Lorencz v Talukdar: Failure to Refer, Wait Times, and Causation

In a medical malpractice case, establishing that the defendant physician breached the standard of care is only half the work. The plaintiff also has to establish that the breach caused the harm. The framework sounds straightforward at the level of general principle. The harder question, the question that decides many cases, is how the trier of fact is supposed to work out what would have happened if the defendant had met the standard. The counterfactual analysis. The reconstruction of a world in which the negligence did not occur.

Lorencz v Talukdar, 2024 SKCA 105, is a clean illustration of what happens when the counterfactual reconstruction defeats the plaintiff even where the breach is clear. The Saskatchewan Court of Appeal dismissed the family’s appeal on November 15, 2024. The trial judge had found that Dr. Sneha Prabha Talukdar breached the standard of care by failing to inquire about and follow up on Mr. Lorencz’s ongoing chest pain at a December 2004 visit. But the trial judge also found that the family could not establish, on the balance of probabilities, that a timely referral to a cardiologist would have occurred quickly enough to prevent Mr. Lorencz’s fatal heart attack the following January. The Court of Appeal affirmed both findings. The appeal was dismissed with costs to the defendant.

The case is doctrinally significant for several reasons. It illustrates the symmetrical operation of the Snell v Farrell “robust and pragmatic” causation framework: the same framework that can save a plaintiff’s case where the precise mechanism is contested can also defeat the plaintiff’s case where the evidentiary record will not support even a robust and pragmatic inference. It articulates the “wait time evidence” framework that applies in failure-to-refer cases. It demonstrates the appellate standard of review for causation findings (the Housen v Nikolaisen palpable and overriding error standard) operating in a context where the plaintiff’s appellate position rested on factual rather than legal grounds. And it sits as a useful counterpoint in the cluster’s growing collection of cases where the same legal framework operates symmetrically — sometimes for the plaintiff, sometimes against.

The facts

The decision under review is the Saskatchewan Court of King’s Bench trial decision in Lorencz v Talukdar, 2022 SKKB 258, which followed a four-day trial in September 2022.

Mr. James Lorencz was 49 years old in late 2004. He had been a patient of Dr. Talukdar’s family medicine practice. The clinical record showed:

• He had completed two Burns index forms (self-administered cardiovascular risk screening tools) at his family medicine visits
• He had seen Dr. Talukdar on December 22, 2004 and had reported “ongoing pain”
• The chart entry for that visit was thin: Dr. Talukdar’s evidence at trial was that she had assumed she knew what the patient was referring to but had not asked specific follow-up questions
• He had also seen Dr. Babchuk
• There was no referral to a cardiologist initiated by Dr. Talukdar during this period
• On January 23, 2005, a month after the December 22 visit, Mr. Lorencz suffered a massive heart attack
• He died several weeks later from the cardiac event and its complications

The family commenced a malpractice action against Dr. Talukdar and Dr. Babchuk, alleging that the failure to investigate the chest pain and to refer to a cardiologist constituted negligence that caused Mr. Lorencz’s death.

The expert evidence

The trial proceeded over four days. The family called Ms. Lorencz and two experts: a family physician who provided standard-of-care evidence (qualified in the field of family medicine) and a cardiologist who provided opinion evidence on whether the death could have been prevented by appropriate medical management.

The defendants called Dr. Talukdar and Dr. Babchuk on their own behalf, plus two experts: a family physician qualified in the standard of care for family medicine and a specialist in internal medicine with expertise in cardiology who addressed both the standard of care and the causation question.

The causation evidence is the operative material for the appeal. The defendants’ internal medicine and cardiology expert (Dr. Sharma) gave evidence that a routine referral appointment with a cardiologist in Saskatchewan could take “weeks or months.” The plaintiffs’ cardiology expert (Dr. Iwanochko) gave evidence on the preventability of the death generally but the record on the timing of any actual referral pathway was limited.

The trial judge’s findings

The trial judge proceeded methodically through the two main issues: standard of care and causation.

On standard of care, the trial judge found a breach. The reasoning, distilled:

• Mr. Lorencz reported “ongoing pain” to Dr. Talukdar on December 22, 2004
• Dr. Talukdar did nothing with that information and did not ask further questions
• Dr. Talukdar perhaps made an assumption about what the patient was referring to
• But “assumptions do not satisfy the standard of care; specific questioning is in order to deal with that which is relayed by the patient”
• The chart entries did not show any form of inquiry or discussion to identify the nature of the pain
• The Burns index forms had information that should have prompted further investigation but did not
• The cumulative effect was a failure to meet the applicable standard

This is a careful, specific articulation of a family physician’s positive obligation to investigate symptoms that are reported. The principle is not that every chest pain produces a cardiac referral. The principle is that a physician who hears about chest pain has an obligation to characterize it: to ask the location, the quality, the radiation, the triggers, the relieving factors, the associated symptoms, and to integrate the patient’s risk profile. Without that characterization, the physician cannot make a defensible clinical decision about whether to investigate further or to refer.

On causation, the trial judge worked through the counterfactual analysis. The reasoning, distilled:

• Had Dr. Talukdar appropriately engaged in differential diagnosis and determined to refer to a specialist, the options would have been an emergent referral or a routine referral
• An emergent referral would only have been made if the patient was exhibiting serious and immediate symptoms during the visit
• The expert evidence did not establish that the symptoms present on December 22, 2004 met the emergent threshold
• In the absence of an emergent indication, the patient would have been referred on a routine basis
• The extent of the evidence at trial was that a routine cardiology referral in Saskatchewan could take “weeks or months”
• There was no precise evidence about when a referral would have been completed even if started in December
• Mr. Lorencz had his heart attack on January 23, 2005 — approximately one month after the December 22 visit
• The trial judge could not conclude on the balance of probabilities that the referral, the investigations, and the medical opinion necessary to prevent the death would have occurred within that one-month window
• Even applying the “robust and pragmatic” approach to causation endorsed by the Supreme Court of Canada in Snell v Farrell, the inference required would amount to speculation

The trial judge accordingly dismissed the family’s claim on the causation analysis, with costs to the defendant.

The Court of Appeal decision

The family appealed the trial judge’s finding on causation. The Court of Appeal addressed the appeal through two questions:

(a) Did the trial judge apply the correct test for determining causation?

The Court of Appeal reviewed the law in detail and concluded that the trial judge did not err in law in his application of the “but for” test. The framework is settled. Clements v Clements, 2012 SCC 32, confirms that but-for causation is the operative standard in negligence cases. Snell v Farrell, [1990] 2 SCR 311, allows the trier of fact to draw a robust and pragmatic inference where direct scientific proof is not available. The trial judge had applied both correctly: he had asked the but-for question, he had considered whether a robust and pragmatic inference could close the evidentiary gap, and he had concluded that it could not because the gap was too wide.

The Court of Appeal observed that Snell v Farrell is not a free pass. The robust and pragmatic approach does not authorize speculation. It permits the trier of fact to infer causation where the evidence as a whole supports the inference even if direct proof is unavailable. Where the evidence does not support even a robust and pragmatic inference, the framework does not save the plaintiff’s case.

(b) Was the trial judge’s finding that Dr. Talukdar’s negligence did not cause Mr. Lorencz’s death the product of a palpable and overriding error?

The Court of Appeal applied the Housen v Nikolaisen, 2002 SCC 33, framework for review of factual findings. The standard for appellate intervention in factual findings is palpable and overriding error: an error obvious on the face of the record that affected the outcome. Mere disagreement with the trial judge’s assessment of the evidence is not enough.

The Court of Appeal found no palpable and overriding error. The trial judge had considered both urgent and non-urgent referral pathways. He had concluded, based on the expert evidence, that the symptoms present at the December 22 visit did not meet the urgent referral threshold. He had concluded, based on the wait time evidence, that a non-urgent referral would not have produced cardiology assessment, investigations, and treatment within the month between December 22 and the January 23 cardiac event. The conclusion was supported by the record. It was not the only possible conclusion, but it was a permissible one. The appellate court accordingly deferred.

The appeal was dismissed with costs in favour of the defendant.

The doctrinal anchors

Several doctrinal points emerge from the case.

Breach without causation. The case is a clean example of a malpractice claim where the standard of care analysis goes in the plaintiff’s favour but the causation analysis defeats the claim. The two elements operate independently. A finding of breach does not automatically produce a finding of causation; each element has to be established on its own. Lorencz is in the same doctrinal family as several cluster cases where the standard of care was met (defeating the claim at the first step) and is doctrinally complementary to plaintiff wins where both elements were established.

For a different example of the same separation operating against the plaintiff, see Williamson v Dr Y (BC Court of Appeal), Papineau v Sharma, and Noel v Hawrylyshyn. For a plaintiff win where both elements were established and the Snell framework saved the case, see Hasan v Trillium Health Centre.

The “wait time evidence” framework in failure-to-refer cases. This is the most operationally significant principle in Lorencz. In a failure-to-refer case, the plaintiff carries the burden of proving not just that the referral should have been made but also that the referral would have produced the necessary clinical action in time to prevent the harm. The proof requires evidence about:

• Whether the clinical picture supported an urgent referral or only a routine one
• What the wait times are for the relevant specialty in the relevant jurisdiction
• What the receiving specialist would have done with the referral and on what timeline
• What investigations or treatments would have followed and how long they would have taken
• Whether the resulting clinical pathway would have intercepted the harm in time

Each of these is an evidentiary requirement. The plaintiff cannot assume the wait time. The plaintiff cannot assume the specialist would have acted quickly. The plaintiff cannot assume the investigations would have produced a particular result on a particular timeline. The plaintiff must adduce evidence on each step.

In Lorencz, the family adduced general evidence on the preventability of the death but did not adduce the specific evidence on the referral pathway timeline. The wait time evidence that came out at trial (a routine cardiology referral in Saskatchewan could take “weeks or months”) came primarily from the defence cardiology expert. The plaintiff’s expert evidence on the timing of the counterfactual was thin enough that the trial judge could not work out, even with the Snell robust and pragmatic approach, whether a referral begun on December 22 would have produced the necessary intervention before January 23.

The principle is broadly applicable. In every failure-to-refer case, the wait time question is central. The plaintiff who does not address it adequately in evidence is at substantial risk of the Lorencz outcome.

The symmetrical operation of Snell v Farrell. The case demonstrates that the “robust and pragmatic” approach to causation is not a free pass for plaintiffs. It is a framework that allows the trier of fact to draw a permissible inference where the evidence supports it. Where the evidence does not support even a permissible inference, the framework does not save the plaintiff.

The cluster now has three cases that illustrate this symmetrical operation:

• Hasan v Trillium Health Centre (plaintiff success): the Snell framework supported the plaintiff’s case where the defendant’s own conduct had created the evidentiary gap
• Noel v Hawrylyshyn (defendant success): the Snell framework did not save the plaintiff’s case where the operative urgency provided a sufficient defence
• Lorencz v Talukdar (defendant success): the Snell framework did not save the plaintiff’s case where the wait time evidence made the counterfactual inference speculative

The pattern is doctrinally important. Plaintiff counsel relying on Snell to bridge an evidentiary gap need to make sure the evidence supports the inference. Where the gap is too wide, the framework provides no help.

Counterfactual reconstruction in failure-to-refer cases. Lorencz is the doctrinal counterpoint to Gumbley v Vasiliou, where the counterfactual reconstruction worked for the plaintiff. In Gumbley, the trial judge could identify the specific call that should have been made (to the on-call intensivist Dr. Warner by 22:00h), the specific clinical response that would have followed (a respiratory therapist transferring the patient, the intensivist following asthma guidelines, intubation within a defined window), and the specific outcome that would have resulted (the catastrophic hypoxic brain injury would have been avoided). Each step of the counterfactual was supported by specific evidence.

In Lorencz, the counterfactual reconstruction required filling in too many blanks: when the referral would have been received, when the cardiologist would have responded, when investigations would have been completed, when interventions would have been initiated. The general wait time evidence (“weeks or months”) was not specific enough to support a finding that the chain of events would have been completed in the month available before the cardiac event. The counterfactual collapsed under the weight of the speculation required.

The lesson for plaintiff counsel: counterfactual reconstruction is the operational task of causation. The more specific the evidence at each step, the stronger the case. General preventability evidence is rarely enough on its own; specific timing evidence is what carries the burden.

The appellate standard for causation findings. The Court of Appeal’s application of the Housen v Nikolaisen palpable and overriding error standard to the trial judge’s causation finding is consistent with the established appellate framework. Causation findings are factual or mixed fact and law. Appellate courts defer to the trier of fact’s assessment unless palpable and overriding error appears on the face of the record.

The implication for appellate strategy: a plaintiff appealing a no-causation finding from trial faces a substantial uphill task. The argument needs to identify specific reviewable error, not just disagreement with the trial judge’s assessment of the evidence. Pure factual disagreements are unlikely to succeed on appellate review.

The emergent versus routine referral threshold. A useful operational point that emerges from the case is the distinction between emergent and routine referral pathways. The emergent pathway is reserved for cases where serious and immediate symptoms are present during the visit. The routine pathway is the default. The thresholds for moving from routine to emergent are clinical judgments calibrated to the specific presenting picture.

For the prescribing or referring physician, the implication is that the choice between emergent and routine is a clinical decision that needs to be made on the evidence available at the time. For the trier of fact in subsequent litigation, the implication is that the choice depends on what the symptoms actually were, not on what they might have been with more thorough investigation. Lorencz illustrates that even where the physician should have investigated more thoroughly (the standard-of-care breach), the resulting hypothetical referral pathway depends on what would have been found, not on what the catastrophic outcome later revealed.

Why the wait time evidence is so important

The wait time evidence question deserves a separate treatment because it operates as the practical hinge of failure-to-refer litigation across Canada.

Specialist wait times in Canada have been a recognized policy problem for decades. The Wait Time Alliance, the Canadian Institute for Health Information, and various provincial health quality bodies all track wait times for major specialties and procedures. In broad terms:

• Cardiology consultations can vary from days (urgent referrals) to many months (routine referrals)
• The Saskatchewan picture that emerged in Lorencz (routine referrals taking weeks or months) is not specific to Saskatchewan; it is roughly consistent with the national pattern
• The Ontario picture is broadly similar; urgent cardiology consultation in major centres can occur within days, routine consultation often takes weeks to months
• The wait time is not just for the consultation but for the investigations and treatments that follow

For failure-to-refer cases, the wait time evidence has several operational implications:

Evidence of urgency thresholds. What clinical features would have triggered an urgent rather than routine referral? Expert evidence is required. The threshold is jurisdictionally and clinically calibrated.

Evidence of actual wait times in the specific jurisdiction at the specific time. The trial judge in Lorencz needed evidence about Saskatchewan in 2004-2005, not generalized wait time policy. The plaintiff needs to adduce evidence specific enough to support the inference.

Evidence of the post-consultation timeline. Even after the consultation occurs, the investigations and treatments take time. The plaintiff’s counterfactual needs to address the whole timeline, not just the initial appointment.

Evidence of the receiving specialist’s likely action. What would the receiving cardiologist (or whoever) have actually done in the counterfactual? Generic preventability evidence is not enough. Specific evidence about the likely clinical pathway is needed.

In practical terms: plaintiff counsel should plan, from the outset of a failure-to-refer case, to adduce specific evidence on each of these dimensions. The general preventability evidence is necessary but not sufficient. The wait time and pathway-specific evidence is what carries the causation burden.

What this means for prospective clients

A few practical observations for families considering a malpractice claim in a failure-to-refer fact pattern.

A breach of the standard of care is necessary but not sufficient. Finding that the physician should have done something differently is the first step. The harder question is whether doing it differently would have changed the outcome. Both questions have to be answered favourably for a successful claim.

The evidence is where the case is won or lost. Specific evidence on the counterfactual is what determines whether the Snell v Farrell robust and pragmatic approach can save the case where direct proof is not available. The work of building the counterfactual is the operational heart of the case.

Wait time evidence is specific to the time and place. A current wait time is not the same as the wait time in the year the events occurred. A wait time in one province is not the same as in another. A wait time for one specialty is not the same as for another. The evidence needs to be calibrated.

Counsel needs to plan the counterfactual from the outset. The investigation phase should include not just the standard-of-care analysis but the counterfactual reconstruction. What would have happened with appropriate care? What evidence supports each step of that reconstruction? The earlier this planning starts, the stronger the case.

Not every preventable death is actionable. This is hard to say but it is the practical reality. A death that could have been prevented with appropriate care is the starting point. A death that the evidence shows would have been prevented with appropriate care, on the balance of probabilities, is what the law requires. The gap between those two is where many failure-to-refer cases fall short. For more on the framework for evaluating these cases, see Suing for Medical Malpractice in Ontario: What You Need to Know and Six Common Misunderstandings About Medical Malpractice in Ontario.

What this means for physicians

A few observations from the other side of the file, for any family physicians or other primary care providers reading this.

Document the investigation, not just the conclusion. A chart entry that records only the physician’s conclusion (“no concerning features”) is much weaker evidence than one that records the specific questions asked and the specific answers given. The trial judge in Lorencz was unable to find that Dr. Talukdar had adequately investigated the reported chest pain because the chart did not reflect any specific inquiries. Whatever Dr. Talukdar may have asked at the visit, the absence of the documentation worked against her.

Treat reported symptoms as a starting point, not an endpoint. “Ongoing pain” reported by a patient is the beginning of a clinical conversation, not the end of it. The chart should reflect the questions that follow: location, quality, radiation, triggers, relieving factors, associated symptoms, severity, frequency, the patient’s interpretation. The “assumptions do not satisfy the standard of care” principle from Lorencz is a clean statement of the duty.

Use the risk screening tools fully. The Burns index forms in Lorencz contained information that should have prompted further investigation. The forms exist to flag cardiovascular risk. Where the forms produce flags and the physician does not act on them, the failure to act becomes a documented standard-of-care breach.

Make the urgency decision explicit. Where a referral is appropriate, the choice between emergent and routine should be a deliberate clinical judgment, articulated in the chart and communicated to the patient. Where the referral is routine, the patient should be informed about the expected wait time and about what to do if symptoms worsen in the interim.

Cluster integration

The “breach without causation” cluster now includes:

• Lorencz v Talukdar (Saskatchewan; failure-to-refer; wait times defeated causation)
• Williamson v Dr Y (BC; anaesthesia; multiple grounds dismissed)
• Papineau v Sharma (Ontario; multiple grounds dismissed)

The Snell v Farrell “robust and pragmatic” framework — symmetrical operation:

• Hasan v Trillium Health Centre (plaintiff success; defendant’s conduct created the gap)
• Noel v Hawrylyshyn (defendant success; operative urgency)
• Lorencz v Talukdar (defendant success; wait time evidence speculative)

The counterfactual reconstruction framework:

• Gumbley v Vasiliou (plaintiff success; specific evidence at each step)
• Lorencz v Talukdar (defendant success; speculation required)

Cardiac and cardiology practice area (new for the cluster):

• Lorencz v Talukdar establishes the failure-to-refer-to-cardiology framework
• Connects to broader cardiac malpractice issues including delayed acute coronary syndrome diagnosis

The wait-times-as-causation-barrier sub-anchor (new):

• Lorencz v Talukdar is the principal cluster authority
• Connects to the Communication Failures commentary which addresses the systemic dimension of referral and continuity failures
• Connects to the Hallway Medicine in Ontario commentary which addresses system pressure as context for SOC analysis

---

Decision Date: November 15, 2024 (Court of Appeal). Trial decision: November 1, 2022 (Court of King’s Bench).

Jurisdiction: Court of Appeal for Saskatchewan (appellate review of Court of King’s Bench decision)

Citations:

• Lorencz v Talukdar, 2024 SKCA 105 (CanLII) — appellate
• Lorencz v Talukdar, 2022 SKKB 258 (CanLII) — trial

Outcome: Trial judge found breach of standard of care by family physician for failing to investigate and refer for chest pain symptoms but dismissed the claim on causation grounds, finding that the plaintiff had not established on the balance of probabilities that a referral would have produced the necessary clinical intervention within the available window before the fatal cardiac event. Court of Appeal dismissed the family’s appeal, finding (a) no error of law in the trial judge’s application of the but-for causation test, and (b) no palpable and overriding error in the factual finding on causation.

Key authorities: Clements v Clements, 2012 SCC 32 (but-for causation); Snell v Farrell, [1990] 2 SCR 311 (robust and pragmatic causation framework); Housen v Nikolaisen, 2002 SCC 33 (appellate standard of review for findings of fact and mixed fact and law); Athey v Leonati, [1996] 3 SCR 458 (multiple causes); Sacks v Ross, 2017 ONCA 773 (what-likely-happened framework in sequential failures).