Prevalence and Correlates of Using Withdrawal Only vs. Condoms and/or Hormonal Birth Control at Last Sex among African-American Adolescent Females

Context: Withdrawal (coitus interruptus) does not protect against STDs and is not highly effective at preventing pregnancy. Few studies have explored differences among adolescent females using withdrawal vs. other forms of STD and pregnancy prevention. Objective: To assess the prevalence and correlates of using withdrawal only vs. condoms and/or hormonal contraceptives. Methods: Sociodemographic, psychosocial and sexual behavior data were collected via ACASI from 701 African-American females, 14-21 years, seeking reproductive health services. Data were obtained prior to randomization in an HIV prevention intervention. Bivariate and logistic regression analyses compared study variables for those who reported withdrawal only at last sex vs. condoms and/or hormonal birth control (pills, patch or depo). Results: A total of 506 (72%) participants used withdrawal, condoms or hormonal methods at last sex; of those, 130 (26%) used withdrawal, and 67 (13%) used withdrawal only. Those who used withdrawal only were: older (18.2 vs. 17.6 years, p=.004), older at first sex (15.2 vs. 14.8 years, p=.035), more likely to have ever tried alcohol (90 vs. 75%, p=.007) and more likely to ever have forced vaginal sex (34 vs. 19%, p=.003) and been emotionally abused (67 vs. 51%, p=.013). In adjusted analyses, only forced vaginal sex (OR: 1.85, p=.048) remained significantly associated with withdrawal only. Conclusions: Withdrawal is common among African-American adolescent females; many use withdrawal as their only form of protection. Better understanding of the factors associated with use of withdrawal only may help to inform the design of STD and pregnancy prevention programs serving African-American adolescents.