Remoteness and causation are two of the five elements a Plaintiff must prove in order to establish the tort of negligence[1]. These two elements are often confused and conflated, and in particular, there is a tendency for the concept of remoteness to be misunderstood. It was argued before the Supreme Court of Canada that at the trial of the action in Hemmings v. Padmore [2], that the trial judge had done exactly that.
In a 100-plus page decision handed down by the Supreme Court of Canada on October 23, 2026, the Supreme Court attempted to clarify the differences between causation and remoteness, and to set out the scope of remoteness of damages. The facts giving rise to this case, its judicial history and the reasons of the Supreme Court of Canada are all extremely complex, and there is no doubt that the meaning of this decision will be debated and argued for years to come. The intent of this article is to set out the key takeaways, and how it will affect tort law in Canada (if at all).
The Facts
The underlying facts to this decision, which arises from a medical malpractice claim surrounding the pregnancy of the Plaintiff Sophie Hemmings, are as complex as they are heartbreaking. Ms. Hemmings had a high risk pregnancy, and due to complications during delivery at Scarborough General Hospital on April 18, 2009, it was necessary for her to deliver her child by way of c-section. The Plaintiff started this procedure under a regional anesthetic, but experienced a panic attack on the operating table. As a result, a sedative, then a general anesthetic was administered, following which the Plaintiff experienced complications, was intubated, and went into cardiac arrest. While the Plaintiff’s child was healthy upon delivery, the Plaintiff was left with significant brain damage as a result of being deprived of oxygen while in cardiac arrest.
As a result, the Plaintiff and her family commenced suit against the following defendants:
- The Plaintiff’s own OBGYN, Dr. Padmore;
- The two OBGYN’s present during delivery, Dr. Peng and Dr. O’Brien;
- The anesthesiologist, Dr. Jamesky, who administered the general anesthetic;
- The hospital at which the delivery took place, with respect to the alleged negligence of a nurse to whom the Plaintiff had spoken ten days before the delivery.
The basis for the claim against Dr. Padmore was that despite the Plaintiff being a high-risk pregnancy, where a c-section and complications at birth were likely, he did not properly warn her of these risks, document these risks in her chart, advise her that termination of the pregnancy was a viable option and determine her wishes in this regard, or prepare an appropriate pregnancy plan in light of these risks.
The basis for the claim against Dr. Jamesky was, amongst other things, that upon the Plaintiff initially experiencing a panic attack, he administered the general anesthetic prematurely, and instead should have first tried other alternatives, including permitting the Plaintiff’s sister, who was in the delivery room, the opportunity to calm her, or waiting for the sedative he had previously administered to take full effect[3].
The basis for the claim against the hospital arose from a phone call the Plaintiff made to the hospital’s labour and delivery unit 10 days before the delivery. In this phone call, the Plaintiff advised she was vomiting. The nurse to whom she spoke, without first looking the Plaintiff’s medical history, told her there was no need for medical attention or for the Plaintiff to come to the hospital. The Plaintiff’s position at trial was that had this nurse looked at her medical history, she would have realized that she was a high risk pregnancy, and would have counselled her to seek medical attention, at which time she would have been induced, and that the complications ultimately experienced during her delivery could have been avoided.
The Decisions of the Trial Judge and the Ontario Court of Appeal
At trial, The Plaintiffs’ claim was dismissed against Drs. Peng and O’Brien, the two OBGYN’s in the delivery room, but liability was apportioned equally between Dr. Padmore, Dr. Jamesky, and the hospital. All three of these Defendants appealed, following which the Ontario Court of Appeal dismissed the claim against the hospital and Dr. Padmore, but held that Dr. Jamesky was wholly responsible for the Plaintiffs’ damages, due to having prematurely admitting a general anesthetic , thus leading to the Plaintiff’s cardiac event. The claim against the hospital was dismissed because the Ontario Court of Appeal concluded that the trial judge did not explain why he preferred the evidence of the Plaintiffs’ expert over the evidence of Hospital’s expert as to what would have happened had she been told to come to the hospital on April 8, 2009. With respect to Dr. Padmore, the Court of Appeal noted that the trial judge had found Dr. O’Brien not responsible for the Plaintiff’s damages on the basis that he had no way of knowing that Dr. Jamesky was going to prematurely administer a general anesthetic, and concluded that the same must have been true of Dr. Padmore.
The Plaintiffs appealed to the Supreme Court with respect to the dismissal of their claim against Dr. Padmore and the hospital, and Dr. Jamesky appealed the finding he was wholly liable. Both appeals were heard together in February 2026. The arguments before the Supreme Court focused on dealt with remoteness and causation, and how the trial judge had conflated the two concepts rather than considering them separately.
As noted above, the Supreme Court of Canada issued its reasons on October 23, 2026, dismissing both appeals, with the result that Dr. Jamesky remained wholly responsible for the Plaintiffs’ damages.
The Supreme Court’s Reasons
The majority reasons of the Supreme Court were written by Justices Rowe and O’Bonsawin, with Justice Jamal concurring. Justice Martin, with Justices Karkatsanis and Moreau concurring, dissented. Justice Wagner, the seventh judge on the panel, wrote a separate dissent.
In considering this matter, Rowe J. and O’Bonsawin stressed that it was critical that, when proceeding with a negligence analysis, causation “remain analytically distinct and be treated separately” from remoteness, which deals with whether the type of harm that befalls the plaintiff is rationally connected to the alleged breach of the standard of care. The court suggested that “causation” should be characterized as “factual causation” and “remoteness” should instead be characterized as “legal causation”.
To further distinguish these two concepts Rowe and O’Bonsawin JJ. noted that (factual) causation involves asking a factual question as to whether the Plaintiff’s conduct caused the Defendant’s injury”[4], is often guided by common sense, and is decided on a balance of probabilities. Factual causation can often be analyzed in more complex situations through a “counterfactual analysis”, which means by considering whether the outcome would have been different had the negligent act had not occurred – i.e. in this case, whether the Plaintiff would have avoided complications in delivery had Dr. Padmore developed an appropriate pregnancy plan, or the nurse whom she spoke to on April 8, 2009 had reviewed her medical records and counselled her to seek medical attention.
Remoteness/legal causation, however, was characterized as “more a question of law and policy than of fact”. Rowe and O’Bonsawin JJ. noted that remoteness has historically remained a difficult concept to grasp, with courts wrestling with the scope of liability that should attach to a party’s negligent conduct and specifically whether this scope should be limited to the specific harm that arose under the circumstances, any probable harm, or any sort of possible harm which might be a natural consequence of the negligent act in question. Based upon their review of the history of the jurisprudence surrounding remoteness, which is reviewed at some length in their reasons, these two judges concluded that the following principles are relevant to the consideration of remoteness:
- A negligent defendant can only be liable for the consequences they could have reasonably foreseen at the time of their breach of the standard of care. For professional defendants, such as the doctors in this case, what is reasonably foreseeable should be assessed based on the skill and knowledge of a reasonable practitioner at the relevant time;
- For the harm in question to be reasonably foreseeable, it must be one with a real risk or probability of occurrence, and certainly more than a “mere possibility”. If a reasonable person in the Defendant’s position would not “brush off the risk of the harm in question as far-fetched”, but would have taken reasonable precautions to avoid or prevent it, this will satisfy remoteness;
- That said, the Defendant need not have foreseen the precise sequence of events leading to the harm suffered by the Plaintiff, just that the general type of harm was something with real risk or probability of occurrence;
- The consideration of remoteness is not an abstract, academic or public policy analysis (as the Plaintiffs in this case seem to have argued vis-à-vis the hospital and Dr. Padmore). Instead, one needs to consider whether the harm in question was a “real risk”, such that a reasonable person would have taken steps to avoid it, as per (2), above.
To summarize, Rowe and O’Bonsawin JJ. stated:
“a defendant can only be liable for a plaintiff’s injuries if the probability that those injuries would occur was something the defendant ought to have considered and taken precautions to prevent, having regard to the degree of harm, and the costs associated with preventing the risk of that harm”.
The author notes that this has been the Canadian law as it pertains to remoteness for almost sixty years, with Canadian courts abiding by Overseas Tankship (UK) Ltd. v. Morts Dock and Engineering Co[5]. and Overseas Tankship (UK) Limited v. Miller Steamship Co.[6] (which anyone who studied Tort law in Canada will better recognize as The Wagon Mound. No. 1, and The Wagon Mound No. 2), hence the title of this article. As such, it appears that this decision should be considered more of a re-statement or clarification of the law of remoteness in Canada, rather than one that changes the law.
The astute reader will also note that the concept of foreseeability is one of the key considerations in an earlier stage of the negligence analysis – that is, of course, determining whether a defendant owes a duty of care to the Plaintiff. Rowe and O’Bonsawin JJ. acknowledge this in their reasons, noting that applying foreseeability a second time at the remoteness stage acts as an additional safeguard to ensure that there is a “sufficient connection” between the defendant’s wrongful conduct and the plaintiff’s “particular injury”. Whether this decision, in an attempt to avoid the conflation of remoteness and causation has instead left the door open for foreseeability and remoteness to be similarly conflated remains to be seen. It appears, then, that even if a Defendant’s act can be found to have caused a Plaintiff’s damages, if the Plaintiffs damages are too remote – i.e. not something that a reasonable Defendant ought to have considered and take steps to prevent – then a Defendant may escape liability.
Having enunciated the foregoing principles, Rowe and O’Bonsawin JJ. go on to consider each of the appeals before the court. In so doing, they concluded that the Court of Appeal correctly applied factual causation in dismissing the Plaintiffs’ claim against the hospital, noting that when one undertakes a counterfactual analysis (i.e. what would have happened had the Plaintiff been told by the nurse to come to the hospital on April 8) there was no evidence before the trial judge that the Plaintiff would have had a successful and uncomplicated delivery, presumably give that the same risk factors for her pregnancy existed on April 8, 2009 as they did when the Plaintiff was induced 10 days later. As such, the SCC concluded that “the relationship between [the nurse’s] breach of the standard of care and [the plaintiff’s] injuries was not sufficiently direct to permit the trial judge to conclude but-for causation without addressing the counterfactual directly”. In other words, all this nurse did was unintentionally divert the Plaintiff down the path which put her in a position to be injured. It was concluded that this was not sufficient to establish factual causation.
Rowe and O’Bonsawin JJ. went on to consider remoteness vis-à-vis both the hospital and the Plaintiff’s own OBGYN, Dr. Padmore. In this regard, they felt that the trial judge did not analyze remoteness in keeping with the principles noted above, as he did not consider whether there was a “real risk” apparent to the nurse to whom the Plaintiff spoke on April 8, 2009, or to Dr. Padmore of anesthetic complications once the delivery commenced. They were not persuaded by the Plaintiff’s submission that in order to analyze remoteness in this manner, as the Court of Appeal did, it would be necessary to define the Plaintiff’s injury so narrowly that to be able to affix these Defendants with liability would require the sort of prescience that could only be obtained through the use of a crystal ball. Instead, Rowe and O’Bonsawin JJ. agreed with the Defendants’ argument that all that was necessary was to consider whether the actual injury – that is, an anesthetic complication leading to a cardiac arrest and an anoxic brain injury in the delivery room– was foreseeable, generally speaking, given the risk factors associated with the Plaintiff’s pregnancy. They concluded in this regard that, given the trial judge’s findings that Dr. Jamesky’s breach of the standard of care was not foreseeable to Dr. O’Brien, despite him being in the delivery room at the time he commenced the c-section, this same risk could not have been reasonably foreseeable to Dr. Padmore eight months earlier when he saw the Plaintiff. The author notes that Rowe and O’Bonsawin’s reasoning in this regard makes sense, insofar as although Dr. Padmore might anticipate the Plaintiff needing a c-section, he could reasonably expect the attending anesthesiologist to do their job in a reasonably competent manner. As such, the Rowe and O’Bonsawin JJ. upheld the Court of Appeal’s dismissal of the Plaintiff’s claim against Dr. Padmore and the hospital.
Dr. Jamesky’s appeal focused on the issue of factual causation, with him raising the argument that if one were to carry out a counterfactual analysis, there would be no evidence upon which the trial judge could properly have concluded that the Plaintiff would have had a successful delivery. Rowe J and O’Bonsawin disagreed with this, stating that there were sufficient facts to conclude that the conversion to general anesthetic prompted the Plaintiff’s cardiac arrest, and that had the Plaintiff been sedated properly and/or calmed down, the c-section could have continued under her regional anesthetic, which means that the Plaintiff would not have needed to be intubated or experience a cardiac arrest. As such, Rowe and O’Bonsawin JJ. stated that the trial judge had considered causation correctly with respect to Dr. Jamesky, and that the Court of Appeal was correct in dismissing the Plaintiffs’ claim against the hospital and Dr. Padmore. Therefore, all appeals before the Court were dismissed.
This brings us to the dissenting opinion of Martin, Karkatsanis and Moreau JJ., written by Martin J. While these three judges agreed that both Dr. Jamesky’s appeal and the Plaintiff’s appeal concerning the hospital were properly dismissed, they concluded that the Plaintiff’s appeal against Dr. Padmore, should be allowed, as the injuries sustained by Ms. Hemmings were not too remote from Dr. Padmore’s negligent conduct[7].
Martin J., in these reasons, embarked upon his own historical analysis of the concept of remoteness in tort. Like the Row and O’Bonsawin JJ, Martin J felt that the concept of causation should only deal with “factual causation”, whereas “legal causation” should be considered separately as part of remoteness, as it speaks to whether the Plaintiff’s injury is too far removed from the alleged negligent act in question. It was noted that when the Supreme Court of Canada last considered remoteness, it concluded that the risk of harm “must be one which would occur in the mind of a reasonable person in the defendant’s position and would not be too far-fetched”. Martin J., like Rowe and O’Bonsawin JJ, as noted above, observed that the concept of remoteness had to be considered in conjunction with duty of care and the “neighbour principal” to ensure that liability is only imposed when “the defendant’s wrong and the plaintiff’s injury can be understood as two sides of the same injustice”. In other words, for an injury not to be too remote, the possibility of these injuries occurring is one of the things that makes the defendant’s conduct unreasonable in the first place. It was noted that this approach is consistent with the aims of “corrective justice”, which underpins the basis of recovery in negligence – i.e. that the person who causes the wrong in question also corrects it. It is also noted that remoteness will require little detailed analysis because the link between the negligent act and the harm caused is obvious, but remoteness becomes more difficult to analyze “at the margins[8]. Lastly, Martin J. concluded that the consideration of remoteness should be characterized as a question of mixed fact and law, rather than of law alone, and that as a result, the standard of appellate review with respect to a trial judge’s consideration of remoteness should be one of “palpable and overriding error” where the trial judge is entitled to deference from an appellate court. In conclusion, Martin J. felt that remoteness should be analyzed using a “risk based” approach, where one considers whether the injury in question falls within the scope of risks which were unreasonable, and as a result negligent, created by the Defendant.
Having regarding for the foregoing, Martin J. concluded that the judgment against Dr. Padmore, the Plaintiff’s own OBGYN, should be restored, to the contrary of the findings of the Ontario Court of Appeal, given that the trial judge was entitled to deference in considering this issue in concluding that it should be reasonably foreseeable to an OBGYN should have foreseen the need for a c-section in a high-risk pregnancy, which potentially gave rise to a risk of anesthetic complications. Martin J. went on to note that remoteness is an “individualized enquiry” for each defendant that need not be reconciled with whether the general type of harm in question was in the reasonable contemplation of another defendant in a different position, at a different time, who was armed with different information. As such, Martin J was of the view that what might have been foreseeable to Dr. O’Brien is not relevant to what might have been foreseeable to Dr. Padmore. Instead, applying the “risk-based” approach to remoteness, Martin J. felt that increased risk of a c-section being necessary, which could result on anesthetic complications and cardiac arrest “fell squarely” within the “class of potential harms” which the Plaintiff could expect to encounter without an appropriate pregnancy plan, or the Plaintiff not being properly advised of the increased risk of her pregnancy, or offered the option to eliminate these risks by terminating her pregnancy. As such, Martin J. concluded that Ms. Hemming’s injuries were not too remote from Dr. Padmore’s breach of the standard of care as her treating OBGYN.
Using the same “risk based” approach to remoteness, Martin J concluded that the Ontario Court of Appeal’s dismissal of the case against the hospital was correct. Specifically, he noted that the symptoms which the Plaintiff reported on April 8, 2009 – that is, frequent vomiting, without any other reported symptoms – do not bear any “apparent relationship” to the Plaintiff having a panic attack on the operating table during the c-section, the subsequent administration of a general anesthetic and the complications arising from same. Instead, the harm that befell the Plaintiff was a whole different set of risks and issues that could not have been apparent to the nurse to whom the Plaintiff spoke.
The final set of reasons delivered in this case came from Chief Justice Wagner. Wagner CJ concluded that none of the Defendants should have been found liable, and that the trial judge made errors that reflected a “serious misunderstanding and misapplication” of factual causation and remoteness, and that the trial judge did not separately consider these two concepts, which resulted in an error reversible on appeal on his part.
Wagner CJ pointed out that the trial judge completed his analysis on remoteness and causation in “only” nineteen (19) paragraphs[9], did not use the words “remoteness” or “factual causation”, and appeared to conflate causation and remoteness by stating that causation is determined by “applying what was reasonably foreseeable”. He also had concerns that the trial judge did not properly appreciate the difference between causation and remoteness.
Specifically, with respect to Dr. Padmore, the trial judge was incorrect to conclude that had this doctor properly warned her of the risks associated with carrying her pregnancy to term, she would not have been injured. While this might on its face satisfy the “but for” test in causation, the trial judge went on to state that a reasonable person might find it foreseeable that the Plaintiff would likely have decided to terminate the pregnancy, thus injecting the “risk analysis” approach for remoteness in to his causation analysis, thus turning the “but for” test from a factual analysis on the balance of probabilities into an analysis where factual causation can be established merely on a “possibility” that the Defendant’s breach may ultimately cause the harm suffered. As, in Wagner’s CJ.’s view, the trial judge had linked an erroneous statement of law to a “key conclusion”, then appellate intervention is necessary. He also felt it that the inconsistency between the trial judge’s finding that the harm sustained by the Plaintiff was not foreseeable to Dr. O’Brien but should have been foreseeable to Dr. Padmore to be a palpable and overriding error.
In terms of the negligence of the hospital, Wagner CJ felt that again, the trial judge mis-stated the legal test for causation, by talking about whether or not the Plaintiff’s injuries were “far-fetched” in terms of the nurse in question’s failure to review the Plaintiff’s medical record and have her seek medical attention on April 8, when causation should be considered on the basis of the balance of probabilities.
Finally, with respect to Dr. Jamesky, it was concluded by Wagner CJ that the trial judge had not properly applied the “but for” test for causation, and gave no consideration at all as to whether the Plaintiff’s injuries were too remote. Instead, he simply concluded that had the Plaintiff not prematurely been given the general anesthetic, she would not have gone into cardiac arrest. Instead, he felt that the evidence before the trial judge, based on the record, was too equivocal to support this conclusion, as there was no evidence that she might not have needed a general anesthetic at all, had alternate measures been used first. Instead, there remained the possibility that the Plaintiff would not have responded to the sedative or being calmed by her sister, and would have needed the general anesthetic in any event – a possibility that the trial judge failed to consider.
That said, Wagner CJ does appear, in his dissent, to agree with the conclusions of the rest of the Court that for an injury to not be too remote, there must be a “real risk” of it happening apparent to the Defendant in question. In this regard, he states in the preamble to his reasons that the task of remoteness is to “identify when a risk is unreasonable so that it will, if it crystallizes into harm, attract liability”.
So…What does it all mean??
In the week or so that these reasons have been in existence, they have been described by others as “dense”, “complex” and “a labyrinth”. All of these descriptions are apt. However, in the author’s view, this case should be treated more as an attempt to clarify the existing law with respect to remoteness, rather than one which materially changes this law. However, it is important to recognize that remoteness and causation are two distinct concepts to be considered separately while carrying out a negligence analysis. Even if a Defendant’s conduct is found, on a “but for” analysis, to have caused the Plaintiff’s damages, unless these damages were reasonably foreseeable to a reasonable person in the Defendant’s position, they will be found too remote, such that the Plaintiff’s claim in negligence will fail.
A review of the judicial history of remoteness shows that the courts have vacillated back and forth between finding negligence for any sort of damage that is a consequent of the negligent act, to finding any damage not within the reasonable contemplation of the Defendant. The trend in in the jurisprudence for at least the past half century appears to have been to follow the Privy Counsel’s reasons in The Wagon Mound (No. 1) and The Wagon Mound (No. 2) and to treat any harm not reasonably foreseeable to the Defendant as being too remote, and in Hemmings v. Padmore, the Supreme Court appears to have followed this approach. As such, this case should be considered akin to “The Wagon Mound (No. 3)”. That said, while remoteness will not be controversial, complex or in dispute in the majority of cases, and is not something that those of us who practice motor vehicle tort law, occupier’s liability or road authority liability will have to consider often, is still an important part of the negligence analysis, particularly for cases which are, as Martin J put it, “at the margins”, such as complex medical malpractice and professional liability cases.
[1] The other three being whether a duty of care is owed, breach of standard of care, and that the Plaintiff has sustained damages
[2] [2026] SCC 36
[3] It was felt the administration of general anesthetic precipitated the Plaintiff’s cardiac event
[4] i.e. “but for the negligence in question, would the harm have occurred?” or in the event of multiple causes, “did a particular defendant’s negligence materially contribute to the harm in question?”,
[5] [1961] AC 288 (Judicial Committee of the Privy Council)
[6] [1967] AC 617 (Judicial Committee of the Privy Council)
[7] However, as Wagner CJ, as will be seen, dismissed the appeal against Dr. Padmore, the majority of the Supreme Court panel heading this matter agreed that he should not be held liable for the Plaintiffs’ damages.
[8] Consider, for example, in a trip and fall where an occupier does not repair a trip hazard, and the Plaintiff falls and hurts themselves. In this situation remoteness is not controversial or difficult to analyze. By comparison, in a case like Mustapha v. Culligan, where the Plaintiff saw a fly in a water cooler jug and went into a psychological tailspin despite not drinking any of the water from the jug. Although the Defendant was negligent in allowing the water jug to be contaminated, the effect on the Plaintiff is arguably not something that Culligan could have reasonably foreseen, nor is it arguably within the scope of unreasonable risk, and instead is too remote from the negligent act in question.
[9] It is not without irony that the author has explained this complex decision in “just” fourteen paragraphs up to this point….












