Proponents of abstinence-only sex education declared victory earlier this month after a study from the University of Pennsylvania found that a program for Philadelphia sixth and seventh graders deterred young teens from having sex.

Abstinence Clearinghouse, an advocacy group, proclaimed on its Web site that the study proved "comprehensive sex ed a big flop."

Not so fast, says the study's lead author, Penn sociologist John Jemmott.

Jemmott's study, published in this month's issue of Archives of Pediatric and Adolescent Medicine, examined an abstinence program that would not have qualified for federal funding during the Bush administration. Those programs required an emphasis on abstaining until marriage, whereas Jemmott's involved no preaching and no denigrating the effectiveness of contraception.

The study followed 662 African American sixth and seventh graders for two years. Some were placed in the abstinence program, others in a comprehensive course that included discussion of abstinence and condom use. Another group participated in a program that dealt only with safer sex, and a final group of control subjects did a workshop on nutrition.

Most of the programs took place in four-hour sessions over two Saturday mornings. The abstinence class included a number of interactive exercises, Jemmott said. For example, the students were asked to think about their hopes five and 10 years in the future. Then they had to consider the consequences of a pregnancy on their plans.

"It's designed to be fun," Jemmott said. "There are games where they can win points, and role-playing and other upbeat activities. There's no preaching, and it's not moralistic."

Some activities got the children thinking about peer pressure and the way it can sway important decisions. When the students were asked to consider the pros and cons of sex, the boys said they expected having sex would make others think more highly of them, while the girls said they thought it would please their boyfriends.

Although the average age of the participants was about 12, more than 23 percent reported they had already had sex at the start of the study.

Jemmott followed the students for two years after the programs.

Of 95 students who said they were virgins at the start of the abstinence training, 33 percent reported that they had sex within the next two years.

By comparison, 41 percent of the virgins in the comprehensive course went on to have sex in the two-year window. For the control group, the figure was 47 percent.

In a sample this size, the difference between the comprehensive class and the abstinence class - 33 percent vs. 41 percent - was not statistically significant, said Jemmott, so it is accurate to say they performed comparably.

But the statistics also led to the finding that only the abstinence class - at 33 percent - got results significantly better than the controls, at 47 percent.

And that has allowed different interpretations from various factions in America's debate over sex education. "On the one hand are people who are concerned that the comprehensive intervention was not significant, and on the other are those who are happy that it wasn't significant," Jemmott said. "Both reactions are wrong."

That's because the study did not prove that a comprehensive approach failed, he said. It had some positive outcomes, he said, such as reducing the number of students who reported multiple partners.

And dozens of other comprehensive programs have passed these kinds of tests. It's misleading, he said, to say that comprehensive sex education "flopped." The group publicizing that assertion, Abstinence Clearinghouse, would not answer questions for this story.

At the same time, Jemmott said that his program differed from other abstinence programs and that its success therefore did not reflect on them. If a program was unproven before, it's still unproven.

What was it about this abstinence program that allowed it to pass the test of statistical significance?

One factor, Jemmott said, was the population targeted. These Philadelphia youngsters were known to be at high risk for having early sex. A program tested recently in Utah could find no significant effects compared with a control group because none of the children in either group said they had had sex.

Proponents of abstinence and sex education alike say they are pleased that someone has found a promising technique to discourage inner-city children from putting themselves at risk of sexually transmitted disease and pregnancy before their 14th birthdays.

"This is top-notch evaluation research conducted by well-respected researchers," said Bill Albert, chief program officer of the National Campaign to Prevent Teen Pregnancy.

According to Albert, studies show that only 8 percent of adults wait until marriage.

Douglas Kirby, of the California-based nonprofit Education Training Research Associates, attributed the Penn team's success to the way it tailors programs to certain populations.

"The Jemmotts have developed a number of programs over the years which have been found to change behavior," said Kirby, referring to Jemmott and his wife, Loretta, a coauthor of the study.

Jemmott said he had limited experience with other abstinence programs, though he was once asked to help evaluate one and said the students heard repeated messages that condoms are not 100 percent effective in preventing pregnancy or diseases such as herpes. He worried that repetition might lead them to believe using condoms was a waste of time.

"The children were significantly lower in knowledge about condoms after the intervention," he said. He considers this a negative side effect that should be taken into account.

Albert said the war between the abstinence and sex-education camps was not a zero-sum game - different approaches could work in different age groups and populations. "This program did what it set out to do. It got kids to delay sex and did it in an honest and respectful way that didn't denigrate contraception," he said.

"We need to buck up and move beyond politics to invest in things that work. While the adults are arguing about all this, teenagers are getting pregnant."

Penn’s Sex-Education Study

Researchers questioned African American participants in grades 6 and 7. Two years later students were asked if they had been involved in any of the following activities during the last three months.

SOURCE: Archives of Pediatrics and Adolescent Medicine