would have made a different decision if properly informed, but also that the same choice would have been made by a reasonable person in the same circumstances. This is called the "modified-objective test." While the decision is evaluated objectively, it does takes into account subjective factors such as the plaintiff's particular qualities and medical situation. The Supreme Court of Canada first introduced the modified-objective test in Reibl v Hughes. (93) Cory J further refined the test in a "duty to disclose" context (as opposed to informed consent) in Arndt v Smith. (94) In that case, a mother who contracted chickenpox during pregnancy was not informed of the risk the infection posed to her unborn child. She later sued for costs relating to her child's resulting disability. Liability hinged on whether a "reasonable person in the plaintiff's position" would have decided to terminate the pregnancy. (95)
Opinion is split in Quebec civil law between the application of a subjective test for causation and an objective test like that outlined in Reibl v Hughes. (96) The subjective test considers causation in relation to the particular patient, and what decision they claim they would have made had they been informed of the risk. The problem with a purely subjective test is its dependence on plaintiff testimony, which may be deceptive or "coloured" a posteriori by the harm they have suffered. (97) In contrast, the objective test only respects reasonable decisions about healthcare. Some argue that this conflicts with the robust modern conception of patient autonomy. (98) Many authors believe that the objective test is incompatible with the principles of civil law, but concede that the objective factors may be considered. (99) While not relying on a formal objective test, judges may achieve this simply by having natural scepticism for unverifiable, uncorroborated testimony. (100)
The first causal link requiring a presumption was whether the risk information would have travelled from Edythe to her daughter Beverly and granddaughter Marcy. On one hand, the trial judge found the evidence did not directly establish that Edythe was open to sharing the information in question with Beverly or Marcy. (101) He also found that there was little discussion about Corey Carin's or Yochai Liss' situations within the Sacks family. (102) On the other hand, he noted that both Edythe Sacks and Marcy White had "good" relationships with Beverly Sacks. (103) From this known fact, Judge Riordan presumed either Edythe or Beverly Sacks would have informed Marcy White, under article 2849 CCQ. (104) The Court of Appeal accepted the trial judge's conclusion at this stage of the causal chain. (105) What remains unclear, however, is how such an analysis would play out in cases where the genetic risk is less serious, or where family communication is more fractured. Judges assessing causation may face real difficulty in determining how information would have travelled through a family had one member been informed. Disclosure of hereditary genetic conditions between family members is by no means a straightforward process. There is evidence that the decision to disclose is complicated by existing patterns of communication, the information already circulating in families, and the probabilistic nature of genetic information. (106) Judicial outcomes may also hinge on what causation test the court adopts. If a subjective test is favoured, what kind of evidence about family relationships and communications will be definitive? Could an objective test suffice in determining whether a reasonable person in the family member's circumstances would disclose the information?
The second challenging link in the causal chain was determining whether Marcy White would have decided to terminate her pregnancy had she been informed of her carrier risk. Here the materialization of risk problematizes plaintiff testimony. In concluding that Marcy White would have terminated the pregnancy, the trial judge considered the following facts: she was highly educated, determined to understand her son Jacob's condition, and decided to terminate a subsequent male pregnancy. (107) Judge Kasirer did not agree. He challenged the reliance on facts occurring after Jacob's birth and, therefore, after the realization of the risk. Ultimately, the trial judge dismissed Audrey Liss' claim as a result of her behaviour after the harm suffered, especially that "she had four more children in Israel without requesting specific testing, or even mentioning Corey's problems to her doctors". (108)
Plaintiffs may also struggle to demonstrate, and courts to determine, the decisions that would have been made in the face of statistical probabilities. The trial judge erred in only adducing testimonial evidence from Marcy that she would have had an abortion if she was certain her son would be affected. For Judge Kasirer, this evidence was insufficient to answer the more nuanced question that should have been posed to Marcy: would she "have decided to end the pregnancy in the absence of a test, only on the basis of a probability?" (109) He calculated that had the mistaken Plott's disease diagnosis been communicated, Marcy would have been under the impression she had a 12.5% probability of having an affected son. (110) There was no evidence, however, of Marcy's tolerance for risk. (111) We agree with Judge Kasirer that Marcy's decision should have been considered against a background of probability rather than certainty. Indeed, uncertainty and probability is the rule rather than the exception in genetics.
The difficult task remains: determining the decision an individual would have made in the face of uncertain odds. How is a plaintiff expected to demonstrate her personal "risk threshold" in the abortion setting, especially if actions and testimony after the birth of an affected child are considered tainted evidence? In addition, while Judge Kasirer's calculations of the statistical risk Marcy faced is commendable, is it definitive? The ethical literature demonstrates that lay interpretation of risk often contradicts theories of Mendelian inheritance: "lay risk is 'not objective, cannot be quantified or measured, and is not static' ..." (112) Marcy White's interpretation of risk may have been deeply coloured by subjective factors, even before she had Jacob. She was connected to a family affected by a serious and misunderstood condition, even if it happened to be geographically and emotionally scattered. How would her previous experiences have affected her decision?
While Watters v White asks us to look into the past, it also gives us an important glimpse into the future. Genetics is rapidly altering the face of both medicine and research. This revolution is bound to challenge the traditional duties of physicians and create challenges for the courts. (113) This case comment discussed three emerging issues Watters v White raises. First, the Quebec Court of Appeals confirmed that the duty to inform is limited to patients, and second, the applicability of the duty to warn in the genetic context. In rare circumstances, physicians may encounter a clear conflict between confidentiality and the interests of relatives in genetic risk information. Discussion about how to resolve this tension has tended to focus on the common law duty to warn. The conditions of the duty to warn, however, may prove too onerous on physicians and too narrow in scope to be embraced in the genetic context. Granting physicians the discretion to breach professional secrecy for compelling reasons offers a promising alternative avenue. Third, we discussed the complexity of the causation analysis in duty to inform cases in genetics. Judges may be called on to make difficult determinations about how genetic risk information is communicated and perceived within families, and how individuals make decisions in the face of probabilities.
These issues are merely the tip of the iceberg. Medical genetics is no longer limited to the study of Mendelian disorders or chromosomal anomalies. (114) The direction of next-generation sequencing technologies is towards the generation of a vast amount of data with uncertain or unknown meaning. (115) In response to innovations in technology, medicine is becoming increasingly personalized. Pharmacogenomics, for example, allows practitioners to tailor drug therapies specifically to an individual's genotype, (116) effectively reducing adverse effects and increasing efficacy. (117) To complicate matters further, genetic tests are no longer limited to medical clinics or hospitals. Consumers can now access genetic tests directly via the web. (118) These myriad developments give us good reason to doubt that Watters v White will be the last word on the duty to inform in genetics.
(1) Liss v Watters, 2010 QCCS 3309 [Liss v Watters],
(2) Watters v White, 2012 QCCA 257, leave to appeal to SCC refused, 34758 (August 30, 2012) [Watters v White].
(3) Online Mendelian Inheritance in Man: An Online Catalog of Human Genes and Genetic Disorders "Pelizaeus-Merzbacher Disease" (28 June 2012), online: OMIM <http://www.omim.org/entry/312080?search=Pelizaeus Merzbacher%20disease&highlight=pelizaeusmerzbacher%20disease# reference32> [OMIM].
(4) Ibid; Watters v White, supra note 2 at para 11; Liss v Watters, supra note 1 at para 6.
(5) Ibid at para 11.
(6) Ibid at paras 20, 139.
(7) Ibid at para 25.
(8) Ibid at paras 26-28.
(9) Gordon V Watters & Naomi Fitch, "Familial Laryngeal Abductor Paralysis and Psychomotor Retardation Citation" (1973) 4:5 Journal of Clinical Genetics 429.
(10) Watters v White, supra note 2 at paras 35-36.
(11) Ibid at paras 20-32.
(12) Ibid at para 31.
(13) Ibid at para 38.
(14) Ibid at para 45.
(15) Ibid at para 41.
(16) Ibid at para 43.
(17) Ibid at para 47.
(18) Ibid at para 50.
(19) Judge Riordan exonerated the Jewish General Hospital, as it was determined that Dr. Fitch was not its agent according to article 1463 of the Civil Code of Quebec (Liss v Watters, supra note 1 at para 157). In addition, the court dismissed Dr. Watters' action against Audrey, judging that she did not have a comprehensive understanding of Corey's diagnosis so as to be able to inform the members of her extended family (ibid at paras 197-220).
(20) Ibid at paras 410-432.
(21) Ibid at para 76.
(22) Ibid at paras 192, 195-196.
(23) Ibid at paras 40-74.
(24) Ibid at para 58.
(25) Ibid at para 68.
(26) Roberge v Bolduc,  1 SCR 374.
(27) hiss v Watters, supra note 1 at para 46.
(28) Ibid at paras 100-04.
(29) Ibid at paras 113.
(30) Ibid at para 174.
(31) Ibid at paras 179-81.
(32) Ibid at paras 192, 195-196.
(33) Ibid at para 196.
(34) Watters v White, supra note 2 at para 71.
(35) Ibid at para 71.
(36) Ibid at para 72.
(37) Ibid at para 77.
(38) Ibid at paras 79-86.
(39) Ibid at para 86.
(40) Ibid at para 95.
(41) Ibid at paras 95-96.
(42) Ibid at para 95.
(43) Ibid at para 102.
(44) Ibid at paras 98-109.
(45) Ibid at para 102.
(46) Ibid at paras 112-117.
(47) Ibid at para 118.
(48) Ibid at paras 119-123.
(49) Ibid at para 124.
(50) Ibid at para 155.
(51) Ibid at para 139.
(52) OMIM, supra note 4.
(53) Watters v White, supra note 2 at paras 129-133.
(54) Ibid at para 145.
(55) Ibid at para 154.
(56) Ibid at para 154;
(57) Presumptions of fact in Quebec law are defined by 2846 CCQ ("A presumption is an inference established by law or the court from a known fact to an unknown fact"), and their parameters given by 2849 CCQ ("Presumptions which are not established by law are left to the discretion of the court which shall take only serious, precise and concordant presumptions into consideration").
(58) Ibid at paras 164, 165.
(59) As described by the trial judge, ibid at para 177.
(60) Suzanne Philips-Nootens, Pauline Lesage-Jarjoura & Robert P Kouri, Elements de responsabilite civile medicate: Le droit dans le quotidien de la medecine, 3d ed (Cowansville Editions Yvon Blais, 2007) at 139-255.
(61) Jean-Louis Baudouin & Patrice Deslauriers, La Responsabilite civile, 7th ed (Cowansville : Editions Yvon Blais, 2007) at paras 2.44-2.55.
(62) Watters v White, supra note 2 at para 121.
(63) Mary Z Pelias, "Duty to Disclose in Medical Genetics: A Legal Perspective" (1991) 39:3 American Journal of Medical Genetics 347 at 351.
(64) Laferriere v Lawson,  1 SCR 541; Philips-Nootens, Lesage-Jarjoura & Kouri, supra note 61 at 146-147; Patrice Deslauriers, "La responsabilite medicale et hospitaliere" in Ecole du Barreau du Quebec, Collection de droit 2012-2013, vol 4 titre 2 (Cowansville : Editions Yvon Blais, 2011) at 130-131.
(65) Turmel v Loisel  RRA 532 at para 71; Quintal v Godin  RRA 502 at paras 115-118; Hopp v Lepp,  2 SCR 192 at 204; Reibl v Hughes,  2 SCR 880 at 885.
(66) Philips-Nootens, Lesage-Jarjoura & Kouri, supra note 61 at 143.
(67) Watters v White, supra note 2 at para 84. The trial judge relied mainly on Safer v Pack, infra note 75.
(68) Ibid at 144. See also Lamarre v Hopital du Sacre-Coeur,  RRA 496 at para 35.
(69) Philips-Nootens, Lesage-Jarjoura & Kouri, supra note 61 at 143 [translated by authors].
(70) Watters v White, supra note 2 at para 122.
(71) Pate v Threlkel, 661 So.2d 278 (Fla 1995) in Lori B Andrews, Maxwell J Mehlman & Mark A Rothstein, eds, "Genetics: Ethics, Law and Policy" 3d (Thomson Reuters, St. Paul, 2010) 450-452.
(72) Ibid at 452.
(74) Safer v Pack 677 A.2d 1188 (NJ Super. Ct. App. Div. 1996) in Andrews, Mehlman & Rothstein, supra note 72 at 453-457.
(75) Molloy v Meier, 679 NW2d 711 (Minn 2004) cited in Andrews, Mehlman & Rothstein, supra note 72 at 460-469.
(76) Ibid at 467.
(77) Watters v White, supra note 2, at para 84, n 12 (citing Gillian Nycum, Bartha Knoppers and Denis Avard, "Intra-familial Obligations to Communicate Genetic Risk Information: What Foundations? What Forms?" (2009) 3 McGill J Law & Health 21 at 37).
(78) Watters v White, supra note 2 at para 84. This is implied by Judge Kasirer, who defines the limited scope of the ratio of Safer v Pack in relation to the facts of his case.
(79) Ibid at para 109; Liss v Watters, supra note 1 at paras 99, 105-114.
(80) Watters v White, supra note 2 at para 111 [emphasis of authors].
(81) Jocelyn Downie, Timothy A Caulfield & Colleen M Flood, eds, Canadian Health Law and Policy, 4th ed (Toronto: LexisNexis Canada, 2011) at 122.
(82) Martin Letendre "Le devoir du medecin de prevenir les membres de la famille d'un patient atteint d'une maladie genetique" (2004) 49:3 McGill LJ 555-591 at 42. Letendre also discusses the similarities between the duty to warn and the civil law "duty to rescue" from article 2, paragraph 2 of the Quebec Charter of Human Rights and Freedoms, RSQ c C-12: "Every person must come to the aid of anyone whose life is in peril, either personally or calling for aid, by giving him the necessary and immediate physical assistance, unless it involves danger to himself or a third person, or he has another valid reason.")
(83) Tarasoff v Regents of the University of California 131 Cal Rptr 14 (Cal Sup Ct 1976) [Tarasoff].
(84) Letendre, supra note 83 at 38.
(85) Smith v Jones,  1 SCR 455, 169 DLR (4th) 385 at para 77. Other Canadian cases discussing the duty to warn include Tanner v Norys  4 WWR 33; Wenden v Trikha (1991), 116 AR 81, aff'd (1993), 14 CCLT (2d) 225; Pittman Estate v Bain (1994) 112 DLR (4th) 258.
(86) Downie, Caulfield & Flood, supra note 82. See also Philips-Nootens, Lesage-Jarjoura & Kouri, supra note 61 at 373. Letendre also points out that the duty to warn is a common law principle that might not be strictly applicable in Quebec (supra note 83 at 583).
(87) Jennifer L Gold, "To Warn or Not to Warn? Genetic Information, Families, and Physician Liability" (2004) 8:1 McGill Journal of Medicine 72 at 76 (concluding that the existence of a clear benefit underlies Canadian decisions on the duty to warn).
(89) Code of Ethics of Physicians, RRQ 1981, cM-9, r 17 [Quebec Code of ethics of physicians].
(90) Ibid, art 20(5).
(91) The Network of Applied Genetic Medicine is a network of researchers developing infrastructure to support basic genetic research in Quebec and its translation into health services and interventions. (See online: RMGA <http://www.rmga.qc.ca/en/index.html>).
(92) Letendre, supra note 83 at 582; Reseau de medecine genetique appliquee, Statement of Principles: Human Genome Research Version 2000 at 10 online: RMGA <http://www.rmga.qc.ca/fr/documents/ Enoncedeprincipesrechercheengenomiquehumaine_fr_000.pdf>.
(93) Supra note 66. See also Paul McGivern & Natalia Ivolgina "Legal Liability In
Informed Consent Cases: What Are The Rules Of The Game?" (2013) 7:1 McGill JL & Health 129 at 136.
(94) Arndt v Smith,  2 SCR 539, 148 DLR (4th) 48, BCLR (3d) 187.
(95) Ibid at para 18.
(96) Philips-Nootens, Lesage-Jarjoura & Kouri, supra note 61 at para 283.
(99) See e.g. Robert P Kouri, "L'obligation de Renseignement en Matiere de Responsabilite Medicale et la 'Subjectivite Rationnelle': Manage de Convenance ou Mesalliance?" (1994) 24 RDUS 345 at 365.
(101) Liss v Watters, supra note 1 at para 165.
(102) Ibid at para 166.
(103) Ibid at para 172.
(104) Supra note 66.
(105) Watters v White, supra note 2 at para 133.
(106) Susan M Cox & William McKellin, "'There's This Thing in our Family': Predictive Testing and the Construction of Risk for Huntington Disease" (1999) 21:5 Sociology of Health & Illness 622.
(107) Liss v Watters, supra note 1 at paras 179-182.
(108) Ibid at para 192.
(109) Watters v White, supra note 2 at para 163.
(110) Ibid at para 164.
(111) Ibid at para 165.
(112) Cox & McKellin, supra note 107 at 641.
(113) See Adrian Thorogood, Bartha M Knoppers, Wybo J Dondorp & Guido MWR De Wert, "Whole Genome Sequencing and the Physician" (2012) 81:6 Journal of Clinical Genetics 511.
(114) See Francis S Collins & Victor A McKusick, "Implications of the Human Genome Project for Medical Science" (2001) 285:5 Journal of the American Medical Association 540.
(115) See Francis S Collins, "Has the Revolution Arrived?" (2010) 464:7289 Nature 674.
(116) See Health Canada, Pharmacogenomics, online: Health Canada <http://www.hc-sc.gc.ca/sr-sr/biotech/about-apropos/pharma-eng.php>.
(117) See John A Robertson et al, "Pharmacogenetic Challenges for the Health Care System" (2002) 21:4 Health Affairs 155.
(118) See Felix W Frueh et al, "The Future of Direct-to-Consumer Clinical Genetic Tests" (2011) 12:7 Nat Rev Genet 511.
Ma'n H. Zawati & Adrian Thorogood *
* Ma'n H. Zawati, LL.B., LL.M., DCL candidate (McGill University), is a lawyer and Academic Associate at the Centre of Genomics and Policy, McGill University. Adrian Thorogood BA&Sc, BCL/LLB candidate (McGill University) is a Research Assistant at the Centre of Genomics and Policy. The authors would like to thank Martin Hetu, Ariane Mallette, George Vardatsikos and Vaso Rahimzadeh for their valuable assistance.
Figure 1: Timeline of Medical Events and Lawsuits 1940s Edythe Sacks, the aunt of Audrey Liss, gives birth to two boys--both die within two months. 1970 Corey Carin is born to Audrey Liss and Frank Carin and after consultation with Dr. Watters is thought to have Plott's disease, a rare sex-linked disorder. Frank Carin brings his son Corey to the hospital to confirm the diagnosis. Both Dr. Watters and Dr. Fitch mention the hereditary implications of the condition. Dr. Watters diagnoses Corey Carin with Plott's disease, and once again explains to Frank the implications of the diagnosis for Audrey. 1973 Dr. Watters and Dr. Fitch publish an article on Corey's case, including a family tree constructed by Dr. Fitch from Edythe Sacks' medical records. 1988 Audrey Liss, who has remarried, gives birth to Yochai Liss, who is affected but not diagnosed with PMD. 2002 Marcy White (Edythe Sacks' granddaughter) gives birth to an affected child, Jacob Trossman. 2003 Jacob Trossman is diagnosed with PMD. In turn, Edythe Sacks and Audrey are diagnosed as carriers. Marcy White becomes pregnant. She terminates her pregnancy after testing for PMD. 2007 Marcy White and her husband Andrew Trossman sue Dr. Watters, Dr Fitch and the Jewish General Hospital to recover losses incurred by Jacob's birth. Audrey Liss sues Dr. Watters to recover losses incurred by Yochai's birth.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||physician's duty to inform patients of genetic conditions leading to wrongful birth with genetic defects; Canada|
|Author:||Zawati, Ma'n H.; Thorogood, Adrian|
|Publication:||Health Law Journal|
|Article Type:||Case note|
|Date:||Jan 1, 2014|
|Previous Article:||An Australian perspective on the doctor-patient relationship: a comparative prelude to examining duties of disclosure in surgical innovation.|
|Next Article:||The distinction between withholding and withdrawing treatment in Rasouli: providing a solution to an ethical problem.|