Interviews: Sex and the American Teen

Dr. Douglas Kirby

Dr. Douglas Kirby on Comprehensive Ed

POV: Could you describe your work with the National Campaign to Prevent Teen Pregnancy, and particularly the study you authored in 2001, Emerging Answers?

Dr. Douglas KirbyDr. Douglas Kirby: For a number of years, I was chairperson of the Effective Programs and Research Task Force for the National Campaign to Prevent Teen Pregnancy. The National Campaign has several task forces, and this one really focuses on research. One of many things that we did was to synthesize all the research that has been conducted in the field that meets certain scientific criteria. Emerging Answers was one of those products. Let me say, though, that although I was the author of it, it was critiqued and reviewed and read by all the members of the Effective Program and Research Task Force, and we intentionally created a task force that included a great diversity of members, in terms of gender, race, ethnicity, and also political persuasion, so there were some members that were conservative, and some that were more liberal. And we basically all agreed on the major conclusions.

POV: What were the conclusions?

Kirby: One is that many studies show that programs that emphasize abstinence as the safest approach, but also encourage those who are sexually active to use condoms and contraceptives do not increase sexual behavior; they do not do harm. They do not hasten the initiation of sex, they do not increase the frequency of sex, and they do not increase the number of sexual partners. In fact, to the contrary, some, but not all, of the programs, delay the initiation of sex or reduce frequency or reduce the number of sexual partners. In addition to that, some of these programs, but not all, also increase condom or contraceptive use. So basically, this is good news, and it's very strong news, very strong evidence, that those programs that emphasize abstinence as the safest approach, but also encourage condom and contraceptive use, those programs do not increase sexual behavior, can reduce sexual behavior, and can also increase condom and contraceptive use.

Education of Shelby Knox - Coronado High School in Lubbock, Texas Coronado High School in Lubbock, Texas

POV: Are those programs considered "abstinence-plus" programs?

Kirby: Yes. But people use different words to describe them. Sometimes they're called abstinence-plus, sometimes people call them comprehensive sex or HIV education programs. "Comprehensive" meaning that they're talking not only about abstinence but also about condoms and contraceptive use.

POV: Did the study find that the successful programs had some characteristics in common?

Kirby: Yes. Among the programs or the curricula that did have a positive impact upon behavior, there tend to be roughly ten to thirteen characteristics, depending on the way you count them. [Read the full list of characteristics in the executive summary of Emerging Answers (PDF).] The effective programs, for example, really focused upon behavior. They talked about sex, they talked about condom and contraceptive use. They also talked about pregnancy and STD and HIV. So they were not real broad programs, but they really talked about and focused on behavior. They gave very clear messages about behavior, and a very clear message was one of the most important criteria. As I mentioned, typically that message was some version of "you should always avoid unprotected sex; abstinence really is the only 100 percent safe approach; if you have sex, you should always use a condom or contraception to prevent STD and pregnancy." A version of that was truly emphasized. The successful programs were also very interactive. They did not consist of having a teacher stand up there and just give [students] didactic material. The effective programs involved youth in a whole variety of activities so that they were engaged and involved. They played games, they did role-playing. They had small-group discussions. They did lots of things in which they were actively involved.

POV: Did you find that the unsuccessful programs had characteristics in common?

Kirby: The ineffective ones, for the most part, just lacked one or more of the ten characteristics. They did not give a clear message, they weren't interactive, they did not really focus on behavior or they focused too much upon knowledge. The effective programs did provide basic information, but they did not primarily provide knowledge. They tried to change personal values, they tried to change perceptions of peer norms. They tried to increase young people's confidence that they could say no to sex or use condoms if they did have sex. In the effective programs there's a lot of skill building, role-playing to say no, role-playing to insist on using a condom.

Something else that should be said about the abstinence-plus programs is that a couple of them actually have an impact for as long as 31 months. That's close to three years, so that's really very encouraging. It is not the case that they can only have an impact in the short term. They can have an impact in the long term if they're well designed and if they have booster sessions after the initial sessions.

POV: Can you give us an example of a program that was successful?

Kirby: One very successful program, for example, is Safer Choices, and it had ten sessions in the 9th grade, ten sessions in the 10th grade, and then it had school-wide activities during all the years, so that young people would receive a clear message and understand it in the 9th grade, it would be reinforced in the 10th grade, and then in the 11th and 12th grade those messages would be reinforced by assemblies, by posters that were put up around campus, by things in the newspapers, et cetera. And that's a good model.

POV: So that's an abstinence-plus program. What did you research tell you about abstinence-only programs?

Kirby: The sad news is that there are very few reasonably good studies of abstinence-only programs, and because there are so few good studies, we really cannot reach any conclusion about them. The Effective Program and Research Task Force created a set of criteria for what should constitute a reasonably good study, and should be included in Emerging Answers. And at that time only three abstinence-only programs met those criteria. Those three programs did not have any positive impact on behavior. But we should not conclude from that that abstinence-only programs do not work. Rather, the appropriate conclusion is that there is very little research, there's very little evidence. And we simply don't know whether or not abstinence-only programs work. Personally, I think that some abstinence-only programs probably are effective at delaying the initiation of first sex, but so far we don't have good evidence telling us which ones.

POV: Why are there so few studies of abstinence-only programs?

Kirby: It's primarily because of the limitation on funding for research. A lot of the existing funding came from a certain title, [a certain category of] federal government funds, and to do good research, to really measure the impact of a program, takes about five to eight years, and it takes a lot of money. And that source of funding limited it to only two or three years, and provided only small amounts of money. Consequently it was just not possible for people to do good research on these programs. That has now changed. There's a very good evaluation being done currently by Mathematica Policy Research on abstinence-only programs, but we don't have the results of that yet. [See related links for update.]

POV: Politically, sex education and government funding are consistently controversial topics. Does that make it more difficult to do the kind of research you're talking about?

Kirby: Well, it does make it a little more difficult to do good research. It makes it more difficult to publish research, particularly negative findings. And when things are so politicized, it makes it hard for researchers to present results saying something didn't work.

POV: What areas or subjects do you feel deserve further research or particular attention?
Kirby: Although the Mathematica Policy Research study, which is a big study, will partially fill the need, there's still a need for other studies [of abstinence-only programs] to be done. Of the hundred studies in the world, a large majority of them, probably 90 of them, deal with abstinence-plus programs. So we need more studies of abstinence-only programs to find out which ones really do work.

POV: Are there some practical implications to what your studies have found?

Kirby: Philosophically -- and speaking now as a citizen rather than just a researcher -- I believe that we should be implementing those programs that are demonstrated to be effective, and it's a real gamble of our taxpayers' dollars to be implementing programs that have not yet been demonstrated to be effective. A lot of money is being used to implement abstinence-only programs that have not yet been evaluated. It's very important to evaluate those programs and then to implement the abstinence-only programs that are effective.

POV: Over the course of your career, what long-term changes have you seen in sex education programs? What things remain constant?

Kirby: It's not easy to change adolescent behavior, and we've certainly made a lot of mistakes over the years. For the first 10, 12 years that I did work in this field, all the programs we evaluated failed to have an impact on behavior. They did other good things, but they didn't change behavior. It was not until roughly around 1988 or 1989 that we had a good study showing that a particular program was effective.

POV: This may be a little outside your specialty, but can you talk a bit about the differences between the United States and other nations in terms of teen pregnancy rates, STD infections, and sex education?

Kirby: It's pretty well known that teen pregnancy rates are much lower in Western Europe than they are here in the United States. The US has the highest teen pregnancy rate in the Western industrialized world. And so many people try to compare, or have examined, why rates are lower in Western Europe than they are here. And, in my mind, the answer is a complex one. It may be the case that they have better sex and HIV education programs. It's certainly the case that in most of the Western European countries they have a more homogeneous population, which has reached greater agreement on what values should be emphasized to young people. Those tend, typically, to be pretty liberal values. But there's much greater agreement upon them than there is in the United States, where we have real polarization. So in Western European countries, they're consistently given a common message, whereas in the US we give conflicting messages.

It's also true that they have access to health care more generally, and that given that access to health care that includes reproductive health care, so they'd be more likely to receive reproductive health care services when they do become sexually active. Poverty is an incredibly important predictor of high teen pregnancy rates. And the Western European countries have greater equality than we do in this country, and there's less poverty there. So that makes a difference. They tend to devote more resources there to young people, more generally, than we do in this country. They're more supportive; they have clearer pathways for them to move from secondary school on into career paths than we do in this country.

POV: When you're studying something as complex as sexual behavior, with so many different influencing factors, how do you try to isolate the effects of a sex education program?

UPDATE: The initial findings of Mathematica Policy Research studies were reported on June 14, 2005 and found that "abstinence education programs increased youth's support for abstinence. The evidence on whether programs raised expectations to abstain is less clear." Download a PDF of the report at the Mathematica Policy Research website.

Kirby: In general, all of those factors fall into 4 different broad groups. One group are biological factors -- things such as age, gender, physical maturity, etc. Even hormone level makes a difference; testosterone level makes a difference. Another broad category is social disorganization and poverty: things like drug use, divorce rates, community crime rates; the use of alcohol and drugs; a whole variety of things associated with social disorganization and disadvantage. The third very important group is values, sexual values, either verbally expressed or modeled by people in the teen's environment: parents' values, perception of peers' values, whether or not their parents gave birth when they were teens, things of that nature. And the last important group is connection to groups that have pro-social values regarding sexual behavior. (By "pro-social" I mean values against sexual risk-taking.)

Parents tend to want their children to behave responsibly, sexually. So if young people are attached to their parents, if they feel close to their parents, they're less likely to have sex and to have unprotected sex. If they're involved in faith communities, which also tend to have pro-social values, they're less likely to engage in sexual risk-taking. If they're attached to school, the same thing is true. So that's the four broad categories. Lots of things have an impact. It's a complex world.

There's no question that parents and media and peers have a huge impact upon young people's sexual behavior. The good news is that parents are part of that list. Parents do have a greater impact on their children, and children's sexual behavior, than parents sometimes realize. So that's good news. But it's also true that media and peers and other factors have a very large impact as well.

POV: What kind of advice would you give to parents or educators?

Kirby: I would encourage parents and educators and others to take a careful look at the research about what we know does and does not work to change sexual behavior -- what is effective, what produces a positive impact on behavior -- and to implement those programs that do have a strong record. That would be my first recommendation: implement effective programs. My second recommendation would be, if you can't do that, then implement programs that have the 10 characteristics of effective programs.

POV: Sex education policy is such a polarizing subject, and so volatile, that curricula can change from year to year even in a single school. Do you have any recommendations for schools on how to best approach these issues?

Kirby: When we do our studies, we have real control over what's implemented. For example, we'll identify 20 schools that agree to participate. We randomly assign ten of them a program that is very carefully implemented with fidelity, and the [other] ten continue doing what they're already doing. And then we measure the impact on behavior over the following three years. That's a good evaluation design. But that's when the study's underway. In a typical school, where there isn't a study, what often happens is that teachers will order a few different curricula, and they will pull activities from different curricula, and kind of do their own thing. And although I can understand why they do that, they end up failing to implement with fidelity a particular curriculum into which a huge amount of thought has gone.

It's also true that schools typically do not allow many classroom periods to be devoted to HIV education or sex education, and consequently there's not enough time to implement some of the more effective curricula. So [I would recommend] allowing more time in the classroom for this topic. We can change behavior. We can reduce teen pregnancies that cause young people to drop out of school. We can reduce STD and HIV rates. [But] we need more time in the classroom. My second recommendation is that we need a process, or oversight, to make sure that teachers really do implement effective curricula with fidelity. Sometimes they start off implementing a particular curriculum with fidelity, but then maybe they go to a conference and they drop some of the old activities and add some new, and then maybe they move away to a new school and a new teacher comes, and a program that was very effective ends up dissipating, even though that was not anyone's intent.

Dr. Douglas Kirby is senior research scientist at ETR Associates, a nonprofit health education organization in Scotts Valley, California. He has served as chair of the Effective Programs and Research Task Force at the National Campaign to Prevent Teen Pregnancy, and is the author of numerous studies of sexual education programs, including Emerging Answers.