Opinion of the Court

Fortis Benefits Insurance Company appeals the district court’s order granting summary judgment against it and in favor of Amy S. Keller (“Keller”) in this action for benefits allegedly due under a group accidental death and dismemberment insurance policy (“AD & D”) issued by For Ms. Keller cross-appeals the court’s dismissal of her state law claims for bad faith under 42 Pa.C.S. § 8371. For the reasons set forth below, we affirm both orders. 1

I.

Inasmuch as we are writing only for the parties who are familiar with this case, we need not set forth the factual or procedural background except insofar as may be helpful. Rather, we will proceed directly to our brief discussion. 2

Fortis’ denial letter stated in part: because Mr. Keller was 6'3" tall, there had to have been a precipitating event. There is insufficient evidence supporting loss of consciousness due to blunt head injury; however, there is substantial evidence that the unconsciousness, which led to his drowning, was caused by a seizure. Having suffered a severe head injury in 1991, Keller was at risk for developing seizure disorder.

Since Fortis both administers and funds the plan at issue here, we must first decide the appropriate amount of deference, if any, to afford Fortis’ decision to deny coverage. Fortis initially approved coverage and then declined coverage. It relied upon the medical expertise of two doctors. One of them could not determine the cause of death and the other ruled it an “accident,” but concluded that the death was “most likely” precipitated by a grand mal seizure. However, other evidence suggested the underlying cause of Mr. *876 Keller’s drowning could not be conclusively determined. In its denial letter, Fortis credited evidence that was consistent with denying payment without explaining why it was ignoring evidence that supported coverage or undermined its conclusion that Mr. Keller’s death was not covered. Accordingly, we believe Fortis’ denial decision is entitled to less deference than would otherwise be the case. 3

A.

Fortis first claims that the death was not an accident, and therefore the policy does not even apply. In the alternative, Fortis argues that if the death was an accident, and the policy does apply, then the “disease exclusion” precludes coverage. However, Dr. Smith, the forensic pathologist to whom Fortis referred the file for an expert opinion, determined that Mr. Keller’s death was “accidental ... due to drowning.” SA 13. The issue has not been whether the drowning death was “accidental,” but whether the precipitating cause was excluded under the policy. Fortis points to its denial letter and relies on the statement that: “there is insufficient evidence supporting loss of consciousness due to blunt head injury.” However, that does not explain how Fortis could rely only upon evidence consistent with denying Keller’s claim while ignoring evidence that supported coverage. Accordingly, we agree that Fortis abused its discretion in determining that Mr. Keller was not killed by an accident.

B.

We must also decide whether Fortis abused its discretion in concluding that Mr. Keller’s death fell within the exclusionary language of its policy. The policy provides that no benefit will be paid “if the loss results directly from ... any physical disease.” The term “physical disease” is not defined in the policy. Fortis chose to apply that term to a grand mal seizure; a one time event that it could then define as a “disease,” thus triggering the exclusion. However, Mr. Keller had no history of seizures following the medical procedure in 1991, and there is no evidence of a latent disease.

The district court explained its rejection of Fortis’ attempt to rely on the exclusion as follows:

The best evidence Fortis can muster is that Keller may have experienced multiple mild seizures and not even known it. Even accepting Dr. Smith’s analysis ... there is absolutely no evidence ... to conclude that Keller suffered from a seizure physical disease dating from 1991. This was an unusual, sudden and unexpected event. The mere potentiality for a seizure sometime in the future is not a disease.

Memorandum Order, Sept. 23, 2003, p. 8. We agree.

C.

For the reasons set forth above, we will affirm.