New Study Lowers Toll of Quake Injuries

A UCLA study has determined that there were only 171 serious injuries and deaths attributable to the 1994 Northridge earthquake in Los Angeles County, a fraction of earlier estimates and a finding that jolted the researchers themselves.

The report, billed as the most in-depth study to date of earthquake casualties--and the first to extensively examine medical records--indicates that only 138 people were hospitalized for earthquake injuries and 33 people were killed. That would mean nearly half of the county’s fatalities--16--were related to the collapse of a single Northridge apartment complex.

Although the fatality findings were not radically different from earlier estimates, the researchers acknowledged that the serious injury findings were. And they contrasted dramatically with public perceptions of the magnitude 6.7 temblor, the researchers said.

Some of the discrepancy is due to UCLA’s conservative approach, which did not count heart attack injuries and deaths. Also, the UCLA study looked only at Los Angeles County, not the less-affected Ventura and Orange counties.

Even so, the numbers are dwarfed by earlier estimates of serious injuries--at least one of which was seven times higher.

Michael E. Durkin, a Woodland Hills public health researcher responsible for a previous, widely cited estimate of more than 1,000 hospitalizations, conceded Wednesday that the UCLA figures are probably more accurate.

“We thought we would find a lot more [serious injuries] based on what we were hearing from the [hospital] staffs,” said Corinne Peek-Asa, one of the lead researchers. But she said the county turned out to be much better prepared for an earthquake than had been thought.

Peek-Asa said the early overestimates may have been attributable to the emotion surrounding the events, to the impressions of hospital staff members run ragged during the crisis, and to preliminary telephone surveys by emergency agencies and others that were not bolstered by review of medical records. Others blamed statistical modeling methods that used flawed assumptions to calculate the estimates.

The UCLA study involved surveying coroner’s records and 78 Los Angeles County hospitals to glean which had treated earthquake victims. At the 16 facilities that reported they did, the team reviewed the medical records of patients admitted within 16 days of the quake.

Earlier reports reflected similar death counts if heart attacks are subtracted, but injury estimates were far higher. Not long after the Jan. 17 quake, the state Office of Emergency Services reported 57 deaths and 9,000 injuries of unspecified degree, but those were in three counties--Los Angeles, Ventura and Orange.

Later, Durkin, who has studied other quakes, ratcheted up the estimates, reporting 72 deaths and 11,846 injuries requiring hospital treatment. According to that estimate, 1,044 people were admitted to hospitals and the others were treated and released.

Durkin, a UCLA alumnus, said Wednesday that he hasn’t seen the UCLA study, but he knows the researchers and considers them competent. While he can’t vouch for the precise numbers, he said, he thinks they are “probably on the right track.”

Durkin said his early estimates, reported in The Times in December 1995, were based on a Red Cross survey of hospitals. He said his subsequent efforts to substantiate the hospitalization numbers indicated that they were too high and probably hovered around 200. One hospital, which he declined to name, had reported 68 earthquake-related admissions, when only seven were related to the temblor, he said.

“Hospitals were over-reporting because they really didn’t have any criteria for reporting” what was earthquake-related, he said. “This [UCLA study] is closer to what I think is correct.”

An official with the Office of Emergency Services, which in 1995 professed to find no problem with Durkin’s findings, said the UCLA study was more focused than earlier surveys, which counted all manner of injuries in the three counties. But the office did not dispute the UCLA findings and, in fact, lauded the effort.

“We haven’t been able to study this [UCLA report] with the attention it deserves, but I would say overall, we are glad to see this kind of a report because these things do make a contribution to health and safety issues,” said Fred Messick, a spokesman for the office.

Others welcomed the findings as a refreshing correction to erroneous perceptions.

“There is a far more scientific basis for the UCLA study” than for earlier reports, said Billie Weiss, director of the injury and violence prevention program for the Los Angeles County Department of Health Services, one of the agencies that funded the study. “This study looked at the records of the people who were hospitalized.”

Weiss said the story might have been much different if the quake, which struck in the wee hours of Martin Luther King Day, had occurred at another time. But “the fact of the matter is, most people fared quite well.”

Peek-Asa agreed. She said that emergency services functioned “extremely well.”

Most of the deaths could not have been prevented by more prompt emergency services, she said. Eighty percent occurred within minutes, mostly as a result of falling buildings or building parts.

Even among those patients who were admitted to hospitals, the injuries were less severe than one would expect, Peek-Asa said. Researchers found few severe head injuries and no “crushed chest” injuries, which are common in earthquakes in developing nations.

She said a lot of the injuries could have been prevented, especially by bolting down furniture. “A lot of people were hit by flying TVs and cabinets,” she said.

Elderly residents were most vulnerable to harm, she said; many tried to respond more quickly than their frail bodies would allow. Elderly or not, many people were injured when they stepped on broken glass.

About a third of the people who were hospitalized delayed treatment for one or more days, increasing the severity of their injuries. Peek-Asa said those hospitalizations might have been avoided if care had been sought more promptly.

The study has been peer reviewed and will be published in June in the International Journal of Epidemiology.