The plaintiffs decedent, Jeremy Prather, was employed by a company that had obtained a Group Insurance Policy from Sun Life which provided accidental death and dismemberment coverage for the company’s employees, in the amount of $92,000 for Prather. The policy limited coverage to “bodily injuries ... that result directly from an accident and independently of all other causes.” The clause we’ve italicized is the focus of this appeal from the district court, which granted summary judgment for Sun Life, which had invoked the clause to deny the payment of death and dismemberment coverage to Prather’s survivor on the'ground that his death had not been the exclusive result of an accident — it had also been the result of “complications from surgical treatment.” Prather’s widow brought this suit “to recover benefits due to [her]” under the plan. 29 U.S.C. § 1132(a)(1).
On July 16, 2013, Prather, age 31, had torn his left Achilles tendon playing basketball. Three days later he met with an orthopedic surgeon to discuss treatment, and of the options offered he chose surgery. He was scheduled to be operated on three days later, July 22. On July 21 he called the surgeon’s office complaining of a swelling in the lower part of his left leg, and that an area of the left calf was both sensitive and warm to the touch. The surgeon told him to elevate the leg. The surgery next day to repair his torn Achilles tendon was uneventful and he was discharged from the hospital the same day. He returned to work and was reported as doing well in a follow-up visit to his surgeon on August 2. But four days later he collapsed at work, went into cardiopulmonary arrest, and died the same day as a result of a deep vein thrombosis (blood clot) in the injured leg that had broken loose and traveled through the bloodstream to a lung, thus becoming a blood clot in the lung — that is, a pulmonary embolism — which caused cardiac arrest and sudden death.
Sun Life’s position is that the pulmonary embolism and ensuing death were consequences not of — at least not entirely of— the accident to Prather’s Achilles tendon, but of the surgery. Of course that means it was also a consequence of the accident, for without the accident there would have been no surgery. But it may have been a consequence of the surgery as well. Indeed it is even possible that had it not been for the surgery, Prather would not have died.
A physician’s assistant employed by Sun Life opined in a “Death Review Claim,” prepared by her as part of Sun Life’s review of Prather’s claim, that deep vein thrombosis and pulmonary embolism are risks of surgery, but that even with conservative treatment, such as immobilization of the affected limb, the insured had an enhanced risk of a blood clot. That was Sun Life’s only medical evidence, however, and it was inconclusive. It was evidence not that Prather’s death was a result not just of the accident but also of independent events — namely the surgery but maybe also or instead the immobilization of his leg before surgery — but that it might have been a partial result of such events — partial because the accident had to have played a role; no accident, no surgery or immobilization, hence no deep vein thrombosis or pulmonary embolism.
Although the insurance contract purports to place the entire burden of proof on the insurance beneficiary — it says the death must have occurred “independently of all other causes,” which Sun Life interprets to require the beneficiary to disprove any possible alternative cause of death— that can’t be right because it would give the insurer carte blanche to reject coverage in a case in which an accident is a conceded cause of death (there is no sug-*735gestión that Prather, a young man, would be dead had he not torn his Achilles tendon), merely because there is some speculative possibility that something else may also have played a role. That would make many accident insurance contracts vacuous, illusory, because often there is an interval between the accident and a resulting injury, and a possibility that something in that interval caused or aggravated the insured’s injury. Since the accident alone— the rupturing of the tendon — may well have caused the blood clot that killed Prather, the insurance company had to present some evidence that the surgery had been a cause of Prather’s death — and it presented none. All it “proved” through the physician’s assistant was that the surgery might have been a cause of Prather’s death; no effort was made to quantify the added risk created by the surgery.
The forensic pathologist who conducted a post-mortem examination of Prather did not attribute his death to the surgery, and the relevant medical literature indicates that there is a significant incidence of deep vein thrombosis (the final trigger of the pulmonary embolism) following the rupture of an Achilles tendon even if the tendon is not operated on.
Deep venous [vein] thrombosis (DVT) is a significant source of morbidity and mortality and is associated with many orthopedic procedures. Previous studies have reported highly variable DVT rates in patients with Achilles tendon rupture undergoing operative and nonoperative treatment. We performed a retrospective chart review for all patients who underwent Achilles tendon repair at our institution from January 2006 to February 2012. Patient data were collected from the electronic medical record system. A total of 116 patients were eligible for the present study. Of these patients, 27 (23.47%) with a surgically treated Achilles tendon rupture developed a symptomatic DVT either while tvaiting for, or after, surgical intervention, with approximately one third of these diagnosed before surgical intervention_
We have shown a high incidence of DVT after Achilles tendon rupture. We recommend a high level of suspicion for the signs and symptoms of DVT during the follow-up period.
Makhdom AM (Asim M.) et al, “Incidence of Symptomatic Deep Venous Thrombosis After Achilles Tendon Rupture,” 52 Journal of Foot & Ankle Surgery 584 (2013) (emphases added).
Sun Life argues unpersuasively that its insurance contract with Prather’s employer gave Sun Life “discretion to decide what evidence was sufficient to demonstrate a disability.” But that would amount to the insurer’s having carte blanche to decide whether or not to honor its contract. The company also argues, again un-persuasively, that “the evidence in this matter makes clear that Mr. Prather’s surgical treatment contributed to his death[,] ... indeed, caused the forming of a blood clot in Mr. Prather’s deep veins.” No, the evidence does not make that clear. All the evidence shows is that his death followed both the surgery and the accident that preceded the surgery. Post hoc is not propter hoc. And we know from the medical literature that an accident alone can create a fatal blood clot and so far as appears could have done so in this case.
Indeed, Prather’s complaint the day before the surgery of a swelling in the lower part of his left leg and that an area of the left calf was sensitive and warm to the touch suggests that the deep vein thrombosis might have formed by then, because swelling, pain, and warmth in the affected limb were all symptoms of DVT. Because Sun Life failed to make any plausible showing that the surgery on Prather’s an*736kle, rather than the accident that necessitated the surgery, caused his death, the judgment in favor of the defendant is reversed and the district court is instructed to enter judgment in favor of the plaintiff. See Senkier v. Hartford Life & Accident Ins. Co., 948 F.2d 1050, 1052 (7th Cir. 1991).