Why are the nearly 3,000 victims of 9/11 missing from an official federal registry of death?
According to the Death Master File -- the official record of 90 million deceased Americans who were issued Social Security cards since 1937 -- there were 6,298 deaths recorded on that awful day in 2001 when terrorists struck the World Trade Center, the Pentagon and a rural area in Pennsylvania.
But since an average of 6,200 Americans die every day, there should have been more than 9,000 deaths recorded for Sept. 11, 2001.
Conspicuous by their absence in the federal file are many prominent victims of the attacks, including New York City Fire Chief Peter Ganci Jr., Fire Department Chaplain Mychal Judge and businessmen Daniel Lewin, founder of Akamai Technologies, and Thomas Burnett Jr. chief operating officer of Thoratec Corp.
"The mystery about 9/11 baffles me," said Beth Givens, executive director of the San Diego-based Privacy Rights Clearinghouse which has received complaints about the accuracy of the death file. "The only things that come to mind are some of the conspiracy theories that we hear out there -- and I don't want to go there."
Conspiracy theorists, indeed, have noticed and are questioning whether the government has told the truth about what happened that day. A video posted on YouTube entitled "Where are the 9/11 Victims?" shows that only 405 people are listed as dying in the state of New York that day.
The Social Security Administration, which oversees the Death Master File, does not have a clear explanation.
"There are several possible reasons," said Social Security spokesman Mark Hinkle. "For example, by law, we cannot make public the death reports we get from certain states. Another possibility is that the death was not reported to us because the person was not receiving benefits or there were no survivors' benefits to be paid on the deceased's Social Security record."
The Centers for Disease Control and Prevention also compiles a mortality registry of information obtained from death certificates, which is one of the sources Social Security uses for the Death Master File. The CDC's registry correctly shows a 3,000-death bump above average for that day in 2001.
Rep. Carolyn Maloney, D-N.Y., has begun inquiries into the reporting failure after she was told of the issue by Scripps Howard News Service. She is asking Social Security Administrator Michael Astrue, and the public health commissioners for New York State and New York City to, jointly, explain how the error occurred.
"Would you explain why individuals killed on 9/11 would be missing from the DMF?" Maloney asked in a joint letter to all three commissioners.
Consumer experts warn that inaccuracies in the Death Master File are a concern for families who need protection from thieves who could profit by assuming the identity of deceased loved ones. Researchers also use the file in a wide variety of medical and scientific studies that could be skewed by inaccurate counts of deaths.
There are many other mysteries in the Death Master File.
Disproportionately more people are listed as having died on either the first or 15th of each month than should be. About 3.6 million people died on the 15th of their month of death, 1.7 million on the first, and an average of less than 1.5 million for all other days. This means that more than 2 million records likely contain the wrong date.
"Social Security receives death reports from other federal government agencies. In the past, these reports included the verified month and year of death, but did not include the day of death," Hinkle said. "In order to process the death information in our systems, we needed to fill in a day of death."
Hinkle said the 15th of the month sometimes was used "as a default day of death" until the precise day of death could be obtained.
Due to clerical errors, the Death Master File also contains the names of thousands of Americans who are still alive, Hinkle said.
Scripps Howard News Service was able to identify 31,931 still living Americans by analyzing back copies of the death file. Forty-one percent of these were listed as having died on the 15th.
These reporting errors are not evenly distributed throughout the nation. A disproportionate number were found in Illinois, Louisiana, Massachusetts, Rhode Island and the District of Columbia. The rate of error was extremely low in rural, lightly populated Western states such as Alaska, Utah, Wyoming, North Dakota and Nevada.
Hinkle did not give an explanation for these discrepancies.
"We make it clear that our death records are not perfect and may be incomplete or, rarely, include information about individuals who are alive," he said. "Because we do not receive reports for all deaths and cannot release all of the reports we do receive, the absence of a particular person (in the Death Master File) does not prove the person is alive. Our error rate is about 0.5 percent."
Hinkle said the Social Security Administration each week reports "erroneous death data" to the United States Computer Emergency Readiness Team, part of the Department of Homeland Security's National Cyber Security Division. He said the administration also "hired a contractor to review all cases of inadvertent exposure of people's information. The contractor has found no patterns of organized misuse and no indications of identity theft."