KS v Willox: Causation by Omission, Regional Standards of Care, and the Causal Chain in a Failed Birth Injury Claim

K.S. was born on October 24, 2000, at 23 weeks’ gestation, weighing just under one and a half pounds. He was born at the Royal Alexandra Hospital in Edmonton after his mother was airlifted from Peace River. By the time of trial he was 15 years old. He had severe cognitive deficits from the loss of white brain matter. He was blind in one eye. He was hypersensitive to external stimuli and could not hold a conversation. He had recently been diagnosed at the Glenrose Hospital with severe Autistic Spectrum Disorder. He would require full-time support for the rest of his life.

A rural general practitioner was found to have breached his standard of care in monitoring the pregnancy. The question was whether that breach caused the harm. The answer was no, and KS v Willox became one of the most searching and difficult causation decisions in Canadian birth injury law.

The Parties and the Procedural History

K.S. is the central plaintiff, represented through his litigation history first by his mother J.S. as next friend, and later (after J.S. withdrew in 2010 and an extended case management process failed to find a replacement) by Tim Wood, a lawyer appointed litigation representative by court order of June 9, 2014. Maria Dudas’s personal claims were discontinued by J.S. on June 24, 2014. K.S.’s father, D.S., played no formal role in the litigation.

The defendants were Dr. David Willox, the Peace River general practitioner who managed J.S.’s pregnancy, and Dr. Brian Muir, the Grande Prairie obstetrician and gynaecologist whom Dr. Willox consulted by telephone on a single occasion. Dr. Willox had retired in 2004 following a brain tumour diagnosis, had undergone four surgeries, continued to experience seizures, and testified that his condition had affected his memory. Dr. Muir had been one of two practising obstetricians at the Grande Prairie Queen Elizabeth II Hospital in October 2000 and its Chief of Surgery from 1999 to 2000.

The trial ran five weeks, heard by Moreau J. of the Alberta Court of Queen’s Bench. The plaintiff appealed; the ABCA (McDonald, O’Ferrall and Veldhuis JJ.A.) heard argument on October 5, 2017, and released its decision on August 21, 2018. The majority (Veldhuis J.A., McDonald J.A. concurring) dismissed the appeal. O’Ferrall J.A. dissented, finding reviewable errors in the causation analysis and an additional breach of the duty to inform, and would have remitted discrete causation issues to trial. The plaintiff sought leave to appeal to the Supreme Court of Canada; the application was dismissed with costs on April 18, 2019.

The Clinical Story: A First Pregnancy in Peace River, October 2000

The Pregnancy and a Concerning Ultrasound

J.S. became Dr. Willox’s patient in 1998. This was her first pregnancy. At her first prenatal visit on July 4, 2000, she told Dr. Willox that her mother had had an incompetent cervix and had undergone a Shirodkar suture (cerclage) for her pregnancies. Dr. Willox recorded “mother incompetent cx Shirodkar suture” in the family history section of J.S.’s prenatal record. He told her that incompetent cervix was not a familial condition, not an inherited predisposition. The pregnancy appeared to progress normally through the summer, apart from minor first-trimester bleeding that resolved.

On October 2, 2000, at approximately 21 weeks’ gestation, J.S. underwent a routine transabdominal ultrasound. The findings were strikingly abnormal for that stage of pregnancy: the internal os was open, with fluid (and the fetal membranes containing the amniotic sac) bulging into the cervical canal. The report flagged the finding as significant. The external os, at the lower end of the cervical canal, was not shown to be open. Dr. Willox called J.S. in for an appointment the next morning.

October 3: The Five-Minute Telephone Consultation

At the October 3, 2000 appointment, Dr. Willox performed a digital pelvic examination and found J.S.’s cervix to be “closed and thick”: the external os was not dilated. J.S. was anxious. She brought up the topic of cerclage. After examining her, Dr. Willox placed a telephone call to Dr. Muir in Grande Prairie. The call lasted approximately five minutes. Dr. Willox relayed the ultrasound results, his clinical examination findings (closed, long and thick cervix), and the fact that J.S. was in her first pregnancy with no prior pregnancy losses. He asked whether J.S. was a candidate for the Shirodkar suture given that she had raised the question herself. Dr. Muir responded “absolutely not.” He recommended an expectant management plan: modified bed rest, abstention from sexual activity, weekly clinical check-ups, and serial ultrasounds. Dr. Willox relayed this advice to J.S., scheduled a follow-up examination, and arranged for a repeat ultrasound on October 25.

Dr. Muir had no personal recollection of the call. He testified as to his standard practice, which was consistent with Dr. Willox’s account. Dr. Muir had not performed emergency cerclage in his own practice; he acknowledged that obstetricians and perinatologists in Edmonton with more experience in the procedure were always available for consultation.

October 13 and 20: Progressive Signs, No Escalation

On the morning of October 13, 2000, J.S. noticed an unusual jelly-like discharge, “like thickened silicone” in her words, which she believed to be part of the mucus plug. She reported this to Dr. Willox at her October 13 appointment. Dr. Willox’s examination found the cervix still closed and thick. He did not make note of the discharge on the prenatal record, did not order an immediate ultrasound (the October 25 date stood), and did not call Dr. Muir or another specialist. The trial judge found, on the evidence, that the discharge was consistent with the passage of part of the mucus plug and was a significant clinical sign of progressive cervical changes, one that warranted specialist consultation.

On October 19 and 20, 2000, J.S. noticed blood in her discharge. She attended Dr. Willox’s office on October 20. Again his examination found the cervix closed and thick. He told her spotting could be normal at that stage, that she would learn more from the October 25 ultrasound, and that he would know more then. No ultrasound was ordered immediately. No specialist was consulted. The trial judge found this to be a second occasion on which specialist consultation was required but not obtained.

October 22: A Crisis

On October 21, J.S. experienced painless twinges she initially attributed to constipation. The following morning she felt something was wrong and went directly to the Peace River Hospital emergency department, arriving at 12:15 p.m. on October 22. Her cervix was 7 centimetres dilated. A contemporaneous ultrasound showed the membranes bulging through the internal os, which was widely open, as was the external os. She was started on medication to slow contractions and flown to the Royal Alexandra Hospital in Edmonton, arriving at 7:18 p.m.

At 8:30 p.m. on October 22, Dr. Okun, the perinatologist on call at the RAH, examined J.S. Her consultation note recorded “possible incompetent cervix” and the observation of hour-glassing membranes. She determined that cerclage was not available because of the position of the placenta, evidence of leaking membranes, and active bloody discharge. She noted “suture next pregnancy.” J.S. recalled asking about the stitch at the RAH and being told she was past the point of no return; everything was already bulging.

Approximately three hours before K.S.’s birth (at around 4:00 a.m. on October 24), J.S.’s temperature rose to 38 degrees C, the first clinical sign of the chorioamnionitis (ChorioA) infection that the placental pathology later confirmed. K.S. was born at 6:56 a.m. on October 24, 2000. The umbilical cord was not inflamed, but pathology showed severe chorioamnionitis of all layers of the extraplacental membranes and very early endovasculitis of fetal surface vessels on the chorionic plate.

K.S.’s Injuries and Condition at Trial

K.S. spent the first five months of his life in hospital. His conditions included patent ductus arteriosus (requiring surgical repair), severe hyaline membrane disease, severe hypotension, white matter brain injury (periventricular leukomalacia), Stage 4 retinopathy of prematurity with retinal detachment of the right eye, acute renal failure, blocked urethral valves, inguinal hernia, and multiple infections. He was discharged on February 22, 2001 with a complex diagnostic picture.

At trial, K.S. was 15 years old, living with his parents in Peace River, enrolled in Grade 9 with a team of educational aides. His IQ was measured below the 0.1 percentile. He had virtually no verbal communication, using a small number of signs and sounds. He used a wheelchair and could manage certain daily tasks with minimal guidance but required continuous supervision. His parents’ home had been modified at birth to be wheelchair accessible. His computer skills were notably developed; his mother described his typing as “phenomenal.” He had been diagnosed in 2015 with severe Autistic Spectrum Disorder and an intellectual disability in the moderate-to-severe range. He did not interact socially in a functional way and reacted severely to sudden or loud noises. Experts agreed he would never be gainfully employed and would require full-time support throughout his life.

The Standard of Care Findings

The Legal Framework: Rural Practitioners and the Locality Principle

Moreau J. addressed the standard of care applicable to a rural general practitioner managing an obstetrical patient in 2000. The standard is presumptively the same as that of an urban counterpart, subject to demonstrated differences in access to facilities, equipment or staff. She noted the narrowing force of national professional examinations and continuing education requirements, citing Crawford (Litigation Guardian of) v Penney for the proposition that the locality rule should be abandoned where modern communication and medical education had produced effective uniformity of standards. Importantly, geographic limitation does not shield a physician from inconvenience, only from genuinely unavailable technology or expertise. In Dr. Willox’s case, there was no suggestion that telephone consultation with Edmonton perinatologists or Grande Prairie obstetricians was unavailable. It was his standard practice.

The standard of care must be assessed at the time of the alleged negligence, not with hindsight. As Moreau J. put it, adapting the observation of Lord Denning: “we cannot look at the 2000 incident with 2016 spectacles.” This principle governed the entire expert evidence analysis. The question was not what a physician ought to have known or done in 2016 when the litigation came to trial, but what the reasonably prudent physician in the relevant position ought to have known and done in October 2000.

Dr. Muir: Standard of Care Met

Moreau J. found that Dr. Muir did not breach the standard of care he owed to J.S. on October 3, 2000. His recommendation of expectant management on the basis of the information relayed to him by Dr. Willox (a patient in her first pregnancy, no prior pre-term losses, clinical examination showing a closed and thick external os, an abnormal ultrasound showing the internal os open with funneling) was a reasonable exercise of clinical judgment. There were no SOGC or ACOG guidelines in October 2000 regarding cerclage based on ultrasound findings. In Edmonton and Northern Alberta in 2000, cerclage was not being performed on the basis of ultrasound findings alone without prior indicators of cervical incompetence. Expectant management was the only reasonable treatment option in those circumstances. Dr. Willox was entitled to rely on Dr. Muir’s expertise and was not required to seek a second opinion or question Dr. Muir’s qualifications.

Dr. Willox: Standard of Care Breached in the Follow-Up Phase

Moreau J. accepted that Dr. Willox met the standard of care through October 3: he saw J.S. within 24 hours of receiving the ultrasound, examined her, and consulted with an appropriate specialist. But she found he breached his duty in what followed. Specifically (para 293):

Dr. Willox was required to schedule serial ultrasounds at least every two weeks after the October 2, 2000 abnormal ultrasound; he did not. He was required to order an immediate ultrasound when J.S. reported the unusual jelly-like discharge at the October 13 visit; he did not. He was required to consult Dr. Muir or another specialist when she reported that discharge; he did not. He was required to order an immediate ultrasound when she reported spotting at the October 20 visit; he did not. He was required to consult a specialist at the October 20 visit; he did not.

The logic was straightforward: the expectant management plan Dr. Muir recommended contemplated serial ultrasounds precisely because progressive changes to the cervix were what mattered. Dr. Willox knew at the time that his clinical examinations could not track internal cervical changes. The progressive symptoms on October 13 and 20 (discharge consistent with the mucus plug, then spotting) were exactly the warning signs that the plan was supposed to detect and escalate. By failing to escalate, Dr. Willox deprived J.S. of the possibility that progressive changes would have been identified and acted upon in time.

The Causation Analysis: Where the Claim Failed

Was the Pre-Term Birth Caused by Incompetent Cervix or Infection?

The defendants’ alternate causation theory was that J.S.’s pre-term delivery was caused by a subclinical infection (ChorioA) rather than an incompetent cervix. If the infection pre-dated the cervical changes, then no treatment of the cervix could have prevented the pre-term delivery.

Moreau J. rejected the infection-first theory (para 361). Applying the robust and pragmatic approach from Snell v Farrell, she found it more probable than not that J.S. had an incompetent cervix that caused the pre-term delivery, with the ChorioA infection contracted afterward through the opening cervix. Key factors: clinical infection was absent at every appointment through October 20; the twinges J.S. experienced were out of proportion to her very advanced cervical dilatation (7 cm), consistent with incompetent cervix rather than infection-induced labour; the hour-glassing of membranes noted by Dr. Okun is a classic presentation of incompetent cervix not seen in active labour; the umbilical cord showed no inflammation, indicating the infection had not yet reached fetal tissues; and the fever appeared only approximately three hours before delivery.

Was Ultrasound-Indicated Cerclage Available in Edmonton and Northern Alberta in 2000?

This was the central fact-finding that determined the outcome of the case. The plaintiff’s theory was that if Dr. Willox had scheduled serial ultrasounds after October 2 (or consulted a specialist after October 13 or October 20), progressive cervical changes would have been tracked, and J.S. would have been offered an ultrasound-indicated cerclage based on those findings.

Moreau J. found that ultrasound-indicated cerclage was not being performed in Edmonton or Northern Alberta in October 2000 for patients without a prior history of pre-term births or other recognized indicators of cervical incompetence (para 272). She accepted the direct evidence of Dr. Chari (a perinatologist practicing in Edmonton) and Dr. Muir that the procedure was not available in those circumstances. The November 2003 Okun survey was a national survey, not a study of actual cases, and not specific to Edmonton or Alberta. Dr. Dansereau’s evidence about practice in Victoria, British Columbia, illustrated different regional practices, not Alberta practice. There were no national guidelines on cerclage based on ultrasound findings in October 2000.

The significance of this finding was absolute: if ultrasound-indicated cerclage was not being performed in Edmonton or Northern Alberta in 2000, then serial ultrasounds or specialist consultation after October 13 or October 20 would not have changed J.S.’s treatment. The only option available to a specialist seeing J.S. in that period would have been to maintain the expectant management plan and watch for J.S.’s condition to evolve to the point where the Williams (1997) definition of incompetent cervix was met.

When Could Incompetent Cervix Have Been Diagnosed?

The Williams Obstetrics definition accepted by the court characterized incompetent cervix as “painless cervical dilatation in the second trimester or perhaps early in the third trimester with prolapse and ballooning of membranes into the vagina.” J.S.’s cervix did not meet this definition at the October 3, 13, or 20 appointments; at each of those dates, her external os was closed and thick on clinical examination. The only time at which her condition fell within the Williams definition was sometime between her October 20 appointment (when the external os was still closed) and approximately noon on October 22 (when she presented at the Peace River Hospital with 7 cm dilatation and hour-glassing membranes).

That was a window of two days, and J.S. was 200 kilometres from an Edmonton hospital where an emergency cerclage could have been performed, with a 24-hour waiting period required before cerclage (to rule out infection) built into the standard management protocol.

Would J.S. Have Received a Cerclage in the Available Window?

Moreau J. found it too speculative to conclude that J.S. would have received an emergency cerclage in the very narrow window of opportunity (para 371). The reasons were cumulative: J.S. resided in Peace River and would have required transport to Edmonton; the window between when her external os started to open (after October 20) and when Dr. Okun determined cerclage was no longer available (8:30 p.m. on October 22) was extremely short; best practice required waiting 24 hours to observe for signs of infection before placing the cerclage, further compressing the window; and by the time J.S. reached the RAH, complications of placental position and leaking membranes had already ruled out the procedure.

The conditional finding on patient consent (para 386), that a reasonable patient in J.S.’s circumstances would have chosen cerclage if offered, did not assist the plaintiff because the court found cerclage would not, on a balance of probabilities, have been offered and performed in time.

Would Emergency Cerclage Have Prevented the ChorioA?

Even proceeding on the assumption that J.S. had received an emergency cerclage in the narrow available window, Moreau J. found it not established that the cerclage would have prevented the ChorioA infection (para 393). The external os would already have been open when the cerclage was placed; there would already have been opportunities for bacterial ascent from the vagina. The 24-hour observation period would have further narrowed the window between the cervix opening and when cerclage could be placed. Had infection set in, the cerclage would have had to be removed. The trial judge found the plaintiff had failed to establish that emergency cerclage on an already-open cervix would have prevented ascending infection in these circumstances.

Would Cerclage Have Extended the Pregnancy and Reduced K.S.’s Injuries?

This part of the causation chain, assessed in the alternative on the plaintiff’s theory, was the only element on which the plaintiff substantially succeeded. Moreau J. found (para 433) that a conservative estimate of pregnancy extension from emergency cerclage was approximately five weeks, to 28–29 weeks’ gestation. At 28–29 weeks, it was more probable than not that K.S. would have avoided the extensive brain damage he suffered (para 485). The risk of serious long-term handicap at the time of his actual delivery (23–24 weeks) was approximately 50%; at 30 weeks it was approximately 10%. Every week counted.

Applied with a 20% negative contingency (15% for significant impairment risk inherent in any 28–29 week birth and 5% for subtle impairments), this counterfactual finding anchored the provisional damages assessment.

What Caused K.S.’s Brain Injury?

The brain injury (periventricular leukomalacia and loss of white matter) was found to have been caused post-natally, not by the ChorioA infection in utero (paras 472–473). Dr. Poskitt (pediatric neuroradiologist) and Dr. MacNab (neonatologist) gave evidence, not contradicted by the defendants’ witnesses, that neonatal infection and hypotension were far more powerful causes of white matter injury in very premature infants than ChorioA. K.S.’s blood test results at birth were negative for bacteria; the umbilical cord was not inflamed; there was no sign of fetal distress in the hours before delivery; K.S. was relatively vigorous at delivery, responded well to resuscitation, and was well-perfused. The white matter damage was bilaterally symmetrical, a systemic insult pattern inconsistent with direct infection and consistent with the blood pressure fluctuations caused by mechanical ventilation of severely immature, stiff lungs. The progressive deterioration of white matter observed in serial ultrasounds (from “perfectly normal” on October 24 and October 30 to showing enlarging ventricles on November 24) was consistent with a postnatal insult occurring in the first week or two of life.

The ASD, which had only been diagnosed in 2015, was found to be causally connected to K.S.’s premature birth and brain injury.

Informed Consent

The plaintiff also advanced a failure-to-inform claim. The trial judge found neither Dr. Muir nor Dr. Willox was required to present ultrasound-indicated cerclage as an available treatment option, because it was not a reasonable option in Edmonton or Northern Alberta in 2000 for a patient in J.S.’s circumstances. A physician is not required to advise patients of fringe treatment alternatives, only of reasonable alternatives. Since the only cerclage that might have been available was an emergency cerclage once the Williams definition was met (after October 20), and that window passed before J.S. reached the RAH, the informed consent claim failed alongside the causation claim.

The majority at the ABCA accepted this analysis without separately addressing the informed consent ground.

O’Ferrall J.A., dissenting, found that the informed consent analysis was infected by the same causation error he identified in the main liability analysis, and additionally found a distinct breach by Dr. Willox arising from J.S.’s specific questions. He cited Hopp v Lepp, confirmed in Reibl v Hughes, for the principle that a physician must answer specific questions put by a patient, even if those questions concern possible risks, regardless of whether the treatment inquired about is what the physician regards as the best option. J.S. had asked Dr. Willox about cerclage at least twice before October 3, and again explicitly at the October 3 appointment where Dr. Willox’s response was: “we don’t need to worry about that, we’re going to wait and see.” The trial judge accepted J.S.’s evidence that she had not received a substantive discussion about the procedure’s risks and benefits and that she would have remembered it if she had. O’Ferrall J.A. held that Dr. Willox’s duty was not exhausted by passing along Dr. Muir’s “absolutely not”: he owed J.S. a more complete explanation of why cerclage was not recommended, what the risks of not pursuing it were, and the option of seeking a second opinion. This was a separate ground of liability that did not depend on cerclage being a “reasonable treatment option” in the Alberta practice landscape of 2000.

The Court of Appeal: Majority and Dissent

The Majority: No Palpable and Overriding Error

Veldhuis J.A. (McDonald J.A. concurring) dismissed the appeal on the standard of palpable and overriding error applicable to findings of fact. The majority identified two core factual findings that determined the causation outcome: that ultrasound-indicated cerclage was not available in Edmonton or Northern Alberta in 2000; and that it was unclear whether emergency cerclage would have prevented the ChorioA infection. Each had been reached by the trial judge through a reasoned choice between competing bodies of expert evidence, exactly what trial judges are required to do. Appellate courts defer to that choice unless it is unreasonable or plainly wrong.

Working through each alleged breach and each additional causation theory advanced on appeal, the majority found that all causation arguments ultimately depended on overturning at least one of the trial judge’s fact findings. None could be overturned. Even if Dr. Willox had ordered more frequent ultrasounds or consulted a specialist after October 13 and October 20, those steps would not have produced cerclage, because cerclage was not being offered in Edmonton for patients in J.S.’s circumstances, and because J.S.’s external os remained closed until sometime after October 20.

O’Ferrall J.A.’s Dissent: Gaps in the Causation Analysis

O’Ferrall J.A. took a different view of the trial judge’s causation analysis. He accepted the findings of fact themselves, including the finding that ultrasound-indicated cerclage was not available in Edmonton in 2000. His concern was that the trial judge had not adequately answered the questions that arose from those findings.

In particular: the trial judge had focused her causation analysis on what would have happened if a specialist had been consulted after October 13. She had not separately considered what would have happened if a specialist had been consulted after October 20, two days before J.S.’s cervix was found to be 7 cm dilated. A referral to an Edmonton specialist on or just after October 20, equipped with successive ultrasounds tracking the progressive changes and a clinical record of the discharge and spotting, might have resulted in a different clinical picture when J.S. arrived at the RAH. The trial judge appeared to have assessed causation as if J.S.’s condition on the evening of October 22 was the fixed reference point; but had she been referred and managed differently in the preceding days, that condition might not have been the same.

O’Ferrall J.A. also found the trial judge’s causation analysis problematic to the extent she had emphasized Dr. Willox’s digital clinical examination findings (cervix “closed and thick”) at the expense of what serial ultrasounds would have revealed. A digital examination can only assess the external os; it gives no information about internal cervical changes. Serial ultrasounds, which the standard of care required, would have provided that information. The trial judge’s conclusion that the expectant management plan would not have changed “based on” what the digital examinations found was therefore circular: the very breach was the failure to obtain the information that would have informed that decision.

On these bases, O’Ferrall J.A. would have remitted two specific causation questions to the trial judge: (1) but for Dr. Willox’s breaches of duty of care or duty to inform, would J.S. have received a cerclage before it was too late?; and (2) if J.S. had received a timely cerclage, would the ChorioA infection necessarily have manifested and, if so, would it have precluded a finding of causation? He was clear that he could not himself answer these questions on the record before the court, because the findings of fact made at trial simply did not provide the necessary stepping stones, and that a directed retrial on discrete issues was the appropriate remedy.

The Provisional Damages Assessment

Although the action was dismissed, Moreau J. assessed damages provisionally against the possibility of reversal on appeal. The assessment is a significant obiter contribution to the quantification of damages for a severely disabled child born at the extreme edges of viability.

The damages were to be assessed on the footing that a 5-week extension to 28–29 weeks’ gestation would have been achieved, with a 20% negative contingency applied to reflect the risk of impairment inherent in any birth at that gestational age (15% for significant impairment, 5% for subtle impairments), though the 15% was ultimately applied only to future earning capacity and the application to other heads was reserved for further argument.

Non-pecuniary damages were assessed at $361,100 (the maximum at the time) on the basis that K.S. had sufficient cognitive awareness to appreciate his disability and be frustrated by it. Future loss of earning capacity was assessed provisionally at $1,765,636 after the 15% negative contingency. Future care costs were assessed provisionally at substantial amounts for accommodation and equipment. A structured settlement was ordered.

The claim for future loss of shared family income was denied. The plaintiff established the first element, that K.S. had lost the capacity to enter into a relationship of financial interdependency, but failed to provide an evidentiary foundation for the second element: the economic value of that lost opportunity. The plaintiff’s expert had not accounted for multiple factors that would affect the quantum, including the possibility of other relationships with financial consequences and the prospect of raising children. The loss was real but the claim was insufficiently evidenced.

An in-trust claim was not advanced. J.S. had signed a release in Dr. Munro’s favour in December 1994 in the Dudas litigation; actually that’s Dudas not this case. In KS v Willox, J.S.’s personal claims were discontinued in June 2014 after she withdrew as litigation representative.

Costs and Supplementary Decisions

The two costs decisions (2016 ABQB 654 and 2017 ABQB 71) document the post-trial cost proceedings. The defendants were successful and entitled to their costs. Moreau J. fixed J.S.’s personal liability for costs from the commencement of proceedings to June 9, 2014, when she was replaced as litigation representative. Tim Wood, appointed by court order, was absolved from costs liability from that date. The bill of costs was assessed in Column 5 with a multiplier of 2, reduced by 25% to account for the differential attributable to J.S.’s personal claim (which would have fallen within a lower column). The case’s complexity, length (five weeks), amount at stake (well over five times the Column 5 threshold), and the need for multiple expert witnesses all justified the Column 5 tariff and a multiplier, while the plaintiff’s partial success on the standard of care issue, the financial hardship of J.S. and D.S. (who would need to support K.S. throughout his lifetime), and Dr. Willox’s breach were acknowledged in setting the multiplier at 2 rather than the defendants’ requested 5.

Expert fees for consultants who were retained but did not testify at trial (Drs. Cave, Iglesias, Liston, and Hahn) were partially allowed. Travel expenses were allowed for client meetings in Grande Prairie and Peace River and for expert consultations, with the exception of a chartered flight to Peace River in January 2010 (to meet with Dr. Willox after his brain tumour diagnosis), which was disallowed in favour of a standard commercial fare.

What KS v Willox Establishes

Principle One: Causation by Omission Requires a Completed Causal Chain

KS v Willox is one of the most detailed applications in Canadian birth injury jurisprudence of the causation-by-omission framework. Where negligence consists of a failure to act, the plaintiff must establish not merely that the defendant should have acted differently, but that the act that should have been taken would, on a balance of probabilities, have prevented the harm. This requires working through the entire hypothetical causal chain, in this case through six distinct links: Would reasonable treatment have led to cerclage being offered? Would J.S. have chosen it? Would it have prevented the ChorioA infection? Would it have extended the pregnancy? Would the extended pregnancy have reduced K.S.’s disabilities? And at each link, each probability conditions the probability at the next.

Dr. Willox’s breach was real. The missed consultations and missed ultrasounds were genuine failures. But causation collapsed at the third link: whether cerclage would have been offered and performed in time. This is the fundamental point for practitioners: proving breach is not enough. The causal chain from breach to harm must be traced with evidentiary particularity.

Principle Two: Regional Medical Practice Evidence Can Defeat National Standards Arguments

One of the most practically significant findings in KS v Willox is the acceptance of evidence that a treatment modality, ultrasound-indicated cerclage, was available and practiced in other parts of Canada in 2000 but was not available in Edmonton and Northern Alberta. This evidence was given by two expert witnesses with direct knowledge of Alberta practice and was preferred over a national survey, over evidence from a British Columbia perinatologist, and over evidence of studies in the literature supporting the treatment’s effectiveness.

The lesson for birth injury plaintiffs is that evidence of “what other obstetricians across Canada were doing” may be insufficient if the defendants can call credible local witnesses with direct knowledge of what was actually being practiced in the relevant community in the relevant year. Where a treatment is emerging nationally but has not been adopted at the relevant time and place, the counterfactual (“she would have been offered the procedure if referred to Edmonton”) is not established merely by proving the procedure was available somewhere. Regional practice evidence, corroborated by national guidelines (or the absence of them), can be decisive.

Principle Three: The Standard Must Be Assessed at the Time of the Events

Moreau J.’s reminder that “we cannot look at the 2000 incident with 2016 spectacles” is a direct application of the principle in ter Neuzen v Korn, that a physician’s conduct must be judged by the knowledge and standards applicable at the time of the alleged negligence. KS v Willox adds particular texture to that principle in cases involving emerging technologies or evolving treatment modalities. Cerclage itself was well established by 2000. What was not established, in Edmonton and Northern Alberta, at that time, was the specific variant of ultrasound-indicated cerclage for patients with no prior pre-term births presenting with an abnormal ultrasound. The plaintiff’s experts described national and international trends toward the procedure. The court found those trends had not translated into Alberta practice in time to be relevant to J.S.’s care.

Principle Four: ASD as a Late-Presenting Consequence of Premature Birth

The KS v Willox trial decision is a notable authority for the proposition that ASD, diagnosed years after birth, can be found as a matter of causation to be connected to extreme prematurity and associated brain injury. K.S. was diagnosed with ASD in June 2015, when he was nearly 15 years old, more than 14 years after his birth. Moreau J. found the causal connection to his premature birth and brain damage established. This is significant for practitioners in ongoing birth injury litigation where the full picture of a child’s disabilities may not emerge until years after the events at issue, and for the purposes of limitation period analysis and damages assessment.

Principle Five: The Duty to Answer Specific Patient Questions (The O’Ferrall Dissent)

Although O’Ferrall J.A. was in the minority, his analysis of Dr. Willox’s duty to inform J.S. about cerclage articulates a principle of general application. Where a patient repeatedly and specifically asks about a treatment option, the physician’s duty is not discharged by conveying a specialist’s one-word refusal. The patient who asks specific questions is entitled to specific answers: what the procedure entails, why it is or is not recommended, what the risks of not having it are, and the possibility of a second opinion. This is drawn directly from Hopp v Lepp and Reibl v Hughes. O’Ferrall J.A.’s point is that whether cerclage was a “reasonable treatment option” in the local medical landscape is relevant to the standard of care in treatment, but it does not fully dispose of the disclosure obligation when a patient has asked specific questions about it. That obligation may exist independently. Practitioners and medical defence counsel should note that this analysis, while dissenting, reflects established doctrine from the Supreme Court of Canada’s informed consent trilogy.

KS v Willox in the Birth Injury Series

KS v Willox is cited in the Birth Injury Claims Guide in this series for its treatment of the causal chain analysis in birth injury cases, the regional standard of care question, and the ASD-prematurity connection. It should be read alongside Cheung v Samra, https://canlii.ca/t/jmzv5 which also involves failure to advance delivery in the face of a developing fetal complication and also resulted in a $14.9 million judgment (after a more direct causal chain was ultimately established on appeal). The contrast is instructive: in Cheung, the causal chain collapsed at the proof stage but was ultimately rescued by the Court of Appeal’s interpretation of the jury’s verdict; in KS v Willox, the causal chain collapsed at the factual level because a key intermediate step (the availability of the treatment that should have been offered) was not established. Together, the two cases illustrate the most common points of failure in birth injury causation analysis.

The case also engages the broader causation framework developed in Snell v Farrell, https://canlii.ca/t/1fstw and Clements v Clements, https://canlii.ca/t/frvld both cited extensively by Moreau J. and the ABCA. The robust and pragmatic approach does not eliminate the requirement of evidence; it simply permits inferences to be drawn where the plaintiff has established a factual foundation. In KS v Willox, the factual foundation for the critical inference, that cerclage would have been performed in time, was found to be absent. The “robust and pragmatic” approach could not bridge that gap.

Quick Reference: KS v Willox

The Five Decisions

• 2016 ABQB 483 — https://canlii.ca/t/gt5bl — Moreau J.; action dismissed; Dr. Willox breached standard of care but causation not established; provisional damages assessed
• 2016 ABQB 654 — https://canlii.ca/t/gvrd3 — Moreau J.; costs: Column 5 multiplier of 2, reduced 25%; J.S. liable to June 9, 2014; Tim Wood absolved thereafter
• 2017 ABQB 71 — https://canlii.ca/t/h4bfn — Moreau J.; costs disbursements: expert fees partially allowed; chartered flight disallowed
• 2018 ABCA 271 — https://canlii.ca/t/htrbz — Veldhuis J.A. (McDonald J.A. concurring): appeal dismissed; O’Ferrall J.A. dissenting: would have remitted causation issues to trial
• 2019 CanLII 32860 (SCC) — https://canlii.ca/t/hzv7h — SCC leave denied with costs; April 18, 2019; 2018 ABCA 271 is final

Key Parties

• Plaintiff: K.S. (infant born October 24, 2000), by litigation representative Tim Wood (appointed June 9, 2014, replacing J.S.)
• Key defendants: Dr. David Willox (Peace River GP); Dr. Brian Muir (Grande Prairie OB/GYN)
• Other defendant (not liability finding): Dr. Okun (RAH perinatologist, not a defendant; perinatologist on duty at RAH who excluded cerclage on October 22)

The Timeline

• October 2, 2000: Routine transabdominal ultrasound at 21 weeks shows internal os open with funneling; abnormal finding suggesting pre-term risk
• October 3, 2000: Dr. Willox examines J.S. (external os closed and thick); 5-minute phone consult with Dr. Muir; cerclage declined (“absolutely not”); expectant management recommended
• October 13, 2000: J.S. reports jelly-like discharge (part of mucus plug per trial judge); cervix still closed; no ultrasound ordered; no specialist consulted (BREACH)
• October 20, 2000: J.S. reports spotting since previous evening; cervix still closed; no ultrasound ordered; no specialist consulted (BREACH)
• October 22, 2000 (noon): J.S. presents to Peace River Hospital emergency, 7 cm dilated; flown to RAH Edmonton; by 8:30 p.m. Dr. Okun rules out cerclage
• October 24, 2000 (4:00 a.m.): Fever, the first clinical sign of chorioamnionitis
• October 24, 2000 (6:56 a.m.): K.S. born at 23 weeks’ gestation; under 1½ pounds

K.S.’s Disabilities

• Severe cognitive deficits; intellectual disability in the moderate-to-severe range (IQ below 0.1 percentile)
• Periventricular leukomalacia (white matter loss); severe Autistic Spectrum Disorder (diagnosed 2015)
• Blindness in right eye; significantly reduced vision in left eye (legally blind)
• Hypersensitivity to external stimuli; no verbal communication; wheelchair-dependent
• Requires full-time support; will never be gainfully employed

The Standard of Care Findings

• Dr. Muir: DID NOT breach standard of care; expectant management was the only reasonable option; cerclage not available in Edmonton or Northern Alberta for first-pregnancy patients without prior pre-term births
• Dr. Willox: BREACHED standard of care, by not scheduling bi-weekly ultrasounds, not ordering immediate ultrasounds after October 13 and October 20 findings, and not consulting specialists on October 13 and October 20

Why Causation Failed

• Ultrasound-indicated cerclage was not being performed in Edmonton or Northern Alberta in 2000 for patients without prior pre-term births, so serial ultrasounds or specialist consultation would not have produced cerclage
• J.S.’s condition did not meet the Williams (1997) definition of incompetent cervix until sometime after October 20; by 8:30 p.m. on October 22 cerclage was no longer available
• The window for emergency cerclage was extremely narrow; J.S. resided 200 km from Edmonton; a 24-hour observation period was required before cerclage; transport logistics made performance in the available window speculative
• It was not established that emergency cerclage would have prevented the ChorioA infection on an already-open cervix

What the Provisional Damages Showed

• Non-pecuniary: $361,100 (maximum at the time)
• Future loss of earning capacity: $1,765,636 (after 15% negative contingency)
• Future loss of shared family income: DENIED; no evidentiary foundation for quantum
• ASD found causally connected to premature birth and brain injury
• Structured settlement ordered

Key Legal Principles

• Causation by omission requires completing the hypothetical causal chain link by link; proving breach is not sufficient; the plaintiff must show each step in the “but for” counterfactual was more probable than not
• Regional evidence of local medical practice can defeat arguments based on national or international trends in treatment; what was “emerging” in Canada in 2000 was not necessarily “available” in Edmonton or Northern Alberta
• The standard of care must be assessed at the time of the negligence, not with the benefit of subsequent guidelines or practice developments
• ASD, even when diagnosed years after birth, can be found as a matter of causation to be connected to extreme prematurity and associated brain injury (relevant for limitation periods and ongoing damages assessment)
• A physician’s duty to answer specific patient questions under Hopp v Lepp and Reibl v Hughes may exist independently of whether the treatment asked about is a “reasonable treatment option”; the O’Ferrall J.A. dissent articulates a potentially broader disclosure obligation

Related Cases in This Series

• Cheung v Samra, 2022 ONCA 195 — https://canlii.ca/t/jmzv5 — IUGR and failure to advance delivery date; causation in perinatal brain injury; jury verdicts in medical negligence; article in this series
• Snell v Farrell, 1990 CanLII 70 (SCC) — https://canlii.ca/t/1fstw — the “but for” test, the robust and pragmatic approach, and inference of causation
• Clements v Clements, 2012 SCC 32 (CanLII) — https://canlii.ca/t/frvld — SCC clarification of the “but for” test; material contribution to risk
• Ter Neuzen v Korn, [1995] 3 SCR 674 — https://canlii.ca/t/1frhk — the standard of care assessed at the time of treatment; professional practice as the baseline standard
• Hopp v Lepp, [1980] 2 SCR 192 — https://canlii.ca/t/1mjv6 — the duty to answer specific patient questions; article in this series
• Reibl v Hughes, [1980] 2 SCR 880 — the modified objective test for informed consent causation; article in this series
• Bailey v Young, 2022 NSSC — birth injury; obstetrical malpractice in Nova Scotia; companion article in this series
• Malinowski v Schneider, 2012 ABCA — https://canlii.ca/t/fslb9 — Alberta obstetrical malpractice; duty of referring physician; article in this series