PROVO — Publishers agonized over lost sales when TV talk show icon Oprah Winfrey stopped endorsing a book each month in 2002.

Now a new Brigham Young University study could open old wounds.

Even casual observers knew regular sel- ections for Oprah's Book Club zoomed up best-seller lists — but they also remained there longer than other titles, according to research done by BYU economist Richard Butler and two of his students.

The combined impact — the "Oprah Effect" — was worth hundreds of millions of dollars.

"Oprah's recommendations had a bigger impact on the sales of books than anything we have previously seen in literature, or seen since," Butler said.

Before Oprah's selection, only 11 of the 45 non-children's books she endorsed on her TV show between 1996 and 2002 had made USA Today's top 150 best-seller list. None had risen higher than 25th.

The first 11 books she picked had been unranked before gaining her stamp of approval, but each rocketed immediately into the top four.

But Oprah's real influence, Butler and his students found, was in her endorsements' staying power. They compared the duration of each of her picks on the list with how long other books spent in the top 150.

Among those most affected were "Poisonwood Bible" by Barbara Kingsolver, which was on the list for 137 weeks, and "Where the Heart Is" by Billie Letts that lasted 98 weeks on the list.

"It's not just an overall average," Butler said. "We're comparing the short-term, median-term and long-term books with Oprah's short-term, median-term and long-term books, and those are really distinct. Her average and below-average picks stayed on the list much longer than the average and below-average best-seller books."

And when Oprah-endorsed titles showed up on paperback best-seller lists a year after being on the hardcover lists, they remained in the top 25 for an average of 6.5 weeks.

The researchers conducted statistical tests to verify that their results were strictly correlated with the Oprah effect and not the product of chance or other variables. The paper was accepted by the academic journal Publishing Research Quarterly.

Butler believes the Oprah effect was fueled by thousands of book clubs around the country, according to an unpublished study his team conducted. One finding of the research, which has been submitted for review to the Journal of Cultural Economics, is that Oprah's selections fared worse during the summer months and in December, when book clubs generally don't meet.