When the stress of a disaster causes death

Friday, Oct. 6, 2017 9:03 PM

Two weeks after Hurricane Irma, the Miami Herald reported that Florida’s unofficial death toll had risen to 75. Depending on the practice philosophies of Florida medical examiners, that figure could go higher.

Obvious hurricane-related deaths include storm-surge drownings, injuries from falling trees or structures and electrocutions from downed power lines. Deaths from natural diseases exacerbated by hurricane conditions are less clear-cut, and medical examiners differ in where they draw the lines.

When Irma knocked out air conditioning at the Rehabilitation Center at Hollywood Hills on Sept. 10, the facility became “a sweatbox,” according to The Associated Press. An air-conditioned hospital was right across the street, but until residents started to die, nobody was transferred. Patients were cooled with “fans, ice and other methods” for three days.

The nursing home claims the building’s temperatures didn’t exceed 80 degrees. Actual recorded temperatures haven’t been released pending a criminal investigation, but if you believe fans kept a Florida building’s summertime temperature at 80 degrees, I have a bridge to sell you.

No medical examiner doubts that disaster-related stress or fear can kill vulnerable people. In order to make determinations about environmental impacts, we weigh factors, including the degree of pre-existing vulnerability, whether the severity of the stress significantly exceeded the normal wear and tear of life and the time interval between the stressor and death.

Eight nursing home residents died Sept. 13, and three more died during the next week. A 12th succumbed more than two weeks after the hurricane.

All died primarily of natural causes (young, healthy people can tolerate a few days without air conditioning). Given their ages and medical conditions, one or more of these people could have died in any given week. But a severe insult to a weakened cardiovascular system can tip that system into a state of progressive failure that’s irreversible and ultimately fatal.

When pre-existing disease and stress interact, there’s no way to determine their relative contributions. There’s certainly no way to say which person would have died regardless of the power outage. So given the circumstances, it’s reasonable to certify that all 12 deaths were hurricane-related.

When pre-existing vulnerability is less severe, timing becomes more important. One medical examiner reported the case of a middle-aged woman who chose to ride out the storm. Shortly after the worst was over, she was found lying dead near the phone. The only autopsy finding was heart disease related to high blood pressure.

Perhaps because the feds pay a funeral benefit when a death is blamed on a natural disaster, the woman’s family insisted that her death was hurricane-related.

The medical examiner polled colleagues, most of whom sided with the family. People with high blood pressure are vulnerable to sudden death. Death occurred shortly after the height of the storm. A death during an ordinary Florida thunderstorm wouldn’t count, but a hurricane significantly exceeds normal weather-related stress.

As the time interval between stress and death lengthens, certainty about causation decreases. Some experts believe that only sudden deaths that occur within an hour of a stressful event should be blamed on that event. But several published studies have found that the incidence of sudden, heart-related death remains significantly elevated for weeks and even months after disasters.

It’s likely that at least some such deaths wouldn’t have occurred absent the stress of the disaster, but we have no idea how many or which ones.

Some disaster-related death certifications are little more than a medical examiner’s best guess.

Dr. Carol J. Huser, a forensic pathologist, served as La Plata County coroner from 2003-12. She now lives in Florida and Maryland. Email her at