Digging for Data

By Elizabeth Bennett  |  Posted 2005-12-06 Print this article Print

With the bodies of 270 Hurricane Katrina victims lying unidentified in a morgue, a database intended to help medical examiners make IDs is coming up short.

Shawn Romero, a database administrator and analyst from the Louisiana Department of Health and Hospitals, knows all too well the limitations of the two databases, especially their insufficient analytics and reporting capabilities.

He devotes several hours a day to extracting records from both VIP databases and uploads them into a Microsoft SQL Server database where he cleanses the records of duplicates, anomalies and misspellings. SAS analytical software corrects data entry errors, such as ZIP codes that staffers accidentally keyed into address fields.

Romero can write queries to find instances of dates that appear more than once in the two VIPs. An analysis of the duplicates could lead to a positive identification based on a birth or wedding date.

But misspellings and multiple spellings of the same name are common in VIP, particularly when more than one person reports that an individual is missing, but the consequences of those typos are significant.

Cataldie uses his own name as an example: "If they were looking for 'Louis Cataldie,' and he were dead in the morgue but they spelled it with a 'y' instead of an 'ie', they wouldn't match me and it wouldn't automatically pull up a list of 'Do you mean this?' or 'Here are some possible matches.'" The VIP database's search is not designed to recognize close but not exact matches, according to Kevin Mallon, a spokesman for FileMaker, a Santa Clara, Calif.-based software developer.

Tom Brondolo, former deputy commissioner at New York City's chief medical examiner's office, says he recognized VIP's limitations shortly after the bodies of victims from the World Trade Center attack arrived at his agency's morgue. There would be 20,000 fragmented remains recovered at the disaster site; VIP had no easy way to electronically link numerous records for body fragments brought to the morgue, Brondolo says.

There were other disadvantages to the database. "VIP's [missing-persons] form was very important," Brondolo says. "But we needed one place to pull all the information together." VIP couldn't be used to track the movement of remains through the morgue, as well as family visits, phone calls, dental and DNA records, death certificates, visitor passes, and other aspects of the identification process.

FEMA counters that VIP has been used successfully for every event in the U.S. with mass fatalities since 1994. "VIP was capable of being used at 9/11 and is continually reviewed," says Debbie Wing, a FEMA spokeswoman. She adds that FEMA was not aware of the Louisiana medical examiner's criticism of the VIP system.

For Cataldie and the hundreds of others working around the clock to identify the dead and return them to their families, the frustration is rising. When asked what his goal is for identifying the remaining bodies at the morgue, the state medical examiner replies, "Yesterday. A month ago. Two months ago.

I want to get that goal done now."

Story Guide:
Identifying the Dead: New Orleans Mortuary Response Team

  • Database of the Dead
  • Digging for Data
  • DMort Base Case

    Next page: DMort Base Case.

  • <1234>
    Senior Writer
    Elizabeth has been writing and reporting at Baselinesince its inaugural issue. Most recently, Liz helped Fortune 500 companies with their online strategies as a customer experience analyst at Creative Good. Prior to that, she worked in the organization practice at McKinsey & Co. She holds a B.A. from Vassar College.

    Submit a Comment

    Loading Comments...
    eWeek eWeek

    Have the latest technology news and resources emailed to you everyday.