FARRELL, Associate Judge:

The question presented is whether a division of this court may and should depart from the standard “more likely than not” test for proximate causation in this suit for negligence based on the defendant-appellee’s alleged failure to screen blood donations adequately. Appellant, conceding that he cannot meet the standard test, urges us to “recognize loss of chance as a protected interest” and to apply that alternate test of causation to the Red Cross’s alleged negligence in this case, in much the same way — he contends — as a division did in a prior medical malpractice case, Ferrell v. Rosenbaum, 691 A.2d 641 (D.C.1997). We conclude that, in urging application of the loss of chance theory to the facts of this case, appellant presses the division’s authority too far. The circumstances of Ferrell differ too markedly from these to provide authority for the departure appellant advocates. Because appellant’s proffered evidence would fail the test of causation followed by our decisions, we affirm the trial court’s grant of summary judgment in favor of the Red Cross.

In July 1982 Calvin Grant (hereafter “Grant” or “appellant”), then age twelve, underwent surgery at Children’s Hospital in Washington, D.C. to repair a congenital heart defect. During the surgery he received five units of whole blood, which had been provided to Children’s Hospital by appellee the American National Red Cross (the “Red Cross”).

All of the five donors whose blood was used on Grant satisfied the blood screening requirements then utilized by the Red Cross. However, in compliance with the *318 Red Cross’s procedures at the time, none of the blood had been tested for alanine aminotransferase (“ALT”) levels. In September 1993, after a liver biopsy, Grant was found to have the hepatitis C virus. He filed a complaint in the Superior Court charging the Red Cross with negligence in not having screened the blood administered to him during the 1982 surgery for ALT. 1 During the litigation, it was determined that one of the five donors of the donated blood had been positive for hepatitis C. At the Red Cross’s request, blood samples from the positive donor and appellant were tested by means of DNA, and it was confirmed that appellant had been infected with the virus during the 1982 transfusion.

In 1982, when Grant underwent surgery, scientists and doctors were aware that besides hepatitis A and hepatitis B there was a form referred to as “non-A, non-B” (or “NANB”) hepatitis. Although today scientists know that most NANB hepatitis is caused by the hepatitis C virus (“HCV”), that virus was not isolated until 1989, and the first test to screen blood for HCV antibodies was not available until 1990. In his suit Grant asserted, nonetheless, that the Red Cross should have tested all donor blood for ALT levels as a “surrogate test” for NANB hepatitis, 2 because blood containing elevated levels of ALT has an increased chance of carrying the NANB hepatitis virus. According to appellant, at the time of his surgery ALT testing could identify a significant portion (up to 40%) of the blood supply infected with the NANB hepatitis, and — he asserted — the Red Cross itself believed that ALT testing might prevent as many as a third of the expected serious cases of NANB hepatitis cases annually, yet made a “business” (or cost-benefit) decision to forgo the testing.

The Red Cross defended by asserting that in 1982, all of the available data and the practice of national blood suppliers counseled against routine screening by ALT donor testing. It proffered evidence that, according to the consensus of leading experts nationwide, ALT testing would not have detected approximately 70 percent of donors infected with the then-unknown viral agent HCV; that the same percentage of the donors excluded on the basis of ALT testing would have been healthy and not affected by that agent; and that as a result routine ALT testing would have annually excluded many thousands of units of healthy blood from donors not carrying hepatitis, while failing to detect the vast majority of donors carrying NANB hepatitis.

Grant responded by conceding that he could not prove by greater than 50% (more likely than not) that he would not have been infected even if ALT testing had been performed. Specifically, he admitted that his expert testimony would be able to establish no more than a 40 percent correlation between ALT levels and infection with the NANB hepatitis, in part because ALT levels fluctuate in individuals. 3 Grant argued nonetheless — as he does on appeal — that a jury should be allowed to *319 decide whether the Red Cross’s negligence in not screening for elevated ALT levels “deprived] him of an opportunity to avoid” the infection he incurred even if that “opportunity” were measured at less than fifty-percent likelihood. Citing decisions of other courts that have applied the so-called “loss of chance” doctrine, he argued that it was “a jury question whether the Red Cross’s negligent failure to test proximately caused Calvin Grant’s injury by increasing his chances of getting NANB infected blood by at least 30%.” The trial court, on the strength of decisions of this court cited by the Red Cross, concluded as a matter of law that Grant had failed to present triable issues of fact on both negligence and proximate causation. It therefore granted summary judgment to the Red Cross.

II.

On review of summary judgment, this court applies the same standard of appraising the evidence as did the trial court. Drejza v. Vaccaro, 650 A.2d 1308, 1312 (D.C.1994). Thus, we will affirm summary judgment if, taking all reasonable inferences in the light most favorable to the non-moving party, a jury could not reasonably find for it under the applicable burden of proof. Nader v. de Toledano, 408 A.2d 31, 42 (D.C.1979). In this negligence action, Grant had the burden of proving both a breach of the standard of care by the Red Cross and a causal connection between the breach and his injury. E.g., District of Columbia v. Wilson, 721 A.2d 591, 597 (D.C.1998). If the proof he offered failed on either score as a matter of law, summary judgment was proper. We do not reach the issue of whether Grant established the relevant standard of care and a breach of it sufficiently to go to the jury, because we agree with the trial court that his proffered evidence on proximate causation failed as a matter of law.

Although the Red Cross cannot fairly be said to have had a physician-patient relationship with Grant, the parties agree that the applicable standards of causation are drawn from our medical malpractice decisions. See Ray v. American Nat’l Red Cross, 696 A.2d 399, 402 (D.C.1997) (analyzing claim of failure of the Red Cross to properly screen blood donations under standards for medical malpractice). Our decisions have heretofore required the plaintiff to prove “a direct and substantial causal relationship between the defendant’s breach of the standard of care and the plaintiffs injuries,” and we have defined that relationship as follows:

The evidence is sufficient to establish proximate cause if the expert 4 states an opinion, based on a reasonable degree of medical certainty, that the defendant’s negligence is more likely than anything else to have been the cause (or a cause) of the plaintiffs injuries.

Robinson v. Group Health Ass’n, 691 A.2d 1147, 1150 (D.C.1997) (internal brackets and quotation marks omitted) (citing in part W. Page Keeton, et al., PROSSER and Keeton on the Law of Torts § 41, at 269 (5th ed.1984), for the principle that the plaintiff “must introduce evidence which affords a reasonable basis for the conclusion that it is more likely than not that the conduct of the defendant was a cause in fact”). The “more likely than not” standard is firmly embedded in our law. See Talley v. Varma, 689 A.2d 547, 553 (D.C.1997); Lasley v. Georgetown Univ., supra note 4, 688 A.2d at 1387; Travers v. District of Columbia, 672 A.2d 566, 570 (D.C.1996); Carmichael v. Carmichael, 597 A.2d 1326, 1330-31 (D.C.1991); Psychiatric Inst. of Washington v. Allen, 509 A.2d 619, 624 (D.C.1986); Gordon v. Neviaser, 478 A.2d 292, 296 (D.C.1984). And see Twyman v. Johnson, 655 A.2d 850, 852-54 & n. 5 (D.C.1995); District of Columbia v. *320 Freeman, 477 A.2d 713, 716 & n. 9 (D.C.1984) (both applying the “more likely than not” test of causation outside the medical malpractice context).

Illustrating application of this principle is our decision in Talley v. Varma, supra, in which the plaintiff sued her physician for near-total loss of her sense of taste after he had treated her with radiation to remove thyroid tissue that remained following previous surgery to remove her cancerous thyroid gland. The plaintiffs medical expert testified that by administering negligently excessive amounts of radioactive iodine (or 1-131), the physician had “increase[d] the risk of complications such as those experienced by [the plaintiff].” 689 A.2d at 551. This court, however, sustained the trial court’s grant of summary judgment to the defendant on the issue of causation, declaring that while the expert’s “testimony on increased risk ... raised the possibility that the injury resulted from the allegedly excessive amount,” the evidence “faded to show that [this] increase in risk from increased dosages of 1-131, made it more probable than not that the ... excessive dose ... caused [the plaintiffs] injury.” Id. at 553 (emphases added). “[M]edical testimony as to the mere possibility of a causal relation,” we stated, “is not sufficient.” Id. (internal citation marks omitted), quoting Sponaugle v. Pre-Term, Inc., 411 A.2d 366, 368 (D.C.1980), and citing Quick v. Thurston, 110 U.S.App.D.C. 169, 172, 290 F.2d 360, 363 (1961) (testimony that “there were two possible theories as to source of [plaintiffs] infection,” one entailing negligence and the other not, was insufficient to create a jury issue on proximate causation).

Grant concedes, as he did in the trial court, his inability to prove that the Red Cross’s assumed negligence more likely than not caused his hepatitis infection, ie., that blood testing for ALT levels would— as a matter of probability — have detected the donor carrying the hepatitis C virus, leading to rejection of that blood donation. Instead Grant urges us to depart from that standard and accept the view of some courts in cases such as this that a plaintiff makes out a triable issue on causation by showing that the defendant’s conduct deprived him of a substantial, though less than fifty percent, chance of a better outcome had due care been exercised. 5 This *321 court’s decisions cited above create a formidable barrier to application of the “loss of chance” doctrine. Indeed, as Grant formulated his position to the trial court— asserting that the Red Cross’s failure to screen for ALT “increas[ed] his chances of getting NANB[-]infected blood by at least 80%” — the argument is virtually the same as the one rejected in Talley v. Varma, supra. Recasting an “increased risk” of injury (Talley) as a “loss of chance” of an injury-free outcome sounds very much like an exercise in semantics.

Grant argues nevertheless that this court has already applied the “loss of chance” doctrine in Ferrell v. Rosenbaum, supra, and thereby recognized it as an accurate formulation of the bedrock “direct and substantial” causal nexus standard. Upon analysis, we do not read Ferrell as deviating from the basic standard of proof of causation by probability. In that case, the plaintiff sued her physician and hospital for misdiagnosing her infant child’s potentially fatal blood disorder of Fanconi anemia. She proffered evidence that the child’s best hope of survival into adulthood had been through a bone marrow transplant from a compatible donor sibling. Indeed, her expert witness would have testified that, according to recent scientific reports, “70 to 90 percent of Fanco-ni anemia patients can be cured of their hematological disease if transplanted with a matched sibling at an early age.” Id. at 651 n. 16 (emphasis added). The plaintiffs theory was that the defendants’ negligence in misdiagnosing the child’s condition deprived her of the opportunity she would have seized — but which she later lost through circumstances — to bear a child or children who could have donated the necessary bone marrow. 6 In reversing summary judgment to the defendants, we acknowledged that “[t]he bare possibility that the Ferrells could have had another child, or children, that could have been a suitable bone marrow donor” would not suffice, id. at 650, and that the plaintiffs burden was to prove that the asserted negligence was a “substantial factor” in causing the injury. Id. (quoting Lacy v. District of Columbia, 424 A.2d 317, 818-19 (D.C.1980)). But we held that, given the proffered testimony that the mother “would have done anything to help [the affected child], including having another child or children,” the “significant” chances that this “would have yielded a suitable donor” for the child, and the even stronger evidence (cited above) of correlation between a transplant and likely cure, the plaintiff had presented a triable issue on whether the alleged negligence “substantially]” contributed to the child’s reduced chances for survival. Id. at 651-52 (citing and relying on conclusion of the court in Daniels v. Hadley Mem’l Hosp., 185 U.S.App.D.C. 84, 93, 566 F.2d 749, 758 (1977), “that there was an ‘appreciable chance’ that [the] patient’s life would have been saved, after [a] bench trial includ[ed] testimony that 75-80% of patients survived if given proper treatment”). 7

*322 The “lost chance” recognized in Ferrell was thus the opportunity for the plaintiff to avail herself of a medical procedure with a high likelihood (a 70-90 percent chance) of success if carried out. No similar claim is made in the present case, given Grant’s inability to offer proof that screening blood for ALT levels would have offered a more than thirty-percent-plus chance of detecting a donor’s hepatitis. Ferrell thus synchronizes with the standard of probability required by our decisions, whereas Grant’s proof does not.

Even if Ferrell were regarded as easing the burden of proof on causation in some medical malpractice cases, however, we are not convinced that a similar relaxation of proof should apply here. In Ferrell the court carefully limited its consideration of lost chance to “a case such as this involving negligent treatment of a potentially fatal condition,” id. at 651, in turn illustrating the situation where ‘“the harm [alleged] appears to have been brought about by two or more concurrent causes.’ ” Id. (quoting Daniels, 185 U.S.App. D.C. at 92, 566 F.2d at 757). In such a case, the lost chance doctrine may well make sense because of the difficulty of differentiating between the consequences of a pre-existing condition and those flowing from the negligent failure to ameliorate it. See, e.g., Hardy v. Southwestern Bell Tel. Co., 910 P.2d 1024, 1026 (Okla.1996) (explaining that the “loss of chance” theory where applied has typically been limited to a case where “negligence increases the risk of harm by aggravating the effect of [a] preexisting condition or risk and/or taking away whatever chance of recovery existed”). Appellant does not claim that the Red Cross failed to diagnose or treat him properly for a pre-existing injury. His claim is that he incurred a new injury— hepatitis C — during treatment for an unrelated condition. Moreover, unlike in the cases he relies on, his injury did not arise from the relationship of patient and physician (or patient and hospital). Grant’s claim that the Red Cross negligently supplied hepatitis-carrying blood is not easy to distinguish from a claim that any provider of supplies or equipment used in medical treatment was negligent in manufacturing or processing the supplies, thereby causing a patient injury. To apply the loss of chance theory to cases such as these would virtually collapse the limitations that our decisions have set to the reach of proximate causation. 8 Cases such as Talley v. *323 Varma, supra, demonstrate that any such relaxation of the standard of proof on causation must be effected by the entire court, and not a division. See M.A.P. v. Ryan, 285 A.2d 810, 312 (D.C.1971).

Grant’s conceded inability to prove that the Red Cross’s assumed negligence more likely than not caused his injury required the entry of summary judgment for the Red Cross. Accordingly, the judgment of the Superior Court is

Affirmed.

. Initially the complaint sued other actors as well, but it was later voluntarily dismissed as to them.

. A surrogate test, while not testing directly for the causative agent of a disease or its antibodies, may reveal a statistical association between a disease and a particular agent.

. Grant himself had shown normal ALT levels on several occasions despite being infected with the HCV. In her deposition Grant’s expert witness, Dr. Johanna Pindyck, acknowledged that Grant’s chance of not being infected would have improved by "at least 30 percent” had ALT testing been used, but that she could not "say for certain whether it would have been greater than that.” This approximated the affidavit of Dr. Thomas Zuck, past president of the Council of Community Blood Centers, that in 1982 there was "only a 30% chance that the implicated donor would have had an elevated ALT level and his blood discarded” as a result of ALT testing. Similarly, a study performed by the National Institutes of Health at about the same time confirmed that ALT testing "would fail to detect about 70% of the blood that would infect recipients with non-A, non-B hepatitis” (Affidavit of Dr. Paul V. Holland).

. The parties agree that medical expert testimony was necessary in this case to establish both negligence and proximate cause. See, e.g., Lasley v. Georgetown Univ., 688 A.2d 1381, 1385 (D.C.1997). *321 (We are unwilling to relax traditional rules of causation and create a new tort allowing full recovery for causing death by causing a loss of less than a 50% chance of survival. In order to demonstrate proximate cause, the burden is on the plaintiff to prove by a preponderance of the evidence that "it is more probable than not that defendant’s act caused his injury.”); Pillsbury-Flood v. Portsmouth Hosp., [128 N.H. 299] 512 A.2d 1126, 1130 (N.H.1986) (rejecting as “ill-advised” a proposed “relaxed” causation standard); Gooding v. University Hosp. Bldg., 445 So.2d 1015, 1020 (Fla.1984) ("We agree with the majority rule ... and hold that a plaintiff in a medical malpractice action must show more than a decreased chance of survival ... plaintiff must show that the injury more likely than not resulted from the defendant’s negligence in order to establish a jury question on proximate cause.”).

. See, e.g., Herskovits v. Group Health Coop., 99 Wash.2d 609, 664 P.2d 474, 478 (1983) ("It is not necessary for a plaintiff to introduce evidence to establish that negligence resulted in the injury or death, but simply that the negligence increased the risk of injury or death”); Thornton v. CAMC, Inc., 172 W.Va. 360, 305 S.E.2d 316, 324-25 (1983) (defendant’s negligent act or omission increased the risk and was a substantial factor in causing the ultimate injury); Hicks v. United States, 368 F.2d 626, 632 (4th Cir.1966) (Prima facie case is made if there was "any substantial possibility of survival” which was lost due to the negligent care); Hamil v. Bashline, 481 Pa. 256, 392 A.2d 1280, 1288 (1978) ("[Ojnce a plaintiff has demonstrated that defendant’s acts or omissions ... have increased risk of harm to another, such evidence furnishes a basis for the fact-finder to go further and find that such increased risk was in turn a substantial factor in bringing about the harm ....”).

The majority of jurisdictions, by contrast, have refused to recognize "loss of chance” as a doctrine lessening required proof of causation. See, e.g., Weymers v. Khera, [454 Mich. 639] 563 N.W.2d 647, 653 (Mich.1997) (affirming summary judgment for physician where plaintiff had only “thirty to forty percent chance" of avoiding injury, and holding that “no cause of action exists for the loss of an opportunity to avoid physical harm”); Jones v. Owings, [318 S.C. 72] 456 S.E.2d 371, 374 (S.C.1995) (“After a thorough review of the 'loss of chance' doctrine, we decline to adopt the doctrine ...."); Kilpatrick v. Bryant, 868 S.W.2d 594, 602 (Tenn.1993) ("the loss of chance [doctrine] is fundamentally at odds with the requisite degree of medical certitude necessary to establish a causal link between the injury of a patient and the tortious conduct of a physician”); Kramer v. Lewisville Mem. Hosp., 858 S.W.2d 397, 405 (Tex.1993) ("The more likely than not standard is ... a fundamental prerequisite to an ordered system of justice.”); Dumas v. Cooney, [235 Cal.App.3d 1593] 1 Cal.Rptr.2d 584, 592 (Cal.Ct.App.1991) ("We therefore decline to establish a more lenient standard of causation ... to account for the theory of lost chance.”); Fennell v. Southern Md. Hosp. Ctr., [320 Md. 776] 580 A.2d 206, 211 (Md.1990)

. The plaintiff’s expert also proffered that the chances of securing a perfectly matching sibling donor increased' with each additional child she had. 691 A.2d at 652 n. 18.

. We earlier had noted that in Daniels the Circuit Court, “applying District of Columbia law, rejected the claim that a plaintiff must conclusively show that the harm would not still have occurred absent the malpractice.” 691 A.2d at 651 (emphasis added). In Dan *322 iels, the court had made what seems the obvious point that "[r]arely is it possible to demonstrate to an absolute certainty what would have happened in circumstances that the wrongdoer did not allow to come to pass.” Daniels, 185 U.S.App.D.C. at 92, 566 F.2d at 757 (emphasis added) (quoted in Ferrell, 691 A.2d at 651). Neither of these statements is inconsistent with the requirement of our decisions of causal proof by the “more likely than not” standard.

. Still another problem with applying the "loss of chance” doctrine is its analytical kinship "with the allocation of damages based on comparative fault, regardless of whether the physician’s negligence is pegged above or below 50 percent in terms of proximate causation.” Kilpatrick v. Bryant, 868 S.W.2d 594, 615 (Term.1993) (Daughtrey, J., concurring in part and dissenting in part). This jurisdiction, of course, has not adopted comparative fault. Instead, as do the courts of Maryland, we follow the rule that:

[traditional tort law is based on probabilities. If a patient had a 49% chance of dying from an injury or disease and if the patient was negligently treated and dies, full recovery will be permitted because, absent the negligence, it was more likely than not that the patient would have survived. Based on the 51% probability of surviving the injury or disease, we exclude the injury or disease as the cause of death. Damages are not reduced by the fact that there was a strong possibility that the patient would have died absent the negligence. Conversely, if the patient had a 51% chance of dying from an injury or disease, and was negligently treated and died, it was probably the pre-existing medical condition, not the negligence, that killed the patient, and there is no recovery. Damages must be proven by a preponderance of the evidence. Damages are not proven when it is more likely than not that death was caused by the antecedent disease or injury rather than the negligence of the physician.

Fennell, supra note 5, 580 A.2d at 214.