Dr. James L. Luke, the city's chief medical examiner, straves off heart disease by taking two asprins and jogging three miles each day.

His deputy, Dr. Brian Blackbourne, an expert on accidents, drives a big, 5,200-pound car, is a seat belt fanatic, and is so haunted by chocking deaths of children that he worries when his own play with toy balloons.

Dr. Douglas Dixon, who does research on gunshot wounds for the medical examiner's office, refuses to open his front door to anyone who has not made an appointment to see him. Marge Bejarano, one of the doctors' secretaries, says she never goes out alone in Washington any more. And Dr. Edward Zimney, another medical examiner, plans to leave his job this summer, explaining, "I just want to get away from dead bodies for a while and see how I feel about it."

More than almost anyone else in Washington, they and others who work at the D.C. medical examiner's office are face to face with the daily toll of city life: the price it exacts in violence, accidents, chronic illness, alcohol and drug abuse. Their data and the remarkable human panorama that unfolds daily tells them a story of everyday hazards, some avoidable, some not.

All of them believe there is no such thing as being safe, and that death can be caused unexpectedly by the most unlikely of objects -- an inhaled toy balloon, a string tied to a baby's pacifier. "We get so many freakish accidents," said autopsy technician Yvonne Williams. "It makes you look for it, almost expect it . . . You just start accepting that anything can happen, and when it does, the surprise is taken out of it. I guess it's being down here and seeing life as a reality -- death as a reality." g

The common denominator of the deaths that come to the medical examiner's office -- a third of all deaths that occur in the District -- is alcohol. "Alcohol is a solvent for that most precious of human attributes, common sense," said Luke, quoting his favorite medical school professor. "And I'm not talking about an alcoholic. I'm talking about one of us. Lots of average people, when drunk, are capable of beating their child, shooting their wife, falling down stairs, running over a neighbor's child or going to sleep with a ciagrette."

Blood tests indicate alcoholic intoxication in half of the District's homicide victims, half of the drivers killed in traffic accidents, a third of the suicide victim, two-thirds of the drowning victims and 80 percent of those who die of exposure.

In people who die from other causes, liver damage from heavy drinking is a frequent finding. "People who come through this office who are over 50 are awfully old," Blackbourne said. "They're old 50s because of the commonness of alcohol."

The medical examiner's office is a natural laboratory for studying conditions in the city and for observing human nature. Since Luke took over in 1971, he has kept carefully cross-referenced records on all deaths, which are now being computerized to make it easier to see significant trends and patterns.

The younger pathologists and technicians in the office say that almost every week they see something new. But Eugene Compton, a beer-bellied,, philosophic man who has picked up bodies for the office for the last 30 years, said nothing surprises him any more. "Most people don't know what is really going on," he said.

For the last decade, the medical examiner's office has kept careful track of the ebb and flow of unnatural deaths in Washington. The figures -- highly variable for some categories, depressingly constant for others -- reveal fascinating patterns of violent deaths and accidents in the city

In spite of widespread public concern about rising crime rates, the number of homicides in the District of Columbia has fallen since the mid-1970s. In 1974, it reached a peak of 299. But in the last five years, it has hovered close to 200 -- about a 33 percent decrease. Last year, there were 209 homicides. No one knows the reason for the decline.

In 1979 -- the most recent year for which a breakdown is available -- 127 of the 195 victims, or 65 percent, were black men between 15 and 59 years old. There were an average of 16 homicides per month, with no apparent seasonal variation. Guns were by far the most common weapon, used in 54 percent of cases.

Compton swears there are more murders over the Christmas holidays and when there is a full moon. After years of crisscrossing the city to pick up victims of violent crime, he believes that no part of the District is safe. "I stay out of the streets at night," he said, "except when I'm working."

Each year a number of victims are people who tried to get a gun away from an armed robber. Others opened their doors to strangers.

"It's an incomprehensible thing to me that an elderly person would let somebody come into his house just because he says he's the paperboy," Dixon said. "I don't think you can let you guard down. It bothers some people to live that way, but to me that's just part of living in the city, like it's hard to find a parking place."

Sucide figures for the District of Columbia havve remained constant over the lasdt decade. "It's constant over the last decade. "It's amazing that every year, we know 75 to 110 people are going to commit sucide in D.C.," Blackbourne said. In 1980, the total was 75. Old coroners' reports show that, despite population growth, there are fewer suicides now than in the late 1930s, when the figure during one year was 168.

According to a 1979 breakkdown, three-quarters of suicides in the District are men, and although the city is 70 percent black, 54 percent of sucide victims are white. Jumping, shooting and hanging are the most common methods, each accounting for about a quarter of the cases. Only 16 percent of the victims took drugs or poison. Alcohol was present in the blood of those who died by shooting or overdose much more often than in other types of suicide.

The popular belief that suicides are more common during the holiday season is false, according to Blackbourne. Suicide "gestures" -- unsuccessful attempts -- do increase, but deaths remain constant at about seven per month.

Deaths from narcotics overdose -- heroin, methadone or a combination -- have fluctuated dramatically over the last 10 years, and are closely tied to the purity and availability of herion on the streets.

Overdose deaths peaked at 82 in 1971, and fell to a low of seven in 1978. Then they began to rise again, and last year there were 62.

Intrigued by the variation, Zimney studied 287 narcotics deaths in the District, spanning an 8 1/2-year period. The victims ranged in age from 13 3 to 62, with an average age of 27. Ninety-three percent were black, and 83 percent were male.

He found that deaths were clustered during certain weeks. When he charted the deaths week by week and compared the pattern to the purity of herion being sold on the streets each week (determined from testing of samples confiscated by police), the two graphs matched almost perfectly. The purer the drug supply, the more overdoses occurred among addicts.

Zimney als found that a significant number of overdose victims died from injecting the drug after a period of abstinence, so that their tolerance to its effects apparently had worn off. One man got out of jail in Connecticut, stopped in New York to buy heroin, and was found dead of an overdose in the men's room of the train when it pulled into Union Station.

The average age of those who died of overdoses last year was 32, five years older than in previous years, an a third of last year's victims had high levels of alcohol in the blood, a rare finding in previous years. "We were wondering whether these people now dying had gone off heroin use back in '77 or '78 when it dried up, and turned to alcohol -- and now, when the stuff is so plentiful, they're shooting up again," but also continuing to drink, he said.

Sudden Infant Death Syndrome, or so-called "crib death," claimed the lives of 22 babies in the District last year. Although the annual totals have been nearly constant since 1975, they were higher in the early 1970s. A study conducted by Luke in the mid-1970s to try to explain the decrease detected a curious phenomenon: the numbers of babies who died alone in their cribs had fallen sharply, but an increasing percentage of sudden infant deaths involved babies who were sharing a bed with a parent or sibling when they died.

He found that the deaths were twice as common in homes where an infant shared a bed as in those where they slept alone. In 1976, 72 percent of sudden infant deaths occurred in bed-sharing situations.

The cause of sudden infant death syndrome is unknown, and Luke said there was no evidence that bedsharing is a factor in all such cases. But he raised the possibility that in some cases, infants might accidentally have been lain on and smothered by someone sleeping in the same bed. The suggestion was supported by the finding that more bed-sharing deaths occurred on weekends, when an adult might take an infant into bed in order to sleep late.

Over the last decade, the number of traffic deaths recorded by the medical examiner's office has ranged from 65 to 98. Deaths from accidents that occur inside the District have been decreasing for the last three years -- last year's total was a record low of 46 -- but the medical examiner's totals are higher because they also include those who die in Washington hospitals after being transported from the scenes of accidents on Maryland and Virginia highways.

Blackbourne, who has served on auto safety standard teams here and in Florida, believes part of the recent improvement in District figures is due to stricter enforcement of driving and pedestrain laws by the police. He said it is not explained by improved automobile design, since the last major strengthening of auto safety standards occurred around 1970.

He routinely inspects the cars involved in fatal crashes in the District to correlate design defects with injuries to occupants. And he strongly disagrees with a report by President Reagan's transition team that concluded no further safety improvements can be made economically.

"There's an awful lot that can be done still . . . and some of it need not be that severe an economic burden," he said. "The sides of cars need a lot of work. I think it would be a tragic mistake if we decide, as unsafe as cars are now, that that's as safe as we can make them."

Last year 145 people died in the District from home and occupational accidents -- a figure that is depressingly similar to annual totals for the last decade. According to a 1979 breakdown, falls are the most common fatal accident -- accounting for about 37 percent of cases -- followed by fires, responsible for about 30 percent of the deaths. Alcohol is found in the blood of about a third of fall and fire victims, and in more than two-thirds of those who die of drowning or exposure.

Blackbourne has paid particular attention to accidental deaths of District children. From 1972 to 1979, there were 288 such deaths of children under 16 -- "the size of one whole elementary school," he said. Half were cause by traffic accidents and house fires, but studying the remainder revealed some seldom-considered safety hazards.

Even though safety standards were set for infants' cribs in 1974, 26 of the deaths were caused by babies' heads or chests becoming wedged in defective cribs, often because slats were broken or a mattress was to small. Blackbourne said many old, unsafe cribs are still in use. "No one ever throws away a crib." he said.

Other deaths occurred because children choked on toy balloons or pieces of hot dog, drowened or scalded themselves during the few moments they were left alone in bathtubs, or accidentally hanged themselves by strings used to tie pacifiers around their necks.

"Accidents . . . have a definite cause," Blackbourne said. "In almost every one, we can think of some preventive measure."

Despite their daily exposure to the toll of accidents and violence, there are people on the mmedical examiner's staff who never wear seat belts, who drink too much or keep handguns at home. Luke said trying to change human behavior, either through persuasion or through public policy, is the hardest part of his job.

"People . . . don't think they're vulnerable, and they are," he said.