OPINION
{1} In this medical negligence appeal we examine the question of causation and recovery of damages based on a so-called “loss of a chance.”1 We discuss the circumstances under which the victim of a negligent misdiagnosis of a preexisting medical condition, like a terminal disease, may recover, at least partially, for wrongful death, even when the victim’s chance of recovery with a correct diagnosis would have been less than 50%. In this case the district court applied an appropriate legal theory of recovery, but it nonetheless directed a verdict for Defendant because there was insufficient evidence at trial to show a causal connection between the misdiagnosis and Plaintiffs death. Based on these deficiencies in the evidence, we affirm the judgment for Defendant.
BACKGROUND
{2} This lawsuit was previously before this Court on a motion for summary judgment. See Baer v. Regents of Univ. of California, 118 N.M. 685, 884 P.2d 841 (Ct.App.1994). The facts set forth in that opinion remain essentially unchanged, and we summarize them here in the light most favorable to Plaintiffs position, as we must do in reviewing a directed verdict for the defense. See Melnick v. State Farm Mut. Auto. Ins. Co., 106 N.M. 726, 728-29, 749 P.2d 1105, 1107-08 (1988) (discussing standard of proof in reviewing directed verdict).
{3} Helmut Baer was employed as a nuclear physicist at Los Alamos National Laboratories (LANL). As part of the regular employment practice at LANL, Baer was required to undergo periodic physical examinations. During his 1985 exam, chest x-rays revealed a lesion in the right lung. A little over one year later, the lesion was interpreted as benign, but the doctor recommended periodic exams and x-rays for the future. As late as 1988, x-rays failed to reveal any changes in the lesion, and another follow-up visit was recommended. In July 1989, Baer was reexamined at LANL by a physician’s assistant, James Pederson, but not by a medical doctor. For the first time no chest x-rays were taken, and Pederson offered Baer no medical advice concerning the lesion. A year later, Baer was independently diagnosed as having large cell carcinoma, and despite subsequent medical treatment, Baer died in October 1991. It was alleged at trial *510that Pederson inadequately reviewed Baer’s medical history and negligently failed to order follow-up x-rays.
{4} Baer’s widow initiated this wrongful death lawsuit as the personal representative of her husband’s estate. Her complaint alleged that during the 1989 examination, Pederson negligently failed to diagnose a cancerous nodule in Baer’s lungs. At trial, Plaintiff presented expert testimony that Pederson had departed from the accepted standard of medical practice in the community in that considering Baer’s medical history, Pederson’s examination should have included a follow-up x-ray. However, Plaintiff did not present evidence that Baer more likely than not had cancer at the time of Pederson’s exam or that the cancer would have been reasonably detectable by the omitted x-ray. On cross examination, Defendant elicited an admission from Plaintiffs expert that it would be “absolutely pure speculation” to say whether an x-ray taken in 1989 would have shown the cancer that was ultimately discovered in 1990.
{5} At the close of Plaintiffs case, Defendant argued that on this state of the record, Pederson’s failure to order an x-ray could not be proximately connected to Baer’s death or the loss of a chance of recovery. The district court agreed and directed a verdict for Defendant, finding in part that “even if ¿Pederson] were negligent, such negligence has not, as a matter of law, been shown to be a proximate cause of [Plaintiffs injuries and damages.” On appeal, Plaintiff argues that (1) Plaintiff is entitled to recover for Baer’s lost chance of survival even if that chance was less than 50%, and (2) there was sufficient evidence of proximate cause connecting Defendant’s medical negligence to Baer’s loss of a chance to submit the case to a jury.
DISCUSSION
{6} This Court has never directly addressed the question of whether a person with a preexisting medical condition, whose chance of recovery is reduced because of medical negligence, can recover for that lost chance if the odds of recovery would have been less than 50% even with the correct medical treatment.
Loss of a Chance
{7} Under established principles of tort theory, negligence must proximately cause an injury for the defendant to be liable for the resulting damages. See Coleman v. Eddy Potash, Inc., 120 N.M. 645, 650, 905 P.2d 185, 190 (1995). Thus, a plaintiff must prove that the defendant’s negligence more likely than not caused the injury, or was a cause of the injury, for which the plaintiff seeks recovery. As a consequence, a plaintiff with a preexisting medical condition such as cancer, whose chance of survival due to the condition is less than 50% (less likely than not) before any negligence occurs, cannot sue a negligent medical care provider for the entire value of the plaintiffs lost life. See Baca v. Baca, 81 N.M. 734, 740, 472 P.2d 997, 1003 (Ct.App.1970) (requiring a probability, not a mere possibility of recovery, to sue in negligence for entire loss of life); see also Alfonso v. Lund, 783 F.2d 958, 963-65 (10th Cir.1986) (citing Buchanan v. Downing, 74 N.M. 423, 394 P.2d 269 (1964) (discussing that in New Mexico proximate cause must be shown as a probability, rather than a possibility, that the plaintiff would have recovered)); Rival v. Atchison, Topeka & Santa Fe Ry. Co., 62 N.M. 159, 167, 306 P.2d 648, 653 (1957). The rule in these cases only measures what a plaintiff cannot recover. These cases do not address the possibility of recovering for a subset of the plaintiffs damages.
{8} The argument that the plaintiff must show that, but for the defendant’s negligence, he had a better-than-even chance of recovery or survival is predicated on the assumption that the only injury for which suit may be brought is the entire value of the person’s life. See Baca, 81 N.M. at 741, 472 P.2d at 1004. If the injury is defined as an indivisible whole, a plaintiff has the burden of proving that the defendant’s negligence was a proximate cause of the entire injury, that is, the victim’s death. Put another way, a plaintiff has the burden to show that the victim had a better-than-even chance of surviving the preexisting condition, assuming appropriate care. Under such an approach, a person with preexisting cancer who did not have a chance of recovery greater than 50% *511could not recover any damages (not even 49% would do), even though he may have been the victim of negligent misdiagnosis that accelerated his death and robbed him of an opportunity to survive. Only the cancer victim with a greater than 50% chance of recovery could sue (and just 51% would suffice), and this time the victim could recover for the entire value of his lost life. Thus, in the preceding hypothetical, if damages are treated as an indivisible whole, the first cancer victim receives nothing and is undercompensated, while the second victim receives everything and is arguably overcompensated. See Jonathan P. Kieffer, The Case for Across-the-Board Application of the Loss-of-Chance Doctrine, 64 Def.Couns.J. 568, 568-69 (1997) [hereinafter Kieffer].
{9} Recognizing this dilemma, Plaintiff does not argue that Defendant’s negligence proximately caused the entire injury, her husband’s death. Due to the severity of the cancer, she could not prove that he had a 51% or better chance of survival, but for the failure to diagnose his disease in a timely manner. Instead, she argues that Defendant’s negligent failure to x-ray her husband’s lungs and detect his lung cancer in a timely manner proximately caused her husband to lose the increased chance of survival that he would have had if his cancer had been detected earlier. There is no dispute about the potentially deadly nature of her husband’s preexisting condition. Plaintiff argues that Defendant’s negligence deprived him of a chance to live; a chance which was measurably greater absent Defendant’s negligence. As we shall see, this “loss of a chance” approach, in which the standard for causation remains the same but the injury is redefined, is consistent with modern tort theory. Rather than defining the loss as a person’s entire life, it is narrowed to the loss of a definable chance to survive. The action is for the chance lost due to the defendant’s negligence, thus taking the preexisting condition into account and excluding it for compensation purposes. The reasonable value of that loss is narrowed as well, representing only a percentage of the value of the entire life.
{10} In his seminal 1981 article, Causation, Valuation, and Chance in Personal Injury Torts Involving Preexisting Conditions and Future Consequence, 90 Yale L.J. 1353, 1378 (1981) [hereinafter King], Professor Joseph H. King, Jr. discusses cases such as this in which negligence cannot be shown to have negated a better-than-even chance of survival, but can be shown to have deprived the patient of a measurable chance of recovery. He maintains that the compensable injury is not the death as such, but the destruction of the chance of survival. Professor King reasons that the loss of a less-than-even chance of survival is worthy of redress for several reasons. He observes that the harshness of denying compensation under such circumstances is arbitrary because a person who has a 51% chance of survival will be compensated fully while one with a 49% chance will receive nothing. He asserts that such arbitrariness “subverts the deterrence objectives of tort law by denying recovery for the effects of conduct that causes statistically demonstrable losses.” Id. at 1377. Additionally, Professor King argues that the failure to compensate under such circumstances gives certain defendants the benefit of uncertainty that would not exist except for their own negligent conduct. Moreover, he warns that the harshness of such a rule creates pressure to manipulate and distort the rules of causation and damages to mitigate against harsh results. See id. at 1376-78.
{11} Professor King’s article has been cited with approval in numerous jurisdictions. See, e.g., United States v. Anderson, 669 A.2d 73, 77 (Del.1995) (stating that the argument for treating loss of a chance as a separate injury gained force after King’s influential article); Holton v. Memorial Hosp., 176 Ill.2d 95, 223 Ill.Dec. 429, 679 N.E.2d 1202, 1210 n. 1, 1213 (1997) (rejecting reasoning requiring a patient to have a greater than 50% chance of survival in order to recover damages); Wendland v. Sparks, 574 N.W.2d 327, 333 (Iowa 1998) (patient could recover for a lost chance of less than 50%); Delaney v. Cade, 255 Kan. 199, 873 P.2d 175, 178 (1994) (stating that the doctrine of loss of a chance gained impetus after publication of King’s article); Smith v. Louisiana Dep’t of Health & Hosps., 676 So.2d 543, 546-47 (La. *5121996) (citing King for principle that defendants are liable for depriving decedent of chance of survival and for percentage probability test for measuring damages); Gordon v. Willis Knighton Med. Ctr., 661 So.2d 991, 999-1000 (La.Ct.App.1996) (patient’s compensation for lost chance should be a percentage of the compensable value of life, proportionate to the lost chance of survival); Robots v. Ohio Permanente Med. Group, Inc., 76 Ohio St.3d 483, 668 N.E.2d 480, 483-84 (1996) (observing that the compensable ’ interest is the loss of a chance of achieving a favorable outcome); Herskovits v. Group Health Coop., 99 Wash.2d 609, 664 P.2d 474, 486-87 (1983) (en banc) (Pearson, J., concurring) (defining the compensable injury as the lost chance of recovery); Zueger v. Public Hosp. Dist. No. 2, 57 Wash.App. 584, 789 P.2d 326, 329 (1990) (adopting the position of Justice Pearson in Herskovits, which was strongly influenced by King).
{12} Although the reasoning may vary from state to state, the majority rule appears to recognize some form of loss of a chance or increased risk of harm in cases analogous to that presented here. See Kieffer at 568. The recognition of the injury of loss of a chance is also consistent with Section 323 and subsection (a) of the Restatement (Second) of Torts (1965) (emphasis omitted), which state that
[o]ne who undertakes, gratuitously or for consideration, to render services to another which he should recognize as necessary for the protection of the other’s person or things, is subject to liability to the other for physical harm resulting from his failure to exercise reasonable care to perform his undertaking, if
(a) his failure to exercise such care increases the risk of such harm[.]
Additionally, we observe that holding a party liable for his or her negligent conduct when that conduct does not cause the entire harm suffered is consistent with Section 50(b)(1) and (2) of the Restatement {Third) of Torts (Proposed Final Draft 1998) (Apportionment of Liability) which addresses the apportionment of liability when damages can be divided by causation:
(b) Damages can be divided by causation when there is a reasonable basis for the factfinder to determine:
(1) that any legally culpable conduct of a party or other relevant person to whom the factfinder assigns a percentage of responsibility was a legal cause of less than the entire damages for which the plaintiff seeks recovery and
(2) the amount of damages separately caused by that conduct.
{13} Only a small handful of states appear to have rejected the theory. See Kramer v. Lewisville Mem’l Hosp., 858 S.W.2d 397, 404 (Tex.1993) (surveying state positions on loss of a chance). We are persuaded that Professor King’s definition of the compensable injury as the lost chance of recovery reflects the better view and that, as a result, such an injury can be compensated without fear of diluting our traditional standard for proximate cause. See Todd S. Aagaard, Notes, Identifying and Valuing the Injury in Lost Chance Cases, 96 Mich.L.Rev. 1335 (1998), Margaret T. Mangan, The Loss of a Chance Doctrine: A Small Price to Pay for Human Life, 42 S.D.L.Rev. 279 (1997), and Kieffer for an overview of the approaches to compensating loss of a chance.
{14} When the injury is defined not as the ultimate injury to the patient, but as the loss of a chance of survival, the standard for proximate cause does not change. In New Mexico, that standard is defined as follows:
A proximate cause of an injury is that which in a natural and continuous sequence [unbroken by an independent intervening cause] produces the injury, and without which the injury would not have occurred. It need not be the only cause, nor the last nor nearest cause. It is sufficient if it occurs with some other cause acting at the same time, which in combination with it, causes the injury.
UJI 13-305 NMRA 1998. In a lost chance case, the injury is the loss of the patient’s chance of survival, whatever that may be. Defining the injury narrowly allows the fact finder to make the proximate cause determination under the normal standard because the question changes from: “Did Defendant’s negligence cause the death?” to “Did Defen*513dant’s negligence cause the loss of a measurable chance of recovery?”
{15} We note that some jurisdictions have compensated plaintiffs for the lost chance of survival by relaxing the standard of proximate cause, requiring “proof of only a substantial possibility that the result complained of would have been avoided but for the tortious conduct.” King, supra, at 1368, n. 53 (citing Clark v. United States, 402 F.2d 950, 953-54 (4th Cir.1968) (adopting substantial possibility standard by implication)); Kallenberg v. Beth Israel Hosp., 45 A.D.2d 177, 357 N.Y.S.2d 508, 511 (1974) (per curiam) (holding jury could find that had decedent been given proper medication, her blood pressure “could have been” controlled and she “might have improved sufficiently” to undergo surgery and recover), aff'd mem., 37 N.Y.2d 719, 374 N.Y.S.2d 615, 337 N.E.2d 128 (1975). We entertain no such suggestion in this case. Under the approach taken in this case, the standards of causation remain the same.
{16} We also observe that courts have often confused the percentage value of a plaintiffs chance of survival — for example, a 25% chance to live — with a percentage analysis of proximate cause — for example, a 25% likelihood that the person’s total injuries were caused by the defendant. See Zueger, 789 P.2d at 329; King, supra, at 1353-55, 1363-64; see also Herskovits, 664 P.2d at 478 (attempting, with only partial success, to clarify difference between causation and valuation). However, the percentage value of the patient’s chance of survival is relevant only to the valuation of the damages that should be awarded. Under this analysis, “[t]he causation inquiry determines whether a defendant should be required to compensate a plaintiff for a loss. The valuation inquiry determines how much compensation is required.” King, supira, at 1389. This analysis, therefore, requires no heightened standard of care and no relaxation of the standard of proximate cause. Instead, it allows compensation for an injury that is divisible from the total injuries suffered by the injured person.
{17} The theory of loss of a chance is consistent with established principies of traditional New Mexico tort law, and it does not require recognition of a new cause of action. New Mexico has previously recognized the liability of negligent actors for enhancing or aggravating preexisting conditions and confined the compensable damages to the degree of enhancement. For example with successive tortfeasors, as in Lujan v. Healthsouth Rehabilitation Corp., 120 N.M. 422, 427, 902 P.2d 1025, 1030 (1995), our Supreme Court determined that “a medical care provider who negligently aggravates the plaintiffs initial injuries ... is liable only for the additional harm caused by the negligent treatment.” In such a situation, the Court reasoned, “[tjhe medical care provider is liable only for the enhanced injury because the total harm is divisible- into separate injuries — that which the patient suffered before being treated by the medical care provider and that which was caused by the medical care provider in the course of treatment.” Id. Similarly, in cases involving negligence that aggravates a naturally-occurring preexisting condition, our Supreme Court has also stated that the defendant is “liable only for the aggravation or acceleration[.]” Hebenstreit v. Atchison, Topeka & Santa Fe Ry. Co., 65 N.M. 301, 306, 336 P.2d 1057, 1061 (1959). A defendant is liable for the injury he or she may cause, and the total injury may be divisible among two or more causes with compensable damages divided accordingly. Whether the divisible injury is the enhanced portion of a preexisting condition, as in Lujan, or the loss of whatever chance for survival the victim may have had, the analysis is the same with respect to causation. In each ease, the tortfeasor is held responsible for the divisible portion of the injury he is proven to have caused, but no more.
{18} Contemporary New Mexico tort law apportions damages according to the percentage of fault, whether the fault is comparative, see Scott v. Rizzo, 96 N.M. 682, 689, 634 P.2d 1234, 1241 (1981) (stating “[{liability based on fault is the cornerstone of tort law”), or successive. See Lujan, 120 N.M. at 427, 902 P.2d at 1030. Thus, it is consistent with established New Mexico law for a defendant to be held responsible for the damages *514he causes, no more under our system of comparative fault, but also no less by virtue of an arbitrary cut-off based upon a 50% chance of survival. Indeed, our Supreme Court has repeatedly emphasized that
[o]ur fault system of recovery ... serves the important social functions of redistributing the economic burden of loss from the injured individuals on whom it originally fell, deterring conduct that society regards as unreasonable or immoral, and providing a vehicle by which injured victims may obtain some degree of compensation and satisfaction for wrongs committed against them and by which society may give voice and form to its condemnation of the wrongdoer.
Trujillo v. City of Albuquerque, 110 N.M. 621, 624, 798 P.2d 571, 574 (1990) (footnote omitted), quoted in Madrid v. Lincoln County Med. Ctr., 1996-NMSC-049, ¶28, 122 N.M. 269, 923 P.2d 1154. Compensating plaintiffs for a lost chance of survival is entirely consistent with the public policy underlying New Mexico tort law.
Proving the Elements of a Lost Chance of Survival
{19} In the current case Plaintiff had to show that Defendant breached a duty of care owed to Baer and that Baer’s lost chance of survival, however measured, was likely caused by that breach. In eases of medical negligence, expert testimony is used
when the trial court reasonably decides that it is necessary to properly inform the jurors on the issues. This includes establishing the standard of care, treatment and information by which the actions of the physician are to be judged, the manner in which he measures up to the standard and whether his alleged acts were the proximate cause of the injuries involved.
Gerety v. Demers, 92 N.M. 396, 411, 589 P.2d 180, 195 (1978). Plaintiffs expert testified sufficiently as to the standard of care and Defendant’s breach of that standard. Plaintiffs expert also testified that at the time Baer’s cancer was diagnosed his tumor was identified as a stage III lesion and was considerably advanced, such that Baer would have had a significantly increased chance of survival if the nodule had been detected earlier, when it was smaller and presumably more operable. Up to this point, Plaintiff seems to have been well on his way toward laying the necessary foundation to establish causation.
{20} However, Plaintiff never presented any such evidence from which a jury could have concluded that an x-ray administered in 1989 would likely have revealed the presence of the cancer that was diagnosed too late in 1990. Such evidence was essential; Defendant had to show that without Defendant’s negligence, Baer’s cancer would likely have been diagnosed earlier, and Baer would have had a measurably increased chance of survival. Without it, the trial court correctly directed a verdict for Defendant, because, as the court stated, “even if defendant were negligent, such negligence has not, as a matter of law, been shown to be a proximate cause of plaintiffs injuries and damages.”
{21} Other jurisdictions that have recognized loss of a chance as a compensable element of damages provide some guidance regarding the evidence Plaintiff might have presented to establish a question of proximate cause for the jury. In Holton, 223 Ill.Dec. 429, 679 N.E.2d at 1213, the Illinois Supreme Court stated that a plaintiff
should be able to present evidence to a jury that the defendant’s malpractice, to a reasonable degree of medical certainty, proximately caused the increased risk of harm or lost chance of recovery.
{22} In Borgren v. United States, 716 F.Supp. 1378, 1381-82 (D.Kan.1989), the United States District Court for the District of Kansas found that the existence of chance could be shown by different kinds of evidence. The Court found sufficient evidence of proximate cause based upon medical opinion as well as evidence based upon the doubling time of a particular type of tumor, and expert testimony of what the likely size of the tumor would have been, to a reasonable degree of medical probability, at the time of the negligent act. In Borgren that evidence was used for the purpose of eliciting medical opinions which estimated the tumor’s severity at the time it should have been detected *515by a competent reading of mammograms. The Borgren court had the mammograms themselves in evidence as well as testimony as to what should have been detected from a competent reading of them. In our case, of course, the negligence was not in reading an x-ray; it was in not taking an x-ray at all. Nonetheless, such doubling-time evidence can be useful, and in fact was elicited at trial from Plaintiffs expert. However, the expert in Plaintiffs ease was unable to connect that doubling-time evidence to an opinion, to a reasonable medical probability, that the tumor would have been of detectable size in 1989 when the x-ray was omitted. In fact, Plaintiffs expert conceded such a conclusion in this case would be speculation.
CONCLUSION
{23} Because Plaintiff in this case never established to a reasonable degree of medical probability that Baer’s cancer would have been detected by an x-ray in 1989, Plaintiff could not show that the failure to x-ray more likely than not caused a reduction in Baer’s chance of survival. Consequently, the district court did not err in directing a verdict for Defendant. We affirm the judgment for Defendant.
{24} IT IS SO ORDERED.
BUSTAMANTE, J., concurs.
FLORES, J., concurs in result.