CHICAGO -- An experimental abstinence-only program without a moralistic tone can delay teens from having sex, a provocative study found.

Billed as the first rigorous research to show long-term success with an abstinence-only approach, the study differed from traditional programs that have lost federal and state support in recent years. The classes didn't preach saving sex until marriage or disparage condom use.

Instead, it involved assignments to help sixth- and seventh graders see the drawbacks to sexual activity at their age, including having them list the pros and cons themselves. Their cons far outnumbered the pros.

The students, mostly 12-year-olds, were assigned to one of four options: eight hour-long abstinence-only classes, safe-sex classes, classes incorporating both approaches; or classes in general healthy behavior, which served as a control group. Results for each class were compared with the control group.

Two years later, about one-third of abstinence-only students said they'd had sex since the classes ended, versus nearly half - about 49 percent - of the control group. Sexual activity rates in the other two groups didn't differ from the control group.

The study was released Monday in the February edition of Archives of Pediatrics & Adolescent Medicine.

Critics of abstinence-only programs have long argued that most evidence shows they don't work. The new study challenges that, but even the authors say the results don't mean more comprehensive sex education should be ignored.

Advocacy groups favoring traditional abstinence-only programs praised the study and said it shows that the Obama administration's move away from funding these programs is misguided.

The administration has focused on programs proven to prevent teen pregnancy. But the study is unlikely to revive enthusiasm for a narrow abstinence approach, and an Archives editorial suggests that it shouldn't.

"No public policy should be based on the results of one study, nor should policy makers selectively use scientific literature to formulate a policy that meets preconceived ideologies," said the editorial by Dr. Frederick Rivara, the journal's editor, and Dr. Alain Joffe, an associate editor.

The abstinence-only program was based on social psychology theories about what motivates behavior. It encouraged abstinence as a way to prevent pregnancy and sexually transmitted diseases, although the researchers didn't collect data on those outcomes.

Psychologist John Jemmott III, the lead author, called the findings surprising given negative results in previous abstinence-only research. Jemmott said the single focus may have been better at encouraging abstinence than the other approaches in his study.

"The message was not mixed with any other messages," said Jemmott, a professor at the University of Pennsylvania who has long studied ways to reduce risky behavior among inner-city kids. He created the four programs for the study with his researcher-wife, Loretta Jemmott.

Valerie Huber, executive director of the National Abstinence Education Program, praised the results and said she hopes they revive government interest in abstinence-only sex education.

When asked if the new study might shape the Obama administration's policy, White House spokesman Reid Cherlin said: "Our approach is to use science and evidence to fund what works, while leaving room for innovation and new thinking. We feel the policy we introduced at the beginning of the administration accomplishes that."

The study was funded by the National Institute of Mental Health and involved 662 black children in Philadelphia.

Monica Rodriguez of the Sexuality Information and Education Council of the United States, an advocacy group favoring comprehensive sex education, said the study doesn't mean other abstinence-only programs would work.

"It's unfair to compare this abstinence-only intervention to the typical abstinence-only-until-marriage program that young people in this country have been put through," she said. These typically portray sex and condom use in a more negative light, she said.

Rodriguez said the program studied might be one approach to try with younger children, but that it probably would be less successful with older, more sexually experienced teens.

Almost one-fourth of the teens studied said they'd already had sex at least once, similar to other studies of urban, mostly black middle school-aged kids.

The classes were taught at schools on weekends. Jemmott said the program might work better if it were taught during regular school hours by the students' regular teachers - an approach he hopes to examine in additional research.