Loss of chance in healthcare litigation is the loss of an opportunity of a better medical outcome due to the breach of a duty care owed by a healthcare professional to their patient. The rationale of the claim was stated by Smith AJA in Gavalas v Singh[1] (“Gavalas“) citing the High Court in Chappel v Hart[2] that, “The law of negligence may be seen as directed to several purposes, but purposes of compensating the injured and promoting reasonable conduct are prominent among them.” Smith AJA held that a loss of chance for a better outcome met the definition of damage suffered by a plaintiff and avoided the “unreasonable result of excessive compensation or no compensation despite the negligence of the defendant.”[3] This position was endorsed by the NSW Court of Appeal in Rufo v Hosking[4] (“Rufo“) wherein Santow JA stated that the lost chance did not need to satisfy the balance of probabilities, i.e. a plaintiff did not need to show there was a 51% chance that they would have obtained a better outcome; all that was needed is for the better chance to be material. [5] His Honour, cited the judgment of Gaudron J in Naxakis v Western General Hospital[6] stating that, adopting an approach in which a plaintiff who proves there was a 51% chance can receive 100% of their loss, yet the plaintiff who only proves 49% receives nothing is at best “rough justice”. [7] The fairest solution is to award full damages for the plaintiff who proves there is a claim on the balance of probabilities and allowing the percentage proved for any plaintiff who proves 50% or below.[8] Hodgson JA added that the “chance must be inherent in the circumstances, not merely an artefact of the way evidence is presented in the case.”[9] This means that if the lost chance is merely the result of difference in expert opinion, then the plaintiff will fail. What is necessary is for the best medical science to say that the treatment had a quantifiable, or material, chance of success.[10]
The judgements in Gavalas and Rufo were rejected by the NSW Court of Appeal in Gett v Tabet[11](discussed below) and Sydney South West Area Health Service v Stamoulis[12] (“Stamoulis”). That case concerned the delay of ten months in the diagnosis of breast cancer. It was agreed by the parties that the delay resulted in an increase in the chance of metastasis from 38% to 42%. This amounted to a 10% increased chance of the plaintiff developing cancer, which it did and the plaintiff was diagnosed as terminal. [13] In the decision Giles JA employed the terminology of “the risk coming home” to the plaintiff.[14] His Honour cited the decision of Spigelman CJ in Seltsam Pty Ltd v McGuiness[15] where the test was formulated in terms of the increased risk being the cause of the relevant damage, “The ‘possibility’ or ‘risk’ that X might cause Y had in fact eventuated, not in the sense that X happened and Y had also happened, but that it was undisputed that Y had happened because of X.”[16] As Giles JA concluded, “An increase in risk gives rise to a possibility of causation, but the possibility does not of itself make out causation. It may be that in particular circumstances the increase in risk will suffice for satisfaction, on the balance of probabilities, that the increase in risk came home”.[17] In other words, it must be established on the balance or probabilities that there was an increased risk, and then that risk must have materialised.[18]
Where the matter concerns purely the loss of chance of a better medical outcome, the High Court in Tabet v Gett[19] comprehensively ruled against damages being available. In that case, the plaintiff suffered a seizure due to an undiagnosed brain tumour, which, along with subsequent treatment, caused brain damage. She argued that, due to Dr Gett’s breach of his duty of care in relation to her diagnosis (initially chickenpox), she was not sent for a CT scan a day earlier, when her father reported problems with her eyes, and that would have revealed the tumour and averted a proportion of her brain damage. While most of the brain damage was held to have been caused by the swelling and the surgery (which was not alleged to have been performed negligently), the trial judge (Studdert J) agreed that Dr Gett had breached his duty to her, and found 25% was due to the events of that day, insofar as if a CT scan had been performed a day sooner, the presence of the tumour would have been revealed and an alternative treatment regime would have been followed, thus causing the plaintiff to lose the chance of a better outcome. His Honour though, did not find that had Dr Gett’s duty been discharged the plaintiff would not have suffered the seizure.[20] Despite a dearth of expert evidence on the point due to the way the trial was conducted, Studdert J held there was a 40% chance of avoiding 25% of the brain damage, and decided that the plaintiff should receive 10% of the amount of the total damage she suffered.[21]
The decision was appealed to NSW Court of Appeal, which held that loss of chance by itself was not actionable damage; a “material increase in risk was not to be equated with material contribution to the injury and that the plaintiff must establish that it was probable that the risk created by the tortfeasor came home”.[22] As noted above, the court also rejected the decisions in Gavalas and Rufo primarily on the basis that such a change in the position of the law as represented by those cases was “a matter of high policy for the High Court.”[23] The court also held that, “no clear limitations have been formulated for the application of the new doctrine” and was contrary to existing understandings of tort law as expressed in the Civil Liability Act 2002 (NSW).[24] In other words, it would be for either the legislature or the High Court adopt and define a cause of action recognising loss of chance in tort. It also held that even if an action for loss of chance did exist, it would not be proved to the extent of 40% on these facts given the lack of expert evidence; at most the two experts who directly dealt with this would only justify a finding of a 15% chance of avoiding the 25% of brain damage.[25] The position in Gett v Tabet was cited with approval in Stamoulis as representing the law in NSW. [26]
The High Court upheld the decision of the NSW Court of Appeal in Tabet v Gett[27]. The leading judgment of Kiefel J firstly noted the evidential difficulties adverted to by the NSW Court of Appeal stating, "the expert medical evidence did not establish the link between the omission of the respondent, with the consequent delay in treatment, and the brain damage which occurred on 14 January, necessary for a finding of causation. There was no evidence as to what harm might have been caused by the delay."[28]
However, even absent such difficulties, the law should not recognise an action for loss of chance. Her Honour’s reasoning was, firstly, the elements necessary to be shown in a case medical negligence are: 1. a duty of care; 2. a breach of that duty; and 3. damage consequent on said breach.[29] Damage is the “gist of the action,” and while in other torts the violation of a legal right is sufficient to import damage, such cannot be the case in negligence actions where injury or actual loss is necessary.[30] The problem with recognising loss of chance is that it would, “require, a degree of precision in the assessment of probabilities which is not part of the more liberal, common sense, approach presently undertaken.”[31] As her Honour points out, “The standard of proof required by the common law already admits of some uncertainty in proof of causation”[32] insofar as it requires a 51% certainty. Logic follows that if recovery could be quantified for less than 51%, then why could it not be done so for greater than 51%, thereby allowing plaintiffs to only recover the percentage they have proved. This is an explicit rejection of the formulation adopted by Santow JA in Rufo noted above.
Gummow ACJ likewise answered with reference to the core principle that claimants in civil actions must prove their case on the balance of probabilities, and to allow recovery for loss of chance would lessen this burden and place it onto defendants.[33] His Honour also adverted to the fact that such a shifting of burdens may encourage defensive medicine, thus resulting in poorer, not better treatment outcomes for patients.[34] Hayne and Bell JJ and Crennan J agreed with the reasoning of Keifel J and Gummow ACJ.[35]
However, Gummow ACJ and Hayne and Bell JJ recognised the above did not preclude that a loss of chance may supply the requisite damage.[36] Indeed, Hayne and Bell JJ employed the language of the risk coming home, or will more probably than not, to illustrate when a loss of chance is actionable.[37] These statements are in line with the position expressed by the NSW Court of Appeal in Stamoulis, that if a risk of harm is proved on the balance of probabilities, then so long as that risk materialises a plaintiff may recover.
The logic advanced by the decisions in Gavalas and Rufo is essentially one based on upholding the duty care owed to patients by healthcare professionals, or in other words, ensuring the consequences of negligence are coextensive with the content of the duty of the negligent party.[38] Such is directly adverted to in the passage from Chappel v Hart cited with approval by Smith AJA in Gavalas. However, as noted by the High Court in Wallace v Kam[39] the duty of care, while usually coextensive with the consequences of negligence, is not always. In this context, does this lead to a gap in tort law in enforcing the duties it imposes? The cases discussed above are illustrative where the patient, who is suffering from a pre-existing condition (such as cancer), receives treatment that does not adequately address the condition thus increasing the risk of harm resulting from it. From a logical perspective, the patient, by seeking treatment from one doctor has placed trust in that doctor that they will be treated to a particular standard. Likewise, a duty of care is voluntarily assumed by the doctor upon agreeing to treat the patient. The doctor has therefore willingly assumed responsibility for the patient’s health on the basis that they will provide an adequate standard of care. There is a trust reposed by the patient in the doctor that invokes certain fiduciary like obligations discussed elsewhere.[40] If a doctor were to perform a procedure negligently that causes injury, he or she would be held liable. But if they were to perform a procedure that was less efficacious than another they should have performed, or not performed a procedure they should have, then is that breach not actionable? If they have assumed responsibility for the care and preservation of a patient’s health, then why are they not liable for failing to so preserve? The answer of the High Court deriving from Tabet v Gett is that, so long as plaintiffs can show the increased risk caused the harm to eventuate, they are liable. This is so even where the harm is caused by the patient’s own illness not being properly managed. Indeed, so long as a risk does “come home”, healthcare professionals will be liable, as was the case in Stamoulis. The damage in both direct injury and loss of chance scenarios involves being in a worse state than otherwise a patient would have been. The difference then, is merely an evidentiary one. In cases such as Stamoulis the loss of chance of a better outcome due to the risk, could be proved on the balance of probabilities, but it could not in Tabet v Gett. In other words, the High Court did not actually rule out actions for loss of chance. All it did was rule out actions for loss of chance that could not satisfy the balance of probabilities.
In light of the above, it should be said that claims for loss of chance (properly understood) are permitted in Australia. But should claims for loss of chance where a plaintiff cannot prove on the balance of probabilities that the risk came home (such as those in Gavalas and Rufo) be allowed? The logic of the High Court in Tabet v Gett and the NSW Court of Appeal in Gett v Tabet on this basis appears to be unassailable. Firstly, such would effectively amount to an usurpation of the Civil Liability Act 2002 (NSW), s 5E that states, “In proceedings relating to liability for negligence, the plaintiff always bears the onus of proving, on the balance of probabilities, any fact relevant to the issue of causation.” Secondly, it is illogical to state that a less than 51% chance of a harm or loss occurring could be compensated, as that would mean conversely, a court would be over 51% certain that it did not. Once this is realised, the idea of allowing claims for loss of chance in the Gavalas and Rufo sense should be rejected, as such involves a nonsensical reformulation of the very basis of civil law in Australia.
Conclusion
There are two differing kinds of actions in Australia known as loss of chance. The first is that evidenced in Stamoulis and has been approved of by both the NSW Court of Appeal and implicitly by the High Court. This action is one where a plaintiff can show, on the balance of probabilities that a harm or loss occurred due to a risk eventuating that would not have otherwise but for the breach of the duty of care. The risk itself need not be 51%, but it does need to be proved to that standard. The second may be termed Gavalas and Rufo loss of chance and has been rejected in Australia. It is understood to mean that, although a plaintiff has not proved on the balance of probabilities that a risk eventuated in loss or harm, a plaintiff may claim to the extent they have proved the risk of that loss or harm existed due to the breach of a duty of care, or inadequate treatment. This does truly represent a shift away from assessing the duty of care as an element in an action for negligence, and towards its breach being its own cause of action irrespective of harm. In the context of tort law, such an action in the guise of loss of chance should not be allowed. Whether breaches of duties of care should be otherwise actionable is, like all decisions relating to the creation of new areas of law, one of policy, but should not abandon burdens of proof.
Endnotes
[1] [2001] VSCA 23, [38].
[2] (1998) 195 CLR 232, 285.
[3] [2001] VSCA 23, [38].
[4] (2004) 61 NSWLR 678.
[5] Ibid 688 [42].
[6] [1999] HCA 22, [30].
[7] (2004) 61 NSWLR 678, 688-689 [44]-[45].
[8] Ibid 688-689 [45].
[9] Ibid 680-681 [10].
[10] Ibid.
[11] [2009] NSWCA 76, [389].
[12] [2009] NSWCA 153.
[13] Ibid [133].
[14] Ibid [29].
[15] [2000] NSWCA 29.
[16] [2000] NSWCA 29, [119].
[17] [2009] NSWCA 153, [37].
[18] Ibid [138]-[141].
[19] [2010] HCA 12.
[20] Tabet v Mansour [2007] NSWSC 36, [193], [306], [376]-[377] (per Studdert J).
[21] [2007] NSWSC 36, [220], [378], [429]-[434].
[22] Gett v Tabet [2009] NSWCA 76, [254].
[23] Ibid [389].
[24] Ibid.
[25] Ibid [245].
[26] [2009] NSWCA 153, [127].
[27] [2010] HCA 12.
[28] Ibid [114]-[117].
[29] Ibid [108].
[30] Ibid [109].
[31] Ibid [150].
[32] Ibid [148].
[33] Ibid [48].
[34] Ibid [59].
[35] Ibid [65]-[69] (per Hayne and Bell JJ); [101]-[103] (per Crennan J).
[36] Ibid [27] (per Gummow ACJ).
[37] Ibid [69].
[38] Wallace v Kam [2013] HCA 19, [26].
[39] [2013] HCA 19 at [26].
[40] Hillston v Bar-Mordecai [2003] NSWSC 89, [32].
