TRACKING OF ANTHRAX LETTER YIELDS CLUES
13 Dec 2002
Source: New York Times, December 7, 2001.
Tracking of Anthrax Letter Yields Clues
By ANDREW C. REVKIN with PAUL ZIELBAUER
At 5:27 p.m. on Oct. 9, amid tons of bills, birthday greetings, credit offers and other mail, an anthrax-laced letter to Senator Patrick J. Leahy zipped into the humming, high-speed machinery at a postal sorting center in Hamilton Township, N.J.
Precisely 20 seconds and 283 items later, an envelope addressed to the Farkas household on Great Hill Road in Seymour, Conn., followed it through the machinery.
The letters took separate routes in the automated system as electronic eyes translated addresses into bar codes and diverted them through different sets of sorters, bins, sacks and ultimately trucks. But anthrax spores that had puffed out of the first envelope during the sorting process evidently settled on the second. And as that letter made its way to Seymour, it retained enough anthrax to mark a route and offer a possible explanation of how anthrax could have reached and killed Ottilie W. Lundgren, a 94-year-old widow who lived a mile from the Farkas home.
A similar possible link has since been drawn between mail sent that same day through the same Hamilton Township hub and the anthrax death of Kathy T. Nguyen, 61, a hospital worker from the Bronx. One letter processed in the sprawling building at almost the same moment as the Leahy letter was traced to Ms. Nguyen's neighborhood, although it has never been found.
Investigators looking into these deaths were able to follow such clues only because of the rapidly increasing automation of the postal system, which sorts 95 percent of flat mail items automatically.
As part of an effort to improve the tracking of first-class mail and thus improve service, bar codes are now applied to the back and front of mail as it is canceled, registering the time and place it enters the system. Each letter gains a digital identity at each sorting hub, one that can be precisely tracked backward or forward.
Until now, investigators say, no one had thought of the data stored on magnetic tapes in the thousands of sorting machines in the 362 regional sorting centers around the country as anything other than a way to check whether a letter had been lost. But the investigators now realize that the data could conceivably allow them to identify and publicize the tens of thousands of other letters that passed through the Hamilton Township center within an hour or so of each of four known anthrax mailings, to Senator Leahy, Senator Tom Daschle, NBC News and The New York Post.
Some officials are pressing for the release of that information, saying it could help people avoid exposure to anthrax. But others say such a step would unnecessarily alarm communities that received mail that only theoretically held anthrax traces and that had caused no disease.
Representative Christopher H. Smith, a Republican whose district includes the Hamilton Township center, says it would be worthwhile to follow the trail of at least a sample of potentially contaminated mailings to test for anthrax bacteria.
But postal officials and experts at the Centers for Disease Control and Prevention say that even this is not necessary. Testing of 284 post offices and mail sorting centers -- those thought to be at greatest risk among the country's tens of thousands of postal centers -- has found only 23 with anthrax, almost all of which had only traces of contamination. And all those sites have been decontaminated with bleach except for the two big sorting centers where the assault began, in Hamilton Township and Washington, hubs that were shut down in October.
The C.D.C. says the fact that surveillance has turned up no anthrax cases since that of Mrs. Lundgren strongly supports the idea that any further risk from the mails is minimal, barring a new round of letters deliberately laced with anthrax.
In an analysis published yesterday in the centers' Morbidity and Mortality Weekly Report, the agency estimated that 85 million letters passed through the Hamilton Township and Washington hubs when both had "widespread environmental contamination" with spores shed by the anthrax letters. Yet no new cases of skin or lung infections have emerged among the 10.5 million people in communities served by those hubs, the report said.
"Despite this very low risk," the report said, "persons remaining concerned about their risk may want to take additional steps such as not opening suspicious mail, keeping mail away from your face when you open it and not blowing or sniffing mail," as well as washing hands after handling letters and making sure to discard envelopes.
The steady increase in postal automation has contributed both to the investigation of the anthrax mailings and, by raising sorting speeds, to the passage of the spores from letter to machine to letter.
The machines, each canceling, coding and sorting some 30,000 pieces of mail an hour, whip letters so quickly, employees and investigators say, that the lines of dozens of them function in a white haze that is generated as the paper crumbles slightly while flying on its way.
When anthrax was added to that process, the result was akin to turning on a blender, Dr. Jeffrey P. Koplan, the director of the disease control centers, said recently.
The resulting secondary contamination has been linked to two skin anthrax infections in New Jersey and is a possible -- but only possible -- explanation for the deaths of Mrs. Lundgren and Ms. Nguyen.
The Postal Service is adding filtered vacuum hoses to its equipment to cleanse the air and has ended cleaning practices in which air hoses were used to clear dust from machines, said Patrick Donahue, its chief operating officer.
The first use of the digital records to trace a tainted letter, postal officials say, was on Oct. 16, the day after an aide to Senator Daschle opened the letter to him at his Capitol Hill office, disgorging a puff of powder. Mr. Donahue said that almost immediately, postal officials began asking for testing of their buildings and employees but "didn't get a lot of cooperation" at the time from health authorities, who were focused on Capitol Hill.
The postal officials, Mr. Donahue said, sought to determine on their own which machine the letter had passed through at the Brentwood Road sorting center, which serves all of Washington. Tim Haney, the manager there, tracked down the codes imprinted on the letter, and those codes identified the machine.
The postal officials also obtained descriptions of the contents of a previous anthrax mailing, to Tom Brokaw of NBC News, and thought they should search that machine for that kind of material.
"We were thinking sugar, salt, thinking something might be lying on the floor," Mr. Donahue said.
Of course, it quickly became apparent that the Daschle letter contained a far finer, and therefore more dangerous, powder. And then came the parade of illnesses in postal workers who had been nowhere near sorting machines, and the death of two of those workers by Oct. 22.
Once the nature of the anthrax preparation became evident, "it was very chilling," Mr. Donahue said.
In looking for the attacker, the F.B.I. and the Postal Inspection Service, using digital printouts, hoped to discern which postal route the letter had come from. They tried to look for any patterns in letters that had spun through the Hamilton Township center about the same time. But no patterns emerged.
The next opportunity to use the data came only after Mrs. Lundgren took ill, before Thanksgiving.
At first, investigators pored over data from the sorting center in Wallingford, Conn., the one closest to her home. But they could find no record of suspicious letters to her, said Dr. David L. Swerdlow, a C.D.C. epidemiologist involved in the work.
The call went out to New Jersey to check printouts from Hamilton Township, on the off chance that something that had passed through there on Oct. 9 had ended up coming her way.
The letter to Seymour quickly came to light. When a team first went to the Farkas home there, no such letter surfaced. But it was found shortly thereafter, and wiped with a swab. One spore took hold in a petri dish and blossomed into an anthrax colony.
The discovery of one spore on a letter to a home a mile from Mrs. Lundgren's, however, does not answer the question of how she could possibly have inhaled a lethal dose of anthrax spores. No contamination has yet been found in her home or at any of the places she visited regularly. So state and federal investigators in Connecticut are still scouring post offices and the Wallingford sorting center for clues.
One hypothesis being tested now, Dr. Swerdlow said, is that a letter contaminated in a sorting machine in New Jersey, Washington or New York City went on to contaminate a postal bin in Wallingford that, in turn, cross-contaminated a letter sent to Mrs. Lundgren's home, in the town of Oxford.
To test that theory, investigators have spent recent days at the Wallingford building, where several sorting machines were found to be contaminated with traces of anthrax last weekend.
Paul Mead, a C.D.C. epidemiologist at the Wallingford center, said the investigative team wanted to complete "a trace forward and a trace backward" of any contaminated mail found there. But he added that tracing postmarked and bar- coded letters was easier than tracing prepaid bulk and commercial mail, which carries no postmark or other "unique identifiers."
"We have to look at other ways to track that mail forward," Mr. Mead said. "We're working with the Postal Service to come up with innovative ways to do that."
Indeed, as tests continued on mail, clothing and other belongings from Mrs. Lundgren's home, federal and state epidemiologists continued their crash course in postal routing and sorting.
The goal is to calculate the chances "that mail she received could have been in the vicinity of mail that could have been contaminated," said Mike Groutt, a C.D.C. spokesman. "They're in the early stages of that," he added.
Dr. Swerdlow said he and his colleagues were also considering a variety of hypotheses, ranging from the highly improbable (that Mrs. Lundgren was the target of a mailing) to the convoluted (the notion of double or triple cross-contamination).
"We're trying to leave no stone unturned," he said.