Suicides accounted for more than half of all violent deaths in Virginia in 2003, with the majority committed by white men with marital troubles and a history of depression, according to a state study to be published today.

The suicide rate was highest among men older than 65, the study found, and one in four of the nearly 800 suicide victims in the state was a veteran of the armed forces.

"We had an idea how [it] would fall out because we have insider information, so to speak," said Virginia Powell, the project's manager for the Department of Health's Office of the Chief Medical Examiner. "But we were taken aback by those numbers."

The 43-page report, to be released by the state health department, said that of the 1,332 violent deaths in the state that year, 60 percent were suicides and 33 percent were homicides. The remaining deaths were divided among unintentional firearm injuries, terrorism and legal interventions such as police-involved killings. Some causes of death never were determined.

A general impression has persisted, Powell said, that young people take their lives far more frequently than senior citizens. But she said the study, funded by the Centers for Disease Control and Prevention with a $1.2 million grant, suggests otherwise upon closely examining the rate among populations, and not just at the tallied figures.

"There are lots of things happening at that stage of life that could trigger a suicide," she said. "There could be a loss of independence, a decline in health, financial problems."

Although jurisdictions in western Virginia tallied the highest suicide rates for their populations, Fairfax County topped the list in sheer numbers, with 45 people taking their own lives in 2003, the first year studied by state officials. Last year's figures are being studied.

Virginia was one of six states initially selected to take part in a five-year pilot program for the National Violent Death Reporting System, a national effort funded through the CDC, said Karen Head, the study's author. Since its inception, she said, the program has expanded to include 16 other states, including Maryland, which has not completed its study. Maryland's report will include statistics from 2003 and 2004, officials said.

The study analyzed the 799 suicides in 2003 in Virginia to determine what risk factors might be present, Head said. It found that three out of four of the violent-death victims were males; blacks were disproportionately more at risk than whites or Hispanics, although whites were more at risk for suicide while blacks were more at risk for homicide; violent deaths were highest in central Virginia and lowest in Northern Virginia; suicide rates were highest among men older than 65 and especially high among men ages 85 and older.

Head said the study highlighted data collected from numerous sources -- autopsies, law enforcement statistics, vital records and forensic examinations -- so that root causes of violence can be better explored and perhaps understood.

"We hope this report brings with it a new awareness and brings to the forefront information that hadn't been available to the general public," Head said.

For example, the report could be helpful, Head said, for health care providers who regularly consult with senior citizens and at mental health centers. The greatest number of suicide victims had undergone treatment for mental illness or depression, while a significant number had substance abuse problems and had previously tried to kill themselves, according to the report.

Of the state's 799 suicides, the report found, 184 victims disclosed their intent to take their lives.

Across the country each year, 50,000 suicides and homicides are committed, with more than 1,200 recorded annually in Virginia. According to the report, firearms were used in 56 percent of the suicides committed in 2003.

In 2003, 435 homicides were committed in the state, according to the report, with more than 65 percent in central Virginia among black men 20 to 24 years old. The majority of the victims were never married, and most were killed on private property such as a house or an apartment. Other common locations included streets, sidewalks and alleys.

"All violent death is premature death," Powell said, "and so of course we're interested in this study in order to prevent that and to do something about the hopelessness and despair caused by the violence."