[No. 84712-6. En Banc.]
Argued February 8, 2011. Decided October 13, 2011
Cheryl R.G. Adamson (of Rettig Osborne Forgette), for appellants.
Christopher H. Anderson (of Fain Anderson VanDerhoef PLLC); Mary H. Spillane (of Williams Kastner & Gibbs); Donna M. Moniz (of Johnson Graffe Keay Moniz & Wick LLP); and Jerome R. Aiken (of Meyer Fluegge & Tenney PS), for respondents.
Bryan P. Harnetiaux and George M. Ahrend on behalf of Washington State Association for Justice Foundation, amicus curiae.
AUTHOR: Justice Susan Owens.
We concur: Justice Charles W. Johnson, Justice Tom Chambers, Justice Mary E. Fairhurst, Justice Debra L. Stephens, Justice Charles K. Wiggins.
AUTHOR: Chief Justice Barbara A. Madsen.
AUTHOR: Justice James M. Johnson.
We concur: Justice Gerry L. Alexander. En Banc ¶1 OWENS, J. -- Linda Mohr suffered a trauma-induced stroke and is now permanently disabled. She and her husband, Charles, claim that negligent treatment by her health care providers diminished her chances of avoiding or greatly minimizing her disability. In other words, they claim that negligence caused Mrs. Mohr a loss of the chance of a better outcome. In Herskovits v. Group Health Cooperative of Puget Sound, 99 Wn.2d 609, 611, 614, 664 P.2d 474 (1983) (Dore, J., lead opinion), this court recognized the lost chance doctrine in a survival action when the plaintiff died following the alleged failure of his doctor to timely diagnose his lung cancer. This case compels consideration of whether, in the medical malpractice context, there is a cause of action for a lost chance, even when the ultimate result is some serious harm short of death. We hold that there is such a cause of action and, accordingly, reverse the order of summary judgment. FACTS ¶2 In Richland, Washington, on the afternoon of August 31, 2004, Mrs. Mohr suffered a hypoglycemic event that caused her to run her car into a utility pole at approximately 45 m.p.h. She was taken by ambulance to the emergency room at Kadlec Medical Center (KMC). Having visible lacerations on her face from the car accident, Mrs. Mohr was given a neurological assessment upon arrival, at around 4:00 p.m., and a computerized tomography (CT) scan of her brain about an hour later. These tests were overseen or authorized by Dr. Dale Grantham, who was charged with Mrs. Mohr's care at KMC on August 31. The results were normal. ¶3 Following those neurological tests, however, Mrs. Mohr reported and was observed to have neurological symptoms, including being wobbly on her feet and having severe pain after being administered pain medication. ¶4 Mrs. Mohr was again transported to KMC by ambulance just after 7:00 a.m. on September 1, 2004, because her husband was concerned that she remained very lethargic through the night. Dr. Brian Dawson was the attending emergency room physician that morning. By around 9:30 a.m., Mrs. Mohr was diagnosed as having a stroke. Specifically, she was first found to have an "evolving infarct . . . in the right middle cerebral artery territory," Clerk's Papers (CP) at 119, which relates to a cause of a stroke. ¶5 Before the transfer, Mrs. Mohr's two physician sons had arrived at KMC to be by her side. They tried to get both Dr. Dawson and then, after her transfer, Dr. Watson to order a CT angiogram. A CT angiogram was not done until 2:30 p.m., after the Mohr sons had Dr. Watson repeatedly paged. Then, although the results were available at 3:27 p.m., Dr. Watson was not located or informed until 4:50 p.m. that the CT angiogram showed a dissected carotid artery. He still did not order anyone to administer anticoagulant therapy, antiplatelet agents, or any other treatment. Dr. Watson had prescribed aspirin around 2:00 p.m. but did not order its immediate administration. ¶6 Mrs. Mohr's sons finally arranged a transfer and transport to Harborview Medical Center. Dr. Watson signed the transfer form as a formality. Only shortly before her transport at 6:00 p.m. on September 1, 2004, was Mrs. Mohr finally given aspirin, though it had to be administered in suppository form because, by then, she could no longer swallow. ¶7 Mrs. Mohr is now permanently brain damaged; a quarter to a third of her brain tissue was destroyed. In particular, the portions of her brain that were damaged are involved with motor control, sensation, and spatial reasoning. ¶8 Mrs. Mohr and her husband filed suit, claiming that Mrs. Mohr received negligent treatment, far below the recognized standard of care. They argue that the doctors' negligence substantially diminished her chance of recovery and that, with nonnegligent care, her disability could have been lessened or altogether avoided. The Mohrs' claim relies, at least in part, on a medical malpractice cause of action for the loss of a chance. In support of their claim, the Mohrs presented the family's testimony, including her two sons who are doctors, and the testimony of two other doctors, Kyra Becker and A. Basil Harris. The testimony included expert opinions that the treatment Mrs. Mohr received violated standards of care and that, had Mrs. Mohr received nonnegligent treatment at various points between August 31 and September 1, 2004, she would have had a 50 to 60 percent chance of a better outcome. The better outcome would have been no disability or, at least, significantly less disability. ¶9 On April 16, 2009, the Benton County Superior Court granted summary judgment for the defendants on the basis that the Mohrs did not show "but for" causation and the hesitancy of the court to expand Herskovits to the facts of this case. The Mohrs appealed, and the Court of Appeals certified the case for our review. ISSUES ¶10 1. In the medical malpractice context, is there a cause of action for a lost chance of a better outcome? ¶11 2. Did the trial court properly grant summary judgment for all defendants under CR 56(c)? ANALYSIS 1. Lost Chance of a Better Outcome [1-3] ¶12 The medical malpractice statute requires the same elements of proof as traditional tort elements of proof: duty, breach, injury, and proximate cause. RCW 7.70.040. Whether there is a cause of action for a lost chance of a better outcome in the medical malpractice context is a question of law, which we review de novo. Berger v. Sonneland, 144 Wn.2d 91, 103, 26 P.3d 257 (2001). The standard formulation for proving proximate causation ¶13 Herskovits involved a survival action following an allegedly negligent failure to diagnose lung cancer. Over the course of a year, Leslie Herskovits repeatedly sought treatment for persistent chest pains and a cough, for which he was prescribed only cough medicine. Id. at 611 (Dore, J., lead opinion). When he finally sought another medical opinion, Herskovits was diagnosed with lung cancer within three weeks. Id. His diagnosing physician testified that the delay in diagnosis likely diminished Herskovits's chance of long-term survival from 39 percent to 25 percent. Id. at 612. Less than two years after his diagnosis, then 60 years old, Herskovits died. Id. at 611. The trial court dismissed the case on summary judgment on the basis that Herskovits's estate, which brought suit, failed to establish a prima facie case of proximate cause: it could not show that but for his doctor's negligence he would have survived because he "probably would have died from lung cancer even if the diagnosis had been made earlier." Id. Though divided by different reasoning, this court reversed the trial court, finding that Herskovits's lost chance was actionable. ¶14 The lead opinion, signed by two justices, and the concurring opinion, which garnered a plurality, agreed on the fundamental bases for recognizing a cause of action for the loss of a chance. The lead opinion explained: To decide otherwise would be a blanket release from liability for doctors and hospitals any time there was less than a 50 percent chance of survival, regardless of how flagrant the negligence. Id. at 614. The plurality similarly noted that traditional all-or-nothing causation in lost chance cases "'subverts the deterrence objectives of tort law.'" Id. at 634 (Pearson, J., plurality opinion) (quoting Joseph H. King, Jr., Causation, Valuation, and Chance in Personal Injury Torts Involving Preexisting Conditions and Future Consequences, 90 YALE L.J. 1353, 1377 (1981)). Both opinions found that "the loss of a less than even chance is a loss worthy of redress." Id. With emphasis, the lead opinion agreed, stating that "'[n]o matter how small that chance may have been--and its magnitude cannot be ascertained--no one can say that the chance of prolonging one's life or decreasing suffering is valueless.'" Id. at 618 (Dore, J., lead opinion) (quoting James v. United States, 483 F. Supp. 581, 587 (N.D. Cal. 1980)). ¶15 The lead and plurality opinions split over how, not whether, to recognize a cause of action. Drawing from other jurisdictions, especially the Pennsylvania Supreme Court's holding in Hamil v. Bashline, 481 Pa. 256, 392 A.2d 1280 (1978), the lead opinion held that the appropriate framework for considering a lost chance claim was with a "substantial factor" theory of causation. The court summarized that once a plaintiff has demonstrated that the defendant's acts or omissions have increased the risk of harm to another, such evidence furnishes a basis for the jury to make a determination as to whether such increased risk was in turn a substantial factor in bringing about the resultant harm. Herskovits, 99 Wn.2d at 616 (additionally noting the Hamil court's reliance on the Restatement (Second) of Torts $ 323 (1965), which provides that one who renders services to another, necessary for the protection of that person, is liable if "his failure to exercise [reasonable] care increases the risk of [physical] harm"). ¶16 Rather than looking to the causation element, the plurality opinion in Herskovits focused instead on the nature of the injury. Herskovits, 99 Wn.2d at 634 (Pearson, J., plurality opinion) ("[T]he best resolution of the issue before us is to recognize the loss of a less than even chance as an actionable injury."). The plurality noted among its concerns about the "all or nothing" traditional tort approach to recovery that it "creates pressure to manipulate and distort other rules affecting causation and damages in an attempt to mitigate perceived injustices." Id. In part, this characterizes what the Herskovits lead opinion does by prescribing that causation in all lost chance cases is to be examined under the substantial factor doctrine. The plurality found it more analytically sound to conceive of the injury as the lost chance. Id. ¶17 Though this court has not reconsidered or clarified the rule of Herskovits in the survival action context or, until now, considered whether the rule extends to medical malpractice cases where the ultimate harm is something short of death, the Herskovits majority's recognition of a cause of action in a survival action has remained intact since its adoption. "Washington recognizes loss of chance as a compensable interest." Shellenbarger v. Brigman, 101 Wn. App. 339, 348, 3 P.3d 211 (2000); see Zueger v. Pub. Hosp. Dist. No. 2 of Snohomish County, 57 Wn. App. 584, 591, 789 P.2d 326 (1990) (finding that the Herskovits "plurality represents the law on a loss of the chance of survival");16 DAVID K. DEWOLF & KELLER W. ALLEN, WASHINGTON PRACTICE: TORT LAW AND PRACTICE $ 4.10, at 155-56, $ 15.32, at 488 (3d ed. 2006) ("Washington courts recognize the doctrine of 'loss of a chance' as an exception to a strict application of the but-for causation test in medical malpractice cases."). In Shellenbarger, the Court of Appeals reversed summary judgment of a medical malpractice claim of negligent failure to diagnose and treat lung disease from asbestos exposure in its early stages. 101 Wn. App. at 342. Expert witnesses testified that had Shellenbarger received nonnegligent testing and early diagnosis, which would have led to treatment, he would have "had a 20 percent chance that the disease's progress would have been slowed and, accordingly, he would have had a longer life expectancy." Id. at 348. The court concluded, "We find no meaningful difference between this and Herskovits' lost chance of survival." Id. at 349. ¶18 Washington courts have, however, generally declined to extend Herskovits to other negligence claims. See, e.g., Daugert v. Pappas, 104 Wn.2d 254, 260-62, 704 P.2d 600 (1985) (declining to apply Herskovits in a legal malpractice claim); Fabrique, 144 Wn. App. at 685 (following Daugert and finding "no authority supporting the application of the 'substantial factor' definition of proximate cause to a negligence or strict liability action involving a contaminated food product"); Sorenson v. Raymark Indus., Inc., 51 Wn. App. 954, 957, 756 P.2d 740 (1988) (distinguishing Herskovits from an asbestos exposure claim that the plaintiff's risk of cancer was increased). Such limitation is common: "[T]he courts that have accepted lost opportunity as cognizable harm have almost universally limited its recognition to medical-malpractice cases." RESTATEMENT (THIRD) OF TORTS: LIABILITY FOR PHYSICAL AND EMOTIONAL HARM $ 26 cmt. n at 356-57 (2010). ¶19 Herskovits has been widely cited as an authority by other state courts and in journal articles for recognizing a cause of action in lost chance cases. See, e.g., Matsuyama v. Birnbaum, 452 Mass. 1, 16, 890 N.E.2d 819 (2008); McMackin v. Johnson County Healthcare Ctr., 2003 WY 91, ¶¶ 16-17, 73 P.3d 1094, 1100, adhered to on reh'g, 2004 WY 44, 88 P.3d 491; Tory A. Weigand, Loss of Chance in Medical Malpractice: The Need for Caution, 87 MASS. L. REV. 3, 9 (2002). Since Herskovits, the majority of states that have considered the lost chance doctrine have adopted it, although with varying rationales. Matsuyama, 452 Mass. at 10 n.23 (listing 20 states and the District of Columbia that have recognized the lost chance doctrine); see Weigand, supra, at 7-10. Several states have rejected the doctrine. Matsuyama, 452 Mass. at 10 n.23 (listing 10 states that have declined to adopt the doctrine). And others have not yet reviewed the issue or have declined to reach the question. Id. [4-6] ¶20 The rationales underpinning the lost chance doctrine have generally been applied the same in wrongful death claims and medical malpractice claims where the ultimate harm is something short of death. See, e.g., Shellenbarger, 101 Wn. App. at 349. In Delaney v. Cade, 255 Kan. 199, 873 P.2d 175 (1994), the Kansas Supreme Court recognized a cause of action for loss of chance of a better outcome. The court observed that many jurisdictions are like Kansas, in that the issue has only come up in a loss of survival case or a loss of a better recovery case. . . . . We have found no authority or rational argument which would apply the loss of chance theory solely to survival actions or solely to loss of a better recovery actions and not to both. Id. at 209-10. But cf. Weymers v. Khera, 454 Mich. 639, 653, 563 N.W.2d 647 (1997) ("we reject scrapping causation (the bedrock of our tort law) in negligence cases where the injury alleged by the plaintiff is something less than death"). ¶21 We note that, significantly, nothing in the medical malpractice statute precludes a lost chance cause of action. In relevant part, chapter 7.70 RCW provides that, in order to prove "that injury resulted from the failure of the health care provider to follow the accepted standard of care," a plaintiff must establish: (1) The health care provider failed to exercise that degree of care, skill, and learning expected of a reasonably prudent health care provider at that time in the profession or class to which he belongs, in the state of Washington, acting in the same or similar circumstances; (2) Such failure was a proximate cause of the injury complained of. RCW 7.70.040. The chapter does not define "proximate cause" or "injury." RCW 7.70.020. ¶22 The principal arguments against recognizing a cause of action for loss of a chance of a better outcome are broad arguments, similar to those raised when Herskovits was decided: concerns of an overwhelming number of lawsuits and their impact on the health care system; distaste for contravening traditional tort law, especially regarding causation; discomfort with the reliance on scientific probabilities and uncertainties to value lost opportunities. See Joseph H. King, Jr., "Reduction of Likelihood" Reformulation and Other Retrofitting of the Loss-of-a-Chance Doctrine, 28 U. MEM. L. REV. 491, 506 (1998); Matsuyama, 452 Mass. at 15 (noting criticisms of the doctrine, namely that it "upends the long-standing preponderance of the evidence standard; alters the burden of proof in favor of the plaintiff; undermines the uniformity and predictability central to tort litigation; results in an expansion of liability; and is too complex to administer") However, none of these arguments effectively distinguish the Mohrs' claim from Herskovits and seem instead to agitate for its overruling. Now nearly 30 years since Herskovits was decided, history assures us that Herskovits did not upend the world of torts in Washington, as demonstrated by the few cases relying on Herskovits that have been heard by Washington appellate courts. ¶23 We hold that Herskovits applies to lost chance claims where the ultimate harm is some serious injury short of death. We also formally adopt the reasoning of the Herskovits plurality. Under this formulation, a plaintiff bears the burden to prove duty, breach, and that such breach of duty proximately caused a loss of chance of a better outcome. This reasoning of the Herskovits plurality has largely withstood many of the concerns about the doctrine, particularly because it does not prescribe the specific manner of proving causation in lost chance cases. Rather, it relies on established tort theories of causation, without applying a particular causation test to all lost chance cases. Instead, the loss of a chance is the compensable injury. ¶24 The significant remaining concern about considering the loss of chance as the compensable injury, applying established tort causation, is whether the harm is too speculative. We do not find this concern to be dissuasive because the nature of tort law involves complex considerations of many experiences that are difficult to calculate or reduce to specific sums; yet juries and courts manage to do so. We agree that [s]uch difficulties are not confined to loss of chance claims. A wide range of medical malpractice cases, as well as numerous other tort actions, are complex and involve actuarial or other probabilistic estimates. Matsuyama, 452 Mass. at 18. Moreover, calculation of a loss of chance for a better outcome is based on expert testimony, which in turn is based on significant practical experience and "on data obtained and analyzed scientifically . . . as part of the repertoire of diagnosis and treatment, as applied to the specific facts of the plaintiff's case." Id. at 17. Finally, discounting damages responds, to some degree, to this concern. ¶25 In Herskovits, both the lead and concurring opinions discussed limiting damages. 99 Wn.2d at 619 (Dore, J., lead opinion), 635 (Pearson, J., plurality opinion). This is a common approach in lost chance cases, responsive in part to the criticism of holding individuals or organizations liable on the basis of uncertain probabilities. RESTATEMENT (THIRD) OF TORTS: LIABILITY FOR PHYSICAL AND EMOTIONAL HARM $ 26 cmt. n at 356 ("Rather than full damages for the adverse outcome, the plaintiff is only compensated for the lost opportunity. The lost opportunity may be thought of as the adverse outcome discounted by the difference between the ex ante probability of the outcome in light of the defendant's negligence and the probability of the outcome absent the defendant's negligence."). Treating the loss of a chance as the cognizable injury "permits plaintiffs to recover for the loss of an opportunity for a better outcome, an interest that we agree should be compensable, while providing for the proper valuation of such an interest." Lord v. Lovett, 146 N.H. 232, 236, 770 A.2d 1103 (2001). In particular, the Herskovits plurality adopted a proportional damages approach, holding that, if the loss was a 40 percent chance of survival, the plaintiff could recover only 40 percent of what would be compensable under the ultimate harm of death or disability (i.e., 40 percent of traditional tort recovery), such as lost earnings. Herskovits, 99 Wn.2d at 635 (Pearson, J., plurality opinion) (citing King supra, 90 YALE L.J. at 1382). This percentage of loss is a question of fact for the jury and will relate to the scientific measures available, likely as presented through experts. Where appropriate, it may otherwise be discounted for margins of error to further reflect the uncertainty of outcome even with a nonnegligent standard of care. See King, supra, 28 U. MEM. L. REV. at 554-57 ("conjunction principle"). ¶26 We find that the Herskovits plurality has withstood the broad policy criticisms raised against it and comports with the medical malpractice statute. We find no meaningful basis to distinguish permanent disability from death for the purposes of raising a loss of chance claim. Accordingly, we hold that Herskovits applies to medical malpractice cases that result in harm short of death and formally adopt the rationale of the plurality opinion that the injury is the lost chance. For the reasons discussed next, as it relates to the facts of this case, we reverse the order of summary judgment. 2. Summary Judgment ¶27 An order granting summary judgment is reviewed de novo. Rivas v. Overlake Hosp. Med. Ctr., 164 Wn.2d 261, 266, 189 P.3d 753 (2008). Summary judgment "shall be rendered forthwith if . . . there is no genuine issue as to any material fact and the moving party is entitled to a judgment as a matter of law." CR 56(c). We review the evidence in the light most favorable to the nonmoving party. Miller v. Jacoby, 145 Wn.2d 65, 71, 33 P.3d 68 (2001). ¶28 Interpreting the facts in the light most favorable to the Mohrs, they have made a prima facie case under the lost chance doctrine that, on August 31 and September 1, 2004, the respondents breached the recognized standard of care for treating a head trauma victim with Mrs. Mohr's symptoms and that their breaches caused Mrs. Mohr a diminished chance of a better outcome. The Mohrs presented the expert testimony of doctors Becker and Harris. Their testimony included opinions regarding breaches of the standard of care: that once given a narcotic, Mrs. Mohr should not have been discharged but observed overnight; that, had Mrs. Mohr been held overnight, her neurological deficits would have been earlier discovered to be a stroke; and that anticoagulants, antiplatelet agents, and general brain protective care reduce the damage caused by strokes. The expert testimony also included information regarding causation, including Dr. Becker's opinion that had Mrs. Mohr "received anti-thrombotic therapy there's at least a 50 to 60 percent chance that things could have had a better outcome. . . . Less disability, less neglect, less . . . of the symptoms of right hemispheric stroke." CP at 225-26. Dr. Harris testified that had Mrs. Mohr received nonnegligent treatment at various points between August 31 and September 1, 2004, she would have had a 50 to 60 percent chance of a better outcome. This included the possibility, according to Dr. Harris, that Mrs. Mohr may have had no disability if she had been properly treated. We find, on this evidence, a prima facie showing of duty, breach, injury in the form of a lost chance, and causation. [7] ¶29 Respondents also argue that the case cannot go forward because the Mohrs have not proved damages. This is a misconception of the requirements of medical malpractice tort law. See RCW 7.70.040. The Mohrs have made a prima facie case of injury: lost chance of a better outcome. [8-10] ¶30 Finally, KMC separately asserts that the trial court's order of summary judgment in its favor should be affirmed because it is not vicariously liable for the negligence of the codefendant physicians. ¶31 Under apparent authority, an agent (e.g., a doctor) binds a principal (e.g., a hospital) if objective manifestations of the principal "cause the one claiming apparent authority to actually, or subjectively, believe that the agent has authority to act for the principal" and such belief is objectively reasonable. King v. Riveland, 125 Wn.2d 500, 507, 886 P.2d 160 (1994). A finding of apparent agency can subject a hospital to vicarious liability for the negligence of contractor physicians or staff working at the hospital. See, e.g., Adamski v. Tacoma Gen. Hosp., 20 Wn. App. 98, 107-08, 579 P.2d 970 (1978). ¶32 KMC and the Mohrs dispute whether the Mohrs could and did reasonably believe that any of the codefendant physicians were employees or agents of KMC. The Mohrs signed a form that included the following language: Patient care is under the control of the patient's attending physician who: is an independent provider and not an employee or agent of the hospital: May request other physicians to provide services during hospitalization (i.e. pathologists, anesthesiologists, radiologists). CP at 107. Without considering the clarity of this language, we note that there are other relevant considerations, including: discharge instructions from the "[KMC] Emergency Department" that included information about treatment by Dr. Grantham at KMC; physician name tags that included KMC with the doctors' names; billing statements from KMC; and identification of Dr. Watson as a "'Hospitalist'" for KMC. Id. at 108, 268-70. It is also informative that KMC's emergency room is an essential part of its operation. See Adamski, 20 Wn. App. at 115. ¶33 In Adamski, the Court of Appeals considered several factors that it found relevant to the question of whether an independent-contractor physician was an apparent agent of the hospital. Id. at 115-16. It stated that "courts generally look to all of the facts and circumstances to determine if the hospital and doctor enjoy such a 'significant relationship' that the rule of respondeat superior ought to apply." Id. at 108. Similarly, the published model jury instructions enumerate seven relevant factors for the determination of apparent agency in the hospital and independent-contractor physician context. 6 WASHINGTON PRACTICE: WASHINGTON PATTERN JURY INSTRUCTIONS: CIVIL 105.02.03 (5th ed. 2005). One factor is "[w]hether the hospital made any representations to the patient, verbally or in writing, regarding their relationship with the physician." Id. However, "no one of [the factors] is controlling." Id. Thus, the notice that the Mohrs received disclaiming an agency relationship between KMC and the treating physicians is but one factor to consider. ¶34 KMC argues that even if there is apparent agency, the hospital is not liable for negligent acts of physicians that it could not control. Cf. McLean v. St. Regis Paper Co., 6 Wn. App. 727, 729-30, 496 P.2d 571 (1972). However, the negligence alleged here concerns the provision of medical services and is well within the scope of the apparent agency relationship alleged between the physicians and KMC. As in Adamski, we find that a hospital may be, depending on the facts found by a jury, liable for the negligence of its contractor doctors, who are held out to be agents of the hospital. Accordingly, we reverse the order of summary judgment. CONCLUSION ¶35 We hold that there is a cause of action in the medical malpractice context for the loss of a chance of a better outcome. A plaintiff making such a claim must prove duty, breach, and that there was an injury in the form of a loss of a chance caused by the breach of duty. To prove causation, a plaintiff would then rely on established tort causation doctrines permitted by law and the specific evidence of the case. Because the Mohrs made a prima facie case of the requisite elements of proof, we reverse the order of summary judgment and remand to the trial court for further proceedings. C. JOHNSON, CHAMBERS, FAIRHURST, STEPHENS, and WIGGINS, JJ., concur. Barbara A. Madsen and James M. Johnson ¶36 MADSEN, C.J. (dissenting) -- A central tenet of tort liability for medical malpractice is that a plaintiff must prove a physician's acts or omissions caused a patient's actual physical or mental injury before liability will attach. The lost chance doctrine adopted by the majority punishes physicians for negligent acts or omissions that cannot be shown to have caused any actual physical or mental harm. Because traditional tort justifications for imposing liability are missing, we should not extend a cause of action for a lost chance of a better outcome as a form of medical malpractice claim beyond its current application. ¶37 Black letter negligence law requires proof on a more probable than not basis that the injury was caused by the negligence of another. The majority holding rests on the fiction that the "injury" is actually the loss of a chance of a better outcome. This is semantic pretense. No matter how the cause of action is described, at the end of the day liability is based on no more than the mere possibility that the physician's negligence has caused harm, a result that conflicts with black letter law that "negligence in the air" is not actionable. ¶38 The majority claims that the tort principles of deterrence and compensation are served by adopting the doctrine. It is incorrect. Deterrence of negligence that does not cause actual harm is a meaningless proposition, and there can be no compensation of injury because the actual injury that occurs may be the result of the preexisting condition. Compensating plaintiffs for preexisting harm is not a legitimate goal of the tort system. ¶39 The majority's holding is also contrary to RCW 7.70.040. If the lost chance doctrine is to be accepted in this state, it should be through action of the legislature, which can consider the numerous public policy questions implicated by the doctrine that the majority never considers and, indeed, is not suitably in a position to consider. ¶40 The lost chance doctrine is also uniquely unfair because only the health care profession is exposed to liability under it. This court, like others, has refused to apply the basic doctrine against members of any other profession. If a lawyer is sued for malpractice, the plaintiff must prove proximate causation of real harm, but this is not true under the lost chance doctrine when a plaintiff sues a physician for negligent treatment that cannot be shown to have proximately caused real harm. The inequity is obvious. Analysis ¶41 It is a fundamental principle that in a medical malpractice action the plaintiff must prove causation of the plaintiff's actual physical (or mental) injury before tort liability will be imposed. To avoid the difficulty posed by this requirement, the majority recognizes a cause of action for which the plaintiff does not have to prove that "but for" the physician's negligence, the injury would not have occurred. Majority at 850-51 (citing Herskovits v. Grp. Health Coop. of Puget Sound, 99 Wn.2d 609, 619, 664 P.2d 474 (1983) (Dore, J., lead opinion); id. at 634-35 (Pearson, J., plurality)). That is, because the majority finds the traditional causation-of-injury requirement to be an insurmountable obstacle, it employs a different concept to anchor a lost chance claim. Majority at 850. The majority simply redefines the injury as the lost chance. With this semantic leap--essentially a fiction--the causation problem is fixed. ¶42 But in reality the problem remains. No matter how the lost chance cause of action is characterized, the plaintiff is freed of the requirement of proving causation because, no matter how the action is described, the end result is that liability is imposed based on possibilities and not on probabilities. See, e.g., Jones v. Owings, 318 S.C. 72, 77, 456 S.E.2d 371 (1995) ("[l]egal responsibility in this approach is in reality assigned based on the mere possibility that a tortfeasor's negligence was a cause of the ultimate harm"); Pillsbury-Flood v. Portsmouth Hosp., 128 N.H. 299, 305, 512 A.2d 1126 (1986) (rejecting plaintiff's reliance on the "loss of a chance" doctrine expressed in Hicks v. United States, 368 F.2d 626 (4th Cir. 1966); the Hicks rule that allows relaxation of the causation requirement where the defendant increased the risk of harm is ill advised; "[c]ausation is a matter of probability, not possibility"). ¶43 The lost chance doctrine contravenes the long-standing rule that a verdict in a medical malpractice action must not rest on "'conjecture and speculation.'" Douglas v. Bussabarger, 73 Wn.2d 476, 505, 438 P.2d 829 (1968) (internal quotation marks omitted) (quoting Glazer v. Adams, 64 Wn.2d 144, 148, 391 P.2d 195 (1964)). A "'possibility'" is not enough. Id. ¶44 Trying to skirt this obstacle by saying that "a plaintiff would still have to establish the loss of chance by a preponderance of the evidence," as the plaintiff argued in Crosby v. United States, 48 F. Supp. 2d 924, 931 (D. Alaska 1999), is not an acceptable excuse because it leads to unacceptable results. As the court in Crosby correctly responded, "[i]f a plaintiff's chance of recovery was reduced from 20 percent to 10 percent, then permitting recovery for that 10 percent loss enables a plaintiff to recover damages even when the plaintiff's actual physical injury was not more likely than not caused by a defendant's alleged negligence." Id. (emphasis added). ¶45 The majority tries to justify the lost chance doctrine on the ground that it serves the tort principles of deterring negligence and compensating for injury when "the ultimate harm is permanent disability." Majority at 856. But as the majority itself explains, these justifications rest on actual physical harm to the plaintiff, "permanent disability" in the majority's own words. But a chance of a better outcome, by definition, is not the same as an actual better outcome because there is no way to establish that any physical harm in fact resulted from the negligent act or omission of the physician. Not only does the doctrine not require proof of "but for" causation, "but for" causation cannot be proved in any event. ¶46 The "deterrence" justification identified by the majority is in fact unrelated to preventing harm-causing negligence. As Benjamin Cardozo famously explained long ago, "'negligence in the air'" is not actionable. ¶47 The Texas Supreme Court aptly observed, when it "reject[ed] the notion that the enhanced deterrence of the loss of chance approach might be so valuable as to justify scrapping [the] traditional concepts of causation," that "[i]f deterrence were the sole value to be served by tort law, we could dispense with the notion of causation altogether and award damages on the basis of negligence alone." Kramer v. Lewisville Mem'l Hosp., 858 S.W.2d 397, 406 (Tex. 1993) (emphasis added). By rejecting the traditional causation in favor of the possible deterrent effect of the lost chance doctrine, the majority imposes liability for damages based on negligence alone--"negligence in the air." ¶48 Moreover, the goal of compensation is not served, either, because there is no way to prove a physician's acts or omissions in fact caused the actual physical harm, rather than the actual harm resulting from the preexisting condition. In fact, under this theory of liability, plaintiffs may be compensated where they suffer absolutely no physical injury as a result of the physician's conduct. Indeed, the Maryland high court has determined that the lost chance doctrine does not result in accurate compensation for any plaintiff's injuries (when the lost chance is less than 50 percent). Fennell v. S. Md. Hosp. Ctr., Inc., 320 Md. 776, 789-90, 580 A.2d 206 (1990). Statistically, if all had received proper treatment, 33 would have lived and 66 would have died. Id. Under the lost chance doctrine, all would be permitted recovery of 33 1/3 percent of the normal value of the case. Id. at 789-90. However, the 33 who would have survived with proper care would be compensated by only 33 1/3 percent of the appropriate damages for the actual injury, i.e., a recovery one-third that which would be necessary to compensate for the actual harm. Id. In the other 66 cases, where the decedents died as a result of the preexisting cancer and not as a result of the negligence, the patients would be overcompensated for actual injury to the extent of the entire one-third recovery. Id. The result, the Maryland court said, is that the lost chance doctrine results in errors in compensation for actual injury in all 99 cases. Id. ¶49 Of perhaps greater importance, in a practical sense, the lost chance doctrine does not conform to RCW 7.70.040. Under this statute, a plaintiff in a medical malpractice action must prove: (1) The health care provider failed to exercise that degree of care, skill, and learning expected of a reasonably prudent health care provider at that time in the profession or class to which he belongs, in the state of Washington, acting in the same or similar circumstances; (2) Such failure was a proximate cause of the injury complained of. RCW 7.70.040. ¶50 The statute provides that a plaintiff must prove the health care provider failed to exercise the requisite degree "of care, skill, and learning" and this failure "was a proximate cause of the injury complained of." RCW 7.70.040. "Injury" in the statute undoubtedly reflects prevailing law stated in O'Donoghue, 73 Wn.2d at 824, that the failure to exercise the required degree of care must be a proximate cause of "the subsequent disability." In other words, the legislature meant an actual physical disability resulting from the failure to exercise proper care, not an amorphous "lost chance" that may well involve no actual disability at all. ¶51 In considering the comparable Alaska statute, which like ours requires a plaintiff to prove the health care provider failed to exercise the proper standard of care and as a "proximate result of this" failure "the plaintiff suffered injuries that would not otherwise have been incurred," ALASKA STAT. $ 09.55.540(a)(3), the federal court in Crosby, 48 F. Supp. 2d at 931, concluded that "the 'loss of chance' theory disrupts traditional causation principles set forth by statute." The court said "AS 09.55.540 clearly and unambiguously requires plaintiffs to establish that a defendant's alleged negligence was more likely than not the cause of injury." Id. The federal court aptly said that, "[t]he statute rejects any presumption of negligence." Id. The court concluded that "[r]ecognizing a 'loss of chance' theory under the circumstances of this case would enable plaintiff to recover even when her injury was not proximately caused by the defendant" and contravene the statute. Id. ¶52 Similarly, the Vermont Supreme Court reached the same conclusion in connection with its comparable state statute, observing that the statutory elements traditionally required that plaintiff produce evidence of a "'reasonable probability or reasonable degree of medical certainty' that the defendant's conduct caused the injury." Smith v. Parrott, 2003 VT 64, 175 Vt. 375, 380, 833 A.2d 843 (quoting Greene v. Bell, 171 Vt. 280, 285, 762 A.2d 865 (2000)). The court said that the "loss of chance theory of recovery is thus fundamentally at odds with the settled common law" codified in the statute. Id. ¶53 The same is true in Washington. Our statute setting out the elements that a plaintiff must prove in a medical malpractice action does not permit a presumption of negligence. It requires proof of proximate cause, not as to a chance of malpractice resulting in possible injury, but as to actual physical injury to the plaintiff. ¶54 If there is to be any change in this law, it should come from the legislature, after appropriate hearings, collection of data, and consideration of competing interests. Only the legislature has the authority to amend the statute. ¶55 Moreover, the legislature is best positioned to consider the myriad of public policy matters implicated by the lost chance doctrine. Among them are concerns about the potential impact on the practice of medicine, the costs of medical malpractice insurance, the costs of medical care, and the costs to society as a whole of compensating an entirely new class of plaintiffs who formerly had no claim under the common law. See Smith, 175 Vt. at 381; Fennell, 320 Md. at 792-95. As one court mentioned, "society is wallowing near the water line with the burdensome and astronomical economic costs of universal healthcare and medical services." Kemper v. Gordon, 272 S.W.3d 146, 152 (Ky. 2008). Malpractice insurance costs are rising and are a part of this financial burden. Id. At the same time, medical science and technology are advancing at a phenomenal pace and our expectations based upon these advancements rise as they advance. Id. But humans must still effectuate the advances, and there are no guarantees notwithstanding our expectations. ¶56 The lost chance doctrine also gives rise to other questions. "For instance, what is a 'late diagnosis'? Does a diagnosis missed this week, but made next week, rise to the level of diminished chance?" Id. What about a case where experts could present "evidence . . . that an MRI misread on Monday, but accurately discerned on Friday, perhaps gives rise to an infinitesimal loss of a chance to recover. Yet, under this doctrine, even a small percentage of the value of human life could generate substantial recovery and place burdensome costs on healthcare providers" that would ultimately be passed on to each person in the jurisdiction. Id. ¶57 What about in the very case before this court, where we are not considering the passage of weeks, or even days, but of hours? ¶58 In addition, even courts rejecting the doctrine have noted "'appealing'" arguments exist in favor of the lost chance doctrine, e.g., id. (quoting Smith, 175 Vt. at 381), and these, too, should be considered by the legislature. ¶59 The ramifications of the majority's opinion are unknown but potentially far-reaching. The majority opinion has the potential to alter health care in this state, as physicians would have to contemplate whether to provide an unprecedented level of care to avoid liability for even a slightly diminished chance of a better outcome. As noted, even a small percentage of chance can equal a substantial award. At the same time, it is no secret that health care insurance coverage is already strained, for those who even have such insurance, and adopting this doctrine cannot help but impact the nature and extent of insurance reimbursement for potential tests and treatments ordered as an eventual result of the majority's decision to expand liability to an unprecedented degree in this state. ¶60 All of these matters are public policy considerations for the legislature. ¶61 Another issue is the inequity of applying the lost chance doctrine in the medical field. As in other states, this court has declined to extend the lost chance of survival doctrine, the specific form set out in Herskovits, to permit suits against other professionals. See Daugert v. Pappas, 104 Wn.2d 254, 704 P.2d 600 (1985) (refusing to extend lost chance doctrine to legal malpractice actions). Courts have questioned the inconsistent application of the doctrine depending upon whether the action is for medical malpractice or other professional malpractice. Smith, 175 Vt. at 381; Gooding v. Univ. Hosp. Bldg., Inc., 445 So. 2d 1015, 1019-20 (Fla. 1984) ("[h]ealth 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" while "[n]o 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"). ¶62 This basic inequity weighs against extension of the doctrine, yet the majority never considers it. In fact, the majority declines to fully consider any of the many reasons why the doctrine should not be accepted. Instead, the majority says that they simply mirror concerns addressed in Herskovits, that Herskovits has not caused any problems, and for the same reasons favoring Herskovits, the lost chance doctrine should be adopted where the ultimate harm is injury short of death. ¶63 I do not share the majority's view that Herskovits has caused no serious harm and therefore it is unlikely that the majority's present opinion will. Nor do I agree that because the majority can find no reason to distinguish the rationale for the decision in Herskovits, this court's hands are essentially tied and we must reach a similar conclusion here. ¶64 First, we have no idea what the impact of Herskovits has been. We do not know how often the case is followed, how often actions brought under it have been settled, or what cases were decided but not appealed. Second, whatever the effect of Herskovits, it is impossible to conclude that effects of the present case will be comparable. If nothing else, the added burdens to society presented by this case will be cumulative to any produced by Herskovits. But in any event, and regardless of Herskovits, we are simply not in a position to casually conclude that there will be little discernible negative impact. We simply do not know, and the court does not represent the branch of government with the capability of weighing all of the policy arguments and other considerations that should be weighed. ¶65 Rather than assume that the issue before us is essentially already determined, as the majority does, this case presents issues and concerns that should be carefully examined before extending the lost chance doctrine and effecting such a sweeping change in the law. The court should not just apply Herskovits to injury short of death, but should instead take the opportunity to examine the issue much more closely. ¶66 Given that the decision whether to extend the lost chance doctrine should belong to the legislature, it is my hope that the legislature will examine this issue. If the legislature concludes that the doctrine should become a part of our state law, then it will be doing so as a duly informed representative body. If not, or if the legislature determines that a different version of the doctrine should be adopted, the legislature can effectively abrogate the majority's holding by amending RCW 7.70.040. The Legislature finds that in those actions founded upon an alleged want of ordinary care or skill the conduct of the responsible party must be shown to have been the proximate cause of the injury complained of. The Legislature also finds that the application of the so called loss of chance doctrine in such cases improperly alters or eliminates the requirement of proximate causation. Therefore, the rule in Jorgenson v. Vener, 2000 SD 87, 616 N.W.2d 366 is hereby abrogated. Similarly, the Michigan legislature effectively rescinded Falcon v. Memorial Hospital, 436 Mich. 443, 462 N.W.2d 44 (1990), when it enacted Michigan Compiled Laws $ 600.2912a(2), which provides: In an action alleging medical malpractice, the plaintiff has the burden of proving that he or she suffered an injury that more probably than not was proximately caused by the negligence of the defendant or defendants. In an action alleging medical malpractice, the plaintiff cannot recover for loss of an opportunity to survive or an opportunity to achieve a better result unless the opportunity was greater than 50%. ¶67 For the reasons stated in this opinion, I dissent. ¶68 J.M. JOHNSON, J. (dissenting) -- The majority improperly extends Herskovits v. Group Health Cooperative of Puget Sound, FACTS ¶69 Mrs. Mohr crashed her car into a utility pole at approximately 45 miles per hour after running into four other vehicles during an accident in which she was driving alone. The Richland Fire Department responded. Mrs. Mohr was treated by emergency medical personnel (EMPs) and brought by ambulance ¶70 Mrs. Mohr was seen in the emergency room by Dr. Dale Grantham. Dr. Grantham and nursing staff noted that Mrs. Mohr had suffered injuries to her head, face, mouth, right forearm, and left leg due to the accident. Dr. Grantham and nursing staff also noted that Mrs. Mohr suffered from diabetes, that her blood sugar was low upon rescue by the EMPs at the crash site, and that she had not been ambulatory at the scene of the accident. ¶71 Dr. Grantham performed a physical exam. During the exam, Mrs. Mohr did not report or demonstrate any acute distress, swelling of the head, numbness, or neck pain. She did not exhibit any motor or sensory deficits. Dr. Grantham ordered blood samples, a finger stick glucose sample, and had Mrs. Mohr taken for x-rays. He also ordered a computerized tomography (CT) scan of her head. The x-rays and CT scan came back normal; they did not show any broken bones, fractures, dislocations, or intracranial injury. ¶72 Mrs. Mohr suffered lacerations to her right eyelid and right hand as a result of her accident. Dr. Grantham sutured these lacerations at 6:36 p.m. He also fed her at this time and noted that she was alert and able to walk to the bathroom, albeit "slightly wobbly on foot." Clerk's Papers (CP) at 91, 94. Another finger stick glucose sample was taken, and a nurse applied antibacterial ointment and dressed Mrs. Mohr's leg wound. ¶73 Dr. Grantham returned at 7:56 p.m. to speak with Mrs. Mohr and her husband. She reported a pain level of "7" on a scale of 1 to 10. Dr. Grantham prescribed Darvocet, a pain medication, and warned Mr. and Mrs. Mohr about its sedative effect. Dr. Grantham noted that Mrs. Mohr was in "good condition, stable condition and improved condition." Id. at 94. The doctor proceeded to give Mrs. Mohr and her husband discharge instructions, telling them to return or contact their physician immediately if her condition worsened or changed unexpectedly, if she did not improve, or if other problems arose. The Mohrs left for their home at 8:20 p.m. ¶74 At 6:32 a.m. the following morning, Mr. Mohr called the Richland Fire Department. Mrs. Mohr was experiencing weakness, a lack of coordination, and nausea. The fire department transported Mrs. Mohr to the emergency room at the same hospital (KMC). She was admitted at 7:11 a.m. ¶75 Mrs. Mohr was seen by Dr. Brian Dawson at 7:16 a.m. She reported weakness and difficulty walking, but no numbness or tingling. ¶76 The results of this CT scan, which came back before 9:30 a.m., were not normal. Instead, it revealed findings that the radiologist thought "may be secondary to evolving infarct which is in the right middle cerebral artery territory." ¶77 The results of the MRI, which came in by 10:32 a.m., led to the discovery of a dissected right internal carotid artery. Dr. Dawson discussed the situation with Dr. Brooks Watson II, and they agreed upon a treatment plan. Mrs. Mohr was transferred to the intermediate care unit at 11:46 a.m., and Dr. Watson prescribed aspirin around 2:00 p.m. ¶78 An urgent ultrasound was performed to rule out carotid dissection in the common carotids, but that procedure could not assess the distal internal carotid artery. For this, a CT angiogram was ordered. The CT angiogram was performed at 2:30 p.m. and confirmed that Mrs. Mohr had a distal dissection of the right internal carotid artery. The findings were discussed with Dr. Watson at 4:50 p.m. ¶79 Dr. Watson discussed the situation with Harborview Medical Center after trying to attempt "neurosurgical input locally." ¶80 One of Mrs. Mohr's sons, a fifth-year resident in diagnostic radiology at the University of Washington, testified at deposition that Mrs. Mohr had lost between one-quarter and one-third of her brain tissue in the period following the accident on August 31, 2004. ANALYSIS ¶81 This case boils down to statutory interpretation. Because RCW 7.70.040 does not provide the cause of action the majority creates, its analysis and result are incorrect. Our legislature has simply not required the impossible of medical caregivers: to guarantee the best possible outcome for patients they help. A. Standard of Review ¶82 Statutory interpretation is a question of law that this court reviews de novo. Berger v. Sonneland, 144 Wn.2d 91, 104-05, 26 P.3d 257 (2001); cf. majority at 850 (citing Berger, 144 Wn.2d at 103). If a statute is plain and unambiguous, its meaning must be derived from the wording of the statute itself. Berger, 144 Wn.2d at 105. Plain words do not require construction. Id. Instead, courts assume the legislature means exactly what it says. Id. Courts should not force a given construction by imagining a variety of alternative interpretations. See id. (quoting W. Telepage, Inc. v. City of Tacoma Dep't of Financing, 140 Wn.2d 599, 608, 998 P.2d 884 (2000)). B. Respondents Are Entitled to Judgment as a Matter of Law: the Mohrs Have Not Established the Statutorily Required Element of Proximate Cause ¶83 The language of RCW 7.70.040 is plain and unambiguous. With respect to the issue raised in this motion for summary judgment, the health care provider's alleged failure to exercise the acceptable standard of care must be a "proximate cause of the injury complained of" before that health care provider may be subject to liability under chapter 7.70 RCW. Proximate cause is a necessary element of proof. RCW 7.70.040. ¶84 A "proximate cause" of an injury is defined as a cause that, in a direct sequence, unbroken by any new, independent cause, produces the injury complained of and without which the injury would not have occurred. Stoneman v. Wick Constr. Co., 55 Wn.2d 639, 643, 349 P.2d 215 (1960). To establish proximate cause, the plaintiff must show both "cause in fact" (that the injury would not have occurred but for the act in question) and "legal causation." Ayers v. Johnson & Johnson Baby Prods. Co., 117 Wn.2d 747, 753, 818 P.2d 1337 (1991). "Legal causation" depends on considerations of "'logic, common sense, justice, policy, and precedent.'" King v. City of Seattle, 84 Wn.2d 239, 250, 525 P.2d 228 (1974) (quoting 1 THOMAS ATKINS STREET, THE FOUNDATION OF LEGAL LIABILITY 110 (1906)). It involves the "determination of whether liability should attach as a matter of law given the existence of cause in fact." Hartley v. State, 103 Wn.2d 768, 779, 698 P.2d 77 (1985). ¶85 The injury complained of in this case is the distal dissection of Mrs. Mohr's right internal carotid artery, which led to a loss of brain tissue. The appellants offer no evidence or testimony, however, that Drs. Grantham, Dawson, or Watson caused this injury. They have not established cause in fact. Consequently, the appellants have not made a showing sufficient to establish the existence of an element essential to their case, and on which they will bear the burden of proof at trial: proximate cause. See Young v. Key Pharm., Inc., 112 Wn.2d 216, 225, 770 P.2d 182 (1989) (quoting Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986)). Thus, there can be no "genuine issue as to any material fact," and the respondents are entitled to a "judgment as a matter of law." CR 56(c); Celotex, 477 U.S. at 322. CONCLUSION ¶86 We should affirm the trial court and answer the question certified to us in the negative. The nonbinding plurality opinion in Herskovits should not be extended to rewrite the medical malpractice statutory scheme adopted by the legislature. Our application of the separation of powers doctrine is not a one-way street. ¶87 Recovery on the basis of "a lost chance of a better outcome" from these targeted medical care providers is highly speculative and places an impossible burden on doctors and hospitals. ALEXANDER, J., concurs with J.M. JOHNSON, J.