Science and Success, Second Edition: Sex Education and Other Programs that Work to Prevent Teen Pregnancy, HIV and Sexually Transmitted Infections
Full Study Report [HTML] [PDF]
Executive Summary [HTML] [PDF]
- Community-based HIV prevention program for use in family medicine and health clinics
- Gender-specific and culturally tailored program
- Four, four-hour interactive group sessions, held on successive Saturdays
- Sessions utilizing poetry and artwork of African American women, role plays, and discussions, demonstrating use of a condom
- Health educator and peer educators (all trained, African American, and female)
- Compensation ($25.00) for travel and child care
For Use With
- Sexually active African American females, ages 14 to 18
- Urban and suburban youth
- Experimental evaluation design with treatment and control conditions in Birmingham, Alabama
- Eligible African American adolescent females (n=522) seeking services at four community health agencies between December 1996 and April 1999 and randomly assigned to treatment (n=251) and control (n=271) conditions
- Data from baseline questionnaire, interview, demonstration of condom use skills, and STI testing
- Follow-up after six and 12 months
- Reduced number of new sex partners
- Reduced incidence of unprotected sex
- Increased condom use
- Long-term: Reduced incidence of STIs
- Long-term: Reduced incidence of pregnancy
SiHLE comes from a Swahili word for beauty and also is an acronym for sisters informing, healing, living, and empowering. This STI/HIV prevention intervention is based on social cognitive theory and theories of gender and power. The program is culturally and gender-specific for African American adolescent women at risk for negative sexual health outcomes. The program’s designers, working in partnership with community African American female teens, developed both the intervention and the study conditions. The intervention consists of four sessions, each lasting four hours and implemented on consecutive Saturdays at a community health clinic, by a trained, female, African American health educator. She is assisted by two female African American peer educators. The peer educators model skills and promote group norms supportive of HIV prevention.
In the four sessions:
- Session one emphasizes ethnic and gender pride. It encourages participants to explore and discuss the joys and challenges of being an African American adolescent female. Participants also acknowledge the accomplishments of African American women through reading their poetry and framing their art.
- Session two raises awareness of HIV risk reduction strategies, such as abstaining from sex, using condoms consistently, and having fewer sex partners.
- Session three uses role-plays and cognitive rehearsal to increase young women’s confidence in their ability to: initiate safer sex conversations with a partner; negotiate safer sex; and refuse unsafe sex. During session three, peer educators also discuss the importance of abstinence and consistent condom use and model condom use skills.
- Session four emphasizes the importance of healthy relationships. The health educator and peer educators lead discussions in how unhealthy relationships can make it difficult to practice safer sex.
From December 1996 through April 1999, recruiters screened 1,130 self-identified African American adolescent females seeking health care services at any of four community health agencies. Of these, 609 (54 percent) met eligibility criteria for the study. Eligibility criteria included: 1) being African American and female; 2) being 14 to 18 years of age; 3) having had vaginal intercourse in the preceding six months; and 4) providing written, informed parental consent. Among those not eligible, nearly 93 percent were not sexually experienced.
Of the 609 eligible adolescents, 522 agreed to participate in the study, completed baseline assessments, and were randomly assigned to treatment (n=251) or control (n=271) conditions. Treatment youth received the HIV intervention. Control youth received a general health promotion program of equal length and duration (four, four-hour sessions). Each participant received $25.00 as compensation for anticipated travel and child care expenses.
Evaluators collected data at baseline and at six- and 12-month follow-up, each time from four sources.
- Participants completed a self-administered questionnaire on socio-demographics and psychosocial aspects of HIV preventive behaviors.
- A trained African American female health counselor then interviewed each participant to assess 1) sexual behaviors; and 2) condom use skills. Self-reported, consistent condom use in the 30 days prior to each assessment was the main outcome measure. Other self-reported sexual behaviors included incidence of protected and unprotected sex and a new partner in the 30 days preceding assessment.
- Participants provided two self-collected vaginal swab specimens, one to test for gonorrhea and chlamydia and the other, for trichomoniasis.
- Self-reported pregnancy and STI incidence (determined by testing) were also assessed.
At baseline, evaluators detected significant differences between the treatment and control conditions in terms of HIV-related sexual behaviors; these were included as covariates in subsequent data analyses. Covariates included: history of douching; gang involvement; alcohol use; nonconsensual sex; depression; having a new partner; desiring to be pregnant; and/or not attending school.
No significant differences were seen on socio-demographic characteristics, condom use, or other outcome measures. For example, the mean age of intervention participants was 15.99; that of control youth was 15.97. Forty-six percent of treatment youth had not completed 10th grade, compared to 49 percent of control youth; 18 and 18.5 percent, respectively, received public assistance; 74 and 72 percent, respectively, lived in a single parent home; 24 percent and 23 percent, respectively, had children. Thirty-eight percent of each group reported using a condom in the past 30 days; three and two percent, respectively, reported unprotected vaginal sex in the past thirty days. At baseline, 19 percent of treatment young women and 16 percent of controls tested positive for chlamydia; six and five percent, respectively, tested positive for gonorrhea; 13 and 12 percent, respectively, tested positive for trichomoniasis.
Of the 251 participants assigned to the HIV intervention, 226 (90 percent) completed the six-month assessment and 219 (87 percent) completed the 12-month assessment. Of the 271 youth assigned to the control condition, 243 (90 percent) completed the six-month assessment and 241 (89 percent) completed the 12-month assessment. No differences in attrition were observed between study conditions at either the six-month or the 12-month assessment. Additionally, evaluators found no differences at either follow-up in: socio-demographic factors; or differences in baseline variables for study condition participants versus those lost to follow-up.
- Increased condom use—Intervention participants were more likely than controls to use condoms consistently in the 30 days preceding the six-month assessment (75 versus 58 percent; P=.06), in the 30 days preceding the 12-month assessment (73 versus 57 percent; P=.02), and during the entire 12 month period (odds ratio 2.01; P=.003). Intervention participants were more likely than controls to report consistent condom use in the six months preceding the six-month assessment (61 versus 43 percent; P=.001) and in the six months preceding the 12-month assessment (58 versus 45 percent; P=.01).
- Reduced incidence of unprotected sex—Intervention participants were significantly less likely than controls to report unprotected sex in the 30 days prior to the six-month assessment (mean difference -1.82 versus 0.27; relative change -50.69; P=.046).
- Reduced number of new sex partners—Intervention participants were less likely than controls to report having a new sex partner in the 30 days preceding follow-up at six months (three versus seven percent; P=.01)and 12 months (four versus six percent; P=.01).
- Reduced incidence of pregnancy—Intervention participants were significantly less likely than controls to report a pregnancy in the six months after baseline (four versus seven percent; P=.04) or in the 12 months after baseline (six versus nine percent; P=.06).
- Reduced incidence of chlamydia—Results of STD-specific analyses over the entire 12-month follow-up period, adjusting for baseline variable and covariates, suggested a treatment advantage in reducing chlamydia infections (OR 0.17; P=.04). There were no observed treatment effects in reducing either gonorrhea or trichomoniasis. Evaluators suggest that the small STI treatment effects are due, in part, to the relatively small number of incident STIs and to missing data for some covariates.
For More Information or to Order, Contact
- Sociometrics, Program Archive on Sexuality, Health & Adolescence: Phone, 1.800.846.3475; Fax, 1.650.949.3299; E-mail,
; Web, http://www.socio.com