The new paperless claims system consists of five big technology pieces, all revolving around a new electronic folder.
In the past, everything about a claim either arrived at the agency on paper or subsequently got printed on paper. All documents were stored inside folders that grew to be five or six inches thick. Folders were mailed around to hearing examiners, medical consultants, appellate judges and others -- and often got lost. Manual photocopying was common, along with the time-consuming task of renumbering documents in each folder for appeals.
In the new system, every document gets stored as a digital file in a central repository. Documents can be transferred as electronic copies to the state agencies that process claims or over the Internet in encrypted form to outside medical experts. Staffers can also copy file collections onto disks and mail them, Gray said. The agency has set up a private Web site and gives special passwords and identification numbers to medical health professionals to submit digital files, such as X-ray images.
In addition, appeals hearings will be recorded and stored as digital audio files in each applicant's electronic folder.
Elaborate software programs are being created to track the digital documents at every stage of the claims review. Unlike regular databases, the information in this repository is stored in many formats and will be accessed with various software programs.
Among the agency's new programs are systems to electronically capture and store incoming information about each applicant's claim, to allow that data to be shared among the many people handling the claim, and to provide custom interfaces for state agencies and hearings offices.
All the programs have to work together smoothly for the system to function, which is no easy feat. Much of the new software is being provided by IBM, which has bet its future on an ability to integrate large repositories of data stored in different formats.
In January, the Social Security Administration started implementing key pieces of its paperless project in certain states. Many states, however, still have to replace their old computers with new machines. So far, 18 states are receiving electronic downloads of claims data, and eight are using the electronic folders, according to Gray. For more than a year, the agency also has been letting people file disability claims at least partially online.
Gray said the agency expects to slice at least 100 days off the time it takes to process claims, which can take two to three years if an appeal is filed.
Already, efforts to digitize data collection have cut five days off the average time required to get an initial ruling, which recently dropped to 97 days.
One challenge, Gray said, is giving staffers speedy access to the data, because moving files over computer networks can be slow. Equally daunting is winning over the people handling claims.
"For 70 years in Social Security we have been working with paper folders,'' Gray said. "For the last seven months, they have been working with electronic folders. It's a change they have to get used to."
The disability project is a bellwether for both electronic government and the software industry's push to offer greater interoperability among programs that previously operated as stand-alone silos -- a big barrier to automation.
"Governments are pushing hard in this direction -- not just federal agencies, but at the state and local level, too," said Connie Moore, research analyst and vice president at Forrester Research. "We are seeing a resurgence of these types of projects at all levels because the technology has continued to advance."
Andrew Warzecha, senior vice president for the Meta Group, agreed that the Social Security Administration has a better chance to go paperless now than in the 1990s, when data-sharing technology was still clumsy: "It is only recently that this category of software has been architected to do these kinds of things."
Of course, who hasn't heard that before? Any way you look at it, the project bears watching for the clues it can offer on whether big vendors like IBM will be able to deliver on their promise to take data-sharing to truly massive levels.