Reducing adolescent pregnancy: Approaches that work

January 1, 1999

Pregnancy-prevention programs that succeed--and many don't--have certain characteristics in common. This review will help you evaluate the adolescent health services available to your patients and may suggest some new ways of looking at the counseling you offer during office visits.

Reducing adolescent pregnancy: Approaches that work

Jump to:Choose article section...Sex and HIV­education programsYouth development programsPutting principles into practice

Pregnancy-prevention programs that succeed--and many don't--havecertain characteristics in common. This review will help you evaluate theadolescent health services available to your patients and may suggest somenew ways of looking at the counseling you offer during office visits.

By Douglas Kirby, PhD.

Adolescent pregnancy remains a significant problem in this country despitesmall decreases in the pregnancy rate during the last six years. About 61%of high school seniors report having intercourse before they graduate fromhigh school, and many teenagers do not consistently use contraceptives properly,exposing themselves to pregnancy or sexually transmitted diseases (STDs).1,2Among 15- to 19-year-old women who rely on oral contraceptives, for example,only 41% take a pill every day. Similarly, among adolescent girls who relyon condoms as a primary method of contraception, only 35% of 15- to 17-year-oldsand 31% of 18- or 19-year-olds use a condom during every act of intercourse.

It's no surprise, therefore, that in 1995 10.1% of all females aged 15to 19, a total of about 890,000 individuals, became pregnant.3An estimated 85% of these pregnancies were unintended. More than 40% ofteenage women in the United States become pregnant before they are 20 yearsold, and many become pregnant a second time before their 20th birthday.4As a result, the birth rate for 15- to 19-year olds in the US in 1995 was56.9 per 1,000, a much higher rate than in other Western industrializedcountries.5

Many of the programs developed during the last two decades to reducesexual risk-taking behavior and pregnancy in adolescents have failed toreduce unprotected sex, let alone pregnancy rates.6 The goodnews is that a growing body of research indicates that two quite differentkinds of programs are effective:

* Sex and HIV­education programs that focus on reducing sexual risk-takingbehavior

* Youth development programs, which may or may not even discuss sexuality.

Many more studies have been done of sex and HIV programs than of youthdevelopment programs. I will review the evidence on both types and attemptto tease out their common characteristics.

Sex and HIV­education programs

These courses discuss not only abstinence but condoms and other methodsof contraception that can protect against STDs or pregnancy. Programs ofthis type include sex or AIDS education during regular school classes, coursesgiven on school campuses after school, and programs conducted in other placesin the community, such as homeless shelters and detention centers. The programsreflect the considerable creativity and differing perspectives of the agenciesthat implement them.

Here are 24 published studies of these programs that meet the followingseven conditions:

* conducted in the US or Canada

* focuses on curriculum-based programs

* targets adolescents of high school age--about 14 to 18 years old

* has experimental or quasi-experimentaldesigns

* uses a sample size of at least 90

* measures program impact on sexual or contraceptive behaviors

* has been published in peer-reviewed journals or volumes orin majorreports

These studies are listed in the box below.

Does education promote sex? Many people are concerned that sex and HIV­educationprograms that talk about condoms and contraception may send the wrong message.They fear that thesecourses appear to sanction sexual intercourse, teachyoung people how to have sex while reducing their chances of contractingan STD or becomng pregnant, or, in general, give young people ideas abouthaving sex. Some of these worries are plausible.

Fortunately, however, evaluations of the programs strongly support theconclusion that sexuality and HIV­education curricula do not increasesexual intercourse, either by hastening the onset of intercourse, increasingthe frequency of intercourse, or increasing the number of sexual partners.The 24 studies examined the impact of 35 different sexuality- and HIV­educationprograms on one or more of these three measures, and no study found a significantincrease on any of them. The data strongly indicate that sex and HIV­educationprograms do not increase sexual activity. These findings are consistentwith evaluations of similar programs in other countries.7

Does education discourage sex? Several of the 24 studies found that sexor HIV­education programs actually decreased one or more of the threemeasures. Six of 19 studies found that the programs being investigated significantlydelayed the onset of intercourse.Six of 13 studies showed a reduction inthe frequency of intercourse. Finally, three of 10 studies found a decreasein the number of sexual partners of program participants.

The studies also suggest that some, but not all, of the programs increasedcondom use or contraceptive use more generally. Eight of the 14 studiesthat examined condom use found that use increased according to at leastone measure. Four of the 10 that looked at contraceptive use found a positiveeffect.

Which matters more, AIDS or pregnancy? Of the curricula that significantlyincreased contraceptive use, a disproportionate number were AIDS educationprograms that increased condom use. Seven of 10 AIDS education programshad a significant effect on condom use. In comparison, one of four sex educationprograms increased condom use and four increased contraceptive use generally.Whether AIDS education programs are inherently more effective than sex educationprograms that cover pregnancy, STDs, HIV, and other topics is not known.It could be that AIDS education programs simply have benefited from betterfunding, better-trained staff, and studies with larger sample sizes, orthat they have some other advantage not related to content over sex educationprograms. For high-risk men in some communities, AIDS is undoubtedly a moresalient threat than pregnancy.

The data also suggest that sex and AIDS education programs may be moreeffective with high-risk youth than with young people at lower risk. Thesefindings may be more statistical than content-related. When large percentagesof youth initiate sex in a short period of time or fail to use condoms consistently,data showing that the intervention group does better than the control groupare easier to come by than when the behavior being investigated is alreadyoccurring at a low rate.

How rigorous are these studies? In the past some investigators, includingme, have questioned the methodologic rigor of these studies.6Two years ago, only three studies of abstinence-only education, sex education,or HIV­education programs used random assignment, large samples, long-termfollow-up, measurements of behavior, and proper statistical analysis, andnone of the three found significant effects on behavior.8­10Fortunately, two very recent studies do meet these tests of rigor. Bothinvestigations clearly indicate that school-based and community sex andHIV­education programs can increase condom or contraceptive use anddecrease unprotected sex.

One of the programs studied, Safer Choices, is a two-year, school-basedprogram for high-school youth designed to prevent HIV infection, other STDs,and pregnancy. Its goal is to reduce the number of youth who have sex andto increase the use of condoms. The Safer Choices intervention is basedon social cognitive and social influence theories and models of school change.11­13It has five primary components: a school health promotion council that focuseson the implementation of the Safer Choices intervention, a 10-session seriesof classes for 9th and 10th graders, peer-implemented school-wide activities,parent education, and school-community linkages. The intervention reducedunprotected sex and increased the use of condoms over a 31-month study period.14

The second program, Be Proud, Be Responsible: Safer Sex, which recruitedparticipants through schools, was implemented on two consecutive Saturdays.A safer-sex intervention indicated that abstinence was the best choice,but emphasized that youngsters who did have sex should use condoms. It wasbased on social cognitive theory, the theory of reasoned action, the theoryof planned behavior, and focus group research with youngsters from the targetpopulation. The intervention included eight highly structured one-hour modules.Both this program and Safer Choices included group discussions, videos,games, brainstorming activities, experiential exercises, and skill-buildingactivities. The curriculum reduced the frequency of sex, increased condomuse, and decreased unprotected sex for the 12 months the program's effectswere studied.15

Common elements. Compared with programs that are not effective at reducingunprotected sex, these two curricula have 10 characteristics they sharewith each other and all successful sex and HIV­education programs (seetable below). These distinguishing characteristics, which rely on variedaspects of effective pedagogy, are similar to those found in programs effectiveat reducing substance abuse.

Youth development programs

These programs offer another way to reduce teenage pregnancy. Insteadof focusing on sexuality, they are more holistic. One of their underlyingprinciples is to help prepare young people for adult life, not just to keepthem problem-free. The programs do this by improving adolescents' life skillsand increasing their opportunities or "life options." Thus, theymay boost young people's motivation to avoid early childbearing by strengtheningtheir belief in their future.

Youth development approaches to reducing adolescent pregnancy are notnew--some professionals recognized the importance of improved life optionsin the early 1980s or even earlier--but interest in them has grown rapidlyin recent years.16 Nonprofit organizations such as the NationalCampaign to Prevent Teen Pregnancy, foundationssuch as the Children's AidSociety, and some federal agencies are funding the development and evaluationof programs using this approach, and some states are redirecting adolescentpregnancy prevention funds to youth development programs.6,17,18

What accounts for this heightened interest in youth development approaches?Several facts have become increasingly clear:

* Sexuality education programs and access to condoms and other formsof contraception, though important, are only one component in an overallstrategy to reduce teen pregnancy.

* Many teenagers are not highly motivated to avoid pregnancy and so willnot consistently use contraception.

* Sexual risk taking has many nonsexual antecedents, such as poor schoolperformance.

* When youth development programs address these antecedents, pregnancyand other risky or deviant behaviors may decrease.

Although studies of youth development programs are far fewer than thoseof sex and HIV­education, the evidence to date suggests that these interventionsmay be effective. Certain programs, both past and present, are particularlynoteworthy for reducing pregnancy or childbearing among adolescents.

Youth Incentive Entitlement Pilot Projects, which targeted youths fromlow-income households who were still enrolled in high school, had a singleprimary component--providing participants with employment.19The adolescents were offered a part-time job during the school year anda full-time job in the summer, and were required to remain in school. Bothjobs were minimum wage.

The Seattle Social Development Project aimed to increase the adolescent'sinteraction with the school and family. It provided teachers with trainingin classroom management and instruction, including interactive teachingand cooperative learning. Parents were trained in developmentally appropriatemethods of parenting; students were taught how to recognize and resist socialpressures to engage in problem behaviors.

The Teen Outreach Program has two major components: (1) weekly classroomor small-group discussions of topics such as understanding personal values,decision-making skills, communication skills, human growth and development,parenting, family relationships, life options, community resources, andvolunteer experiences; and (2) volunteer service in the participants' schoolor community.20 The discussion sessions also give students anopportunity to develop a relationship with an adult facilitator. Duringthe school year, participants meet in the small groups for approximatelyan hour each week and perform a minimum of 20 hours of volunteer service.

The American Youth and Conservation Corps provides education and workexperience within the context of community services. Corps members workfull time in small teams in a variety of community projects, which typicallylast for two to 12 weeks. They may assist nursing homes, for example, helprenovate low-income housing, or complete environmental projects. The hostagencies often help train and supervise the corps members. The program hasbeen shown to reduce pregnancy among African-Americans.21

The Quantum Opportunities Project was a multiyear program starting inthe ninth grade and continuing through high school.22 It includededucational activities such as tutoring, computer-based instruction, andhomework assistance; service activities, including community-service projects,assistance at public events, and paid jobs; and development activities--acurriculum on life and family skills, and college and job planning.

The Children's Aid Society Program in New York City is one of the mostcomprehensive youth development programs ever offered.23,24 Basedon the philosophy that it is important to influence many facets of youths'lives over a continuous and prolonged period, it has numerous components.Participants and their parents are offered comprehensive education on familylife and sex. Teens also have access to medical and health services, includingreproductive health services; dance and dramatic productions; a varietyof sports; a job club; staff tutoring and help with homework; and ongoingrelationships with adults in the program who care about them. In addition,many of the program participants who complete high school are guaranteedcollege admission.

Common elements. These youth development programs are quite diverse,which suggests that many types are effective at reducing adolescent pregnancy.Although it is too early to know which program characteristics are criticalto reducing unprotected sex and pregnancy, the features they have in commonsuggest what makes them successful.

All the programs increase participants' opportunity to interact withadults--through jobs, a closer relationship with schools, weekly meetingswith adults about important topics, and other activities. Survey researchclearly indicates that such connections reduce teen pregnancy.25The programs also curtail participants' discretionary time. This may decreaseteens' opportunities for engaging in unprotected sex or spending time withpeers who are not in the program. Finally, all successful programs had thepotential to increase adolescents' belief in a bright future. They offerhope for doing meaningful work, getting a job, or being successful in schooland may increase the motivation to avoid early childbearing.

Putting principles into practice

Sex and HIV­education programs and youth development programs arecomplementary. One type of program focuses on the sexual antecedents ofsexual risk taking, such as sexual beliefs, attitudes, and norms; the otheraddresses nonsexual antecedents, including connections to adults, schoolperformance, employment, or belief in a good future. Neither type of programeliminates all sexual risk-taking behavior, but both have a significantimpact on sexual risk taking or pregnancy. It therefore seems likely thatthese programs will be more effective in combination than separately andthat both types of program can be effective components in larger, more comprehensiveefforts to reduce sexual risk taking and pregnancy.

All young people need not participate in both types of programs, however.Some youth know how to use contraception effectively but may not be motivatedto avoid pregnancy. Others lack knowledge or skills but want to avoid pregnancy,have relationships with adults, and feel hopeful about the future. Pediatriciansare in an excellent position to steer their patients toward appropriateprograms, and to support these programs in their communities. And duringoffice visits, they can reinforce two key findings of this review: to preventpregnancy, facilitate the use of condoms and other forms of contraceptionby sexually active youth, and strengthen the motivation to postpone childbearingby helping young people believe in a successful future.


1. Centers for Disease Control and Prevention: Trends in sexual riskbehaviors among high school students--United States, 1991­1997. MMWR1998;47(36):749

2. Alan Guttmacher Institute: Sex and America's Teenagers. New York,NY, Alan Guttmacher Institute, 1994

3. Henshaw SK: US Teenage Pregnancy Statistics. New York, NY, Alan GuttmacherInstitute, 1998

4. Hayes CD (ed): Risking the Future: Adolescent Sexuality, Pregnancy,and Childbearing, vol 1. Washington, DC, National Academy Press, 1987

5. Ventura SJ, Anderson RN, Martin JA, et al: Births and deaths: Preliminarydata for 1997. National Vital Statistics Reports, 47(4).Hyattsville, MD,National Center for Health Statistics, 1998

6. Kirby D: No Easy Answers: Research Findings on Programs to ReduceTeen Pregnancy. Washington, DC, National Campaign to Prevent Teen Pregnancy,1997

7.Grunseit A, Kippax S, Aggleton P, et al: Sexuality education and youngpeople's sexual behavior: A review of studies. J Adolesc Res 1997;12(4):421

8. Kirby D, Korpi M, Barth RP, et al: The impact of the Postponing SexualInvolvement curriculum among youths in California. Fam Plann Perspect 1997;29(3):100

9. Kirby D, Korpi M, Adivi C, et al: An impact evaluation of SNAPP, apregnancy- and AIDS-prevention middle school curriculum. AIDS Preventionand Education, 9 (Suppl A)1997;44

10. Thomas B, Mitchell A, Devlin M, et al: Small group sex educationat school: The McMaster teen program, in Miller B, Card J, Paikoff R, etal: Preventing Adolescent Pregnancy. Newbury Park, CA, Sage Publications,1992

11. Bandura A: Social Foundations of Thought and Action. Englewood Cliffs,NJ, Prentice Hall, 1986

12. McGuire W: Social Psychology, in Dodwell PC (ed): New Horizons inPsychology. Middlesex, England, Penguin Books, 1972

13. Marsh D, Brown E, Crocker P, et al: Building Effective Middle Schools:A Study of Middle School Implementation in California Schools. Los Angeles,CA, School of Education, University of Southern California, 1988

14. Coyle KK, Basen-Enquist KM, Kirby DB, et al: Short-term impact ofa multi-component school-based HIV, other STD and pregnancy prevention program(in press)

15. Jemmott JB, Jemmott LS, Fong GT: Abstinence and safer sex: A randomizedtrial of HIV sexual risk-reduction interventions for young African-Americanadolescents. JAMA 1998;179(19):1529

16. DryfoosJG: A new strategy for preventing unintended teenage childbearing.Family Planning Perspectives 1984;16(4):193

17. Philliber S: Personal communication, December 1, 1998

18.Catalano RF, Hawkins JD, Kosterman R, et al: Long-term effects ofthe Seattle social development project: Implications for theory and practice.Meeting of the Society for Research on Adolescence, San Diego CA, February27, 1998

19. Olsen RJ, Farkas G: The effects of economic opportunity and familybackground on adolescent cohabitation and childbearing among low-incomeblacks. J Labor Economics 1990;8:341

20. Allen JP, Philliber S, Herrling S, et al: Preventing teen pregnancyand academic failure: Experimental evaluation of a developmentally-basedapproach. Child Development 1997;64(4):729

21. Jastrab J, Masker J, Blomquist J, et al: Evaluation ofNational andCommunity Service Programs: Impacts of Service: Final Report on the Evaluationof American Conservation and Youth Service Corps. Cambridge, MA, Abt Associates,1996

22. Hahn A, Leavitt T, Aaron P: Evaluation of the Quantum OpportunitiesProgram (Q0P): Did the Program Work? Waltham, MA, Center for Human Resources,Brandeis University, 1994

23. Carrera M, Dempsey P: Caffera/Dempsey Replication Programs: 1993­94Summary of Client Characteristics and Outcomes. Accord, NY, Philliber Associates,1994

24. Kaye J, Philliber S: Comprehensive adolescent pregnancy preventionprograms: Are they working? New York: The Children's Aid Society, 1995

25.Resnick MD, Bearman PS, Blum RW, et al: Protecting adolescents fromharm: Findings from the National Longitudinal Study on Adolescent Health.JAMA 1997; 278(10):823

By Douglas Kirby, PhD.
THE AUTHOR is a Senior Research Scientist at ETR Associates, Santa Cruz, CA.