Americans angered by the government's collection of millions of telephone records may be surprised to learn that other public and private agencies already gather information that can be used to track patterns of illness or behavior.

The concept, known as statistical data mining, uses formulas and computers to assess patterns of activity.

Grocery stores, for example, can use information from electronic swipe cards to better control their inventories, assess buying patterns and target promotions.

In exchange for a discount, "you give up a little bit of your privacy," said Dr. Artur Dubrawski, director of the Auton Lab at Carnegie Mellon University, which has been a leader in researching new approaches to data mining.

USA Today reported Thursday that the National Security Agency obtained information from telephone companies about calls made by millions of Americans. The disclosure renewed concerns about civil liberties and questions about the legal underpinnings for the government's actions.

U.S. Sen. Wayne Allard, R-Colo., said the NSA was using the data to analyze calling patterns in order to detect and track suspected terrorist activity, according to information provided to him by the White House.

"Telephone customers' names, addresses and other personal information have not been handed over to NSA as part of this program," he said.

President Bush on Thursday sought to assure Americans their civil liberties were "fiercely protected."

Barry Steinhardt, director of the technology and liberty project for the American Civil Liberties Union, noted yesterday that the ACLU has a suit pending in federal court in Michigan related to NSA intercepts of communications without obtaining a warrant.

"But some of us were surprised to see the sheer breadth of this," he said of the disclosure that the government had been collecting information on the phone calls of millions of Americans.

The Auton Lab, part of Carnegie Mellon's School of Computer Science, has not been involved in the NSA effort, Dr. Dubrawski said yesterday.

He said the government could have used data mining to identify potential contacts made by suspected terrorists, or even the probability that others might be terrorists.

The Auton Lab has collaborated with a number of other government agencies and Fortune 500 companies, he said, noting there are many potential benefits from data mining -- for example, understanding trends in the stock market or the benefits and side effects of certain prescription drugs.

"We are flooded with data," he said. "What we really need is information on which we can act and make decisions and discoveries."

In one project, Carnegie Mellon researchers used data mining to help analyze consumer complaints to more quickly identify outbreaks of foodborne illness, he said.

The technology also has been used to analyze sales of over-the-counter medications, with the idea that some outbreaks of illnesses might be spotted before patients using those treatments become sicker and go to the doctor, Dr. Dubrawski said.

Trends also have been analyzed in emergency room treatment, he said. While they could assist in early detection of a bioterrorist threat, he said public health departments have become increasingly interested in the technique to help predict flu outbreaks and other problems.

In those efforts, researchers have been careful to obtain information in ways that protect individual privacy, Dr. Dubrawski said. He noted, for example, that pharmacy purchasing patterns have been analyzed by ZIP code rather than by individual stores.

Mr. Steinhardt of the ACLU acknowledged that data mining can be put to good use but said more regulation is needed.

"The truth is, there are more permissions to use the data than there are regulations," he said.

Dan Donovan, a Giant Eagle spokesman, said the company only requires a name and address from customers seeking to obtain a Giant Eagle Advantage Card. Other information, such as Social Security, driver's license and phone numbers, and date of birth, is optional.

Data from the program is used to identify and understand target audiences, personalize direct mail and electronic communications to customers and provide them with offers and measure the effectiveness of targeted programs, he said.

He noted that the company does not sell customers' personal or purchase information unless they authorize the company to do so and shares the information only to offer customers special savings, to allow them to complete transactions or to comply with subpoenas or other legal requirements.