In response to a query about why he wanted to climb the world’s tallest mountain, British explorer George Mallory responded, “Because it is there.” As Mallory and others have learned, the lure and mystique of mountain climbing are not without risk of injury, including death. The controversy in this case is whether a climbing death on Mt. Kilimanjaro falls within the meaning of the terms “accidental injury” and “disease” as used in a life insurance policy. Although this inquiry is seemingly straightforward, it turns out that, in the context of insurance law, unraveling the meaning of these simple words is no easy task.

Luanne Chale brought this breach of contract suit against Allstate Life Insurance Company (“Allstate”) after it declined to pay her a $100,000 accidental death benefit under her husband’s life insurance policy. Because her husband’s death from high altitude edema falls within the definition of “accidental injury” and does not fit with the “disease” exclusion of the policy, Mrs. Chale was entitled to summary judgment. Accordingly, we reverse the district court’s grant of summary judgment in favor of Allstate.

Background

The dispute between Mrs. Chale and Allstate arises from the tragic mountain climbing death of her husband, Bernard Chale (“Chale”). Each year, thousands of adventurers like Chale ascend Mt. Kilimanjaro, hoping to reach the 19,430 ft. snow-capped summit of Africa’s tallest mountain.

Although there are many risks in climbing, ranging from avalanches to crevasses, high altitude itself presents a risk. At elevations above 8,000 ft., decreased atmospheric pressure reduces the concentration of oxygen in the air enough to make it noticeably difficult to breathe. Medicine for Mountaineering & Other Wilderness Activities 221-22 (James A. Wilkerson, M.D., ed., 5th ed. 2001). At these heights, experienced climbers allow themselves time to acclimatize because altitude sickness can set in when a person ascends faster than the body can adjust. Mountaineering: The Freedom of the Hills 396 (Don Graydon, ed., 5th ed. 1992). Rapid ascent upsets the body’s chemistry, and can result in headaches, nausea, fatigue, dizziness, loss of appetite, and vomiting. Medicine for Mountaineering at 230-32. As the syndrome progresses in severity, a climber may develop high altitude pulmonary edema (HAPE), characterized by fluid in the lungs, and high altitude cerebral edema (HACE), characterized by swelling of the brain. Id. at 232-33.

Victims of HAPE often experience extreme fatigue, breathlessness, coughing, rattling breaths, chest tightness, congestion, and may turn blue. Id. at 233-34. HACE causes confusion and lethargy. Id. at 232. HAPE and HACE are an exceptionally dangerous duo when they attack together. Id. at 233. Unless a decompression chamber or pressurized oxygen is available, descent to lower altitudes is the only way to treat these conditions. Norbert F. Voelkel, High-Altitude Pulmonary Edema, 346 The New England Journal of Medicine 1606 (May 23, 2002). Recovery following descent is usually rapid and complete. Medicine for Mountaineering at 233, 235. Left untreated, HAPE and HACE can cause death in a matter of hours. Death from HAPE and HACE is relatively rare and not a common or expected outcome of a high altitude adventure. See id. at 234 (describing the risk of developing HAPE after rapid ascent to 12,000 ft. as 0.5%).

Despite these risks, many climbers find the allure of the mountains iiresistible. Mt. Kilimanjaro, whose summit straddles the border of Tanzania and Kenya, is an especially popular destination because its ascent requires no technical skills. See, e.g., Harald Lange, Kilimanjaro: The White Roof of Africa 157-58 (1985). Even novice peak baggers can hire professional outfitters to guide them along the multi-day trek, as Chale did in December 2000.

Chale began his climb on December 23, 2000, at 4,600 ft. Over the next four days, he and his climbing party ascended to their final camp at 17,000 ft., where the air is about half as dense as it is at sea level. See Michael Ward, Mountain Medicine: A Clinical Study of Cold and High Altitude, 4 (1995). Just after midnight on December 27, the party set out for the summit, splitting up into a faster and slower group. At 11:30 AM, on its way down, the faster group passed Chale, who was headed for the top. Up to that point, Chale had experienced some fatigue and breathing problems, but no other symptoms of HAPE or HACE. A few hours later, Chale died, after having reached the summit. A post-mortem report revealed that Chale had been afflicted with “severe edema and congestion of the lungs and brain”- — -in other words, HAPE and HACE.

Before his death, Chale held a life insurance policy issued by Allstate. Under the policy, his wife was to receive $100,000 upon his death and an additional $100,000 if his death was “solely from accidental injury” and not due to “disease [or] infirmity of the body or mind.” Neither “accidental” nor “disease [or] infirmity” were defined in the insurance policy.

After Chale’s death, his wife notified Allstate afid demanded payment. Allstate paid her $100,000 in straight death benefits under the policy but not the additional $100,000 accidental death benefit. Mrs. Chale then filed suit in Oregon state court, and Místate successfully removed to federal court. The parties filed cross-hnotions for summary judgment. The district court granted Allstate’s motion, denied Mrs. Chale’s motion as moot, and entered a judgment of dismissal with prejudice.

DlSCÜSSION

I. Applicable Law and Standard of Review

Oregon law governs this inquiry; if no Oregon Supreme Court decision is directly on point, our task is to predict how the court would decide the case. S.D. Myers, Inc. v. City and County of San Francisco, 253 F.3d 461, 473 (9th Cir.2001). We invoke the traditional standard of review on summary judgment: the district court’s dismissal on summary judgment is reviewed de novo and the evidence must be considered in' the light most favorable to the non-moving party. See Padfield v. AIG Life Ins. Co., 290 F.3d 1121, 1124 (9th Cir.2002).

II. The “Accidental Injury” Provision

We first consider the parties’ dispute over the interpretation of the term “accidental injury” in the fife insurance policy. As with any contract, our goal is to “ascertain the intention of the parties.” Hoffman Constr. Co. v. Fred S. James & Co., 313 Or. 464 , 836 P.2d 703, 706 (1992) (quoting Totten v. New York Life Ins. Co., 298 Or. 765 , 696 P.2d 1082, 1086 (1985)).

Initially, we look to the policy as a whole. See id. As the drafter of the contract, Allstate could have explicitly defined “accidental” or “accident” any way it wished. Botts v. Hartford Accident & Indem. Co., 284 Or. 95 , 585 P.2d 657, 660 (1978). Because it did not, and because the rest of the contract sheds no additional light on the parties’ intent, Allstate “must accept the common understanding of the term by the ordinary member of the purchasing public.” Id.; see also Pope v. Benefit Trust Life Ins. Co., 261 Or. 397 , 494 P.2d 420, 421 (1972); Finley v. Prudential Life & Cas. Ins. Co., 236 Or. 235 , 388 P.2d 21, 26 (1963).

The next step, Oregon law instructs, is to seek clarification from the plain meaning of the term. See Hoffman Constr. Co., 836 P.2d at 706 . This inquiry is equally unavailing, as we do not presume to discover a workable definition of “accident” where the Oregon Supreme Court found none. See St. Paul Fire & Marine Ins. Co., Inc. v. McCormick & Baxter Creosoting Co., 324 Or. 184 , 923 P.2d 1200, 1212 (1996) (finding the dictionary definition of “accident” broad enough to encompass the definitions offered by both parties).

The lack of a commonly accepted, singular meaning of “accident” has given Oregon courts a number of occasions to grapple with the task before us today. Carving out a universally applicable definition is no mean feat. Cf. Botts, 585 P.2d at 660 (“There are probably not many words which have caused courts as much trouble as ‘accident’ and ‘accidental.’ ”). Notably, the Oregon Supreme Court has acknowledged the futility of such an undertaking. The court’s approach in Botts treats the legal interpretation of “accident” as malleable to the facts of a given case. Thus, “[i]n situations in which ‘accident’ or ‘accidental’ are not defined in the policy, it is for the court to decide the definition which is properly applicable to the particular factual situation.” Id. at 661.

Although Botts makes clear that ours is a moving target, we are not entirely without guidance. It is well established that the proper focus of the inquiry is the injurious result, not the conduct leading to that result. See Fox v. Country Mut. Ins. Co., 327 Or. 500 , 964 P.2d 997, 1006 (1998) (where the insured was a voluntary passenger in a friend’s car when the friend intentionally drove it off a cliff in order to collect insurance money, the court focused on the insured’s state of mind with respect to sustaining the injury (death), not his state of mind with respect to placing himself in a position where he risked the injury (getting in the car)); see also North Clackamas Sch. Dist. v. Oregon Sch. Bds. Assn. Prop. & Cas. Trust, 164 Or.App. 339 , 991 P.2d 1089, 1092 (1999) (reaffirming that “the meaning and determination of ‘accident’ focuses not on conduct, but on result”). Hence, we must focus on whether Chale’s contracting HAPE and HACE and his resulting death were accidental. Whether his act of climbing Mt. Kilimanjaro was intentional — which it certainly was — plays no part in this inquiry.

At the broadest level, an “accidental” event must contain at least some kernel of chance or abnormality. See St. Paul Fire & Marine Ins. Co., 923 P.2d at 1213 (defining “accident” as an “incident or occurrence that happened by chance” (quoting Finley, 388 P.2d at 26 )). Under some circumstances, the insured’s state of mind is relevant to this inquiry. If Chale’s purpose was to bring about his death, or if he “engaged in an act so certain to cause ... [his death] that the court will say that ... [he] intended the harm,” his widow cannot recover the accidental death benefit. See Fox, 964 P.2d at 1005 .

Intentional injury is, of course, antithetical to accidental injury. In various contexts, Oregon cases have expressed this concept in different ways, requiring accidents to be “unexpected and unintended,” Pope, 494 P.2d at 423 (internal quotation marks omitted), to happen “without design and contrary to intention and expectation,” Finley, 388 P.2d at 26 , or be “unforeseen, unexpected, unintended or the like,” Safeco Ins. Co. v. House, 80 Or.App. 89 , 721 P.2d 862, 866 (1986). Most of these definitions simply exclude from coverage as accidents events that are “intentional” under the ordinary tort meaning of intentional as purposeful or done with knowledge of substantially certain outcome. However, the additional concept of foreseeability appears to have created a stumbling block for litigants.

Foreseeability is but one factor that can help courts “effectuate the reasonable expectations of the insured.” Botts, 585 P.2d at 660. It is also a term loaded with vagaries. As the Botts court pointed out, some unforeseeable injuries, such as when an elderly person dies in her sleep of old age, are not commonly considered accidental. Id. Yet it is equally true that some perfectly foreseeable injuries are universally considered accidental.

Consider, for example, a mountaineer who falls into a crevasse while traveling on a glacier. Climbers know that crevasses are a danger associated with glaciers; yet when apparently solid snow crumbles underfoot and causes the mountaineer to fall to an icy death, no one would dispute, and even Allstate concedes, that this perfectly foreseeable event would be classified as an accident. Automobile travel provides a far more mundane example. Every automobile passenger is aware of the risk of a car accident, yet no one would argue that a run-of-the mill collision is not an “accident.” As these examples illustrate, if the litmus test for foreseeability turns on mfere awareness of the possibility of risk, a legal rule that excludes all foreseeable events from “accident” coverage would yield absurd results.

Thus, the district court erred by equating “accidental” to “reasonably unforeseeable” where, in its view, an injury “may be reasonably foreseeable even if it is not certain or substantially certain that it will occur in every instance.” Chale v. Allstate Life Ins. Co., No. CV 01-1622-BR, slip op. at 13 (D.Or. May 31, 2002). In premising its ruling on Mrs. Chale’s failure to “produce[ ] any evidence to establish a reasonable person endeavoring to climb Mt. Kilimanjaro would be unaware of the risk of HAPE and HACE, id. at 12 (emphasis added), the district court turned Oregon law on its head. Urging us to affirm this faulty reasoning, Allstate argues that, in an analogous situation, someone who voluntarily walks across a field during an electrical storm would not have suffered an “accidental injury” if he were struck by lightning because the risk was foreseeable. This illustration is perhaps the most convincing evidence of the flaw in Allstate’s argument. Any interpretation of the insurance policy that assumes that a reasonable insured person would not consider such a scenario to be accidental strains credulity.

At the very least, such an interpretation flouts the firmly established rule that ambiguous terms are to be construed against the insurance company. Hoffman Constr. Co., 836 P.2d at 706 ; I-L Logging Co. v. Mfrs. & Wholesalers Indem. Exch., 202 Or. 277 , 275 P.2d 226, 232 (1954); see also North Clackamas, 991 P.2d at 1091-92 . Application of this rule is appropriate because the use of “accidental” in Chale’s policy was reasonably susceptible to more than one meaning. See Botts, 585 P.2d at 661 (contemplating many different definitions of “accident”).

Mrs. Chale argues that, as in Fox , the “accidental injury” term in the policy should be construed to include Chale’s death because he had not “expected or intended to” contract HAPE and HACE. We agree. Chale contracted an unusual and extremely serious medical condition while engaging in the activity of mountain climbing. Even if he was or should have been aware of the risk of HAPE and HACE, his chances of falling victim to these conditions were slim. Because Chale was struck by altitude-induced edema much in the way someone might be struck by lightning, his death falls within the definition of “accidental injury” contemplated by any reasonable purchaser of Allstate’s life insurance policy.

In reaching this conclusion, we are persuaded by the reasoning in Paulissen v. United States Life Ins. Co., 205 F.Supp.2d 1120 (C.D.Cal.2002), where, under California law, the insured’s death by HAPE was “accidental” because “[djeath from HAPE cannot be said to be a common or expected result of a trek at high altitudes.” Id. at 1128 . Allstate urges us to ignore Paulis-sen due to claimed irreconcilable differences between Oregon and California law. Specifically, Allstate complains that California recognizes a difference between policy language that provides coverage for “accidental death” and language that provides coverage for death resulting from “accidental means,” whereas Oregon long ago rejected this distinction. Compare Paulissen, 205 F.Supp.2d at 1127-28 (applying California law’s distinction “between policies that cover ‘accidental death’ and those that cover death by ‘accidental means’ ”), with Botts, 585 P.2d at 659-60 (“lay[ing] the distinction to rest” in Oregon), and Pope, 494 P.2d at 421 -22 & n. 1 (criticizing “[tjhe distinction between injury by ‘accidental means’ and ‘accidental results from intended means’ ” as “logo-machy” incomprehensible to the average insurance policy purchaser).

Allstate’s protest rings hollow. When the Oregon Supreme Court abolished the remaining “vitality of the means-result distinction” in Botts , its purpose was to broaden coverage for insurance policy holders bound by “accidental means” language — not, as Allstate would have it, to constrict “accidental injury” coverage. See Botts, 585 P.2d at 659-660 (expressing its “doubts as to whether the ordinary purchaser would expect the concept of ‘accident’ to have a different meaning depending upon whether the policy purports to require accidental means or accidental results” but explaining that its elimination of the narrow “accidental means” construction was not strictly necessary because the policy at issue did not contain an “accidental means” 'requirement). Nonetheless, Allstate argues that although the means-result distinction no longer exists, we must still “determine whether there was an accident.” It insists that we look no further than Chale’s intentional choice to reach high altitudes, which was undisputedly not accidental. In other words, Allstate’s test for “whether there was an accident” is to identify whether the “means” of the injury- — Chale’s choice to climb Mt. Kilimanjaro- — was accidental.

Accepting Allstate’s position would not only require Oregon law to re-embrace the abolished means-result distinction, but would also require us to read “accidental means” in place of “accidental injury” into the policy. Because the proper focus is on the injury that befell Chale, Paulissen’s rationale is particularly apposite. There, Paulissen, the insured, died while trekking in Nepal at altitudes similar to the ones experienced by Chale. Like Paulissen, Chale’s “death was caused by accident because it was an unusual or unanticipated result flowing from a commonplace cause.” Paulissen, 205 F.Supp.2d at 1128 .

III. The “Disease” Exclusion

We next consider whether the policy’s “disease” exclusion is applicable. Because Chale’s death was “accidental,” Mrs. Chale is entitled to the $100,000 accidental death benefit unless Allstate can surmount its burden to show that the death was caused by “disease” or “infirmity.” See Stanford v. American Guar. Life Ins. Co., 280 Or. 525 , 571 P.2d 909, 911 (1977) (“The insurer has the burden of proof that the loss is excluded.”).

We note, as an initial matter, that “disease” and “infirmity” are properly treated as synonymous. Hutchison v. Aetna Life Ins. Co., 182 Or. 639 , 189 P.2d 586, 590 (1948). Although the definitions of these terms are exceptionally broad in common usage, see Paulissen, 205 F.Supp.2d at 1129-30 (remarking that they “would include the slightest and most temporary ailment”), they must be “narrowly construed in the context of insurance policies.” Id. at 1130 ; see also Todd v. Occidental Life Ins. Co., 208 Or. 634 , 303 P.2d 492, 495 (1956) (“In construing a contract of insurance, the courts will give as favorable a construction for the benefit of the insured as is cognizable in the words used to prevent forfeiture.”). Under Oregon law, we must limit our construction of the term “disease” in Chale’s policy “only to some ailment or disorder of an established or settled character to which the insured is subject.” Hutchison, 189 P.2d at 590 ; see also Todd, 303 P.2d at 495 .

Allstate does not dispute that Chale did not have HAPE or HACE before ascending to high altitudes, or that rapid descent would have alleviated his symptoms. These afflictions thus can hardly be described as any more “established” or “settled” than, for example, hypothermia or suffocation. See Paulissen, 205 F.Supp.2d at 1130 (holding deceased’s HAPE “was not a disorder of a somewhat established or settled character” but “a mere temporary disorder ... arising from sudden and unexpected derangement of the system” (internal quotations and alterations omitted)). It is true that someone who is freezing to death or drowning certainly suffers from a physiological abnormality. But in neither case could the affliction be considered so “established” or “settled” as to amount to “disease.” For both disorders, the onset is sudden and the outcome either death or rapid rescue from the brink of death — just like HAPE and HACE.

Nor does Allstate suggest that Chale had any other “established” or “settled” disorder that contributed to his death. Instead, it rests its argument upon two faulty premises. First, Allstate views as dispositive its medical expert’s uncontested use of the word “disease” to describe HAPE and HACE. Unfortunately for Allstate, resolution of this issue goes beyond medical semantics. Whether HAPE and HACE are “diseases” or “infirmities” as contemplated by the insurance policy is a question of law. See id., 205 F.Supp.2d at 1127 n. 8. Although testimony from medical experts can help inform the legal decision maker about the nature of these afflictions, it does not dictate the proper legal interpretation of this policy term. See id. This is the province of courts rather than doctors.

Allstate’s second argument reveals the weakness of its position. It contends that because “there is no evidence of any ... accident,” the “diseases” of HAPE and HACE directly caused Chale’s death. In other words, it ties its argument on the “disease” issue to an outcome in its favor on the “accidental injury” issue. But because Chale’s death was “accidental,” Allstate’s effort to bootstrap is unavailing. The district court should have granted summary judgment to Mrs. Chale on her claim for recovery of accidental death benefits under the insurance policy. Considering only the evidence before the court at the time Mrs. Chale moved for summary judgment, no genuine issue of material fact remains to be litigated. Accordingly, we reverse the district court’s grant of summary judgment to Allstate and its denial of summary judgment to Mrs. Chale.

IV. The Motion to Strike

In her reply declaration, Mrs. Chale included the following statements: that Chale was a “fit individual” and “healthy and fit”; that Chale never took mountaineering classes and was not aware he could die as a result of altitude sickness; and that other people survive climbs and exposure to altitudes “greater than that at which Mr. Chale experienced edema.”

The district court granted Allstate’s motion to strike this evidence based on Oregon Court Rule LR 56.1(e), which provides that “the court has no independent duty to ... consider any part of the court record not otherwise referenced in the separate concise statements of the parties.” Because the stricken material was not included in Mrs. Chale’s Concise Statements of the Facts, the district court did not abuse its discretion. See Golden Gate Hotel Ass’n. v. City and County of San Francisco, 18 F.3d 1482, 1485 (9th Cir.1994). Exclusion of these statements does not, however, change the outcome of the case.

Conclusion

Chale’s premature death from HAPE and HACE falls squarely within the bounds of Allstate’s Accidental Death Benefit Rider. Thus, the judgment of the district court is reversed and remanded for entry of judgment for Mrs. Chale in accordance with this opinion.

REVERSED.

. John Bartlett, Familiar Quotations 876 (14th ed. 1968). . Although Fox interprets a policy provision controlled by a state statute, at minimum its definition sets a floor for denominating an event as accidental.