If a driver slams his car into a tree and dies with a blood-alcohol content three times the legal limit, is that an accident?

The federal appeals court in Boston faced that question in a Rhode Island case involving a Connecticut man who died in a 2002 crash in Johnston. And the judges couldn’t agree on the answer.

Two members of a three-judge panel said the man did not die as a result of an “accident” for purposes of his accidental death and dismemberment insurance policies. But the dissenting judge said the man’s death “can be termed nothing other than an unfortunate accident.”

In their dueling opinions, the judges drew comparisons to Russian roulette, smoking cigarettes and “death from autoerotic practice.”

Stephen Stamp, 37, of Litchfield, Conn., died in August 2002 after his 1996 Camaro veered off Route 295 and crashed into a wooded area in Johnston.

Earlier in the day, Stamp, a maintenance manager for Exxon Mobil Chemical Company, had attended a meeting for employees at a resort in Westbrook, Conn. The meeting included presentations in the morning, a boat cruise in the afternoon and dinner in the evening, according to the court decision.

At 5:20 p.m., Stamp spoke with his wife, Karen Stamp, confirming that after dinner he planned to drive to his parents’ home near Providence. She and their young daughter planned to join him the next day to celebrate his brother’s 40th birthday. Stamp had several drinks during the cruise and dinner. But coworkers and his wife said he did not seem impaired, according to the decision.

At 9:20 p.m., Stamp called a friend in Rhode Island, hoping to visit him, but the friend declined, saying he needed to go to bed early. The friend said Stamp was upbeat and did not sound intoxicated. Cell-phone records show Stamp was in the Providence area at 9:58 p.m., but apparently he didn’t go directly to his parents’ house. Defendants Metropolitan Life Insurance Co. and ExxonMobil suggest Stamp stopped and drank some more. Stamp’s hand had a stamp on it that read “copy,” and MetLife suggests he went to a bar and had his hand stamped at the door.

Whatever the case, Stamp’s car struck a tree in Johnston just after midnight, and he was pronounced dead at the scene. His blood alcohol content was .265 — more than three times Rhode Island’s legal limit of .08, according to the decision. The autopsy report said the cause of death was “multiple injuries due to blunt force trauma” and noted “acute ethanol intoxication” as a significant finding. The report classified the manner of death as “accidental.”

As the beneficiary of her husband’s life-insurance policies, Karen Stamp submitted a claim for benefits. MetLife paid her claim for basic life-insurance benefits, but denied her claims under his accidental death and dismemberment coverage. She appealed to ExxonMobil, which had the power decide all coverage matters. But ExxonMobil denied the appeal, concluding his death was not the result of an accident.

“The risks flowing from driving while intoxicated are completely within the control of the participant,” the denial letter said. “While it is true that certain behavior that increases risk [such as skiing and horseback riding] would not result in loss of coverage, [driving while intoxicated] can be distinguished because it unreasonably increases the risk associated with a normally safe activity by interfering with an individual’s ability to perceive and respond to risk.”

Karen Stamp filed a lawsuit in U.S. District Court in Providence, but Senior U.S. District Judge Ronald R. Lagueux dismissed it.

IN AN APPEAL to the 1st U.S. Circuit Court of Appeals, her lawyers said the chance of dying from driving while intoxicated is 1.19 percent, so “it cannot be reasonably concluded by either common sense or legal analysis that death resulting from driving under the influence of alcohol is ‘highly likely.’ ”

But in the majority opinion, Circuit Judge Kermit V. Lipez said such statistics are meaningless in this context because they don’t take into account Stamp’s level of intoxication. “The risk of being involved in a fatal crash rises as blood alcohol levels rise,” he wrote.

Lipez noted that in a key 1990 case — Wickman v. Northwestern National Insurance Co. — the 1st Circuit ruled that a man did not die accidentally when he climbed over a guardrail on the side of a railroad bridge and fell. The court said “he reasonably should have expected serious injury.” And the court rejected the idea that his death had to be considered accidental unless the man “actually expected to die, essentially that he specifically intended to commit suicide.”

Lipez said the 1st Circuit’s reasoning in the Wickman case has been widely accepted by other federal courts. And he noted that rather than focusing on probabilities, the court had analyzed “what a reasonable person would perceive to be the likely outcome of the intentional conduct.”

“From a statistical standpoint, the likelihood of dying from a single round of Russian roulette is 16 percent — one in six,” Lipez wrote. “In fact, those who play Russian roulette have a decent chance, statistically speaking, of not being injured. Nonetheless, such a death would not be publicly regarded as an accident because the mortal risk associated with playing Russian roulette is patently obvious to any reasonable person.”

Even if there was evidence that extremely drunk people have a “better than even chance of arriving home safely,” Lipez said he would have ruled the same way because, “It was not arbitrary for the plan administrator here to conclude that Mr. Stamp’s decision to drive while grossly intoxicated qualifies as overwhelmingly and disproportionately risky conduct.”

Lipez noted MetLife’s medical department had advised the plan administrator that a blood alcohol content of .265 would cause “lethargy, stupor, combativeness, incoherency and vomiting” in a “sporadic drinker.” And an internal Mobil memo said “blackouts are likely” with blood alcohol content of .20.

IN DISSENT, Circuit Judge Juan R. Torruella said Stamp’s decision to drive “was unlike the game of Russian roulette” because “the point of that game is to see if you are lucky enough to live.”

Torruella said the key question is not whether Stamp was highly intoxicated, but whether he intended to kill himself by becoming intoxicated and getting behind the wheel. He said there is no evidence that Stamp was distraught, and he noted that a few hours before Stamp died he had upbeat conversations with his wife, coworkers and a friend.

“At most, Mr. Stamp miscalculated and misjudged his ability to drive that fateful evening,” Torruella wrote. And Stamp bought insurance to protect himself and his family against such “unforeseen accidents,” he said.

Torruella said the court should focus on an insured person’s expectations. And he cited a 1995 decision, by the 5th U.S. Circuit Court of Appeals, involving a Texas man who died from “autoerotic asphyxiation,” which the court defined as “the practice of limiting the flow of oxygen to the brain” in order to “heighten sexual pleasure.” The 5th Circuit said, “Death was not a normal expected result of the behavior.”

Likewise, Torruella said, “Much more often than not, driving while under the influence has a non-fatal outcome.”

He said smoking cigarettes is responsible for 87 percent of lung-cancer deaths. “Yet it would be highly unusual [for] the administrator, or this court, to conclude that anyone who smokes is engaging in a suicidal act and is thus excludable from coverage under this policy,” he wrote.

“Drunken driving is, to be sure, a foolish and reckless act,” Torruella wrote. “However, when an insured [person] dies as a consequence of his driving while intoxicated, his death should still be considered accidental because a reasonable person would not view death as a natural and probable consequence of drunk driving.”

efitzpat@projo.com