Panel consists of Justices COHEN, BRISTER,* and SMITH.**
EN BANC OPINION
TERRY JENNINGS, Justice.Jennings writes for the majority of the en banc Court, joined by Chief Justice Schneider and Justices Hedges, Taft, Nuc-hia, and Radack. Justice Brister concurs in the judgment of the en banc Court. Justice Cohen joins sections I through IV(D) of the majority opinion, does not join section IV(E), and dissents from the judgment of the en banc Court, joined by Justices Mirabal and Smith.
In this personal injury suit for unseaworthiness under the general maritime law and for negligence under the Jones Act,1 appellant, Coastal Tankships, U.S.A., Inc. (Coastal), appeals from a jury verdict awarding damages to appellee, Florence Anderson, administratrix of the estate of Morris Anderson.2 Coastal presents six issues for our review. In its first and second issues, which are dispositive of the appeal, Coastal contends that (1) the trial court abused its discretion in allowing Anderson’s medical expert to testify that *595exposure to the chemical naphtha3 caused Anderson’s bronchiolitis obliterans organizing pneumonia (BOOP) and, thus, (2) there was legally insufficient evidence of medical causation to support the jury’s verdict. We reverse and render.
We note at the outset, in regard to these two issues, that our primary role is to ascertain whether the trial court abused its discretion in finding reliable, and thus in admitting, expert testimony under Texas Rule of Evidence 702. Helena Chem. Co. v. Wilkins, 47 S.W.3d 486, 499 (Tex.2001); Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713, 718-19 (Tex.1998). If expert testimony is not reliable, it is not evidence. Merrell Dow Pharmaceuticals, Inc. v. Havner, 953 S.W.2d 706, 721-29 (Tex.1997). Due to the confusion regarding the appropriate way to analyze the reliability of the expert opinion in this case and the application of Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993), and its Texas progeny, a thorough review of the facts of this case and the applicable case law is necessary.
I. Facts and Procedural Background
Coastal owned and operated the steam tanker S.S. Coastal Manatee. Anderson joined the crew of the Coastal Manatee on July 6, 1994 as a utility hand in the steward’s department. He was healthy and physically capable of performing his duties when his voyage began. Before then, he had suffered no chronic illnesses, pneumonia, or respiratory problems; had never smoked; had been physically active; and had passed Coastal’s health examination. During the voyage, the Coastal Manatee twice loaded naphtha, which the vessel carried for approximately 11 days. Viewed in the light most favorable to Anderson, both times naphtha was loaded, he and other crew members smelled particularly strong chemical fumes throughout the ship. Anderson and at least one other crew member, radio electronics officer Butch Smith, smelled fumes in their quarters. The fumes once drove Smith out of his room to sleep elsewhere.4 Anderson soon began having headaches, shortness of breath, nausea, dizziness, shoulder stiffness, and tightness in his chest. Anderson was sent to his quarters to rest for at least five days, and his condition worsened until he was so ill that he could no longer perform his duties. None of the other approximately 29 crew members aboard the Coastal Manatee became ill from the fumes or developed BOOP, and only one to three other crew members felt occasional nausea. On September 23,1994, Anderson had to be taken ashore aboard a motor launch. When he came ashore, Anderson was sick, had labored breathing, and looked like he “was caving in.”
A few days later, Anderson went to the hospital under the care of Dr. Eugene Brown, who referred Anderson to pulmonary specialist Dr. David Miller. Dr. Miller became Anderson’s treating physician and diagnosed Anderson with BOOP. The evidence in the light most favorable to Anderson shows that, despite objective testing indicating his respiratory health *596gradually improved, he felt symptoms associated with BOOP until his death of prostate cancer, which was unrelated to his BOOP.
Anderson sued Coastal for his personal injuries, alleging negligence under the Jones Act and unseaworthiness under general maritime law. The case was tried to a jury. Over the objection of Coastal, the videotaped deposition testimony of Dr. Miller, Anderson’s sole testifying expert on medical causation, was introduced into evidence.
Dr. Miller, who had been board certified in pulmonary disease and internal medicine for 15 and 20 years respectively, testified that he saw “lots of ... common inhalation type problems” in his practice. Dr. Miller stated that Anderson told him he had inhaled a “gaseous” material (naphtha) while working aboard the Coastal Manatee and related the acute symptoms he had suffered.5 Dr. Miller noted that Anderson’s oral history was consistent with Dr. Miller’s experience of those injured by chemical exposure. Dr. Miller ordered chest x-rays and lung-function tests. After reviewing the lung-function test results, examining Anderson, and reviewing Anderson’s medical records, Dr. Miller initially diagnosed Anderson with pneumonia secondary to chemical inhalation. The lung x-rays showed abnormal areas, which were also consistent with Anderson’s symptoms and with injury from chemical inhalation. Dr. Miller then ordered a surgical biopsy, which revealed that Anderson had BOOP. In the end, Dr. Miller diagnosed that Anderson had sustained a “significant inhalation injury,” consistent with the inhalation of fumes aboard the vessel, and opined that “chemical inhalation” had caused Anderson’s acute lung problems. In reaching this conclusion, Dr. Miller ruled out several other possible causes of Anderson’s BOOP by considering Anderson’s work and personal history, temporal proximity of the fumes and symptoms, medical records, and test results.6
On cross-examination, Dr. Miller conceded that BOOP is usually idiopathic.7 Nonetheless, Dr. Miller believed there was a causal relationship between the chemical exposure and Anderson’s BOOP, based on his examination, the test results, the timing of the illness, and Anderson’s history.8 *597Furthermore, although Dr. Keith Wilson, Coastal’s own medical-causation expert, agreed that BOOP is idiopathic about a third of the time, he conceded that BOOP has some recognized causes, including some toxic fume exposures.9 Dr. Wilson also testified that chemical pneumonia, which Dr. Brown concluded Mr. Anderson first had, can cause BOOP.
However, Dr. Miller was not aware of any literature associating BOOP with exposure to chemicals other than oxides of nitrogen, which he conceded are found “everywhere.” Similarly, Dr. Wilson testified that he was unable to find any literature connecting BOOP to naphtha exposure. Dr. Wilson also testified without dispute that the medical literature associates BOOP with only limited chemical agents, such as nitrogens of oxide, which naphtha does not contain. Additionally, Dr. Miller acknowledged (1) he was not a toxicologist, (2) he did not know how much naphtha Anderson was exposed to or for how long, and (3) his diagnosis presumed that there was a causal relationship between naphtha exposure and BOOP.
Aside from the above expert testimony, the only other evidence of causation in the record was (1) Dr. Brown’s statements in the medical records, the discharge summary, and other notes; (2) the naphtha material safety data sheet (MSDS);10 and (3) lay testimony that Anderson was healthy before inhaling the fumes, but ill very soon afterwards.
The jury found Coastal negligent and the vessel unseaworthy and awarded Anderson $1,254,500, including damages for past physical pain and mental anguish.
II. Standard of Review
We, as an appellate court, may not disturb a trial court’s determination that a witness is or is not qualified as an expert unless a clear abuse of discretion is shown. Gammill, 972 S.W.2d at 718-19; Hernandez v. State, 53 S.W.3d 742, 750 (Tex App.-Houston [1st Dist.] 2001, no pet.). We will not conclude that a trial court abused its discretion simply because we would have ruled differently in the same circumstances or the trial court committed a mere error in judgment. See Hernandez, 53 S.W.3d at 750. The test is not whether the facts present an appropriate case for the trial court’s action in the opinion of the reviewing court; rather, we gauge an abuse of discretion by whether the trial court acted without reference to any guiding rules or principles. E.I. du Pont de Nemours & Co. v. Robinson, 923 S.W.2d 549, 558 (Tex.1995). Thus, a trial court enjoys wide latitude in determining whether expert testimony is admissible. Hernandez, 53 S.W.3d at 750; see Hon. Harvey Brown, Procedural Issues Under Daubert, 36 Hous. L.Rev. 1133, 1159 (1999) [hereinafter “Brown—Procedural”].
III. Expert Testimony
An expert may testify on scientific, technical, or other specialized subjects if the *598testimony would assist the fact finder in understanding the evidence or determining a fact issue. See Tex.R. Evid. 702.
A. Daubert
In Daubert, the Supreme Court held that the Frye11 “general acceptance” test was no longer applicable under the new Federal Rules of Evidence and offered some “general observations.” 509 U.S. at 593, 113 S.Ct. at 2796. These “general observations” set new standards for the admissibility of expert testimony. The Court held that “the trial judge must ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable.” 509 U.S. at 589, 113 S.Ct. at 2795 (emphasis added).
Although “not presum[ing] to set out a definitive checklist or test” in considering the admissibility of expert scientific testimony, the Daubert Court offered the following as a “flexible” inquiry to be made by the trial court:
1. whether the theory or technique “can be (and has been) tested”;
2. whether the theory or technique “has been subjected to peer review and publication”;
3. what is “the known or potential rate of error” for any tests or techniques; and
4. whether there is “general acceptance” in the relevant scientific community.
509 U.S. at 593-94, 113 S.Ct. at 2796-97 (citations omitted).
B. Daubert’s Texas Progeny
Emphasizing the importance of trial courts’ “scrutiniz[ing] proffered evidence for scientific reliability when it is based upon novel scientific theories, sometimes referred to as ‘junk science,’ ” but not limiting its holding to novel theories, the Texas Supreme Court adopted the Daubert analysis in regard to Texas Rule of Evidence 702. Robinson, 923 S.W.2d at 554. Although noting that the factors the trial court may consider in making its determination will differ in each case, the Robinson court offered its own version of the Daubert inquiry, which included, but was not limited to, the following factors:
1. the extent to which the theory has been tested;
2. the extent to which the technique relies upon the expert’s subjective interpretation;
3. whether the theory has been subject to peer review;
4. the technique’s potential rate of error;
5. whether the underlying theory or technique has been generally accepted as valid by the relevant scientific community; and
6. the nonjudicial uses that have been made of the theory or technique.
Robinson, 923 S.W.2d at 557.
Daubert’s test was adopted in almost identical form by the Court of Criminal Appeals.12 For example, after having previously utilized a Dauberesque analysis in Kelly v. State, 824 S.W.2d 568 (Tex.Crim.App.1992),13 the Texas Court of Criminal *599Appeals embraced Daubert in reversing and remanding an appellate court decision that upheld the exclusion of expert testimony on eyewitness identification. Jordan v. State, 928 S.W.2d 550, 555-56 (Tex.Crim.App.1996). The court cited the following “list of nonexclusive factors that could affect a trial court’s determination of reliability”:
1. the extent to which the underlying scientific theory and technique are accepted as valid by the relevant scientific community, if such a community can be ascertained;
2. the qualifications of the expert(s) testifying;
3. the existence of literature supporting or rejecting the underlying scientific theory and technique;
4. the potential rate of error of the technique;
5. the availability of other experts to test and evaluate the technique;
6. the clarity with which the underlying scientific theory and technique can be explained to the court; and
7. the experience and skill of the person(s) who applied the technique on the occasion in question.
Id. at 554 & n. 6 (citing Kelly, 824 S.W.2d at 573) (emphasis added). This inquiry is “substantively identical to the inquiry mandated by the Supreme Court ... in Daubert .... ” Nenno v. State, 970 S.W.2d 549, 560 (Tex.Crim.App.1998), overruled on other grounds, State v. Terrazas 4 S.W.3d 720, 727 (Tex.Crim.App.1999).
We have noted that our highest courts have repeatedly emphasized that the pertinent, suggested inquiries in assessing the reliability of expert testimony are applied flexibly and are not exclusive or required,14 Hernandez, 53 S.W.3d at 752 (citing Daubert, 509 U.S. at 593-94, 113 S.Ct. at 2797 and Robinson, 923 S.W.2d at 557). A trial court may make other inquiries, instead of or in addition to those noted in Daubert, Robinson, and Jordan, that are germane to an expert’s qualifications and field of expertise in determining the reliability of the proffered evidence. Hernandez, 53 S.W.3d at 752. The “methods of proving reliability will vary, depending upon the field of expertise.” 15 Id.
Both the Texas Supreme Court and the Texas Court of Criminal Appeals have also held that non-scientific expert testimony (ie., that involving technical or other specialized knowledge) must also meet the reliability requirement of Daubert/Robinson/Jordan. Gammill, 972 S.W.2d at 718; Nenno, 970 S.W.2d at 562. Both courts have further noted, however, that the Dau-bert/Robinson/Jordan inquiries for assessing the reliability of scientific evidence may not always apply to other kinds of expert testimony. Gammill, 972 S.W.2d at 726; Nenno, 970 S.W.2d at 562; accord *600Helena Chem., 47 S.W.3d at 499 (citing Gammill for proposition that Robinson’s inquiries may not apply to certain testimony).
In Nenno, the Court of Criminal Appeals pointed out that the Daubert/Robin-son/Jordan inquiries “do not necessarily apply outside the hard science context.” Id. at 561 (emphasis added). The court emphasized that “methods of proving reliability will vary, depending upon the field of expertise,” and recognized that,
[w]hen addressing fields of study aside from the hard sciences, such as the social sciences or fields that are based primarily upon experience and training as opposed to the scientific method, Kelly’s requirement of reliability applies but with less rigor than to the hard sciences. To speak of the validity of a “theory” or “technique” in these fields may be roughly accurate but somewhat misleading.
Nenno, 970 S.W.2d at 561 (emphasis added).16
Acknowledging that “hard science methods of validation, such as assessing the .potential rate of error or subjecting a theo*601ry to peer review, may often be inappropriate for testing the reliability of fields of expertise outside the hard sciences,” the Nenno court offered an appropriately tailored translation of the Daubert inquiry to areas outside hard science, as follows:
1. whether the field of expertise is a legitimate one;
2. whether the subject matter of the expert’s testimony is within the scope of that field; and
3. whether the expert’s testimony properly relies upon and/or utilizes the principles involved in the field.17
Id. 18
Mindful of these guiding rules and principles, we address Coastal’s first and second issues.
IV. Analysis
In its first issue, Coastal contends the trial court abused its discretion by allowing Anderson’s medical expert, Dr. David Miller, to testify that naphtha exposure caused Anderson’s BOOP. Specifically, Coastal claims Dr. Miller’s methodology was unreliable. At trial, Coastal timely objected to Dr. Miller’s opinion testimony on causation. Once Coastal objected, Anderson had the burden to show that Dr. Miller’s methodology was sound. See Gammill, 972 S.W.2d. at 718.
A. Applicable Causation Burdens and Principles
Two burdens of proof apply. Under the Jones Act, Anderson had to show that Coastal’s negligence played any part, “even the slightest,” in producing Anderson’s BOOP. E.g., Maritime Overseas Corp. v. Ellis, 971 S.W.2d 402, 406 (Tex.1998). For his unseaworthiness claim, Anderson had to show that the un-seaworthy condition played a substantial part in bringing about or actually causing Anderson’s BOOP and that his BOOP was either a direct result or a reasonably probable consequence of the Coastal Manatee’s unseaworthiness. E.g., Offshore Pipelines v. Schooley, 984 S.W.2d 654, 663 (Tex.App-Houston [1st Dist.] 1998, no pet.).
Two causation principles also apply. First, in regard to the causation issues presented in this case, we are mindful that toxic-tort case law is instructive by analogy. One common way of understanding toxic-tort causation is to break it down into general causation and specific causa*602tion.19 General causation asks whether a substance is capable of causing a particular injury in the general population; specific causation asks whether that substance caused a particular individual’s injury. See Havner, 953 S.W.2d at 714; Neal v. Dow Agrosciences L.L.C., 74 S.W.3d 468, 472 (Tex.App.-Dallas 2002, no pet. h.) (designated for publication) (quoting Havner and applying distinction); Minn. Mining & Mfg. Co. v. Atterbury, 978 S.W.2d 183, 191 (Tex.App.-Texarkana 1998, pet. denied) (same); Sanders at 110.20 Proving one type of causation does not necessarily prove the other, and logic dictates that both are needed for a plaintiff in a toxic-tort suit to prevail.21 See Atterbury, 978 *603S.W.2d at 199-200, 203; Sanders at 110.22
Second, expert testimony is required to prove causation in this case, contrary to Anderson’s claim. To establish causation in a personal injury case, a plaintiff must prove the defendant’s conduct caused an event and that that event caused the plaintiff to suffer compensable damages. See Burroughs Wellcome Co. v. Crye, 907 S.W.2d 497, 499 (Tex.1995). The causal link between the event sued upon and the plaintiffs injuries must be shown by competent evidence. Morgan v. Compugraphic Corp., 675 S.W.2d 729, 731 (Tex.1984). A jury may decide the required causal nexus between the event sued upon and the plaintiffs injuries when (1) general experience and common sense will enable a lay person fairly to determine the causal nexus, (2) expert testimony establishes a traceable chain of causation from injuries back to the event, or (3) expert testimony shows a probable-cause nexus. Weidner v. Sanchez, 14 S.W.3d 353, 370 (Tex.App.-Houston [14th Dist.] 2000, no pet.); Blankenship v. Mirick, 984 S.W.2d 771, 775 (Tex.App-Waco 1999, pet. denied).
The evidence was undisputed that BOOP has several possible causes and is idiopathic a third of the time. General experience and common sense simply do not enable a fair understanding of causation under these circumstances; accordingly, expert testimony is required.23 See Lenger v. Physician’s Gen. Hosp., Inc., 455 S.W.2d 703, 706 (Tex.1970) (noting that “when the proof discloses that a given result may have occurred by reason of more than one proximate cause, and the jury can do no more than guess or speculate as to which is, in fact, the efficient cause, the submission of such choice to the jury has been consistently condemned by this court and by other courts.”); Ins. Co. of N. Am. v. Myers, 411 S.W.2d 710, 713 (Tex.1966) (holding that, when coupled with insufficient expert testimony, lay testimony and factual circumstances did not show injury aggravated cancerous tumor); Allen v. Roddis Lumber & Veneer Co., 796 S.W.2d 758, 763 (Tex.App.-Corpus Christi 1990, writ denied) (holding lay affidavit created no fact issues on cause of formaldehyde emissions); Hernandez v. Tex. Employers Ins. Ass’n, 783 S.W.2d 250, 252-53 (Tex.App.-Corpus Christi 1989, no writ) (holding expert testimony needed to determine cause of asthma, which had uncertain causal nature); Orkin Exterminating Co. v. Davis, 620 S.W.2d 734, 736-37 (Tex.Civ.App.-Dallas 1981, writ ref'd n.r.e.) (holding that, because expert testimony *604implied various possible causes, lay jury could not determine whether injuries were caused by accident); Ill. Employers Ins. of Wausau v. Wilson, 620 S.W.2d 169, 172 (Tex.Civ.App.-Tyler 1981, writ ref'd n.r.e.) (holding lay testimony alone did not suffice to connect injury to particular infection, given disease’s complexity and time between injury and loss); Porter v. Whitehall Labs., Inc., 9 F.3d 607, 612 (7th Cir.1993) (requiring expert testimony on whether medicine caused condition leading to renal failure and applying Daubert to that testimony).
B. The Reliability of Dr. Miller’s Differential Diagnosis
Here, because Dr. Miller’s medical causation opinion testimony falls outside the category of “hard science,” and not all of the Daubert/Robinson/Jordan inquiries thus make sense when applied to his opinion testimony, we turn to Nenno’s translation of the these inquiries. We therefore consider
1. whether Dr. Miller’s field of expertise is a legitimate one;
2. whether the subject matter of Dr. Miller’s testimony is within the scope of that field; and
3. whether Dr. Miller’s testimony properly relies upon and/or utilizes the principles involved in the field.
See Nenno, 970 S.W.2d at 561.
The trial court would not have abused its discretion if it impliedly found that Dr. Miller met the first two Nenno inquiries. As a pulmonologist, Dr. Miller’s field of expertise is certainly legitimate. His testimony further revealed that he has extensive experience in diagnosing and treating acute lung injuries, including those caused by chemical exposure. His causation opinion testimony concerned Anderson’s BOOP and was within the scope of his field of expertise.
In regard to his reliance upon and utilization of the principles of his field, Nen-no’s third inquiry, Dr. Miller used a “differential diagnosis,” sometimes referred to as “differential etiology.” This is a clinical process whereby a doctor determines which of several potential diseases or injuries is causing the patient’s symptoms by ruling out possible causes — by comparing the patient’s symptoms to symptoms associated with known diseases, conducting physical examinations, collecting data on the patient’s history and illness, and analyzing that data — until a final diagnosis for proper treatment is reached. See Neal, at 473 n. 3 (quoting Atterbury); Atterbury, 978 S.W.2d at 199-200, 203; Kannankeril v. Terminix Int’l, Inc., 128 F.3d 802, 807 (3rd Cir.1997). Differential diagnosis is “the basic method of internal medicine” and enjoys widespread acceptance in the medical community. See, e.g., Heller v. Shaw Indus., Inc., 167 F.3d 146, 156 (3rd Cir.1999) (quoting In re Paoli)) In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 755, 758 (3rd Cir.1994). Generally speaking, when properly conducted the technique has important non-judicial uses, is generally accepted as valid by the medical community, and has been subjected to use, peer review, and testing. See Robinson, 923 S.W.2d at 557.
Even with all the advances of medical science, the practice of medicine remains an art. A properly conducted and explained differential diagnosis is not “junk science.” If a differential diagnosis provides a sufficient basis on which to prescribe medical treatment with potential life-or-death consequences, it should be considered rehable enough to assist a fact finder in understanding certain evidence *605or determining certain fact issues.24
In answer to Nenno’s third inquiry, Dr. Miller’s opinion was based on what the trial court could reasonably have concluded was a properly explained and conducted differential diagnosis, which is in itself a reliable and widely accepted methodology. Thus, we hold that the trial court did not abuse its discretion if it impliedly found Dr. Miller gave a reliable differential diagnosis under Daubert/Robinson/Jordan.25
*606C. The Limits of What Dr. Miller’s Differential Diagnosis Can Prove
Our holding that the trial court did not abuse its discretion if it impliedly found Dr. Miller offered a rehable differential diagnosis does not end our inquiry, however. The question remains as to what Dr. Miller’s differential diagnosis can prove.
The federal courts are split on whether a differential diagnosis, without more, can reliably show both specific and general causation under Daubert in the toxic-tort context. See Heller, 167 F.3d at 154-55 (recognizing disagreement); see generally Sanders at 121 (“Nevertheless, substantial disagreement still remains on what constitutes a rehable differential diagnosis. Two dimensions highlight this disagreement: (1) whether one must ‘rule in’ the putative cause before ‘ruling out’ other causes; and (2) whether temporal order alone — that the cause preceded the effect — is sufficient to support the causal attribution.”); Hon. Harvey Brown, Eight Gates for Expert Witnesses, 36 Hous. L.Rev. 743, 849 (Fall 1999) [hereinafter “Brown-Gaies ”] (recognizing split); Jack E. Karnes, Establishing the Standard for a Physician’s Patient Diagnosis Using Scientific Evidence: Dealing with the Split of Authority Amongst the Circuit Courts of Appeal, 15 B.Y.U. J. Pub.L. 1 (2000).
One branch of federal case law indicates that a properly conducted and explained differential diagnosis, by itself or with little else, could sufficiently establish causation for a toxic-tort plaintiff to prevail.26 *607These courts apparently view a differential diagnosis as establishing both general and specific causation (even if they do not always use those precise terms). See generally Sanders at 124 (discussing and collecting cases). Anderson relies on cases like lege
In contrast, other federal courts indicate that a differential diagnosis does not, by itself, establish causation sufficiently for a toxic-tort plaintiff to prevail. See generally id. at 122-24 (discussing and collecting cases). Rather, these courts also require some type of “hard science’ on the issue of general causation (even if they do not always use that precise term), i.e., whether the chemical can cause the illness generally, before allowing in differential-diagnosis testimony.27 They likely impose this re*608quirement because a differential diagnosis presumes, but does not solely establish, that the chemical is capable of producing harmful effects generally. See Brown-Gates at 858 (citing Hall v. Baxter Healthcare Corp., 947 F.Supp. 1387, 1413 (D.Or.1996)). That is, a differential diagnosis can only “rule out,” but does not itself “rule in,” the possible causes of an illness. See Cavallo v. Star Enters., 892 F.Supp. 756, 771-72 (E.D.Va.1995), aff'd in part on same grounds, rev’d in part on other grounds, 100 F.3d 1150, 1159 (4th Cir.); Sanders at 122-24 (“The ‘rule in before rule out’ position of Cavallo presumes that, at least in toxic tort cases, a differential diagnosis, no matter how well done, can rarely prove general causation by itself. The Cavallo position has been repeated in numerous cases.”) (collecting cases); Brown-Gates at 858 (citing Cavallo). Coastal relies on cases like these.
At least one Texas court has concluded, in the context of a toxic-tort suit for negligence and product liability, that a doctor’s differential diagnosis by itself would, at best, be evidence of specific causation only:
Another form of scientific evidence is known as differential diagnosis. This involves a history of the patient and a physical examination, and refers to the process a physician would undertake to eliminate other generally known causes of illnesses and injuries to diagnose a specific cause of a particular patient’s illness or injury. The Texas Supreme Court has stated, “an expert’s assertion that a physical examination confirmed causation should not be accepted at face value.” [Havner ] at 720. Clearly, an expert’s recitation that he has examined a patient and has done a history of the patient and has concluded that X caused the patient to suffer with Y would not be sufficient to support causation. If the physician explained the exact methodology that he used in arriving at the conclusion, including discussing the exact other causes that have been ruled out and the generally accepted literature that he relied upon in making that conclusion, the differential diagnosis evidence could be sufficient to prove specific causation. Even though some courts have held that differential diagnosis is a valid form of evidence to support general causation, it most likely is not sufficient under current standards promulgated by the Texas Supreme Court. This statement is made based upon what appears to be the court’s total reliance on objective evidence and its disdain for any form of subjective analysis.
Atterbury, 978 S.W.2d at 199.28
Although Atterbury was a toxic-tort case, we follow the reasoning of the Atter-*609bury court and federal cases like it because we believe those cases properly utilize Texas’s guiding rules and principles regarding the reliability of expert testimony.29 Under Atterbury’s logic, if it is proven that a chemical generally causes a particular illness (general causation), only then would a properly conducted and explained differential diagnosis be relevant to show that the chemical caused the illness in the plaintiff (specific causation). See id., 978 S.W.2d at 200; accord Raynor v. Merrell Pharms., Inc., 104 F.3d 1371, 1375-76 (D.C.Cir.1997) (holding doctor’s causation opinion based on differential diagnosis was properly deemed inadmissible, when, among other things, no reliable evidence first proved general causation). This is because a differential diagnosis presumes that chemical X can cause condition Y generally, but does not itself so prove.
Dr. Miller’s testimony provided nothing outside his differential diagnosis except the temporal proximity of the naphtha fumes to Anderson’s symptoms. Dr. Miller admitted that he was not a toxicologist, did not know what chemicals comprised naphtha, and had read no literature connecting naphtha and BOOP.30 Although both Dr. Miller and Dr. Wilson testified that the lack of naphtha/BOOP studies did not necessarily preclude a causal relationship, a lack of scientific evidence cannot excuse imposing liability without proof of causation. See Austin, 25 S.W.3d at 292 (citing Havner, 953 S.W.2d at 727-28). Both medical experts further testified that BOOP was idiopathic about a third of the time, with the exception of several identified causes, none of which was known to be naphtha. Accordingly, Dr. Miller’s differential diagnosis, no matter how reliable the trial court impliedly found it to be, could at most show specific causation. See Atterbury, 978 S.W.2d at 199, 200-01; accord Raynor, 104 F.3d at 1375-76; Cavallo, 892 F.Supp. at 771-72 (excluding expert’s testimony when supported only by differential diagnosis, but lacking principles and methods of toxicology); see also Sanders at 122-24, 130 (“The opinions that require the [differential diagnosis] expert to ‘rale in before ruling out’ are, in our view, correct.”); Brown-Gates at 849-66. Dr. Miller’s differential diagnosis merely presumed that naphtha could be a source of BOOP generally, then eliminated other possible sources of Anderson’s BOOP until only naphtha was left. Thus, because Dr. Miller’s diagnosis simply presumed that naphtha could cause BOOP, that diagnosis, without more, would not be reliable to show general causation, i.e., that naphtha can cause BOOP generally. See, e.g., Atterbury, 978 S.W.2d at 199, 200-01.
D. The Result
The very nature of Dr. Miller’s differential diagnosis would limit that diagnosis to proving specific causation. Furthermore, Anderson failed to meet his burden to show that Dr. Miller’s diagnosis could also be reliable as to general causation. In the toxic-tort context, a plaintiff must establish general causation for a differential diagnosis to be relevant to show *610specific causation. See, e.g., Atterbury, 978 S.W.2d at 200; Raynor, 104 F.3d at 1376. Accordingly, the trial court 'will have abused its discretion in admitting Dr. Miller’s differential diagnosis, and we will sustain Coastal’s first issue, if the remaining record does not also contain general-causation evidence that is reliable under Daubert/Robinson/Jordan.
We note that our dissenting colleagues join the opinion thus far; however, they part ways on the remaining issue of whether the record contains reliable general-causation evidence.
E. The Sufficiency and Reliability of the General Causation Evidence on Which Anderson Relies
In its second issue, Coastal claims there was legally insufficient evidence of medical causation to support the jury’s verdict. We review the remaining record for sufficient evidence of general causation. We apply the reliability test as articulated in Daubert/Robinson/Jordan in our sufficiency review as well.31 Austin, 25 S.W.3d at 284-87 (citing Havner, 953 S.W.2d at 712).
Anderson argues there was legally sufficient evidence of causation under the Jones Act’s “featherweight” causation standard. Anderson further argues other evidence in the record constituted legally sufficient evidence of causation. Anderson refers us to “evidence that came from the various doctor’s [Dr. Brown’s] reports, the exhibits (particularly the MSDS), and the lay testimony [of timing].”
First, the Jones Act’s “featherweight” causation standard cannot transform no evidence into some evidence. The proper focus is not on the causation burden of proof, but on whether the expert opinion testimony is rehable in the first place. As noted above, if expert opinion testimony is unreliable, it is no evidence, not even a feather’s weight. See Havner, 953 S.W.2d at 712, 713 (“The testimony of an expert is generally opinion testimony. Whether it rises to the level of evidence is determined under our rules of evidence, including Rule 702.... If the expert’s scientific testimony is not rehable, it is not evidence.”).
Second, Anderson did not present Dr. Brown (or his differential-diagnosis notes) as his causation expert, and Dr. Brown’s notes came in, along with all the medical records, as exhibits to Dr. Miller’s deposition testimony presented at trial. Even if Dr. Brown’s differential diagnosis (as stated in the medical records, discharge summary and other notes) could nonetheless be considered expert testimony, it, like Dr. Miller’s, could show at most specific causation.
Third, the evidence presented here revealed that BOOP has several possible causes and is often idiopathic. General experience and common sense simply do not enable a fair understanding of either specific or general causation under these circumstances. Accordingly, expert testimony was required. Anderson offered no additional expert testimony showing that, generally, naphtha can cause BOOP. Simi*611larly, no expert established that Dr. Brown’s diagnostic records, the naphtha MSDS,32 or any other non-expert evidence on which Anderson now relies met the appropriate Daubert/Robinson/Jordan inquiry as to general causation. It is this evidentiary gap that is fatal to Anderson’s claims.
Put another way, if Anderson had produced only Dr. Brown’s differential diagnosis (as stated in the medical records, discharge summary, and other notes), the naphtha MSDS, and Dr. Wilson’s statement that chemical pneumonia can cause BOOP, would he have carried his Daubert/Robinson/Jordan burden? He would not. Cf. Havner, 953 S.W.2d at 717-28 (overturning jury verdict based on various experts’ causation opinions and explaining type of specific, detailed showing needed to carry one’s burden under Robinson and Rule 702). How, then, can those same evidentiary fragments survive the sufficiency challenge made here and below? They cannot. It was Anderson’s burden to provide expert testimony linking these evi-dentiary pieces together through the lens of Daubert/Robinson/Jordan. See Gammill, 972 S.W.2d at 718 (proponent’s burden); Allen, 796 S.W.2d at 763 (expert testimony required). Anderson failed to meet this burden.
Accordingly, the trial court abused its discretion in admitting Dr. Miller’s specific-causation testimony when the record also lacked general-causation testimony by an expert applying inquiries like those under Daubert/Robinson/Jordan. See Atterbury, 978 S.W.2d at 199-200, 203 (holding toxic-tort plaintiff must prove both specific and general causation to prevail). Thus, there was legally insufficient evidence to support the jury’s finding that Anderson’s naphtha exposure “played any part, even the slightest” in causing his BOOP. See Ellis, 971 S.W.2d at 406.
We sustain Coastal’s first issue. Because we sustain Coastal’s first issue, we must also sustain its second issue.
Because Coastal’s first and second issues are dispositive of the appeal, we need not address the remaining issues.
Y. Conclusion
The United States Supreme Court, the Supreme Court of Texas, and the Texas Court of Criminal Appeals have repeatedly emphasized that the suggested inquiries to determine the reliability of an expert’s opinion are to be used flexibly and are not exclusive. Daubert, 509 U.S. at 593-94, 113 S.Ct. at 2797; accord Kumho Tire Co., 526 U.S. at 141-42, 150-51, 119 S.Ct. at 1171, 1175-76; Robinson, 923 S.W.2d at 557; Nenno, 970 S.W.2d at 561.
We emphasize that a trial court may also consider other factors, not previously suggested, that are germane to an expert’s qualifications and field of expertise in determining the reliability of proffered expert opinion testimony. The “methods of proving reliability will vary, depending upon the field of expertise.” Nenno, 970 S.W.2d at 561. A trial court, “in discharging its duty as gatekeeper must determine how the reliability of particular testimony is to be assessed.” Hernandez, 53 S.W.3d at 752 (citing Gammill, 972 S.W.2d at 726). Trial courts, functioning in their role as gatekeepers, have wide latitude in admitting reliable expert testimony and excluding unreliable expert testimony. We respect the discretion of trial court judges in performing their gatekeep-ing function and will not disturb their rul*612ings on the reliability of expert testimony unless it appears from the record they acted without reference to pertinent guiding rules or principles. See Robinson, 928 S.W.2d at 558.
Although we respect the discretion of trial courts in performing their gatekeep-ing function, we must conclude, in this case, that the trial court abused that discretion in admitting Dr. Miller’s specific-causation testimony when there was no general-causation testimony by an expert. Thus, there was legally insufficient evidence in this case that exposure to naphtha played any part, “even the slightest,” in causing Anderson’s BOOP.
We reverse and render judgment in favor of Coastal.
Justice Brister dissented from the panel’s decision to affirm the trial court’s judgment.
En banc consideration was requested. Tex.R.App. P. 41.2(c).
A majority of the Court voted for en banc consideration of the panel’s decision. See id.
The en banc Court consists of Chief Justice SCHNEIDER and Justices COHEN, MIRABAL, HEDGES, TAFT, NUCHIA, JENNINGS, RADACK, KEYES, BRISTER, WILSON, and SMITH 33.Justice JENNINGS, writing for the majority of the en banc Court, joined by Chief Justice SCHNEIDER and Justices HEDGES, TAFT, NUCHIA, and RA-DACK. See Tex.R.App. P. 47.5.
Justice BRISTER concurring in the judgment of the en banc Court. See id.
Justice COHEN, joined by Justices MIRABAL and SMITH, joining only sections I through IV(D) of the en banc Court’s majority opinion and dissenting from the judgment of the en banc Court. See id.
Justices KEYES and WILSON not participating. See id.
The Honorable Scott Brister, who became Chief Justice of the Fourteenth Court of Appeals on July 16, 2001, continues to participate by assignment for the disposition of this case, which was submitted on May 7, 2001.
The Honorable Jackson B. Smith, retired Justice, Court of Appeals, First District of Texas at Houston, participating by assignment.
. 46 App.U.S.C. § 688 (2002).
. For simplicity’s sake, we refer to both Mr. Anderson and his wife, acting as his adminis-tratrix, as "Anderson.”
. "Naphtha” is “a name originally applied to an inflammable volatile liquid (a constituent of asphalt and bitumen) issuing from the earth in certain localities; now extended to most of the inflammable oils obtained by dry distillation of organic substances, esp. coal, shale, and petroleum.” The Compact Oxford English Dictionary 1146 (2d ed.1991).
. Although Smith moved toward the part of the vessel housing Anderson's room to escape the fumes, Smith testified that he could still smell some fumes where he had moved, although the air was better.
. The record does not support Justice Bris-ter’s implication, in his concurrence, that Anderson was untruthful or litigation-motivated in relating his history to Dr. Miller. It is undisputed that Dr. Miller was Anderson’s treating physician, who conducted his diagnosis as part of Anderson’s treatment shortly after Anderson's coming ashore. Furthermore, the record does not support Justice Brister's conclusion that "Anderson's BOOP was diagnosed by biopsy, not by process of elimination” and that "this was no diagnosis.” The biopsy ultimately confirmed BOOP, but only after the history, examination, and other tests narrowed down the field, as discussed below. Therefore, Dr. Miller's differential diagnosis was an essential part of both Anderson’s treatment and the ultimate diagnosis of BOOP.
. Dr. Miller ruled out (1) viral and bacterial pneumonia because Anderson had had no fever (a disputed fact, which we must resolve in Anderson’s favor); (2) tuberculosis because Anderson stated he had not been exposed to that disease; (3) asthma, cancer, "wet lung syndrome,” "acute respiratory distress syndrome,” and smoking-related injury because of Anderson's medical history and tests; and (4) asbestosis, despite Anderson's having been exposed to asbestos for about 30 years, because the biopsy did not support that disease.
. "Idiopathic” means "a disease of unknown cause.” Stedman's Medical Dictionary 848 (26th ed.1995).
. As the immediately preceding discussion of Dr. Miller's diagnosis shows, that diagnosis did not, as Justice Brister’s concurrence claims, rely "heavily” or solely on Anderson's truthfulness. Rather, the doctor used Anderson’s oral history as a starting point for *597objective medical tests. In any event, it is well-settled that statements made for purposes of medical diagnosis and treatment are deemed so reliable that they are excepted from the hearsay rule. See TexR. Evid. 803(4).
. Dr. Wilson testified that other identified causes of BOOP are infections, tissue transplantation, inflammatory conditions (like arthritis), and some medications.
. The MSDS indicated that naphtha could irritate mucous membranes and the respiratory tract and act as an asphyxiant; could, upon overexposure, lead to headache, nausea, drowsiness, fatigue, pneumonitis, pulmonary edema, and central nervous system depression; and could cause stomach irritation, unconsciousness, congestion, and hemorrhaging of the lung and internal organs.
. Frye v. United States, 293 F. 1013 (1923).
. We note that the supreme court found the reasoning of the Court of Criminal Appeals persuasive in developing Robinson’s reliability test. See E.I. du Pont de Nemours & Co. v. Robinson, 923 S.W.2d 549, 556 (Tex.1995) ("We are persuaded by the reasoning in Daubert and Kelly [v. State, 824 S.W.2d 568 (Tex.Crim.App.1992)].”).
. Although Kelly involved novel scientific evidence, the court later concluded that the standard established in that case applied to all scientific evidence, whether novel or not. *599Hartman v. State, 946 S.W.2d 60, 62-63 (Tex.Crim.App.1997).
. The factors a trial court will find helpful in determining whether the underlying theories and techniques of the proffered evidence are scientifically reliable "will differ with each particular case.” Robinson, 923 S.W.2d at 557. Thus, rulings on the reliability of proffered expert testimony are necessarily case specific.
. Justice Brister, in his concurrence, confuses Daubert’s reliability requirement with creating "standards we must apply.” Justice Brister’s rigid application of the Daubert/Rob-inson/Jordan inquiries has been expressly rejected by our higher courts. Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 141-42, 150-51, 119 S.Ct. 1167, 1171, 1175-76, 143 L.Ed.2d 238 (1999); Helena Chem., 47 S.W.3d at 499 (citing Gammill; Gammill, 972 S.W.2d at 726; Nenno, 970 S.W.2d at 562). Thus, we are not following our "own brand-new standards.”
. For this particular holding, Nenno relied in part on the panel opinion of the United States Court of Appeals for the Fifth Circuit in Moore v. Ashland Chem., Inc., which was vacated by the en banc court. 126 F.3d 679, 685-89 (5th Cir.1997), vacated en banc, 151 F.3d 269 (5th Cir.1998), cert. denied, 526 U.S. 1064, 119 S.Ct. 1454, 143 L.Ed.2d 541 (1999); see also Nenno, 970 S.W.2d at 561. That reliance does not undermine Nenno’s holding, however. In the original Moore opinion, the panel held that the district court had erred in excluding, under Federal Rule of Evidence 702, a doctor's differential-diagnosis based opinion that the plaintiff’s exposure to chemical gasses caused the plaintiffs disease because that opinion was "well grounded in the principles and methodology of his field of clinical medicine," i.e., was a reliable differential diagnosis. Id. at 689-90, 694-96, 701-02, 703. In so holding, the panel opinion noted that "clinical medicine (as opposed to research and laboratory medical science) is not a hard science discipline; its goals, subject matter, conditions of study and well developed methodology are sui generis and quite different from that of hard science and its methodology.” Id. at 682; see also id. at 688-90. The Moore court thus a recognized that the Daubert inquiries are flexible and may, but do not necessarily, apply outside the hard-science context. In overruling the panel and holding that the trial court correctly excluded the doctor's opinion under Daubert, the en banc Moore court expressly "rejected the position that application of the Daubert factors is unwarranted in cases where expert testimony is based solely on experience or training.” Moore, 151 F.3d at 275 n. 6. As discussed below in section IV(C), the en banc Moore opinion correctly reflects Texas law in its ultimate disposition, i.e., effectively holding that a differential diagnosis by itself, without further scientific evidence that "rules in” the chemical as a cause, is usually not reliable to prove general causation in the toxic-tort context. See id. at 277-79. However, the en banc Moore opinion does not reflect Texas law to the extent that the opinion rigidly applies the Daubert inquiries to a field (clinical diagnosis, not a "pure” hard science) in which those inquiries are ill-suited. See Moore, 151 F.3d at 275 n. 6. In contrast, both our state courts of last resort have explicitly held that the specific Daubert/Robin-son/Jordan inquiries for assessing the reliability of scientific evidence may not always apply to other kinds of expert testimony. See Gammill, 972 S.W.2d at 726; Weatherred v. State, 15 S.W.3d 540, 542 (Tex.Crim.App.2000) (holding Nenno’s test applies to “soft" science); Nenno, 970 S.W.2d at 561-62; see also Helena Chem., 47 S.W.3d at 499 (Robinson’s inquiries might not apply to certain testimony). In fact, less than a year after the en banc Moore opinion issued, the Supreme Court expressly rejected the position, espoused by the en banc Moore court, that the specific Daubert inquiries necessarily apply rigidly to all areas of expertise. Kumho Tire Co., 526 U.S. at 141-42, 150-51, 119 S.Ct. at 1171, 1175-76; accord Fed.R.Evid. 702 advisory committee's notes (2000) (noting that (1) Daubert inquiries are neither exclusive nor dispositive and cannot always apply to every type of expert testimony and (2) amended rule is broad enough for consideration of any or all Daubert inquiries when appropriate).
. The Nenno court clarified that it was not attempting "to develop a rigid distinction between 'hard' science and 'soft' sciences, or nonscientific testimony” because “the distinction between various types of testimony may often be blurred.” Id. at 560-61. Rather, the court noted that its observations applied flexibly and to all types of expert testimony and were merely an appropriately tailored translation of the Daubert inquiries. Id. at 561. We thus consider Nenno’s third inquiry broad enough to encompass any of the Daubert and Robinson inquiries, and any other inquiry not listed in those cases, that is appropriate to the expert and the expert’s field.
. See also Henderson v. State, 77 S.W.3d 321, 324-25 (Tex.App.-Fort Worth 2002, no pet. h.) (designated for publication) (following Nenno, noting that specific Daubert inquiries do not necessarily apply outside hard-science context, and declining to apply those specific inquiries in holding trial court did not abuse discretion in admitting expert testimony involving “clinical medicine”); In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 758 (3rd Cir.1994) ("Unfortunately, most of the Daubert factors ... are of limited help in assessing whether the methodologies of [two doctors performing clinical diagnoses] are reliable (i.e. scientifically valid). We find that the final Daubert factor, the existence of standards controlling the technique's operation, is slightly more helpful in deciding whether the methodologies of the plaintiffs’ doctors are reliable.”).
. Justice Brister's criticism in his concurrence that this Court has created a "new general causation rule” is without merit. Our observations in regard to general and specific causation in toxic-tort-type cases are taken directly from Havner: "Sometimes, causation in toxic tort cases is discussed in terms of general and specific causation. General causation is whether a substance is capable of causing a particular injury or condition in the general population, while specific causation is whether a substance caused a particular individual's injury.” Id., 953 S.W.2d at 714 (citations omitted). These observations are nothing new, nor do they create a bright-line rule on how to prove causation. They merely state the obvious. In fact, the model of general and specific causation is widely accepted among federal courts of appeals. We note that Justice Brister quotes one sentence from a law review article in support of his position, but this sentence is taken out of context, and the article supports this Court’s analysis:
Cause-in-fact in toxic tort cases is usually thought of as two separate issues: general causation and specific causation. [Fn. 13: Arguably, these are separate issues in all tort cases. However, the general causation issue is often obvious. That cars striking trees at 60 mph might cause injury to occupants is not a point requiring expert testimony.] ....
It would be a mistake to argue that causal issues in toxic tort cases are fundamentally different from those presented in other tort cases. However, toxic tort cases do differ in degree in several significant ways. First, often there is causal ambiguity. The level of exposure ... is often uncertain. Evidence of a relationship between the substance and the injury is often uncertain. The timing between the exposure and the disease may be suspect.... Second, there is a fundamental problem of multiple causation. ... Other substances [then asbestos], however, do not cause unique injuries, and substances that do cause signature diseases may also cause others.... Third, ... [p]laintiffs may find it difficult to prove that a particular injury was the result of the defendant’s substance or another cause.” Joseph Sanders, Julie Machal-Fulks, The Admissibility of Differential Diagnosis Testimony to Prove Causation in Toxic Tort Cases: the Interplay of Adjective and Substantive Law, 64 Law & Contemporary Problems 107, 110-11 (Aut.2001) (citations omitted) [hereinafter "Sanders”].
. See also S.W. Ref. Co. v. Bernal, 22 S.W.3d 425, 440 (Tex.2000) (Baker, J., concurring, joined by Hecht, J.) (in discussing personal-injury class-action suit arising from refinery fire, noting problems of creating class for "general causation” issues, and another for "specific causation” issues); Austin v. Kerr-McGee Ref. Corp., 25 S.W.3d 280, 288, 292 (Tex.App.-Texarkana 2000, no pet.) (using and applying both terms in analyzing toxic-tort causation); Texas Workers' Comp. Ins. Fund v. Lopez, 21 S.W.3d 358, 364-65, 366 (Tex.App.-San Antonio 2000, pet. denied) (using and applying both terms in analyzing personal injuries alleged to result from sandblasting).
. The supreme court recognizes that it is possible that a toxic-tort plaintiff may not be able to find reliable direct evidence of specific causation. See Havner, 953 S.W.2d at 715. A plaintiff in such a situation may be able to prove specific causation circumstantially by taking general-causation evidence, such as epidemiological studies, and showing he is similar to the studies’ subjects. See id. at 720; Atterbury, 978 S.W.2d at 191 (quoting Hav-ner): Sanders at 110. Thus, depending on the circumstances and the proof available to a plaintiff, similar evidence, if properly adapted, may possibly be used to prove both general and specific causation. However, both aspects of causation must still be proved.
. We disagree with Justice Brister’s position, in his concurrence, that we need not discuss specific and general causation and Dr. Miller’s differential diagnosis. The parties each rely heavily on cases that weigh in on opposite sides of the differential-diagnosis debate. That debate, and some of those cases in particular, are analyzed in terms of general and specific causation. It is the parties' reliance on those cases, and our determining why we agree or disagree with those cases, that requires us to address these issues. Additionally, to decide this case, we must resolve whether Dr. Miller made a differential diagnosis and, if so, whether that diagnosis is good for nothing (Coastal’s position) or good for everything (Anderson's position). That is the heart of the parties' dispute. That dispute can be resolved only by discussing the specific/general-causation aspect of the differential-diagnosis debate.
. Anderson's reliance on Morgan, in which the supreme court held that lay testimony sufficiently showed chemical fumes caused an injury, is misplaced. Lay testimony sufficed in Morgan because a default judgment left no evidentiary dispute as to whether “general experience and common sense” could allow a lay person to determine causation. 675 S.W.2d at 733. Here, in contrast, both parties’ experts testified that BOOP is sometimes idiopathic or at best has multiple causes that are hard to determine. Moreover, Morgan predates Robinson and Havner.
. See In re Paoli, 35 F.3d at 758 ("[Differential diagnosis involves assessing causation with respect to a particular individual. This merely makes it a different type of science than science designed to produce general theories; it does not make it unreliable science.”) (emphasis added); Benedi v. McNeil-P.P.C., Inc., 66 F.3d 1378, 1384 (4th Cir.1995) ("We will not declare such methodologies invalid and unreliable in light of the medical community’s daily use of the same methodologies in diagnosing patients.”); Sanders at 120, 121 ("Moreover, no opinion we know has concluded that differential etiology, when properly performed, is inadmissible.... Courts accept the general validity of the technique of differential diagnosis.”); see also Green v. Texas Workers’ Comp. Ins. Facility, 993 S.W.2d 839, 844 (Tex.App.-Austin 1999, pet. denied) ("We begin with the observation that medicine is as much art as science, and clinical medicine necessarily evolves from a process of trial and error. As such, clinical medicine has traditionally been granted some leeway in a determination of admissibility. A physician may not always be able to isolate to a mathematical certainty why human beings react in certain ways or why certain ailments respond to particular cures. This does not necessarily render such diagnoses and treatments ineffective or unreliable. Unfortunately, we have yet to reduce all of life’s mysteries to a discrete set of scientific principles.”); cf. Fed.R.Evid. 703 advisory committee's notes (1972) ("Thus a physician in his own practice bases his diagnosis on information from numerous sources and of considerable variety, including statements by patients and relatives, reports and opinions from nurses, technicians and other doctors, hospital records, and X rays.... The physician makes life-and-death decisions in reliance upon them. His validation, expertly performed and subject to cross-examination, ought to suffice for judicial purposes.”).
. Justice Brister’s criticism, in his concurrence, that our holding is that "differential-diagnoses are always reliable” and we create a "new differential-diagnosis rule” is without merit. Our holding is not that Dr. Miller’s differential diagnosis was reliable or that a differential diagnosis is always reliable. Our role is not to determine reliability at all. Rather, our role is to ascertain whether the trial court abused its discretion in finding Dr. Miller’s causation opinion testimony reliable. It appears Justice Brister would substitute his judgment for that of the trial court and hold Dr. Miller’s causation opinion testimony was unreliable.
Moreover, we disagree with Justice Bris-ter’s suggestion, in his concurrence, that a differential diagnosis cannot support any aspect of legal causation when applied to an illness with some unknown causes. Such a bright-line rule would be unduly restrictive in a world in which many things are not or cannot be known with absolute certainty. We acknowledge that "[a] differential diagnosis that fails to take serious account of other potential causes may be so lacking that it cannot provide a reliable basis for an opinion on causation.” Westberry v. Gislaved Gummi AB, 178 F.3d 257, 265 (4th Cir.1999); c.f., Cooper v. Smith & Nephew, Inc., 259 F.3d 194, 202 (4th Cir.2001) (outside toxic-tort context, relying on Westberry for same). However, a specific-causation conclusion based on a differential diagnosis " 'should not be excluded because [a medical expert] has failed to rule out every possible alternative cause of a plaintiff's illness.' The alternative causes suggested by a defendant 'affect the weight that the jury should give the expert’s testimony and not the admissibility of that testimony’. ...” Westberry, 178 F.3d at 265 (quoting Heller; citations omitted); accord Heller v. Shaw Indus., Inc., 167 F.3d 146, 156 (3rd Cir.1999) (holding trial court should not have excluded differential diagnosis on basis that it failed to rule out every possible alternative cause, not merely obvious ones); Sanders at 121 (“Moreover, most courts would agree ... that a failure to account for all possible causes does not render an expert opinion based on differential diagno*606sis inadmissible.”); cf. Purina Mills, Inc. v. Odell, 948 S.W.2d 927, 936 (Tex.App.-Texarkana 1997, pet. denied) ("Disputed facts may be established by circumstantial or direct evidence. Absolute certainly is not required. Nor must the plaintiff exclude every other possibility. All that is required before there can be a finding of ultimate fact is proof of a causal connection beyond the point of conjecture or mere possibility.”); Cooper, 259 F.3d at 202 (citing Westberry for same).
. See, e.g., Turner v. Iowa Fire Equip. Co., 229 F.3d 1202, 1208-09 (8th Cir.2000) (although holding differential diagnosis there unreliable, stating that a proper and reliable differential diagnosis would have sufficiently shown causation even without published studies, etc.); Heller, 167 F.3d at 154-55 (concluding thorough differential diagnosis could support general causation even if no published studies existed in support, if supported by strong temporal relationship); Zuchowicz v. United States, 140 F.3d 381, 385, 389-90 (2nd Cir.1998) (holding one doctor's differential diagnosis alone, plus temporal proximity of overdose and illness, sufficiently supported causation fact finding); McCullock v. H.B. Fuller Co., 61 F.3d 1038, 1043-44 (2nd Cir.1995) (differential diagnosis, plus doctor's review of MSDS and various scientific and medical treatises (none of which said chemical’s fumes caused injury at issue), supported trial court's decision to admit expert’s testimony); see also Sanders at 124 ("Contrary to these [‘rule in before rule out'] holdings, several courts appear either to reject the requirement that one must first rule in before ruling out or to reject the requirement that plaintiff’s expert cite clear evidence that the substance in question can cause injuries at the dose levels experienced by the plaintiff.”) (collecting cases); id. at 127-28 ("As in the case with ruling in before ruling out, however, a number of courts have permitted experts to testify as to specific causation based on little more than temporal order. Not surprisingly, perhaps, many of these cases are the same ones adopting the minority position on ruling in before ruling out.”) (collecting cases). Cf. Cooper v. Smith & Nephew, Inc., 259 F.3d 194, 200 (4th Cir.2001) (although there holding, outside toxic-tort context, differential diagnosis unreliable, noting reliable differential diagnosis is valid foundation under Rule 702 for medical-causation testimony); Hardyman v. Norfolk & W. Ry. Co., 243 F.3d 255, 261-65 (outside toxic-tort context, differential diagnosis was acceptable method of proving causation when plaintiff's experts also testified that dose/response relationship, etc., could not be quantified and had not been scientifically studied). Although there was at least some evidence of, or no dispute about, general causation in them, further cases also contain language supporting the broad use of differ*607ential diagnoses in proving causation. See Westberry, 178 F.3d at 262-65 (differential diagnosis, plus temporal proximity and MSDS, supported trial court’s decision to admit expert's causation testimony concerning talc’s effect on plaintiff’s sinus condition; however, defendants did not dispute that high talc levels could irritate sinus); Kannankeril, 128 F.3d at 808-09 (holding trial court erred in excluding expert’s testimony when opinion supported by differential diagnosis, temporal proximity, and reading of standard textbooks and references; however, "widely accepted scientific knowledge of nature of” chemical’s toxicity existed); Sanders at 130 (noting that some evidence of general causation existed in some of these cases).
. See Meister v. Med'l Eng’g Corp., 267 F.3d 1123, 1129, 1131 (D.C.Cir.2001) (affirming JNOV for defendant in breast-implant case when verdict was based, in part, on treating doctor’s differential diagnosis, which in turn was based on timing and review of unreliable case reports that were contrary to "voluminous epidemiological evidence”); Black v. Food Lion, Inc., 171 F.3d 308, 312-14 (5th Cir.1999) (reversing trial court's admission of expert testimony that was little less than differential diagnosis because the expert's testimony did not “bear the necessary indicia of intellectual rigor,” whether measured under Daubert or under lesser test trial court applied); Moore, 151 F.3d at 273, 278-79 (affirming trial court’s exclusion of testimony by expert who conducted apparent differential diagnosis, relied on temporal proximity and the MSDS, and relied on study involving another chemical without explaining that chemical’s connection to chemical at issue; stating, "In the absence of an established scientific connection between exposure and illness, or compelling circumstances ... the temporal connection between exposure to chemicals and an onset of symptoms, standing alone, is entitled to little weight in determining causation.”); Raynor v. Merrell Pharrns., Inc., 104 F.3d 1371, 1375-76 (D.C.Cir.1997) (holding trial judge properly deemed inadmissible doctor’s causation opinion based on differential diagnosis, when, among other things, no reliable evidence first proved general causation and, in fact, "overwhelming body of contradictory [general-causation] epidemiological evidence” existed); Cavallo v. Star Enters., 892 F.Supp. 756, 771-72 (E.D.Va.1995) (excluding expert’s testimony when supported only by differential diagnosis, but lacking principles and methods of toxicology), off d in part on same grounds, rev’d in part on other grounds, 100 F.3d 1150, 1159 (4th Cir.); see also Sanders at 122-24 ("The ‘rule in before rule out' position of Cavallo presumes that, at least in toxic tort cases, a differential diagnosis, no matter how well done, can rarely prove general causation by itself. The Caval-lo position has been repeated in numerous cases.”) (collecting cases); id. at 126 (“Post-Daubert opinions that apply the Daubert factors frequently take the position that determinations of causation based solely on temporal order should be excluded for lack of reliability”) (collecting cases); Brown-Gates at 850-66. Anderson correctly notes that at least some judges in Moore and Cavallo considered the questions presented in those cases "close” and that the appellate courts there held the respective trial courts' rulings were within the trial courts’ discretion, the opposite of what Coastal seeks to do here. See Moore, 151 F.3d at 278-79 (affirming trial court's exclusion of expert testimony); id. at 279 (Bena-vides, J., concurring) (because issue was "close one,” concluding trial court would not have erred in admitting the evidence, either); Cavallo, 100 F.3d at 1159 (noting admissibility question was "close one,” but deferring to trial court's discretion to exclude testimony); *608see also Green, 993 S.W.2d at 844 (assuming without deciding that trial court's excluding treating physician’s causation testimony, based primarily on what appears to be differential diagnosis, was not abuse of discretion, but noting issue was close one). However, the Fifth Circuit Court of Appeals has since held a trial court abused its discretion in admitting causation testimony based on little more than a differential diagnosis. See Black, 171 F.3d at 312-14.
. See also id. at 200, 201-02 (concluding one doctor's differential diagnoses, and second doctor’s causation opinion based on temporal proximity, constituted some evidence of specific causation, but no evidence of general causation); accord Brown-Gaf&s at 849-59 ("Indeed, it is now clear that differential diagnosis alone will not necessarily make a clinical physician’s causation testimony admissible.”; "A treating physician's differential diagnosis does not address general causation. On the contrary, a differential diagnosis 'assumes that general causation has been proven for the list of possible causes it eliminates.... Thus, in the absence of general causation evidence, courts have excluded, or found insufficient, physician's causation testimony that was based upon a *609differential diagnosis.") (collecting cases); Sanders at 130 ("The opinions that require the [differential diagnosis] expert to 'rule in before ruling out' are, in our view, correct.”).
. We recognize that Atterbury and many of the cases holding similarly involved more latent or long-term cause-and-effect issues than exist here. Nonetheless, we find the Atterbury holding the most analogous to the situation here.
. We note that Dr. Wilson testified without dispute that no scientific studies linking naphtha and BOOP existed, although some studies linked a limited number of other chemicals to BOOP.
. We observe that proving general causation in the toxic-tort or analogous context will usually involve the use of “hard science.” See Atterbury, 978 S.W.2d at 199-203 (in toxic-tort context, applying Daubert/Robin-son/Jordan inquiries to assess reliability of plaintiffs' experts' opinions on general causation); Cavallo, 892 F.Supp. at 771-72 (in toxic-tort context, requiring that expert opinion on general causation be derived from "scientifically valid methodology” and thus requiring expert to apply principles and methods of toxicology on same). It is for this reason that the specific Daubert/Robinson/Jordan inquiries are usually more pertinent to proof of general causation in this context. We note, however, that the observations in this footnote are not intended to create a bright-line rule.
. Additionally, because there is no evidence of what tests were conducted in compiling the MSDS, the MSDS has limited value in determining causation. See Moore, 151 F.3d at 278.