EX PARTE TRINITY INDUSTRIES, INC.

HOUSTON, Justice.

This case presents a question regarding the proper test for determining causation in workers' compensation cases and a question regarding the proper standard for reviewing workers' compensation cases.

On September 2, 1992, Vallie J. Cunningham, who worked as a punch press operator, was rendered permanently and totally disabled as the result of a stroke he suffered while at the workplace of his employer, Trinity Industries, Inc. On December 4, 1992, Cunningham sued Trinity, seeking workers' compensation benefits. The trial court, after a bench trial, entered a judgment for Cunningham, "find[ing] that [Cunningham had] satisfied the dual burden of proving legal and medical causation." Trinity appealed, arguing, among other things, that Cunningham had failed to present substantial evidence of either legal or medical causation. The Court of Civil Appeals affirmed. Trinity Industries, Inc. v. Cunningham, 680 So.2d 253 (Ala.Civ.App.1995). We granted certiorari review to determine whether, in affirming, the Court of Civil Appeals had erred in overruling caselaw regarding the test for causation in "nonaccidental" injury cases and whether that court had misconstrued the "substantial evidence" standard of review.

Cunningham's duties required him to continually lift pieces of metal weighing 15 pounds; he would place a piece in the punch press machine, punch it, remove it, and pick up the next piece.1 On the day he suffered the stroke, Cunningham began his work around 7:00 a.m., and he continued to work until around 11:30 a.m., taking only one 15-minute break. After he left his machine around 11:30, Cunningham walked to the bathroom, sat down on the toilet, and while sitting there experienced symptoms of a stroke. He was later informed that he had suffered a stroke, and that he was an undiagnosed hypertensive. Before the stroke, Cunningham had never experienced high blood pressure.

We have repeatedly stated that an employer is not the absolute insurer of an employee's health and should bear only the costs of compensating employees for accidents that arise out of and in the course of their employment.2 See, e.g., Reynolds Metals Co. v. Gray, 278 Ala. 309, 178 So.2d 87 (1965). Section 25-5-31, Ala.Code 1975, provides:

Whether an accidental injury "arises out of" the claimant's employment is basically a question of whether there is a causal relationship between the claimant's performance of his or her duties as an employee and the complained-of injury. Determining whether a causal relationship has been established between the performance of the claimant's duties as an employee and the complained-of injury is especially difficult and troublesome when the complained-of injury was not produced by some sudden and traumatic external event.3 For simplicity, we will refer to such events as "nonaccidental" injuries. More than 50 years ago, in Pow v. Southern Constr. Co., 235 Ala. 580, 180 So. 288 (1938), this Court held that the term "accident arising out of employment" included more than just incidents in which injuries arise from sudden and traumatic external causes. In Pow, the Court allowed recovery to the family of a person who had contracted and had subsequently died from pneumonia, after having been forced by his employment duties to work outside in very wet conditions. However, the Court did not hold that every injury that could conceivably be argued to have been linked to a person's employment would be compensable. Instead, the Court held that a claimant in such a situation must prove:

Pow, 235 Ala. at 584, 180 So. at 290 (quoting Gulf States Steel Co. v. Christison, 228 Ala. 622, 154 So. 565, 569 (1934)).

The Court of Civil Appeals, in City of Tuscaloosa v. Howard, 55 Ala.App. 701, 318 So.2d 729 (Ala.Civ.App.1975), grappled with the very difficult problem of determining when heart attacks and other similar physical ailments of a "nonaccidental" nature, which, like pneumonia, can and do occur independently of on-the-job risks, "arise out of" the claimant's employment and, therefore, are compensable under our workers' compensation statutes. That court, following the general development of workers' compensation law nationwide, adopted and further refined the two-part causation test set out in Pow v. Southern Constr. Co.4 In order to establish "legal causation," which is the first prong of the causation standard that the Howard case set out, the injured employee had to show that "the performance of the duties for which he [or she] is employed ... expose[d] [him or her] to a danger or risk materially in excess of that to which people not so employed are exposed [ordinarily in their everyday lives]." Howard, 55 Ala.App. at 705, 318 So.2d at 732. Once a claimant establishes "legal causation," he or she then must establish "medical causation," the second prong of the Howard standard, by producing evidence tending to show that the particular exposure to risk proven in regard to prong one "was in fact [a] contributing cause of the [complained-of] injury." Howard, 55 Ala.App. at 706, 318 So.2d at 732.

In its review of this case, the Court of Civil Appeals overruled Howard, based on its conclusion that the holding in Howard "was contrary to our Supreme Court's holding in [Southern Cotton Oil Co. v. Wynn, 266 Ala. 327, 96 So.2d 159 (1957)]." Trinity, 680 So.2d at 257. In Wynn, this Court allowed recovery for a claimant who was "stricken with a cerebral hemorrhage" while he was shoveling coal for his employer. The Court of Civil Appeals correctly pointed out:

Trinity, 680 So.2d at 256.

The causation standard applied in Howard is not inconsistent with this Court's holding in Wynn. A claimant does not have to show any "unusual strain or overexertion" in order to satisfy the first prong of the Howard standard. Rather, to establish "legal causation," one seeking redress under the Workers' Compensation Act for "nonaccidental" injuries need only establish that the performance of his or her duties as an employee exposed him or her to a danger or risk materially in excess of that to which people are normally exposed in their everyday lives. Clearly, a person shoveling coal is exposed to dangers materially in excess of the dangers to which we all face in merely living. The Howard "legal causation" standard was intended to act, and does act, to prevent employers from being unfairly saddled with the cost of being made the absolute insurer of an employee's health. Without a "legal causation" standard, a person who becomes ill or dies because of a natural cause, such as an aneurysm or slipping into a diabetic coma, unrelated to any job-related risk, would be able to recover under our workers' compensation statute merely because he or she was lucky enough to have the disabling event resulting from that natural cause occur at the place of employment or just after the employee has left the place of employment. Such a result was not intended by the legislature when it enacted our workers' compensation law. See, e.g., Reynolds Metals Co. v. Gray, 278 Ala. 309, 178 So.2d 87 (1965); B.F. Goodrich Co. v. Martin, 47 Ala.App. 244, 253 So.2d 37, cert. denied, 287 Ala. 726, 253 So.2d 45, 46 (1971).

The confusion as to whether the Court of Civil Appeals in Howard overstepped the bounds set by our earlier cases such as Southern Cotton Oil Co. v. Wynn is a product of the evolutionary nature of Alabama's workers' compensation law. The Court in Wynn stated that "in order to show [causation], the plaintiff must establish a logical causal connection between his work and the injury." 266 Ala. at 331, 96 So.2d at 162 (emphasis added). In Howard, the Court of Civil Appeals, continuing the on-going effort of Alabama's appellate courts to chart the boundaries of our workers' compensation law and to create coherent and workable standards, enunciated a unified standard5 of causation for all "nonaccidental" injury cases. The Court of Civil Appeals in Howard added form to the "logical connection" causation test set forth by this Court in Wynn, by slightly modifying the Pow standard in order to create a general standard for "nonaccidental" injury cases. The Court of Civil Appeals in Slimfold Mfg. Co. v. Martin, 417 So.2d 199, 201-02 (Ala.Civ.App.1981), cert. quashed, 417 So.2d 203 (Ala.1982), aptly stated:

After much research, we must conclude that Howard sets out the correct causation standard for all workers' compensation claims relating to "nonaccidental" injuries.6

Cunningham's injury occurred on September 2, 1992; therefore, this case is governed by the new Workers' Compensation Act, Ala. Acts No. 92-537, which became effective May 19, 1992. The new Act provides that "[i]n reviewing the standard of proof set forth herein and other legal issues, review by the Court of Civil Appeals shall be without a presumption of correctness." Ala.Code 1975, § 25-5-81(e)(1). It further provides that "[i]n reviewing pure findings of fact, the finding of the circuit court shall not be reversed if that finding is supported by substantial evidence." Ala.Code 1975, § 25-5-81(e)(2).

Although the new Act does not define "substantial evidence," this Court has defined that term as it is used in Ala.Code 1975, § 12-21-12(d), which was enacted before the new Workers' Compensation Act and which pertains to "all civil actions brought in any court of the State of Alabama"; this Court has defined the term "substantial evidence," as it is used in § 12-21-12(d), to mean "evidence of such weight and quality that fairminded persons in the exercise of impartial judgment can reasonably infer the existence of the fact sought to be proved." West v. Founders Life Assurance Co., 547 So.2d 870, 871 (Ala.1989). We need look no further for an adequate, applicable definition. See, also, Ala.Code 1975, § 6-5-542(5); and Clements v. Dr. John Alvan Stewart, P.C., 595 So.2d 858, 861 (Ala.1992) (quoted with approval in Campbell v. Williams, 638 So.2d 804 (Ala.), cert. denied, ___ U.S. ___, 115 S.Ct. 188, 130 L.Ed.2d 121 (1994)), wherein the Court stated that the difference between the definition of "substantial evidence" as that term is used in § 6-5-542(5) and the definition of that term as it is used in § 12-21-12(d) is a "difference without a distinction." Therefore, under the applicable standard of review, we will not reverse the trial court's finding of fact if that finding is supported by substantial evidence—if that finding is supported by "evidence of such weight and quality that fair-minded persons in the exercise of impartial judgment can reasonably infer the existence of the fact sought to be proved." West; § 12-21-12(d). To the extent that Whitsett v. BAMSI, Inc., 652 So.2d 287 (Ala.Civ.App. 1994), and its progeny are inconsistent with this standard, they are overruled.

Trinity argues that Cunningham failed to produce substantial evidence as to the causation element of his workers' compensation claim. As we have stated above, in order to establish causation in a workers' compensation case where the injury is nonaccidental, meaning that the injury was not caused by a sudden and unexpected external event, a claimant must satisfy a two-part causation test by producing substantial evidence establishing both (a) legal causation and (b) medical causation. See City of Tuscaloosa v. Howard, 318 So.2d 729 (Ala.Civ. App.1975); and Ex parte Moncrief, 627 So.2d 385, 388 (Ala.1993) (citing Hammons v. Roses Stores, Inc., 547 So.2d 883 (Ala.Civ.App. 1989)) ("for an injury to be compensable under the Workers' Compensation Act, the employee must establish both legal and medical causation"). Merely showing that there is a close spatial or temporal relationship between the injury and the place or time of the claimant's performance of his or her job is not in itself always sufficient to satisfy either of the two prongs of Alabama's workers' compensation nonaccidental injury causation test.

After reviewing the evidence submitted at trial, we think it is clear that Cunningham produced substantial evidence tending to show legal causation—that in performing his duties as a punch press operator he was exposed to "danger or risk materially in excess" of that danger to which all persons are ordinarily exposed in their everyday lives. Howard, 318 So.2d at 732. Cunningham produced more than adequate evidence to establish that he was exposed to cardiovascular stress in the operation of his punch press "materially in excess" of the base-line level of risk to which we all are exposed in merely living. However, a much closer question exists as to whether Cunningham produced substantial evidence tending to establish "medical causation," the second prong of the Howard nonaccidental injury causation standard.

In order to establish medical causation, a claimant like Cunningham must produce substantial evidence tending to show that the exposure to risk or danger proven in step one of the Howard test "was in fact [a] contributing cause of the injury" for which benefits are sought. Howard, 318 So.2d at 732. "Whether the claimant has satisfied this test ... must be determined on a caseby-case basis." Ex parte Price, 555 So.2d 1060, 1062 (Ala.1989). We further stated in Price that a claimant need not always produce medical expert testimony as to the issue of medical causation in order to carry his burden of proof. Id. In determining whether substantial evidence was produced at trial as to medical causation, we must look to see whether "evidence [was produced at trial] of such weight and quality that fair-minded persons in the exercise of impartial judgment [could] reasonably infer that the risk or danger proven to have existed "was in fact [a] contributing cause of the injury."

Cunningham had to produce substantial evidence linking his stroke to the cardiovascular stress he was subjected to in running his punch press. Because the nature and origin of strokes is obviously beyond the understanding of the average person, this case falls within the Price exception, so that medical evidence in the form of expert medical testimony or excerpts from learned medical treatises was necessary to establish medical causation. See Charles Gamble, McElroy's Alabama Evidence, § 128.10(1), pp. 346-47 (4th ed. 1991) ("It goes without saying that there are certain medical matters which are subject only to expert testimony and are outside the understanding of the lay witness."). As shown from the testimony at trial, doctors and medical researchers themselves disagree vigorously as to what can actually cause the onset of a stroke. See also Charles J. Frankel, James G. Zimmerly, and Richard M. Patterson, eds., Lawyers' Medical Cyclopedia of Personal and Allied Specialties, Vol. 5A, § 34.27a(E) (Supp.1984 & 1995). The question for this Court then becomes whether, when viewed in the light most favorable to Cunningham, as it must be viewed under the applicable standard of review, the expert medical testimony presented by Cunningham in conjunction with his other evidence constitutes "substantial evidence" of "medical causation," as those terms are defined above.

Dr. Gordon Kirschberg, a neurologist testifying for Cunningham, testified as follows:

In order to establish "medical causation" in a workers' compensation case, a claimant need only produce substantial evidence tending to show that the exposure to risk proven in step one of the Howard twopart causation test, in this case cardiovascular stress, "was in fact [a] contributing cause of the injury," in this case the stroke. Howard, 318 So.2d at 732 (emphasis added). Cunningham was not required to produce substantial evidence indicating that the exertion was the only factor contributing to the onset of the stroke. Although Dr. Kirschberg's testimony seems to be weak and at points contradictory, he does conclude that the cardiovascular stress, associated with the strenuous work that Cunningham had been performing for four and one-half hours before collapsing while on a break, could have precipitated the onset of Cunningham's stroke.

266 Ala. at 331, 96 So.2d at 162 (emphasis added). In rejecting the employer's assertion that Wynn's evidence of causation was insufficient, this Court stated: "[T]he circumstances of the plaintiff's injury, considered in conjunction with the medical testimony, support a fair inference that the injury was caused by the exertion of [Wynn's] work." 266 Ala. at 332, 96 So.2d at 163. Although Wynn was decided before the legislature adopted a substantial evidence standard for workers' compensation actions, Wynn`s reasoning is still valid. In the context of certain injuries or diseases, the origin of which in general can be scientifically linked to certain strong risk factors or to certain stimuli but the origin of which as to any one person cannot be scientifically determined with certainty, medical evidence of causation in a workers' compensation case, whether in the form of testimony or treatise excerpts, need only show that the work-related risk could have been a precipitating factor in bringing about the onset of the disease.7

If we were to require a higher level of certainty from medical experts to establish medical causation in cases like this one, we would effectively be removing diseases like stroke from the list of compensable diseases. We would also be encouraging claimants to seek out experts of lower quality and of less integrity. Cunningham's expert testified honestly and forth-rightly. Because of the peculiar nature of strokes, no expert could have legitimately testified that any one factor definitely caused Cunningham's stroke. When we consider Dr. Kirschberg's expert opinion testimony in conjunction with the circumstances of the injury, we must conclude that Cunningham produced substantial evidence of medical causation.

Although the Court of Civil Appeals applied an incorrect rationale, it reached a correct decision. Its judgment is affirmed.

AFFIRMED.

SHORES,* KENNEDY, COOK, and BUTTS, JJ., concur.

HOOPER, C.J., and MADDOX, J., dissent.

HOOPER, Chief Justice (dissenting).

Ala.Code 1975, § 25-5-31, provides for compensation to be awarded for personal injury or death "caused to an employee by an accident arising out of and in the course of his employment." The key question in this case was whether the employee's stroke arose out of the course of his employment.

Since 1975, the test in Alabama for determining whether a worker's job caused certain health conditions has been a two-fold determination based on both a legal test and a medical causation test. City of Tuscaloosa v. Howard, 55 Ala.App. 701, 318 So.2d 729 (Ala. Civ.App.1975). This two-part test has been used most frequently in regard to heart attack claims, but it has also been used in other cases where there was the strong possibility that the injury to the employee had occurred because of pre-existing latent health problems and was not truly work-related. Hellums v. Hager, 360 So.2d 721 (Ala.Civ.App. 1978).

The Court of Civil Appeals opinion in this case explicitly overruled Howard. See Ex parte Trinity Industries, Inc., 680 So.2d 253 (Ala.Civ.App.1995). While I must commend the majority of this Court for attempting to uphold Howard, its opinion has the same practical effect as overruling Howard. This problem requires me to dissent.

Several aspects of the majority opinion cause this problem. One of these aspects is the following statement regarding the facts: "Before the stroke, Cunningham had never experienced high blood pressure." However, neither party claims Cunningham had never experienced high blood pressure. Cunningham had neglected to have his blood pressure checked. The medical testimony indicated he was "undiagnosed hypertensive."

Howard involved a fireman who suffered a heart attack. The fireman suffered chest pains while doing maintenance labor at a fire station, and 11 days later had a heart attack that required hospitalization. The trial court awarded compensation to the fireman. The Court of Civil Appeals in Howard held that the trial court was to determine "if the job caused the injury or death." Howard, at 732, 318 So.2d 729. Howard set out a twopart causation test—legal and medical. The legal causation test: "If in the performance of the duties for which he is employed an employee is exposed to a danger or risk materially in excess of that to which people not so employed are exposed, and an injury occurs, such injury may legally be determined to have arisen from his employment...." Howard, at 732, 318 So.2d 729. (Emphasis added.) The medical causation test: "that such exposure, though operating with or upon other exposure common to all, was in fact, the contributing cause of the injury." Howard, at 732, 318 So.2d 729. (Emphasis added.) In order for the employer to be liable, the employee must satisfy both tests.

The majority opinion describes the Howard test for legal causation as imposing on a worker seeking compensation the burden of showing "that the performance of his or her duties as an employee exposed him or her to a danger or risk materially in excess of that to which people are normally exposed in their everyday lives." It added, "Clearly, a person shoveling coal is exposed to dangers materially in excess of the dangers to which we all face in merely living." 680 So.2d at 267. However, Howard's legal causation test requires that the risk be higher for the plaintiff than for someone not so employed, not higher than for someone "merely living." Compared to "merely living," being employed anywhere doing virtually anything would probably expose the employee to a higher risk. As Cunningham's expert witness, Dr. Kirschberg, testified at trial, "Any sort of physical or excessive mental activity can [raise blood pressure]." Dr. Kirschberg also stated: "[Y]ou can have elevated blood pressure causing or having some cause of influence on the possibility of stroke." Thus, the Howard legal causation test is effectively eliminated when comparing an employee's risk at work to that of "merely living."

The majority opinion restates the medical causation test of Howard. The Howard test is "that such exposure [to a workplace risk], though operating with or upon other exposure common to all, was in fact, the contributing cause of the injury." Howard, at 732, 318 So.2d 729. The majority today restates that test: "[that] the particular exposure was in fact [a] contributing cause of the [complained of] injury." 680 So.2d at 266-67. Thus, the majority today changes the medical causation test from a question of whether the "exposure" is "the cause" or "the contributing cause" to a question whether it is "a cause." The practical implications of this change are huge. Requiring the plaintiff to prove that the work was "the cause" or "the contributing cause" puts the burden on the plaintiff to offer substantial evidence to support a finding that the employment was the leading cause of the injury. The standard stated today—"a cause"—requires evidence only that the employment was one factor among several contributing factors. This change in the Howard legal and medical causation tests eliminates any differentiation between a person's lifestyle and his or her employment. In other words, a person with a family history of heart disease, who smokes two packs of cigarettes a day, and who never exercises could say, "I was at work" (that would prove an exposure above "merely living"); "I had a heart attack while at work." (Being "at work" would be easy to prove as a factor in a subsequent heart attack because the plaintiff almost certainly would have suffered more stress while at work.) The standard as stated in the majority opinion means that a person who, while at work, suffers from the ill effects of any disease or condition could demand workers' compensation from his or her employer. It makes the employer "the insurer" of the employee.

In 1992 the Legislature stated the standard of review for determining the amount of evidence required to support "pure findings of fact" in workers' compensation actions. § 25-5-81(e)(2). The majority today quotes from Southern Cotton Oil Co. v. Wynn, 266 Ala. 327, 331, 96 So.2d 159, 162 (1957), and then states: "Wynn `s reasoning is still valid." 680 So.2d at 271. In Wynn, this Court applied the "scintilla" standard of review for determining the evidence required to support findings of fact. The majority today applies the reasoning of Wynn in discussing the evidence required to find causation.

In addition, Cunningham's injury occurred on September 2, 1992; therefore, this case is governed by the Workers' Compensation Act, as amended by Ala.Acts 1992, Act No. 92-537, which became effective May 19, 1992. Section 25-5-81(c) provides:

(Emphasis added.) Section 25-5-81(c) defines "clear and convincing" proof:

Arguably, the stroke that Cunningham suffered resulted from "gradual deterioration." The majority does not address the application of § 25-5-81(c) to this fact situation.

Section 25-5-81(e) provides:

An appellate review of the standard of proof is de novo, under § 25-5-81(e). Therefore, the evidence requirement of § 25-5-81(c) must be met in order to uphold the trial court's decision in favor of Cunningham. The majority holds that the trial court's decision is to be upheld if it is supported by "substantial evidence." However, "substantial evidence" is not the standard of proof set by the Legislature in § 25-5-81(c). "Substantial evidence" is the standard of review of the trial court's "pure findings of fact." The Legislature has provided that generally in workers' compensation cases the standard of proof is "preponderance of the evidence." In cases "involving injuries which have resulted from gradual deterioration or cumulative stress disorders," the standard of proof is "clear and convincing evidence." I am not convinced the trial court applied the correct standard of proof in this case.

Cunningham's injury is not a typical accident-related injury. As even Cunningham's expert testified: "[T]he most common causes [of a stroke] are atherosclerosis or so-called hardening of arteries and hypertension or high blood pressure, but other things also contribute greatly, either via those two or by themselves. For example, smoking is a major contributor." Cunningham was a heavy smoker, he was overweight, and he was an undiagnosed hypertensive. The gradual deterioration of Cunningham's health, based on these non-work-related factors is the better explanation for the cause of his stroke. Therefore, the standard used for injuries caused by "gradual deterioration" appears to be the more appropriate standard. "Clear and convincing" evidence means evidence that, when weighed against the evidence in opposition, produces "a firm conviction as to each essential element of the claim and a high probability as to the correctness of the conclusion." Ala.Code 1975, § 25-5-81(c). In this case, the doctor that treated Cunningham stated that Cunningham's stroke was not related to his work. The only expert witness provided by Cunningham gave very weak testimony in support of Cunningham's claim. Dr. Nasrollah Eslami, Cunningham's treating physician, testified that he did not believe Cunningham's employment at Trinity contributed to his stroke. Dr. Eslami testified: "See, a stroke, it doesn't happen—it doesn't happen as a result of one factor in a short period of time. It is a disease process that goes on, like having, for example, [a] heart attack. You have a disease process that is slowly progressive in causing changes in the vessel walls[.]" Dr. Eslami further testified that he did not think that working with a punch press and lifting 15 pounds of metal, which was a routine that Cunningham was accustomed to, would have caused extremely high blood pressure: "I don't think that would have caused the extremely high blood pressure."

Dr. Gordon Kirschberg, the plaintiffs' expert witness on whose testimony the majority opinion relies, testified that the work might have been a factor. Dr. Kirschberg, in answer to the question whether physical activity, such as the work in this case, has a role in causing or initiating a stroke, stated: "That is something which nobody really knows.... [Y]ou can have elevated blood pressure causing or having some cause of influence on the possibility of stroke." Dr. Kirschberg testified: "[Y]ou give one last push to the blood pressure [and] it may cause some spasm or little clots to fall off and that may, in fact, be the cause of the stroke."

Not only is the "clear and convincing" standard of proof not met in this case, but the "preponderance of the evidence" standard is not met either. The treating doctor stated that the work had no relationship to the stroke, while Cunningham's expert witness gave weak testimony that barely supported Cunningham's claim.

It should be kept in mind that in nonaccident cases, expert testimony supporting a finding of causation is required. The plaintiff's expert said, "[N]obody really knows." In other words, his statements are only theory, not established medical facts. Further, the expert said "possibly" or "maybe" as to whether that theoretical view applied in Cunningham's case. Dr. Kirschberg testified: "Well, yeah, I want to be very clear about it so I'm not misquoted, but I don't think anybody has any proof that if you elevate your blood pressure there's a definite link between that and having a stroke...." This evidence is not "clear and convincing." Neither does it meet a "preponderance of the evidence" standard.

One can argue that because of the nature of strokes a worker will never be able to prove that his or her employment caused the stroke. This argument may be true, especially when considered in light of the testimony provided by medical experts in this case. However, workers' compensation claims are governed by statute. If the statute has set the standard of proof too high for nonaccidental injuries, then it is the duty of the Legislature to alter the statute. It could be that the Legislature considered the medical fact that non-accident-related injuries, e.g., strokes, are normally caused by many factors over the course of a person's life. Medical experts themselves have not come to a firm opinion as to the cause of strokes. They can happen anytime, anywhere. How can we hold an employer liable for a disease caused by such an indeterminate and questionable source? It is possible the Alabama Legislature had this medical fact in mind when it amended the Workers' Compensation Act in 1992?

The majority opinion effectively eliminates the two-part test of Howard. Also, it has, for all practical purposes, repealed the standard of proof requirements that the Legislature established under § 25-5-81. The potential for recovery on invalid workers' compensation claims, based on the holding of the majority opinion, is unlimited. Now, virtually anyone who has a heart attack, a stroke, or any other nonaccidental injury while at work can recover under the Workers' Compensation Act. I would reverse the judgment of the Court of Civil Appeals.

MADDOX, Justice (dissenting).

The law concerning the burden of proof in workers' compensation cases involving heart attacks and strokes was set out by the Court of Civil Appeals over 20 years ago in City of Tuscaloosa v. Howard, 55 Ala.App. 701, 318 So.2d 729 (Ala.Civ.App.1975). The standard set out in Howard was later reaffirmed in Newman Brothers, Inc. v. McDowell, 354 So.2d 1138 (Ala.Civ.App.1977). In Howard, the Court of Civil Appeals held that there were two standards of causation in workers' compensation claims involving heart attacks: legal causation and medical causation. The employee must present evidence as to both aspects of causation or else he or she has failed to meet the burden of proof.

I agree with the Chief Justice's interpretation of the medical causation test set out in Howard, as he explains it in his dissenting opinion. It appears to me that Howard requires the allegedly harming exposure at work to be "the contributing cause of the injury" upon which the workers' compensation claim is based, not just "a contributing cause." The significance of this difference is that the exposure caused by the employment must be more than one of many different factors causing a compensable injury; it must be the final or the main cause resulting in the "nonaccidental" injury.8

Section 25-5-81(c), Ala.Code 1975, states that "injuries which have resulted from gradual deterioration or cumulative physical stress disorders" will be "compensable only upon a finding of clear and convincing proof." The injury involved here is a stroke. Thus, the question before the Court, based on the record, is whether a stroke is an injury of sudden onset or an injury resulting from a "gradual deterioration." If it is the latter, the statute states that the injury will be "compensable only upon a finding of clear and convincing proof."

The Legislature did not elect to state what it meant when it used the phrase "injuries which have resulted from gradual deterioration," but this phrase presents a medical question. In Howard, the Court of Civil Appeals said the following:

55 Ala.App. at 705-06, 318 So.2d at 732 (emphasis added). Applying the Howard test to the facts of this case, I am unable to find legal evidence to meet either the legal test or the medical test of causation. Stated differently, the record reflects no direct evidence as to whether the stroke was a result of a gradual deterioration or was a sudden onset. Consequently, I cannot agree that the injury is compensable.