delivered the Opinion of the Court.
We granted certiorari to review the decision of the Colorado Court of Appeals in
Carroll v. CUNA Mut. Ins. Society,
I.
This case arises from Lyman Carroll’s claim for insurance benefits under a group accidental death and dismemberment policy issued by CUNA and covering Mr. Carroll’s wife, Marie Carroll, as an “Insured Person.” The insurance policy provided benefits for “bodily injury caused by an accident occurring while the Group Policy is in force as to the Insured Person and resulting directly and independently of all other causes in loss covered by the Group Policy.” Upon review of the circumstances of Mrs. Carroll’s death, CUNA refused payment. CUNA asserted that Mrs. Carroll’s death was not caused by an “accident” and did not result “directly and independently of all other causes,” as required by the terms of the insurance policy. Mr. Carroll then filed the present suit in El *748 Paso County District Court. He requested a declaratory judgment establishing his entitlement to benefits for Mrs. Carroll’s death under the terms of the insurance policy. The case was tried to the court without a jury.
The circumstances of Mrs. Carroll’s death are taken from the testimony of Mr. Carroll at trial. On March 22, 1990, Marie Carroll was a sixty-seven year old woman of generally good health who suffered from hypertension. 1 At approximately 9:00 in the evening of March 22, Marie Carroll and her husband began to engage in sexual intercourse. A short time later, as Marie Carroll approached orgasm, she leaned forward and fell to the floor. When Mrs. Carroll was arising from the floor she stated that she felt as if her head had exploded and that it was burning. She told her husband she was suffering from a severe headache.
That night and the next day Mrs. Carroll had trouble verbalizing and continued to feel unwell. Mrs. Carroll, however, went about her regular activities throughout the day. At approximately 9:00 in the evening on March 23,1990, Mr. Carroll took Mrs. Carroll to the hospital. The attending physician determined that Mrs. Carroll was suffering from a massive hemorrhage in the left side of her brain. Later that evening, Marie Carroll slipped into a coma. The next morning, after examination, Mrs. Carroll was declared brain dead.
Dr. Randall Bjork, a neurologist, testified at the trial regarding the medical causes of Mrs. Carroll’s death. He stated that Marie Carroll had died from a massive intracere-bral hemorrhage. According to Dr. Bjork, Mrs. Carroll suffered this hemorrhage due to the rupture of an aneurysm 2 in her brain. The rupture occurred in part as a result of Marie Carroll’s elevated blood pressure during intercourse. Dr. Bjork explained that hemorrhages occurring at the point of orgasm are a known phenomenon in the medical literature and occur at a frequency of approximately one in every 300,000 people each year. Dr. Bjork further testified that the aneurysm probably had been present in Mrs. Carroll’s brain for many years prior to the time of the rupture. He described an aneurysm as something like a “powder keg” or “time bomb.” Almost anything, including ordinary events such as interviewing a babysitter or playing chess, can trigger its rupture. Dr. Bjork also testified that Marie Carroll’s existing hypertension could have increased the likelihood that the aneurysm would rupture and the amount of hemorrhage resulting.
After Mr. Carroll had presented his case, CUNA moved for a judgment in its favor on the ground that Mr. Carroll had not proved his case by a preponderance of the evidence. The trial court granted CUNA’s motion. The trial court found that Mrs. Carroll suffered from hypertension and an aneurysm in her brain prior to the incident in question. In addition, the trial court found that on the evening of March 22, 1990, during intercourse, Mrs. Carroll sustained elevated blood pressure on top of her existing hypertension. The court further found that the ruptured aneurysm occurred as part of Mrs. Carroll’s orgasm and not as a result of her fall. The trial court determined that the rupture of the aneurysm had caused Mrs. Carroll’s death and that the rupture could have occurred at any point during normal, reasonable, everyday activity. After noting that the insurance policy did not define the term “accident,” the trial court relied on the case of
Bobier v. Beneficial Standard Life Ins. Co.,
Mr. Carroll appealed the trial court’s judgment to the Colorado Court of Appeals. The court of appeals affirmed the trial court but did so on the basis of a different rationale. The court of appeals acknowledged that the
Bobier
definition of accident was controlling and that the definition could encompass the circumstances of the present case. The court of appeals noted, however, that the insurance policy, besides requiring that the injury be caused by an accident, also required that the injury “result directly and independently of all other causes.”
Carroll,
Mr. Carroll sought review of the court of appeals’ decision by this court. We granted certiorari to determine:
Whether the court of appeals erred in interpreting the clause in an accidental death and dismemberment insurance policy, “directly and independently of all other causes,” to preclude coverage for any injury or death that is “due, even in part, to a preexisting bodily infirmity” such as Mrs. Carroll’s condition.
We now hold that the court of appeals’ construction of the phrase “directly and independently of all other causes” to preclude coverage when the injury or death is due even in part to a preexisting bodily infirmity was unduly narrow. Instead, we conclude that this phrase means that the accident must be the predominant cause of injury in order for the injury to be compensable. Because the trial court’s findings establish that Mrs. Carroll’s death was predominantly caused by her preexisting aneurysm, we hold that the insurance policy did not cover Mrs. Carroll’s death. Therefore, we affirm the court of appeals’ judgment but do so by adopting a different construction of the language of the policy.
II.
The accidental death policy that CUNA issued to Mr. Carroll defines a covered “injury” as follows:
Injury means bodily injury caused by an accident occurring while the Group Policy is in force as to the Insured Person and resulting directly and independently of all other causes in loss covered by the Group Policy.
For an injury to be compensable under this policy it must be caused by an accident and result directly and independently of all other causes. We first address whether Mrs. Carroll’s injury was caused by an accident. 3
The CUNA policy does not define the term “accident.” Relying on the Colorado Court of Appeals’ decision in
Bobier v. Beneficial Standard Life Ins. Co.,
When interpreting insurance contracts, we strive to define terms in accordance with their commonly accepted meanings.
Reed v. United States Fidelity and Guaranty Co.,
Courts have had difficulty crafting a definition of the term “accident” that will fairly apply to all contingencies. 4 When considering this problem some courts have attempted to distinguish between accidental means and accidental results. The basis for this distinction is the relationship between cause and effect. Under the accidental means test, the precipitating cause of the injury must be accidental or unintended. This test encompasses one common understanding of the term accident as an unexpected action or event, such as a slip or a fall, that then causes injury. Under the accidental results test, only the injury, not the precipitating cause, must be unexpected for the injury to be accidental.
Courts strictly applying the accidental means test have held that death or injury does not result from an accident or accidental means within the terms of an accident insurance policy where it is the natural result of the insured’s voluntary act, unaccompanied by anything unforeseen except the death or injury.
See, e.g., Smith v. Continental Casualty Co.,
Commentators have criticized the distinction between accidental means and accidental *751 results as illusory and contrary to the normal expectations of the average policy holder. The basis for the criticism is the difficulty of distinguishing between voluntary and involuntary acts. Appleman states the difficulty as follows:
Almost every action we take has some element of design; if we drive an automobile upon the highway, where another collides with us, could this not have been foreseeable, or at least within the realm of potential so far as the “means” were concerned? Every instance where a person walks, or jumps, and slips or falls, comes within a like category. To permit a rigid construction of such expressions is to permit a deception to be practiced upon the public,....
1A Appleman § 368, at 492.
The modem trend is to reject the distinction between accidental means and accidental results when considering whether a particular death or injury is accidental.
See
John D. Ingram and Lynne R. Ostfeld,
The Distinction Between Accidental Means and Accidental Results In Accidental Death Insurance,
12 Fla.St.U.L.Rev. 1, 9 (1984) (hereinafter, Ingram and Ostfeld). Under this view, the unexpected consequences of an individual’s behavior provide the “accidental element” for purposes of an insurance policy.
5
See, e.g., Taylor v. John Hancock Mut Life Ins. Co.,
11111.2d 227,
The rejection of the distinction between accidental means and accidental results was first articulated by Justice Cardozo in his dissent in
Landress v. Phoenix Mut. Life Ins. Co.,
An effect which is the natural and probable consequence of an act or course of action is not an accident, nor is it produced by accidental means. It is either the result of actual design, or it falls under the maxim that every man must be held to intend the natural and probable consequence of his deeds. On the other hand, an effect which is not the natural or probable consequence of the means which produced it, an effect which does not ordinarily follow and cannot be reasonably anticipated from the use of those means, an effect which the actor did not intend to produce and which he cannot be charged with the design of producing under the maxim to which we have adverted, is produced by accidental means....
Landress,
In
Reed v. United States Fidelity and Guaranty Co.,
we adopted Justice Cardozo’s reasoning in
Landress
and rejected the distinction between accidental means and accidental results. 176 Colo, at 572,
The Colorado Court of Appeals in
Bobier v. Beneficial Standard Life Ins. Co.,
There was some dispute in the evidence as to what caused the aspiration. The insurance company presented evidence that the insured had cardiac arrhythmia, and that the cardiac arrhythmia caused an arrhythmic seizure which then caused the aspiration.
Bobier,
We reaffirm our adherence to the definition of accident as set forth in footnote two of Justice Cardozo’s dissenting opinion in Landress and adopted by us in Reed. See supra at pp. 751-752. The distinction between accidental means and accidental results is too illusory to be useful. The ordinary purchaser of an insurance policy is probably unaware of the distinction between these concepts. An unanticipated or unusual result flowing from a commonplace cause may be an accident. Although this definition is broad and could in the extreme bring about results inconsistent with common understanding, we believe that insurance companies wishing to avoid the consequences of the broadly inclusive definition of accident in Reed must include limiting language readily comprehensible by the ordinary purchaser. This position is further bolstered by the fact that the definition of accident that we reaffirm today has been the law in this state for over twenty years. Insurance companies have had ample time to devise language that properly balances the actuarial risk of liability with the insurance premiums they charge consumers. Thus, we hold that a voluntary act that causes an unforeseeable, unintended, or unexpected result can be considered an accident.
Therefore, we agree with Mr. Carroll that Mrs. Carroll's death was accidental under the terms of CUNA’s insurance policy. The trial court found that during intercourse Mrs. Carroll experienced elevated blood pressure on top of her preexisting hypertension. The trial court also found that the rupture of Mrs. Carroll’s aneurysm occurred as part of the orgasm she experienced during sexual intercourse. The clear implication of these findings is that Mrs. Carroll’s voluntary participation in sexual intercourse with her husband contributed in part to her death. Death was certainly not an expected, intended, or foreseeable result of intercourse. We therefore find that under the definition of accident adopted in Reed and reaffirmed in this opinion, Mrs. Carroll’s death was an accident. Thus, Mr. Carroll has satisfied the first requirement for recovery under the insurance policy.
Ill
We now consider the second requirement for recovery under the CUNA insurance policy. Under the insurance policy, to be compensable, the injury not only must be caused by accident but also must result directly and independently of all other causes in loss covered by the policy. The court of appeals interpreted the phrase “directly and independently of all other causes” to preclude coverage when the injury or death is due even in part to a preexisting bodily infirmity.
Carroll,
In
Maguire,
the insured suffered from a recurring mental illness diagnosed as schizophrenic reaction, paranoid type. The insured was sitting in his house with a shotgun across his lap. When his wife approached, he threatened her with the gun. She then called the police.
Id.
at 176,
The court of appeals’ construction in
Ma-guire
of the phrase “directly and independently of all other causes” to require that the accident be the predominant cause of loss is consistent -with our opinion in
Reed v. United States Fidelity and Guaranty Co.,
In a strict or literal sense, any departure from an ideal or perfect norm of health is a disease or an infirmity. Something more, however, must be shown to exclude the effects of accident from the coverage of a policy. The disease or the infirmity must be so considerable or significant that it would be characterized as disease or infirmity in the common speech of men.
Reed,
176 Colo, at 574,
Thus, the court of appeals’ holding that the insurance policy in this case precludes liability any time the injury or death is due even in part to a preexisting illness or disease sweeps too broadly. The court of appeals’ holding would deny recovery even if the disease is a remote cause of the injury. This is contrary to the expectations of the average policy holder and to the policies discussed in Reed. See also 1A Appleman § 362, at *755 489-90. We hold, therefore, that the court of appeals’ decision in this case excessively limits recovery under accidental injury and death insurance policies. 11 The proper interpretation of the clause “directly and independently of all other causes” is that contained in Maguire. Benefits under accidental death or injury policies of the type at issue here are recoverable as long as one can show that the accident is the predominant cause of the loss.
Other jurisdictions have interpreted the phrase “directly and independently of all other causes” in accidental injury insurance policies to mean that the accident must be the predominant cause of the injury.
12
See, e.g., Quesinberry v. Life Ins. Co. of North America,
Other courts have interpreted the phrase “directly and independently of all other causes” to require that the accident be the direct or proximate cause of the injury.
See, e.g., INA Life Ins. Co. v. Brundin,
In what seems to us a preponderance of American jurisdictions, the test is whether the accidental injury, as contrasted with the contributing disease or bodily condition, is the proximate or predominant cause of the disability or loss, sometimes additionally qualified as the active, efficient, dominant, originating, or direct cause.
Mahon,
We, therefore, interpret the expression “directly and independently of all other causes” to require that the accident be the predominant cause of injury. The court of appeals’ statement that this phrase precludes recovery whenever the injury is due, even in part, to a preexisting bodily infirmity is incorrect.
IV.
We now consider whether the trial court properly granted judgment for CUNA *756 at the conclusion of the plaintiff’s case in light of our holding that to recover under the insurance policy Mr. Carroll must prove by a preponderance of the evidence that the predominant cause of Mrs. Carroll’s death was an accident. In its findings and conclusions, the trial court wrote:
The Court also finds, based on the testimony of plaintiffs expert, that the rupture of this preexisting aneurysm on March 22, 1990 caused Mrs. Carroll’s death and that the rupture could have occurred at any point during any kind of normal, reasonable everyday activity....
[[Image here]]
The Court finds that these facts do not constitute an accident under the definition in Bobier because the rupture of the aneurysm and resulting death do not “flow” from a commonplace cause. The rupture of the aneurysm is a totally separate intervening cause which could have occurred at any time. It does not flow in any sense of the imagination from the commonplace cause of intercourse.
Although the trial court based its analysis on the definition of accident in Bobier, the main focus of its findings is that there was little causal relationship between Mrs. Carroll’s sexual activities and her death. The trial court states that “the aneurysm did not ‘flow" from a commonplace cause” and that the rupture “could have occurred at any time.” The main cause of Mrs. Carroll’s death was her preexisting aneurysm and hypertension. Sexual intercourse was merely coincident with the rupture and not the predominant cause.
We therefore conclude that although the trial court and the court of appeals misinterpreted the law, the findings of the trial court demonstrate that it reached the correct result in entering a judgment for CUNA. The evidence even when considered in the light most favorable to Mr. Carroll supports the trial court’s findings, which in turn establish that the predominant cause of Mrs. Carroll’s death was her preexisting aneurysm and hypertension. Mr. Carroll has not sustained his burden of proving by a preponderance of the evidence that Mrs. Carroll’s death was caused “directly and independently of all other causes.” We therefore affirm the judgment of the Colorado Court of Appeals but do so based on reasoning different from that of the court of appeals.
Notes
. Hypertension denotes abnormally high blood pressure. Webster’s Ninth New Collegiate Dic-tionaiy 593 (9th ed. 1989).
. Webster's Dictionary defines an aneurysm as “a permanent abnormal blood-filled dilation of a blood vessel resulting from disease of the vessel wall.” Webster’s Ninth New Collegiate Dictionary 85 (9th ed. 1989).
. Although this issue is not comprehended within the issue on which we granted certiorari, it is necessary to consider it for a full resolution of the case. The insurance policy provision at issue in the present case must be read as a whole in order to understand the meaning of its parts. In order to interpret whether an injury was caused by accident directly and independently of all other causes, one must first determine what constitutes an "accident” under the policy. Both parties recognized this necessity and argued the issue in their briefs and at oral argument.
. Some courts have eschewed the task of developing a comprehensive definition of the term "accident.” Rather, these courts have applied the "common understanding” of that term to a particular set of facts in order to determine whether the situation was covered by an accidental injury insurance policy. In
Bolts v. Hartford Acc. & Indent. Co.,
the Oregon Supreme Court stated that "[t]he problem arises from an erroneous impression that there is one all-encompassing definition of 'accident' or 'accidental' without regard to the particular factual circumstance in which the meaning of the terms is brought into question.”
. In determining whether a certain result is accidental, it is customary to look at the casualty from the point of view of the insured, to see whether, from the insured's point of view, it was unexpected, unusual, and unforeseen. 1A Apple-man § 360, at 452-53.
. Some courts that have abolished the distinction between accidental means and accidental results may make an exception in cases involving overexertion or regular daily activities.
See
Ingram and Ostfeld at 23;
see also Benante v. Allstate Ins. Co.,
. The court also suggested other sequences of events and causation that a jury could have found and that wotdd have led to particular *753 conclusions as to whether Mrs. Bobier’s death was accidental.
. CUNA agrees with the predominant cause standard as well. It argues, however, that by applying this standard the result reached by the court of appeals should be affirmed. See infra part IV.
. Another panel of the court of appeals indirectly adopted this same interpretation in
Bobier,
. Courts group limitation clauses in accidental injury liability insurance policies into two categories. The first is similar to the clause at issue in this case and is referred to as a "limited coverage clause.” The second is similar to the clause in
Reed
and generally states that the injury in question shall not be caused wholly or in part, directly or indirectly, by any bodily or mental disease, defect or infirmity. Courts have referred to this kind of clause as an "exclusionary clause" and consider it to be more restrictive of coverage than the limited coverage clause.
See Mahon v. American Cas. Co.,
. This is especially so because CUNA has used the less restrictive "limited coverage clause” in its policy rather than the more restrictive “exclusionary clause.” See supra p. 754 and note 10.
. Some courts have interpreted the phrase "directly and independently of all other causes” to preclude liability whenever the loss results in part from a preexisting condition.
Winchester v. Prudential Life Ins. Co. of America,
