EDMONDSON, J.

Robert C. Smith, Jr., Monnet, Hayes, Bullis, Thompson & Edwards, Oklahoma City, Oklahoma, for Appellants.

Hilton H. Walters, R. Gene Stanley, Rife Walters Stanley & Natarajan LLP, Oklahoma City, Oklahoma, for Appellees.

Opinion

EDMONDSON, J.

¶ 1 In a medical malpractice action we are asked to review orders excluding testimony from plaintiffs' two expert witnesses and a summary judgment granted to defendants based upon the excluded testimony. We conclude the testimony should not have been excluded. We reverse the orders of the District Court excluding the testimony and granting summary judgment, and remand the cause for further proceedings consistent with the Court's opinion.

¶ 2 Mrs. Nelson went to the Emergency Department of St. Mary's Regional Medical Center seeking medical assistance at 7:20 p.m. on the evening of July 21, 2006. The emergency room physician, Dr. Vaughan, ordered diagnostic tests, diagnosed an incarcerated hernia with possible bowel obstruction, and attempted to reduce the hernia. Dr. Vaughan telephoned Dr. Shepherd, Mrs. Nelson's internist and primary care provider. Dr. Shepherd instructed Dr. Vaughan to telephone Dr. Shreck, a surgeon. Dr. Shreck came to the hospital, reduced Mrs. Nelson's hernia, and she was admitted to the hospital.

¶ 3 One of the tests order by Dr. Vaughan was a CT scan. The CT scan showed free air in Mrs. Nelson's abdomen and required immediate surgery. The CT scan results were faxed to the hospital at 1:50 a.m. on the morning of July 22nd, but neither Dr. Vaughan or Dr. Shreck saw the report at that time.

¶ 4 The medical record indicates Dr. Shreck reduced Mrs. Nelson's incarcerated hernia by manipulation. Mrs. Nelson became septic, went into septic shock during the morning of July 22nd, and she had a cardiac arrest while being prepared for surgery to address a perforated or dead bowel. She was resuscitated. After the surgery, Mrs. Nelson was given dopamine and Levophed to raise and control her blood pressure. At 3:00 p.m. on July 22nd, Dr. Shepherd switched Mrs. Nelson's medication to vasopressin. At approximately 11:00 p.m., Mrs. Nelson's blood pressure started to fall, her pulse became unstable and she died.

¶ 5 A medical malpractice action was brought against Mrs. Nelson's medical providers for her last illness. Two defendants, Dr. Shepherd and Enid Medical Associates, moved to exclude the proposed testimony of plaintiffs' two expert witnesses. They argued each witness had not provided legally proper testimony on the issue of the cause of Mrs. Nelson's demise because the testimony did not satisfy the requirements of Daubert v. Merrell Dow Pharmaceuticals, Inc. , 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). The two defendants also sought summary judgment because the causation element of the malpractice claim action was missing from plaintiffs' claim.

¶ 6 The trial court ruled inadmissible the testimony from plaintiffs' two expert witnesses and granted summary judgment to the two defendants. The trial court made an express determination that there was no just reason for delay and expressly directed the filing of a final judgment. The plaintiffs appealed and the Court of Civil Appeals affirmed the trial court's order. This Court granted plaintiffs' petition for certiorari.

Appellate Review Standard for Summary Judgment and a Daubert Order Excluding Testimony on Causation

¶ 7 The standard for appellate review of a summary judgment is de novo and an appellate court makes an independent and nondeferential review testing the legal sufficiency of the evidential materials used in support and against the motion for summary judgment. Summary judgment is proper when a party is entitled to judgment “as a matter of law” based upon the submitted evidentiary materials.

¶ 8 Plaintiffs' action is based upon allegations that the two defendants proximately caused the injuries. A medical malpractice claim, like all negligence claims, contains three elements: (1) a duty owed by the defendant to protect the plaintiff from injury, (2) a failure to properly exercise or perform that duty and (3) plaintiff's injuries proximately caused by the defendant's failure to exercise the required duty of care.

¶ 9 Evidence of causation necessary for a negligence action, proximate cause, is usually an issue of fact to be determined by a jury; and proximate causation “becomes a question of law for the court only when there is no evidence from which a jury could reasonably find a causal nexus between the act and the injury.” If a defendant establishes there was no legally cognizable causal connection between the defendant's conduct and the injuries suffered by the plaintiff, then the issue of causation becomes a question of law, and a defendant is entitled to summary judgment as a matter of law.

¶ 10 Defendants' combined motion for summary judgment argued: “Plaintiffs cannot establish causation, an element of negligence, against Dr. Shepherd. Therefore, Dr. Shepherd is entitled to summary judgment.” Defendants supported this statement referencing the trial court's previous determination that Dr. Russell's testimony was inadmissible upon application of the principles in Daubert v. Merrell Dow Pharmaceuticals, Inc. , supra .

¶ 11 In Christian v. Gray we explained a clear abuse of discretion appellate standard applies when we review a decision on the admissibility of expert testimony, and a clear abuse of discretion may be shown by an error of law or an error of fact: “An abuse of discretion occurs when a court bases its decision on an erroneous conclusion of law or where there is no rational basis in evidence for the ruling.” A trial court determination that no fact exists of record to support the issue of fact submitted for resolution is a determination of an issue of law and requires a de novo review. Thus, a trial court determination that no fact exists in the trial court record, i.e., a complete absence of proof, to support the reliability of a particular expert for the purpose of admission of that expert's opinion presents an issue for de novo review.

¶ 12 In summary, we use a nondeferential appellate standard and review de novo a trial court's order granting summary judgment, and we use a nondeferential appellate standard and review de novo a trial court's Daubert order which determines the lack of facts supporting the reliability of a particular expert's opinion for the purpose of admission at trial.

The Daubert Challenge and the Record

¶ 13 The Oklahoma Evidence Code, 2702, provides: “If scientific, technical or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training or education may testify in the form of an opinion or otherwise, if: 1. The testimony is based upon sufficient facts or data; 2. The testimony is the product of reliable principles and methods; and 3. The witness has applied the principles and methods reliably to the facts of the case.” An expert's opinion must be “based on what is known,” i.e. facts and data, that are then used as part of a reliable method in forming an opinion.

¶ 14 Plaintiffs' response to the Daubert motion included Dr. Russell's opinion, and various articles and portions of depositions. His opinion relies on various published peer-reviewed articles. One of these is an article stating doses of vasopressin greater than 0.04 units/minute have been associated with decreases in cardiac output and cardiac arrest. Mrs. Nelson received a vasopressin “fixed dose” of 0.20 units/minute which was not tapered during infusion, although the order had been given by Dr. Shepherd to taper her dose. Dr. Russell testified that “the use of vasopressin in septic shock is off label ... [and] many of the drugs we use in intensive care are by some definition off label ... [and this results in dosing off label] because there's no label for vasopressin in septic shock.” Dr. Russell did not object to Mrs. Nelson receiving vasopressin, he objected to the dosage she received and that it was not a tapered dosage.

¶ 15 Defendants argue the recommended dosage for vasopressin in the published articles relied on by Dr. Russell are overly cautious concerning potential cardiac complications and not based upon science. Attached to plaintiffs' response is a statement by Dr. Russell stating that it is not speculation that a 0.2 units/min dose of vasopressin can cause cardiac arrest. He stated that “the precise dose of vasopressin that causes coronary vasoconstriction in humans is difficult to know, but that based on my studies and the literature the dose is in the range of 0.08 to .1 units/minute.” He relies upon a 2001 published study, and statements in this study are challenged by defendants. Defendants argue the opinions by Dr. Russell and Dr. Sheena are mere speculation and are not based upon scientific research or proper clinical observation. Defendants' motion is accompanied by photocopies of several articles in their Appendix of Literature.

¶ 16 Dr. Russell relies upon his clinical experience (observation), a published report based upon a review of the literature for treatment of septic shock patients, published studies utilizing case reports, an animal study, a published manual stating some effects of vasopressin, a document on vasopressin representing the source of its information as a “package insert,” a document authored by an individual with the academic credential of a “Pharm. D.,” and to some extent the additional studies cited in these publications.

¶ 17 Defendants challenge Dr. Russell's reliance on an article: “Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock,” with fourteen authors for the Surviving Sepsis Campaign Management Guidelines Committee, sponsored by eleven different medical professional organizations, and published by the Society of Critical Care Medicine in Critical Care Med , as well as in Intensive Care Medicine . The methodology of this article was based upon “a systematic review of the literature” for the purpose of developing “management guidelines for severe sepsis and septic shock,” and the study includes a recommendation for vasopressin dosage. This article states “Doses of vasopressin >0.04 units/min have been associated with myocardial ischemia, significant decreases in cardiac output, and cardiac arrest.” This article grades its recommendations and evidence and qualifies its statement concerning vasopressin as supported “by level IV or V evidence:” “nonrandomized, historical controls and expert opinion” and “Case series, uncontrolled studies, and expert opinion.” Dr. Russell participated in some of the research reported in this article, including an article published in Intensive Care Med. in 2001, an article which defendants challenge as unscientific if used for adopting its observation that dosages in excess of 0.04 units/minute can cause or create cardiac complications.

¶ 18 A portion of the trial court's order not challenged in this proceeding states “For the purpose of this motion, ... the Vasopressin dosage administered by Dr. Shepherd was excessive for Mrs. Nelson's ailment and a negligent act .” (Emphasis added). Attached to plaintiffs' response to Dr. Shepherd's Daubert motion is a portion of Dr. Shepherd's deposition where he explained: “I think the dose I wrote was a mistake. The question, I guess, is whether or not it caused harm.” Thus, while defendants appear to agree with recommendations in the literature for the proper dosage of vasopressin in patients similar to Mrs. Nelson, they disagree whether statements in the literature associating a high dose with adverse cardiac effects are conclusions based in science.

¶ 19 One of the resources relied on by Dr. Russell is a 2004 animal study involving Yorkshire pigs and performed at a university hospital in Pittsburgh. The abstract of the study states in part, “The data indicate that the safe dose range for exogenous vasopressin in septic shock is narrow and support the current practice of fixed low-dose administration, generally 0.04 units/min and in no case exceeding 0.1 units/min.” Dr. Russell cites an American Heart Association manual with a Pharmacology Summary Table stating “Precautions/Contraindications” for vasopressin : “Potent peripheral vasoconstrictor. Increased peripheral vascular resistance may provoke cardiac ischemia and angina. Not recommended for responsive patients with coronary artery disease.” Dr. Russell does not rely on only case studies, or only animal studies, or only his clinical experience as a physician who studies this topic, or only his understanding of, and argument concerning, the pharmacological properties of vasopressin, but on all of these areas.

¶ 20 Defendants filed an appendix of literature containing 10 articles, of three of which Dr. Russell is one of the authors, one in which Dr. Russell is responding to comments on a published study, and one involving a controlled infusion of vasopressin at 0.04 units/minute and which supports Dr. Russell's conclusion on a beneficial dosage.

¶ 21 In the remaining five articles, three do not expressly contradict Dr. Russell's testimony on causation, but appear to have been used to show what is generally accepted concerning vasopressin. The first, Beale and Hollenberg, et al., (2004), states “there is still inadequate understanding as to the mechanisms and potential therapeutic risk/benefit ratio of the use of vasopressin in septic shock. At this stage, vasopressin should only be used as part of properly constructed clinical trials until more information is available.” Beale and Hollenberg's conclusion may be simply stated, don't use vasopressin unless in a clinical trial because its mechanisms are not fully understood.

¶ 22 The second, an article by J.C. Russell and P.J. Glover (2002), reviews the published studies and repeats Dr. J.A. Russell's observation that some cardiac arrests occurred in patients who had received “doses greater than 0.05 U/min.” The article also states “Although heart rate may slow, bradycardia has not been reported in septic patients with low dose infusions.” These authors note: “There has been some concern expressed over the possibility of excessive vasoconstriction caused by vasopressin. Pharmacological doses have been shown to cause significant coronary and mesenteric ischaemia and it has been suggested that vasopressin should be used cautiously, if at all, in patients with symptomatic coronary artery disease.” This study also states that data concerning the clinically important adverse effects of vasopressin are lacking due to small study sizes and wide patient variability.

¶ 23 The third article, by Klinzing & Simon (2003), notes Dr. Russell's study which “found a decrease in cardiac index when vasopressin was given in doses 0.04 IU/min.” These authors note that “vasopressin has well-known vasoconstrictive properties in the splanchnic area.” They found: “Globally vasopressin caused a significant decrease in heart rate, cardiac output, and oxygen uptake.” The reduction in global oxygen delivery was partially compensated by increased oxygen extraction. These authors stated “a substantial reduction in cardiac output” was observed with a vasopressin dosage, “a mechanism for increased fractional splanchnic flow with the decreased cardiac output during vasopressin infusion is also unknown.”

¶ 24 The last two articles contain a critical comment on Dr. Russell's conclusion on causation of serious cardiac complications from vasopressin. The first, Torgersen and Dnser (2010), involved research where different doses were administered to two groups, one group of 25 persons received 0.067 and the other group of 25 persons received 0.033 IU/minute dose. They state Dr. Russell's 2001 article “did not prove a causative relationship between the occurrence of adverse events and VP doses >0.04 IU/min.” However, their opinion on Dr. Russell's failure of proof is based upon their characterization that Dr. Russell's study was “uncontrolled” and “observational.” They do state that results of their study and one by Luckner and Mayr (Crit Care Med 35:2280–2285) “are in contrast to the findings of a cases series” study authored, in part, by Dr. Russell. It appears the distinction being drawn by the data reported by Torgersen and Dnser involves the beneficial use of a dosage at 0.066 or 0.067 IU/min in the Torgerson/Luckner studies instead of Dr. Russell's recommended dosage of 0.04 IU/minute and less.

¶ 25 Torgersen and Dnser also state that certain baseline differences existed between their 0.033 IU/minute group and their 0.067 group, and “these baseline differences do not allow drawing firm clinical conclusions on the effects of the two AVP dose regimes on heart function.” Torgersen and Dnser's study had 50 patients with 25 receiving the higher dose, and Dr. Russell's study involved 50 patients receiving various amounts of vasopressin and cardiac arrests suffered by four patients who had a dosage “more than 0.05” units per minute.

¶ 26 The second article is a 2010 published study by authors Bauer and Lam, and which reviews recently published studies, including “the Vasopressin and Septic Shock Trial (VASST) and its subgroup analyses.” Although the article opines, “A major challenge in the attempt to determine which dose of arginine vasopressin should be used is that the true physiologic arginine vasopressin response during septic shock has not been fully elucidated,” it concludes “since arginine vasopressin doses as low as 0.01 unit/minute have yielded both physiologic replacement levels of vasopressin and blood pressure increase, it may be a reasonable starting dose for patients.” This recommended dose is based upon certain physiological responses.

¶ 27 Bauer and Lam challenge the conclusion “Doses of vasopressin >0.04 units/min have been associated with myocardial ischemia, significant decreases in cardiac output, and cardiac arrest ” since this study [by Dr. Russell and others] did not include a “matched cohort.” They conclude: “These data suggest that the occurrence of adverse effects of arginine vasopressin in general and with arginine vasopressin doses above 0.04 unit/minute may be lower than previously hypothesized.” Bauer and Lam cite two published studies, one of which was co-authored by Dr. Russell and published in the New England Journal of Medicine in 2008, and the other article was published in Intensive Care Med in 2006.

¶ 28 Dr. Russell disagrees with Bauer and Lam's criticism of his 2001 study and argues that his 2008 “randomized controlled trial of vasopressin ” excluded patients with acute coronary syndromes and or severe heart failure, and adverse reactions to high dose vasopressin could be observed in patients with a history of heart disease because of the reported adverse effect of decreased cardiac output associated with a high dose: “Thus, when vasopressin is infused at a safe, low dose of up to 0.03 units/minute, there was not an increased risk of cardiac arrest, emphasizing again Dr. Russell's research and recommendations repeatedly in his studies of the importance of the use of low dose vasopressin infusion in septic shock.” Indeed, some of the publications used by Defendants support Dr. Russell's conclusion for a vasopressin dosage of 0.04 units/minute or less in circumstances of septic shock.

¶ 29 In summary, defendants argued “it is speculation that Vasopressin causes a decreased cardiac index and cardiac arrest.” Klinzing and Simon state an observed reduction in cardiac output when vasopressin is used, although they do not explain the causal mechanism. J.C. Russell and P.J. Glover state that “Pharmacological doses have been shown to cause significant coronary and mesenteric ischaemia....” Beale and Hollenberg's conclusion may be simply stated, don't use vasopressin unless in a clinical trial because its mechanisms are not fully understood. Torgersen and Dnser criticize Dr. Russell's conclusion associating higher dosages of vasopressin with cardiac complications, but decline to draw “firm clinical conclusions” on the effects of the two studied dose regimes on heart function. Dr. Russell disagrees with some of the conclusions of Bauer and Lam.

Application of Daubert and General Causation

¶ 30 In Christian v. Gray, we explained causation is often divided into general causation and specific causation in some controversies involving allegations of injury resulting from a person's exposure to a harmful substance. General causation is whether a substance is capable of causing a particular injury or condition in the general population, while specific causation is whether that substance caused the particular individual's injury. Defendants raised the issue of general causation by challenging the opinion of plaintiffs' witnesses that vasopression is a coronary vasoconstrictive or that it causes a decreased cardiac index or cardiac arrest. When an external agent is the alleged cause of injury, an expert witness typically “demonstrate[s] that the medical and scientific literature provides evidence that in some circumstances the exposure under consideration can cause the outcome under consideration,” (thus showing general causation); and the next step is to “apply this general knowledge to the specific circumstances of the case at hand, incorporating the specifics of exposure, mitigating or exacerbating influences, individual susceptibilities, competing or synergistic causes, and any other relevant data.”

¶ 31 Generally, a trial court “should focus on the experts' methodology rather than the conclusions that they generate.” But because conclusions and methodology are not entirely distinct from one another, “[a] court may conclude that there is simply too great an analytical gap between the data and the opinion proffered,” i.e. , whether the expert has unjustifiably extrapolated from an accepted premise [or data] to an unfounded conclusion. Federal courts have stated a trial judge is not required to be a scientist, but is required to determine whether the expert's method in reaching a conclusion is “scientifically sound” “and that the opinion is based on facts which sufficiently satisfy [Federal] Rule 702's reliability requirements.”

¶ 32 The trial court found significant that “Dr. Russell's opinion is not supported by any published articles.” The trial court noted that Dr. Russell “relies heavily” upon a published study he authored which recommended a fixed specific dose of vasopressin for patients with severe septic shock. The trial court found legally significant the statement in the article that the study does not “draw any conclusions regarding the effect of vasopressin on mortality in severe septic shock.” The trial court stated that Dr. Russell has “not tested his theory,” and his study has been criticized. The trial court did not comment on studies which rely on information in Dr. Russell's studies. This analysis by the trial court addresses general causation.

¶ 33 Defendants' argument on general causation is that there is simply too great an analytical gap between the data and the opinion proffered on vasopressin causing cardiac complications. They cite the Tenth Circuit opinion in Hollander v. Sandoz in support of their argument. In Hollander, an witness gave an opinion on bromocriptine, but the “generic assumption that bromocriptine behaves like other ergot alkaloids carries little scientific value.” In Hollander, a witness opined in support of plaintiff's action that an active ingredient in the drug caused an increase in blood pressure, but it was held necessary for the witness to give a pharmacological explanation why animal studies showed decreases in blood pressure from this same ingredient, and this explanation had not been tested. In Hollander, the use of case reports to show general causation was rejected, in part, because of the “scant number” of case reports showing injury compared to the number of persons who had used the drug. Hollander does not require epidemiology toxicologic studies to satisfy the scientific reliability of an expert witness on the issue of causation.

¶ 34 Evidence of general causation may take different forms utilizing different methods of analysis, one of which may be the form of peer-reviewed published studies using various analytical methods. Of course, a professional publication “is not a sine qua non of admissibility; it does not necessarily correlate with reliability.” Dr. Russell's reliance upon published professional peer-reviewed studies is one methodology which satisfies Daubert and 12 O.S. § 2702 when it is based upon studies which comport with the dictates of good science, as opposed to an opinion based upon “junk science.” In Daubert, the Supreme Court developed a four-pronged but flexible test to determine the scientific validity and thus the evidentiary relevance and reliability of the principles that underlie a proposed submission: (1) peer review and publication; (2) the known or potential rate of error; (3) general acceptance; and (4) testing a theory by attempting to find evidence to disprove it (falsification).

¶ 35 A nonexhaustive list of accepted methodologies for scientifically determining general causation of an injury from a toxic and external substance includes epidemiology and in vivo and in vitro toxicologic studies as well as chemical analysis of the substance and adverse case reports . These methods have greater and lesser degrees of scientific reliability for conclusions determining causation due to varied techniques for acquiring the data and the type of data collected. For example, while two authors place “chemical structure analysis” and “adverse case reports” at “the bottom of the scientific probity barrel,” they do recognize that these two methods “are marginally relevant to the question of general causation.” The Reference Manual on Scientific Evidence states case reports may be all that is available, and while causal attribution based on case studies must be considered with caution, “such studies may be carefully considered in light of other information available.” Generally, toxicology models based upon animal studies may be used to determine toxicity in humans, within certain limitations, including the limitation of extrapolating results to a different species. The third edition of the Reference Manual on Scientific Evidence explains when considering the presence or absence of risk factors “physicians will often use any type of evidence that might support causation, for example, biological plausibility, physiological drug effects, case reports, or temporal proximity to exposure.” Additionally, “Although physicians use epidemiological studies in their decisionmaking, 'they are accustomed to use any reliable data to assess causality, no matter what their source' because they must make care decisions even in the face of uncertainty.”

¶ 36 Dr. Russell's opinion is not voiced contra mundum , against the world. Some do not agree with his conclusions and object to one of his studies lacking a matched cohort. The defendants' view of the nature of scientific proof appears to be that it must speak with one voice to satisfy the “general acceptance” element to Daubert and be truly scientific. In a post-Joiner case, the U.S. Court of Appeals for the First Circuit has noted “[t]he mere fact that two experts disagree is not grounds for excluding one's testimony;” and the First and Third Circuits have noted “[W]itnesses may be competent to testify as experts even though they may not, in the court's eyes, be the ‘best’ qualified. Who is ‘best’ qualified is a matter of weight upon which reasonable jurors may disagree.” This Court has made similar observations. While a court must assess science and not merely witness qualifications, it “need not weigh or choose between two legitimate but conflicting scientific views.” Two expert witnesses may each rely upon good science and reach different conclusions.

¶ 37 Dr. Russell's opinion is based upon published case studies, an animal study, clinical experience, and a pharmacologic description of vasopressin which, according to his opinion, give consistent information in support of his opinion on high-dose vasopressin. Dr. Russell's opinion concerning cardiac complications with vasopressin associated with a reduction in cardiac output and coronary ischemia is based upon literature in the record before us. The trial court stated that Dr. Russell is “highly credentialed,” “has been involved in research on Vasopressin as it relates to septic shock,” and “presented numerous papers on the topic and presented talks at international meetings.” Dr. Russell gave an opinion based upon his own published research prior to Mrs. Nelson's injury. In the Ninth Circuit Court of Appeals after remand in Daubert, the appellate court made the following observation.

... experts whose findings flow from existing research are less likely to have been biased toward a particular conclusion by the promise of remuneration; when an expert prepares reports and findings before being hired as a witness, that record will limit the degree to which he can tailor his testimony to serve a party's interests. Then, too, independent research carries its own indicia of reliability, as it is conducted, so to speak, in the usual course of business and must normally satisfy a variety of standards to attract funding and institutional support. Finally, there is usually a limited number of scientists actively conducting research on the very subject that is germane to a particular case, which provides a natural constraint on parties' ability to shop for experts who will come to the desired conclusion. That the testimony proffered by an expert is based directly on legitimate, preexisting research unrelated to the litigation provides the most persuasive basis for concluding that the opinions he expresses were “derived by the scientific method.”

Daubert v. Merrell Dow Pharm., Inc. , 43 F.3d 1311, 1317 (9th Cir. 1995) cert. denied , 516 U.S. 869, 116 S.Ct. 189, 133 L.Ed.2d 126 (1995).

Some do not agree with Dr. Russell's conclusions while others use his studies in conjunction with their own for explaining the possible roles of vasopressin. While some may not agree with Dr. Russell's conclusions, his method of combining arguably consistent case studies, including those predating Mrs. Nelson's injury, animal studies, evidence of known pharmacologic attributes of a drug, and all in the absence of contrary scientific studies or other evidence of greater reliability, satisfy Daubert's reliability standard for showing general causation. Dr. Sheena's opinion concerning a high dosage of vasopressin and its relationship to general causation and cardiac complications is based upon his clinical experience, the same literature as Dr. Russell and Dr. Russell's opinion. Dr. Sheena's opinion satisfies Daubert on the issue of general causation.

Application of Daubert and Specific Causation

¶ 38 Specific causation is the cause of the particular individual's injury. In this case, specific causation of Mrs. Nelson's injury is based upon Drs. Russell and Sheena's opinions on internal causation (the underlying physiological mechanisms producing observed signs and symptoms) and external causation (the relationship between environmental factors (such as chemical exposure or a medication) and the illness actually suffered by Mrs. Nelson. In summary, is the dosage of vasopressin the specific cause Mrs. Nelson's injury?

¶ 39 Dr. Russell testified that the dose of vasopressin given to Mrs. Nelson “certainly could be coronary vasoconstrictive.” He was asked if the dose “was vasoconstrictive,” and he replied that “we can't measure that directly,” and then replied “it's probable, yes.” He explained vasopressin has a dose range when it changes from being vasodilatory to vasoconstrictive on the coronary arteries. He also explained the very high dose which would lead to vasoconstriction of end organs, which would increase the workload on the heart, with potential cardiac side effects such as arrhythmias, and cardiac arrest due to asystole or profound bradycardia. He stated his opinion that “the relatively sudden onset of the arrhythmias leading to profound bradycardia leading to essentially asystolic cardiac arrest was contributed to in a significant manner by the very high dose of Vasopressin.”

¶ 40 The trial court characterized Dr. Russell's medical opinion on causation as a legally insufficient “educated guess.” The trial court pointed to Dr. Russell's testimony that arrhythmia and bradycardia may occur, (1) in the absence of a high dose of vasopressin, or (2) on a normal dose of vasopressin, or (3) on a normal dose of dopamine or Levophed, or (4) as a result of septic shock. However, Dr. Russell testified the dose of dopamine Mrs. Nelson received would have a significantly less vasoconstrictive effect than the high dose of vasopressin she received. He testified Dr. Shepherd's use of vasopressin as a medication for Mrs. Nelson was proper. However, he also testified the vasopressin dose she was prescribed and received was too high and was sustained for too long a period of time.

¶ 41 On the issue of specific causation the trial court states Dr. Russell has not ruled out other potential causes of Mrs. Nelson's death. Dr. Russell testified that as patients “stabilize” their dosage of norepinephrine is tapered and then stopped, and then the dosage of vasopressin is tapered and then stopped. He stated one of his concerns was that although there was an order to taper the vasopressin, “there was a fixed dose, as I reviewed it, which remained at that dose, as I understood the chart, for the duration that it was infused.” He explained the “standard dose” for septic shock patients at Dr. Russell's hospital. He was asked why he thought the use of vasopressin caused Mrs. Nelson to have a cardiac arrest. He stated, “first, the very high dose which would lead to vasoconstriction of end organs, which would increase the workload on the heart, would potentially cause cardiac side effects such as arrhythmias that were noted, and that does lead to and has been accompanied by cardiac arrest due to asystole or profound bradycardia. So my interpretation is that the relatively sudden onset of the arrhythmias leading to profound bradycardia leading to essentially asystolic cardiac arrest was contributed to in a significant manner by the very high dose of vasopressin.” He was then asked and replied affirmatively that similar conditions may arise in the absence of vasopressin with a patient in septic shock, or with a patient administered Levophed or dopamine.

¶ 42 He was asked if his opinion was speculation on the proper dosage of dopamine and vasopressin and their potential vasoconstrictive effects. He replied his opinion was a judgment based upon clinical responses to doses of dopamine at a specific amount compared to “usual doses of vasopressin,” “and then looking at animal studies, looking at higher doses of vasopression and higher doses of dopamine.” He stated “In humans we don't directly measure vasoconstriction, and so we have indirect interpretation of other findings. So when we say ‘vasoconstriction,’ we don't have a test to go and measure it.” He further explained data on humans includes studies on dose responses in sepsis patients to different vasopressors and measurements are taken for change of blood pressure, cardiac output, “and change in other vasopressor requirements” “which give us an indirect interpretation of what the vasoconstriction activity is.”

¶ 43 Dr. Sheena testified Mrs. Nelson's death was caused by “a combination of septic shock and the overdose of vasopressin.” He quantified Mrs. Nelson's chance of survival immediately post-op as over fifty percent. He stated that a few hours after post-op her chance of survival had been improving because “her clinical parameters had improved.” He testified all decisions made by Dr. Shepherd were appropriate except for the dosage of the vasopressin : “he ordered an overdose of vasopressin in treating the patient.” The trial court pointed out Dr. Sheena's statements on potential causes of arrythmias other than a high dose of vasopressin. However, the trial court also stated that “For the purpose of this motion, this court assumes Dr. Sheena is an experienced and credentialed physician, and that the Vasopressin dosage administered by Dr. Shepherd was excessive for Mrs. Nelson's ailment and a negligent act.” The trial court stated: “Daubert requires a showing that Dr. Sheena used a reliable, analytically appropriate method, that has been tested and subjected to peer review, has a known rate of error, and is widely accepted within the medical community. Dr. Sheena offers no proof on these points and nothing to rule out that Dopamine, Levophed, septic shock, or some other natural cause was the sole cause of Mrs. Nelson's death.” The trial court concluded that Dr. Sheena's opinion was mere ipse dixit and his proposed testimony was inadmissible.

¶ 45 Dr. Sheena is employed as an emergency room physician at a university medical center and a heart hospital, and once a month teaches medical residents, interns, and medical students at Baylor University Medical Center. He is also employed as a “hospitalist,” a physician who admits patients to a hospital from an emergency room, and also treats them during their hospital stay, including patients in an intensive care unit.

¶ 46 He stated his experience working as a physician in small hospitals. His opinion in this case was based upon his experience, education, and his review of Mrs. Nelson's “entire chart.” He stated that in his medical practice “I see people with incarcerated hernias all the time” and he has also treated patients with a perforated bowel and sepsis. However, his experience treating patients with septic shock secondary to a perforated bowel was no more than twenty-five patients in his career.

¶ 47 He stated his experience in prescribing vasopressin for his patients. He opined concerning Mrs. Nelson's dosage of vasopressin : “vasopressin at those dosages frequently causes. . . decrease[d] blood flow to the splanchnic circulation. It also affects the heart, and in this case, I think it caused her to have a rhythm disturbance....” At this point in his deposition, Dr. Sheena was not asked what authority he was relying on for his opinion on the proper dosage of vasopressin, but was asked whether rhythm disturbances could be caused by dopamine, Levophed, or septic shock as well as the dosage of vasopressin Mrs. Nelson received. He subsequently stated the range for a “standard dose for vasopressin.” He stated vasopressin is a vasoconstrictive, and the appropriate dosage will depend upon the patient, and there are relatively higher appropriate dosages in certain circumstances when a patient is not in septic shock. He was not asked to explain his authority for this “standard dose.” He was asked how he calculated Mrs. Nelson's percentage chance of survival and how it changed during her hospital stay. He was not asked to explain his opinion on the degree or percentage of contribution to her death which he attributed to the specific dosage of vasopressin she received.

¶ 48 He stated that a few hours after post-op, and prior to being administered vasopressin, her chance of survival had been improving because “her clinical parameters had improved.” He explained, “she had had surgery, and she had finally gotten a couple doses of antibiotics ... she was in a situation where she was improving as of this point that they started the vasopressin, and you can see how that led to a prompt worsening of her conditioning ... her clinical worsening had to do with the vasopressin.” He stated her chance of survival after surgery was over 50%, and it continued to improve, based on certain clinical parameters, during the five or six hours post-op before the vasopressin began to be administered. He stated that her condition even improved following the initial dose of vasopressin. He concluded that seven hours after the initial dose the high dosage created a cardiac rhythm disturbance.

¶ 49 The trial court stated Dr. Russell's opinion did not rule out other potential causes of Mrs. Nelson's death because Dr. Russell stated the possibility of Mrs. Nelson experiencing a cardiac arrest in the absence of the dose of vasopressin she received. Similarly, the trial court stated Dr. Sheena “offers no proof on these points and nothing to rule out that Dopamine, Levophed, septic shock, or some other natural cause was the sole cause of Mrs. Nelson's death.” The trial court's error on specific causation shown by differential diagnosis is that the opinion need not “rule out” every other potential cause, i.e., vasopressin need not be the sole cause for the opinion testimony to be admissible.

¶ 50 The Reference Manual on Scientific Evidence notes that “many cases involving issues of external causation have involved witnesses who testify having arrived at an opinion on cause through a process of ruling out or eliminating other causes, a process frequently referred to by the courts and witnesses as ‘differential diagnosis' or ‘differential etiology.’ ” The U.S. Court of Appeals for the Tenth Circuit has explained the method of “differential diagnosis” in the medical context “is a common method of analysis, and federal courts have regularly found it reliable under Daubert.” Differential diagnosis is a method where the expert makes a “determination of which of two or more diseases with similar symptoms is the one from which the patient is suffering, by a systematic comparison and contrasting of the clinical findings,” i.e. , a determination of cause.

¶ 51 Plaintiffs' witnesses used the method of differential diagnosis to evaluate data and then make an opinion on the cause of Mrs. Nelson's demise. In summary, a physician performs a differential diagnosis by first “ruling in” all scientifically plausible causes of the plaintiff's disease or injury, and then “ruling out” the least plausible causes of disease or injury until the most likely cause remains. Defendants argued the testimony of plaintiffs' experts possessed the quality of speculation because they could not (1) “rule in” vasopressin as a coronary vasoconstrictive or that it causes a decreased cardiac index or cardiac arrest, and (2) rule out other potential causes of cardiac arrest which were specific to Mrs. Nelson. While the first objection is one of general causation, the latter raises the issue whether the high dose of vasopressin was a specific cause of the injury.

¶ 52 Most if not all elementary textbooks on statistics explain a statistical truism that correlation is not causation, and we have explained, in the context of an individual as opposed to a group, an opinion may not exclusively use the related post hoc ergo propter hoc (after this, because of this) reasoning to show causation. But the fact that correlational evidence cannot definitively establish causality does not mean that it may not be some evidence of causality. An expert's opinion may rely on a temporal relationship between an alleged cause and subsequent injury as one factor to show causation.

¶ 53 The facts specific to a patient's exposure to an external cause, in this case a high dosage of vasopressin and the doses of dopamine and Levophed and other medications, are facts specific to a particular individual, and are part of Dr. Russell's analysis and Dr. Sheena's analysis of specific causation. Medical diagnosis “is not an exact science,” “physicians make probabilistic judgments on a day-to-day basis,” and they must usually consider and assess alternative causal models before accepting a particular model as the preferred explanation. Dr. Russell and Dr. Sheena considered dopamine and Levophed and discounted them as causes for Mrs. Nelson's cardiac arrest. They considered alternative causes and offered a reasonable explanation as to why they still believed that the defendants' actions were a substantial factor in causing the injury.

¶ 54 Dr. Sheena testified concerning his experience in prescribing vasopressin for his patients. He testified Mrs. Nelson's death was caused by “a combination of septic shock and the overdose of vasopressin.” Dr. Sheena was critical of Dr. Shreck in not finding the CT scan results, starting Mrs. Nelson on antibiotics sooner, and a delay in her surgery of “a four-hour delay, about.” Dr. Sheena's only criticism of Dr. Shepherd was the vasopressin dosage. Dr. Sheena testified the dosage was a contributory cause to Mrs. Nelson's injury, with her decreased chance of survival based upon the vasopressin dose.

¶ 55 In Robinson v. Oklahoma Nephrology Associates, Inc., we explained “A defendant whose conduct contributed to cause a plaintiff's injury is liable for the injury even if his conduct was not sufficient by itself to cause the injury.” Classification of tortfeasors based upon the type of cause each contributed to a plaintiff's injury is nothing new, and is found in our opinions discussing joint and concurrent tortfeasors, as well as well-known legal treatises on causation. In Christian v. Gray, we noted the use of the differential diagnosis methodology to isolate the “most probable” cause or the “most likely” cause of injury.

¶ 56 Because the trial court viewed Dr. Sheena's testimony as inadmissible without him “ruling out” other causes of Mrs. Nelson's cardiac arrest, the trial court did not properly address the scope of Dr. Sheena's opinion on contributory causes of the cardiac arrest including the vasopressin dosage. The trial court ruled Dr. Sheena's testimony was inadmissible because it failed to rule out four potential causes, and one of these expressly stated by the trial court was septic shock. Dr. Sheena's testimony specifically and expressly “ruled in” septic shock as a contributing cause and gave his opinion how Mrs. Nelson's septic shock should have been treated upon a more timely viewing of her CT scan and the medical response to such a viewing, such as administration of antibiotics.

¶ 57 Of course, classifying Dr. Sheena's statements as asserting more than one cause for Mrs. Nelson's injury does not exempt his opinion from the requirements of Daubert. In response to the defendants' Daubert motion, plaintiffs noted Dr. Sheena's experience in following “guidelines for the treatment of septic shock and the use of vasopressin in the treatment of septic shock as a last resort, ... [and these guidelines] were established by four medical groups, the Society of Critical Care Medicine, the American College of Chest Physicians, the European Society of Internal Medicine, and the American College of Emergency Physicians,” of the last of which Dr. Sheena is a member.

¶ 58 Defendants argue Dr. Sheena does not routinely work as a hospitalist and has not treated a sufficient number of patients similar to Mrs. Nelson in a critical care setting. They argue his treatment of patients in an emergency room is not a critical care setting for the purpose of administering vasopressin. They do not discuss how the clinical experience of a physician with many years of clinical experience treating similar patients in the emergency room of small and teaching hospitals would not provide him or her with knowledge of the potential cardiac complications of high dose vasopressin. They do not challenge his experience as an emergency room physician or the propriety of his dosage standard for administering vasopressin in an emergency room. They fault him for relying upon some of the same research as Dr. Russell. They argue a failure on Dr. Sheena's part to show a temporal relationship between the high dose of vasopressin and the cardiac arrest because vasopressin begins to take effect within a few minutes of being administered.

¶ 59 There is more to this latter point on a temporal relationship than what is expressly reviewed in defendant's Daubert motion. Dr. Russell stated he had a particular concern about the high “fixed dose” of vasopressin which remained at that fixed dose and was not tapered for the duration of infusion, although the order had been given by Dr. Shepherd to taper the dose. Dr. Russell also testified concerning a published article discussing “a temporal relationship with cardiac effects” grouped “in a four-hour window” after administration of vasopressin and reactions which could occur “later than that.” At 3:00 p.m., Mrs. Nelson's medication was switched to vasopressin, and at approximately 11:00 p.m. she died. Dr. Russell testified that “for a number of hours” prior to her death she had poorly perfused extremities, and he opined she suffered from vasoconstriction of end organs with cardiac side effects leading to her death.

¶ 60 The admissibility of an opinion by physician expert witness on the issue of external causation of a patient's injury is not predicated on the opinion definitively ruling out every potential cause other than the one alleged by a plaintiff. The Reference Manual states as follows.

At some level, most diseases have multiple host and environmental factors that contribute to their presence. A commonly held misconception is that the presence of a nontoxic or other toxic cause for a condition automatically excludes a role for the toxin being considered as an external

cause ... The converse can also be true ... two toxic agents have been found to interact in a synergistic manner so that their combined effects are much greater than event he sum of their individual effects.

Even if causal factors do not interact synergistically, several may contribute in an incremental fashion to a disease and should not be assumed to be mutually exclusive. Accordingly, the common statement that “alternative causes of disease must be ruled out” before causation is attributed can be more accurately refined to say that “the role of other causes must be adequately considered.”

The Federal Judicial Center, Reference Manual on Scientific Evidence , 476 (2d ed.2000) (material and citations omitted).

Opinions cited by the Reference Manual in support of this quotation include federal Circuit Court opinions explaining a physician's differential diagnosis is not required to rule out every possible cause to determine a cause of a patient's symptoms, and the existence of possible alternative causes “goes to weight and not admissibility” of the physician's opinion. Circuit opinions after publication of the Reference Manual have reached a similar conclusion. These opinions are consistent with our opinions such as Robinson v. Oklahoma Nephrology Associates, Inc. , supra, and our explanation that a defendant whose conduct contributed to cause a plaintiff's injury is liable for the injury even if his or her conduct was not sufficient by itself to cause the injury. Plaintiffs were not required to provide expert testimony that vasopressin was the sole cause of Mrs. Nelson's injury.

¶ 61 Drs. Russell and Sheena did not testify the vasopressin dose was the sole cause of Mrs. Nelson's cardiac arrest. Dr. Russell testified that the maintenance of the high dose contributed in a significant manner to the cause of Mrs. Nelson's cardiac arrest. He also stated “in general” a patient may suffer a cardiac arrest while receiving standard doses of Levophed and dopamine, “but in the case of Mrs. Nelson the vasopressin was the cause of the arrest because the doses of both dopamine and Levophed were decreasing before Mrs. Nelson's cardiac arrest and so dopamine and Levophed were not the cause of cardiac arrest.” His testimony was that Mrs. Nelson's chart showed some post-operative improvement and her symptoms showing a worsening post-operative condition occurred after vasopressin had been administered. Drs. Russell and Sheena testified on the high dose of vasopressin being the most probable cause or contributory cause to Mrs. Nelson's death. We conclude the testimony of Drs. Russell and Sheena are admissible on the issue of specific causation. Their opinions satisfy the requirements of Daubert and 12 O.S. § 2702, for both general causation and specific causation.

Conclusion

¶ 62 The summary judgment granted to the defendants, David Shepherd, M.D., and Enid Medical Associates, Inc., was based upon the trial court excluding the testimony of Drs. Russell and Sheena. We conclude their testimony should not have been excluded. We reverse the orders of the District Court excluding their testimony and granting summary judgment. The opinion of the Court of Civil Appeals is vacated, and the judgment of the District Court is reversed and the cause is remanded to the District Court for further proceedings consistent with the Court's opinion.

¶ 63 CONCUR: REIF, C.J., WATT, EDMONDSON, COLBERT, and GURICH, JJ.

¶ 64 DISSENT: WINCHESTER and TAYLOR, JJ.

¶ 65 NOT PARTICIPATING: KAUGER, J.

¶ 66 DISQUALIFIED: COMBS, V.C.J.