*1597Opinion
—In a medical malpractice action predicated upon delayed diagnosis and treatment of lung cancer, the jury rendered a verdict in favor of plaintiff Kenneth Dumas and against defendants David Cooney, M.D. and Alfred Spivack, M.D. It awarded plaintiff damages of $321,400. The trial court entered judgment accordingly, and defendants appeal. Defendants contend that the trial court erred by instructing the jury on the theory of lost chance. We agree and conclude that the error was prejudicial. We therefore reverse the judgment and remand for a new trial.
Background
In 1986, following an annual physical examination, Dr. Spivack diagnosed lung cancer in plaintiff. A surgeon then removed a tumor and part of plaintiff’s right lung. Plaintiff underwent treatment, but the disease recurred in May 1989. Plaintiff died on November 26, 1989, six days after the jury rendered its verdict.1
Plaintiff’s theory at trial was that defendants, who were partners, negligently failed to diagnose his cancer in 1984. In September of that year, plaintiff was treated by Dr. Cooney for a separate illness and referred to a radiologist for a chest X-ray. The radiologist reported to defendants that there existed something of unknown significance in plaintiff’s lung and recommended further study in three months’ time. According to plaintiff, defendants failed to communicate the radiologist’s recommendation, and he therefore did not undergo another X-ray until his physical examination in 1986.
Plaintiff’s expert witness opined that plaintiff’s tumor was a “stage 3A” tumor in 1986 but a “stage 1” tumor in 1984. He explained that a patient having a tumor resected at stage 1 has about a 67 percent chance of being alive and free of disease in five years. He continued that approximately 67 percent of those patients who survive disease-free for five years will survive disease-free for ten years which would be considered a cure for the form of lung cancer suffered by plaintiff.
Plaintiff’s expert also testified that a patient having a tumor resected at a more advanced condition known as stage 2 has about a 33 percent chance of *1598being alive and disease-free in five years. He estimated that plaintiff’s chance of a five-year disease-free survival was 15 to 20 percent.
On cross-examination, the witness acknowledged that only about 10 percent of patients diagnosed with the form of lung cancer suffered by plaintiff are diagnosed at stage 1. He agreed that there are relatively few five-year survivors because there are few stage 1 patients and the mortality rate of lung cancer is nearly identical to the incidence because of the low rate of cure.
Defendants’ principal expert witness testified that all lung cancer is essentially incurable because it spreads before the primary tumor can be detected. He opined that plaintiff’s tumor began spreading in 1974 and could not have been detectable until relatively late in its biological life.
Plaintiff divided his argument to the jury into three parts: liability, causation, and damages. He argued causation as follows: “Now, I’d like to address the question of causation in this case, and the major issue regarding causation is if the resection of the tumor and the removal of the right upper lung had occurred in 1984, or shortly after three months from September of 1984 . . . would it have had a beneficial effect on the length and quality of [plaintiff’s] life?
“Now, the plaintiff’s evidence in this area relies on the statistical studies and the medical literature that forms the basis for the staging system and the basis for the step by step treatment of lung cancer in patients at that time. . . .
“The testimony is that if [plaintiff] had been diagnosed and resected in 1984 that in [the expert’s] opinion [plaintiff] probably would have had a stage I with no positive nodes. . . .
“So even if the lymph nodes were positive in ’84 . . . there is also an opportunity to have a stage II rather than a stage III, which is a better prognosis for [plaintiff].
“Now, [the expert] then testified as to what the prognosis would be for various stages. He said that if [plaintiff] was a stage I, that his prognosis would have been about 67 percent, about two-thirds for a five year survival disease free. . . .
*1599“Now, he mentioned again if it was a stage II he would have one in three chances to go five years disease free, and at stage III, which [plaintiff] was, it left him with a prognosis of 15 percent or less going five years disease free. And we do know from the evidence that in less than two-and-a-half years recurrence was documented in [plaintiff’s] case back in April.”
Plaintiff continued his argument on causation by emphasizing evidence which discredited defendants’ expert witness and then argued damages as follows: “Now, I would like to make a few comments on damages, and my comments in this case are going to be relatively brief . . .
“A case like this is something of a departure from the usual injury case. Usually in an injury case something occurs and a direct injury follows. In a case like this we are not talking about the defendants being responsible for the plaintiff’s cancer. That has never been anything we tried to imply in any way in our testimony. What we are talking about is the effect of a delay in diagnosis. You are going to receive a legal instruction concerning how this applies to the damage part of the case, and that instruction will read as follows:
“Damages for negligence in providing medical care may be based on evidence that it is a reasonable medical probability that the plaintiff would have been benefitted by the possible cure, possible lengthening of his life, and/or improved personal comfort from more prompt diagnosis and treatment.
“Basically what this instruction is saying that with this kind of testimony and this kind of evidence you can draw an inference, and the inference, of course, will be drawn on the basis of probability. And if you draw that inference, then you can award damages for the probability that the plaintiff would have been benefitted for the possible cure, possible lengthening of his life and/or improved personal comfort from more prompt diagnosis and treatment.
“Now, this approach I think is reflected in two areas of damages we are going to talk about. ...”
Plaintiff then outlined claims for past and future job-related economic benefits and past and future pain and suffering.
Defendants began their argument on causation as follows: “Our position will be that unfortunately even if, even if the physicians were negligent that *1600there is no causation in this case. The injury is due to a most unfortunate disease that [plaintiff] has that is not curable. And I think you have heard all of the evidence, and I will put it all in perspective I hope for you when we get to that. Unfortunately, that is exactly what it is. There is no causation here.”
Defendants concluded their argument on damages with the following remarks: “One last item in terms of the damages. You are going to be told by the court that you are going to be allowed if you find liability to compensate [plaintiff] for any loss of chance. And again it is the same standard. You have to prove it by the preponderance of the evidence, and so on. I submit to you that that is something you are going to have a hard time evaluating. How do you put a value on that? You have to use your judgment. If you find there is liability but loss of chance is, say, 10 percent and you say the damages are a million dollars, is the damage award a hundred thousand? What is it?”
The trial court instructed the jury in the language of BAJI No. 2.60 (plaintiff has the burden of proving by a preponderance of the evidence that defendants’ negligence was a legal cause of damage to plaintiff) and BAJI No. 3.76 (legal cause is a cause which is a substantial factor in bringing about the injury). It then gave 10 instructions on the subject of negligence and 10 instructions on the subject of damages. The instruction on damages which plaintiff’s counsel read to the jury during argument was given at plaintiff’s request over defendants’ objection.
The jury rendered its verdict by a 10 to 2 vote. The award consisted of $90,000 for past pain and suffering, $20,000 for future pain and suffering, $150,000 for past loss of earnings, and $100,000 discounted to a present value of $61,400 for loss of future earnings.
The Instruction
The issue before us concerns the propriety of giving the jury the instruction assailed by defendants.
The trial court told the jury: “Damages for negligence in providing medical care may be based on evidence that it is a reasonable medical probability that the plaintiff would have benefitted by possible cure, possible lengthening of his life, and/or improved personal comfort from more prompt diagnosis and treatment.”
Defendants focus upon the concept of “possibility” and contend that the charge permitted the jury to find for plaintiff on the theory of lost chance, a *1601theory heretofore unrecognized in California that permits recovery even though the evidence shows no more than a better-than-even chance that a defendant caused a plaintiff’s loss.2
Plaintiff counters that (1) the instruction was a mere instruction on damages rather than one which relaxed the traditional burden required to prove causation, (2) if it was error to give the instruction the error was not prejudicial, or (3) this court should adopt the theory of lost chance and approve the instruction.
Derivation of the Instruction
We observe that plaintiff adapted the instruction from one at issue in Pulvers v. Kaiser Foundation Health Plan, Inc. (1979) 99 Cal.App.3d 560 [160 Cal.Rptr. 392]. That case was a wrongful death action predicated upon a doctor’s delayed diagnosis of the decedent’s condition. The trial court gave the following instruction at the plaintiffs’ request: “ ‘Liability for negligence in providing medical care may be based upon an inference that it is a reasonable medical probability that the decedent would have been benefited, .. . by possible lengthening of his life . . . even if no cure would have resulted from more prompt diagnosis and treatment.’ ” (Id. at pp. 565-566.) The jury rendered a verdict for the plaintiffs, but the trial court granted a new trial on the ground that the instruction was erroneous. On plaintiffs’ appeal from the order, the court reversed. The defendants’ principal argument in support of the trial court’s conclusion was that the use of the word “liability” in the instruction removed from the jury’s consideration the issue of negligence and imposed an absolute liability. The court explained: “The jury was carefully instructed, at length, on the definition of negligence *1602in a malpractice case. The instruction followed that elaborate definition and did no more than to tell the jury that ‘liability for negligence' could be found under the circumstances set forth. We cannot accept defendants’ theory that a reasonable jury would have construed the instruction as negating all of the earlier instructions. The obvious meaning of the language used told the jury that, if they found negligence they could find causation from the inference set forth in the rest of the instruction.” (Id. at p. 566, fn. omitted.)
Thus, the issue raised in Pulvers was whether an instruction on causation operated to direct a finding on negligence. Since the principal issue before us is whether an instruction on causation or damages correctly states the law, Pulvers does not assist us. Nor does the case from which the instruction in Pulvers was derived.
In Cullum v. Seifer (1969) 1 Cal.App.3d 20 [81 Cal.Rptr. 381], the trial court granted the plaintiff’s motion for a new trial in a medical malpractice action predicated upon delayed diagnosis. On defendant’s appeal from the order, the court affirmed holding that sufficient evidence supported the order. It acknowledged that the evidence on proximate cause was weak, but noted that expert testimony indicated “with reasonable medical probability that the disease had not affected other lymph areas prior to [tire plaintiff’s first consultation with the doctor], and that ‘a delay would make [the plaintiff] need more treatment.’. . . The delay itself prevented more certain proof by plaintiff that had treatment been instituted sooner, her disease would have been arrested or cured. It seems only reasonable to find that an inference may be drawn that plaintiff and her condition fell within that group wherein prompt treatment would have ‘confined [her disease] to one place’ and ‘improved the prognosis.’ ” (Id. at p. 26.) The court concluded: “We cannot narrowly construe the testimony relative to early prognosis and prompt treatment. We observe testimony throughout the record . . . which, if believed, would support the inference that it is a reasonable medical probability that the plaintiff would have been benefited, i.e., by possible lengthening of her life and/or her personal comfort—even if no cure would have resulted from more prompt diagnosis and treatment. These are matters of injury proximately caused by the delay, and support the granting of a new trial.” (Id. at pp. 27-28, fn. omitted.)
Although Cullum tacitly supports plaintiff’s theory that the instruction before us is one on damages and not proximate cause, our analysis of the question reveals that the theory of lost chance is implicated no matter how an instruction in the language of Cullum is characterized. Again, since lost chance was not raised in Cullum, the case provides no guidance.
*1603The Not-better-than-even Chance
Liability for medical malpractice is predicated upon a proximate causal connection between the negligent conduct and the resulting injury. (Budd v. Nixen (1971) 6 Cal.3d 195, 200 [98 Cal.Rptr. 849, 491 P.2d 433].) “[C]ausation must be proven within a reasonable medical probability based upon competent expert testimony. Mere possibility alone is insufficient to establish a prima facie case. [Citations.] That there is a distinction between a reasonable medical ‘probability’ and a medical ‘possibility’ needs little discussion. There can be many possible ‘causes,’ indeed, an infinite number of circumstances which can produce an injury or disease. A possible cause only becomes ‘probable’ when, in the absence of other reasonable causal explanations, it becomes more likely than not that the injury was a result of its action. This is the outer limit of inference upon which an issue may be submitted to the jury.” (Jones v. Ortho Pharmaceutical Corp. (1985) 163 Cal.App.3d 396, 402-403 [209 Cal.Rptr. 456].)
“Proof of causation is particularly difficult in cases where the results complained of are such as might normally be expected to follow from the original disease or injured condition, as is the case when a plaintiff complains of the defendant doctor’s failure to cure, rather than of any positive effects of mistreatment. Plaintiffs in such cases are faced with the difficulty of obtaining and presenting expert testimony that if proper treatment had been given, better results would have followed.” (Annot., Medical Malpractice: “Loss of Chance” Causality (1987) 54 A.L.R.4th 10, 17.)
In this type of case where the testimony establishes a probability of a better result (usually the patient’s survival) absent a doctor’s negligence, a finding for the plaintiff is consistent with existing principles of proximate cause. (See Herskovits v. Group Health Co-Op. (1983) 99 Wn.2d 609 [664 P.2d 474, 485] (cone. opn. of Pearson, J.); see e.g., Burford v. Baker (1942) 53 Cal.App.2d 301 [127 P.2d 941] [where failure to timely diagnose hip injury resulted in bone deformity, proximate cause was established by testimony that prompt treatment would have greatly enhanced the probability of a good result].)
On the other hand, in cases where the evidence indicates that there is less than a probability, i.e., a 50-50 possibility or a mere chance, “that proper conduct on the part of a medical practitioner would have produced results different from those complained of, the courts have been faced with the question whether the evidence is sufficient to establish the causation component essential to the plaintiff’s right of action.” (Annot., Medical Malpractice: “Loss of Chance” Causality, supra, 54 A.L.R.4th at pp. 17-18.)
*1604The most fundamental argument against the traditional rule of causation as it operates in these types of cases is that it bars tort recovery “on behalf of the survivors of many potentially terminal patients, no matter how blatant the health care provider’s negligence. Through negligence, a physician or other health care provider could reduce a patient’s chances of survival from as high as fifty percent to, for example, ten percent, and yet remain unanswerable in the law of tort.” (Perez v. Las Vegas Medical Center (1991)_ Nev. _ [805 P.2d 589, 591].)
When faced with a lost chance and the traditional “all or nothing” prospect, some courts have allowed recovery by adopting a relaxed standard of causation permitting recovery when the plaintiff establishes a “substantial possibility” of causation. (See Fennell v. Southern Maryland Hosp. (1990) 320 Md. 776 [580 A.2d 206, 210-211], and cases cited therein at fn. 4.) This approach effectively creates a new tort with a more lenient threshold of causation as its distinguishing feature. (Ibid.) “Courts adopting the relaxed causation/new cause of action approach continue to award ‘all or nothing’ damages, and when the plaintiff establishes a ‘substantial possibility’ that the doctor’s negligence caused the death, there is full recovery. The result is that relaxing the rules of causation merely improves the plaintiff’s odds of receiving all rather than nothing.” (Id. at p. 210.)3 On the other hand, courts have also expressly retained the traditional rule of causation in medical malpractice cases when confronted with the lost chance argument. (See Fennell v. Southern Maryland Hosp., supra, 580 A.2d at p. 215; see cases cited in Falcon v. Memorial Hosp. (1990) 436 Mich. 443, fn. 11 [462 N.W.2d 44, 64] (dis. opn. of Riley, C. J.).)
Other courts addressing the issue have adopted the lost chance theory first suggested by Professor King in King, Causation, Valuation, and Chance in Personal Injury Torts Involving Preexisting Conditions and Future *1605Consequences (1981) 90 Yale L.J. 1353. (See e.g., Perez v. Las Vegas Medical Center, supra, 805 P.2d at pp. 592-593; Falcon v. Memorial Hosp., supra, 462 N.W.2d at pp. 56-57; see cases cited in Fennell v. Southern Maryland Hosp., supra, 580 A.2d at p. 212, fn. 8.) Under this approach, the injury to be redressed is not the plaintiff’s death or failure to have a better result, but the loss of a chance at survival or a better result. “ ‘[A] claim under that theory does not involve the creation of a new tort, but rather involves a redefinition of damages involved in the claim. Traditional principles of law relating to duty, breach, causation, and burden of proof remain the same—what changes is the acceptance of the concept that damages may be recovered for the loss of a chance of survival where that chance is substantial and can be identified and quantified (and thus valued) without resort to conjecture or speculation.’ ” (Fennell v. Southern Maryland Hosp., supra, 580 A.2d at p. 212, citing Weimer v. Hetrick (1987) 309 Md. 536 [525 A.2d 643, 653] (cone. opn. of McAuliffe, J.).)
Professor King explains the concept as follows: “Causation has for the most part been treated as an all-or-nothing proposition. Either a loss was caused by the defendant or it was not. Inexplicably, the all-or-nothing approach of the causation inquiry has been allowed to slip its analytical moorings, influencing the identification and valuation of losses in cases involving preexisting conditions and claims for future consequences. A plaintiff ordinarily should be required to prove by the applicable standard of proof that the defendant caused the loss in question. What caused a loss, however, should be a separate question from what the nature and extent of the loss are. This distinction seems to have eluded the courts, with the result that lost chances in many respects are compensated either as certainties or not at all.
“To illustrate, consider the case in which a doctor negligently fails to diagnose a patient’s cancerous condition until it has become inoperable. Assume further that even with a timely diagnosis the patient would have had only a 30% chance of recovering from the disease and surviving over the long term. There are two ways of handling such a case. Under the traditional approach, this loss of a not-better-than-even chance of recovering from the cancer would not be compensable because it did not appear more likely that [sic] not that the patient would have survived with proper care. Recoverable damages, if any, would depend on the extent to which it appeared that cancer killed the patient sooner than it would have with timely diagnosis and treatment, and on the extent to which the delay in diagnosis aggravated the patient’s condition, such as by causing additional pain. A more rational approach, however, would allow recovery for the loss of the chance of cure even though the chance was not better than even. The probability of long-term /survival would be reflected in the amount of damages awarded for the *1606loss of the chance. While the plaintiff here could not prove by a preponderance of the evidence that he was denied a cure by the defendant’s negligence, he could show by a preponderance that he was deprived of a 30% chance of a cure.” (King, Causation, Valuation, and Chance in Personal Injury Torts Involving Preexisting Conditions and Future Consequences, supra, 90 Yale L.J. at pp. 1363-1364.)
Professor King criticizes the traditional rule of causation as it operates in the area of lost chance. He first claims the all or nothing concept is arbitrary because it treats the better-than-even chance as a certainty and gives no credit to the conceptually nigh even chance. Second, he believes the rule is contrary to the deterrent objectives of tort law because it denies recovery for the effects of conduct that causes statistically demonstrable losses. Third, he considers the rule a catalyst for distorting other rules affecting causation and damages in an attempt to mitigate perceived injustices. And fourth, he deems the rule is unfair because the imponderables of chance must be grappled with only because of the defendant’s conduct. (King, Causation, Valuation, and Chance in Personal Injury Torts Involving Preexisting Conditions and Future Consequences, supra, 90 Yale L.J. at pp. 1376-1378.) He likens a lost chance to the loss of a lottery ticket, an asset he accepts as having some value notwithstanding the prospect of winning with it is not better than even. (Id. at p. 1378.)
Relaxed Causation for a Lost Chance
The issue whether a relaxed standard of causation should be adopted in California was addressed in Simmons v. West Covina Medical Clinic (1989) 212 Cal.App.3d 696 [260 Cal.Rptr. 772].
Simmons was an action for wrongful birth and wrongful life brought by a mother and child after the child was born with Down’s Syndrome. The action was predicated upon a physician’s negligence in failing to test for the risk of Down’s Syndrome in time to abort the pregnancy. The trial court granted the defendants’ motion for summary judgment on the basis that the evidence as to causation established that there was only a 20 percent chance that the test would have detected a risk of Down’s Syndrome. The court affirmed on the basis that the traditional rule of causation negated recovery.
The plaintiffs in Simmons raised the specter of lost chance as follows: “Plaintiffs contend their low statistical probability of avoiding the harm should not prevent them from recovering damages for their lost chance of terminating the pregnancy. Plaintiffs assert a jury could consider their low statistical probability of avoiding the harm, but only for the purpose of calculating damages. They assert it is unfair and unjust to apply the traditional *1607medical probability test of causation to cases such as this, where the compensable injury is a lost chance rather than a tangible physical injury.” (Simmons v. West Covina Medical Clinic, supra, 212 Cal.App.3d at p. 704.)
The court acknowledged the authorities in foreign jurisdictions that had adopted the lost chance theory in medical malpractice actions. It then distinguished between actions for negligently reducing a patient’s chance of survival and depriving a woman of the chance to abort a genetically defective child. It viewed the former type of case as one where the patient’s condition could be attributable to a host of factors in addition to the physician’s negligence with the attendant difficulties in proof and the latter type of case as one where the hereditary condition was not caused by the physician’s negligence. It concluded that the justifications for a relaxed standard of causation were not present in the latter type of case.
The court nevertheless addressed the merits of the plaintiffs’ contention as follows: “Under the facts of this case, we decline to establish a more lenient standard of causation. To do so would be contrary to sound logic, legal precedent, and public policy. It would unwisely encourage costly and unreasonable overtesting and overtreatment for defensive purposes. Physicians would find it necessary to place the requirements of the legal system before the needs and the finances of the patient. In addition, the physicians’ increased exposure to liability would adversely impact already high medical malpractice premiums, resulting in an upward spiral of consumer costs. The uncertainty fostered by such a ruling would undoubtedly open the proverbial floodgates of our overburdened judicial system, [f] We refuse to expand the circle of liability by abandoning established tort law principles of causation where there is only a mere possibility of detecting the genetic defect. We do not wish to intrude upon the Legislature’s task of weighing such matters of public policy, and leave to it the function of deciding whether to provide a remedy for those genetically defective children and their parents who are unable to prove to a reasonable medical certainty that medical negligence deprived the mother of the chance to terminate her pregnancy.” (Simmons v. West Covina Medical Clinic, supra, 212 Cal.App.3d at pp. 705-706.)
Plaintiff construes the opinion of the Simmons court on lost chance as dictum because it was unnecessary given the threshold conclusion that the rationale for relaxed causation did not apply to wrongful birth and life cases.
The court’s remarks, however, specifically addressed an issue raised by the plaintiffs in that case.
*1608In any event, we agree with Simmons and believe it is equally applicable to the preexisting condition arena of medical malpractice as recognized by Gooding v. University Hosp. Bldg., Inc. (Fla. 1984) 445 So.2d 1015. Gooding was a wrongful death case alleging medical malpractice predicated upon a negligent diagnosis. In concluding that the theory of loss of chance to survive is inconsistent with the concept of proximate cause, the court observed: “Relaxing the causation requirement might correct a perceived unfairness to some plaintiffs who could prove the possibility that the medical malpractice caused an injury but could not prove the probability of causation, but at the same time could create an injustice. Health care providers could find themselves defending cases simply because a patient fails to improve or where serious disease processes are not arrested because another course of action could possibly bring a better result. No other professional malpractice defendant carries this burden of liability without the requirement that plaintiffs prove the alleged negligence probably rather than possibly caused the injury. ... We cannot approve the substitution of such an obvious inequity for a perceived one.” (Id. at pp. 1019-1020, fn. omitted.)
We therefore decline to establish a more lenient standard of causation in medical malpractice cases to account for the theory of lost chance.
We recognize that plaintiff disputes that we would in fact encourage a proliferation of defensive medicine, an escalation of medical costs, and an unwarranted expansion of liability exposure if we relaxed traditional causation requirements. But this very disagreement on such weighty matters of public policy militates against judicial tampering with the long-standing meaning of causation in deference to legislative consideration of the issue.
Lost Chance as the Injury
“A good argument can be made that damages ought to be recoverable when, due to a doctor’s negligence, a patient loses a substantial, though less than probable, chance of survival. There are also some arguments against expanding damages in medical malpractice cases to include damages for loss of chance of survival.” (Fennell v. Southern Maryland Hosp., supra, 580 A.2d at p. 212.)
For example, the lost chance theory produces more statistical errors than a traditional analysis.
“Because loss of chance recovery is based on statistical probabilities, it might be appropriate to examine the statistical probabilities of achieving a ‘just’ result with loss of chance damages. . . . ffl To compare the two *1609rules, assume a hypothetical group of 99 cancer patients, each of whom would have had a 331/3% chance of survival. Each received negligent medical care, and all 99 died. Traditional tort law would deny recovery in all 99 cases because each patient had less than a 5[1]% chance of recovery and the probable cause of death was the pre-existing cancer not the negligence. Statistically, had all 99 received proper treatment, 33 would have lived and 66 would have died; so the traditional rule would have statistically produced 33 errors by denying recovery to all 99. HQ The loss of chance rule would allow all 99 patients to recover, but each would recover 331/3% of the normal value of the case. Again, with proper care 33 patients would have survived. Thus, the 33 patients who statistically would have survived with proper care would receive only one-third of the appropriate recovery, while the 66 patients who died as a result of the pre-existing condition, not the negligence, would be overcompensated by one-third. The loss of chance rule would have produced errors in all 99 cases.” (Fennell v. Southern Maryland Hosp., supra, 580 A.2d at pp. 212-213.)
In addition, the lost chance theory has troubling implications.
If lost chance “is just in the context of a medical malpractice action, it would be equally just and applicable in such actions involving other professions, including the legal profession. For example, if a disgruntled or unsuccessful litigant loses a case, and it could be shown through expert testimony that there was a forty percent chance of winning the case, but the lawyer’s negligent efforts reduced the chance of winning by some degree, the litigant would be able to pursue an action based upon the loss of chance doctrine .... [T]he ‘injury’ would be the loss of chance as opposed to the unfavorable verdict.” (Perez v. Las Vegas Medical Center, supra, 805 P.2d at p. 599, fn. 3 (dis. opn. of Steffen, J.).)
Finally, the lost chance theory is, in reality, focused on causation rather than injury and damages.
A lost chance of survival or a better result cannot be proven unless and until the death or adversity occurs. Otherwise, in the case of a loss proven by statistics alone, a court could be placed in a position of awarding damages without injury, e.g., in the case of a patient who “miraculously recovers despite the negligence and unfavorable odds.” (Fennell v. Southern Maryland Hosp., supra, 580 A.2d at p. 213.) The true injury in lost chance cases is therefore the death or adversity. (Ibid.) Thus, “when we strip away the rhetoric, damages are really being awarded for the possibility that the negligence was a cause of the death.” (Ibid.) Stated another way, “[t]he recognition of a lost chance as a cognizable injury is necessarily based on the reasoning that but for the defendant’s negligence, the plaintiff might *1610possibly have avoided an adverse result Thus, recognition of lost chance as a recoverable interest contradicts the very notion of cause in fact.” (Falcon v. Memorial Hosp., supra, 462 N.W.2d at p. 65 (dis. opn. of Riley, C. J.).)
The damage label plaintiff places upon the instruction given in this case is therefore of no moment. The instruction allows for damages for possible cure and possible lengthening of life and/or improved personal comfort from more prompt diagnosis and treatment. California law does not permit tort compensation for a mere possibility. Redefining lost chance as a new form of injury simply does not diminish that the theory radically alters the meaning of causation.
We also note that Professor King’s criticisms of the traditional rule of causation are not unassailable.
“Professor King aptly characterizes a lost chance as a ‘raffle ticket’ destroyed by the defendant’s negligence. . . . King advocates compensation for ‘statistically demonstrable losses,’ [citation] so that a person deprived of a forty percent chance of survival should be compensated for forty percent of the compensable value of his life. [Citation.] Thus, tort law is transformed from a compensatory system to a payout scheme on the basis of a statistical chance that the defendant caused the plaintiff’s death. It is no answer that full compensation based on less than a certainty that a patient would have survived is overcompensation. Professor King criticizes the probability standard of causation because, in his view, it treats the better-than-even chance as a certainty, ‘as though it had materialized or were certain to do so.’ [Citation.] Clearly, causation can never be proven to a certainty; the law settles for less in determining that a defendant should be held liable for damages to a plaintiff. Thus, Professor McCormick describes the preponderance of the evidence standard of proof in terms of ‘probability’: [f] ‘The most acceptable meaning to be given to the expression, proof by a preponderance, seems to be proof which leads the jury to find that the existence of the contested fact is more probable than its nonexistence[.] Thus the preponderance of the evidence becomes the trier’s belief in the preponderance of probability.’ [Citation.] [][] McCormick notes that some courts are ‘shocked at the suggestion that a verdict, a truth-finding, should be based on nothing stronger than an estimate of probabilities.’ [Citation.] This statement reveals the very foundation of the tort system. Imperfect as it may be, our legal system attempts to ascertain facts to arrive at the truth. To protect the integrity of that goal, there must be some degree of certainty regarding causation before a jury may determine as fact that a medical defendant did cause the plaintiff’s injury and should therefore compensate the plaintiff in damages. To dispense with this requirement is to abandon the truth-seeking function of the law. Professor King is willing to do so in his attempt to *1611compensate for the precise magnitude of any lost chance. Professor King’s criticism of the more likely than not standard for causation, like the lost chance theory itself, is based on the erroneous premise that it is the purpose of tort law to compensate for lost chances. But tort law should not operate by the same principles that govern lotteries and insurance policies. If the acts of the defendants did not actually cause plaintiff’s injury, then there is no rational justification for requiring defendants to bear the cost of plaintiff’s damages.” (Falcon v. Memorial Hosp., supra, 462 N.W.2d at pp. 65-66, fns. omitted (dis. opn. of Riley, C. J.).)
The debate on lost chance is vigorous, and we favor the established rule of tort causation. More fundamentally, however, we believe that “[s] weeping modifications of tort liability law fall more suitably within the domain of the Legislature, before which ¿11 affected interests can be heard and which can enact statutes providing uniform standards and guidelines for the future.” (Rowland v. Christian (1968) 69 Cal.2d 108,121 [70 Cal.Rptr. 97 [443 P.2d 561, 32 A.L.R.3d 496] (dis. opn. of Burke, J.).)
We therefore conclude that the trial court erred by giving the jury in this case the instruction on lost chance.
Prejudice
“The precise question, therefore, which we must decide is whether this error requires a reversal of the judgment. Generally speaking if it appears that error in giving an improper instruction was likely to mislead the jury and thus to become a factor in its verdict, it is prejudicial and ground for reversal. [Citation.] To put it another way, ‘[w]here it seems probable that the jury’s verdict may have been based on the erroneous instruction prejudice appears and this court “should not speculate upon the basis of the verdict.” ’ [Citations.] As we observed in Butigan v. Yellow Cab Co. (1958) 49 Cal.2d 652, 660-661 [320 P.2d 500, 65 A.L.R.2d 1], ‘The determination whether, in a specific instance, the probable effect of the instruction has been to mislead the jury and whether the error has been prejudicial so as to require reversal depends on all the circumstances of the case, including the evidence and the other instructions given. No precise formula can be drawn.’ [Citations.]” (Henderson v. Harnischfeger Corp. (1974) 12 Cal.3d 663, 670-671 [117 Cal.Rptr. 1, 527 P.2d 353].) Circumstances to be considered include the degree of conflict in the evidence on the critical issues; whether respondent’s arguments to the jury may have contributed to the misleading effect; whether the jury requested a rereading of the instructions; the closeness of the jury’s verdict; and the effect of other instructions in remedying the error. (LeMons v. Regents of University of California (1978) 21 Cal.3d 869 , 876 [148 Cal.Rptr. 355, 582 P.2d 946].)
*1612Plaintiff concedes that the crucial issue as to prejudice is legal cause. He acknowledges that the evidence was in sharp conflict and argues that this conflict demonstrates the jury resolved the issue in a manner consistent with the traditional rule of causation. This follows, so the argument goes, because the evidence favoring plaintiff was that plaintiff was at stage 1 in 1984 and the evidence against plaintiff was that he was incurable by 1984. Plaintiff continues that had the jury accepted defendants’ evidence on causation it would have rendered a defense verdict so the jury necessarily accepted his evidence. Since he had a 67 percent chance of long-term disease-free survival absent defendants’ negligence in his stage 1 condition, plaintiff concludes that the verdict does not implicate the lost chance concept and is consistent with the traditional rule of causation.
We disagree with this analysis because the evidentiary conflict was not so black and white as plaintiff portrays.
Plaintiff’s expert conceded that very few lung cancer patients are diagnosed at stage 1. He also told about the less-than-even chances for long-term disease-free survival of patients diagnosed at stages 2 and 3. Thus, the jury could have rejected the expert’s testimony that plaintiff was at stage 1 in 1984, believed that plaintiff’s cancer was at a more advanced stage which offered a less-than-even-chance of long-term disease-free survival, and still found for plaintiff under the lost chance theory.
Under this circumstance, we should not speculate upon the basis of the verdict. (LeMons v. Regents of University of California, supra, 21 Cal.3d at p. 876.)
Moreover, the likelihood that the jury was misled into rendering a verdict on the lost chance theory was enhanced by plaintiff’s argument which emphasized the erroneous instruction and suggested that a finding that plaintiff was at stage 2 in 1984 would not preclude recovery.
That the jury was otherwise properly instructed on proximate cause does not detract from our analysis. Where two instructions are inconsistent, the more specific charge, in this case the charge on lost chance, controls the general charge. (LeMons v. Regents of University of California, supra, 21 Cal.3d at p. 878.)
We conclude that the probable effect of the instruction at issue was to mislead the jury. Giving the instruction was therefore prejudicial error.
*1613Disposition
The judgment is reversed and the matter remanded for a new trial. Defendants are awarded costs on appeal.
Capaccioli, Acting P. J., and Cottle, J., concurred.
Respondent’s petition for review by the Supreme Court was denied March 12, 1992. Mosk, J., Panelli, J., and Kennard, J., were of the opinion that the petition should be granted.