Pregnancy Experience and Use of Emergency Contraception Among Young Women in the Bay Area

Pregnancy Experience & Use of Emergency Contraception Among Young Women in the Bay Area Shira Rutman June 2006 Significance Three and a half million unintended pregnancies occur each year in the United States. (Henshaw 1998) Emergency contraception [EC] has the potential to prevent 50% of these pregnancies and 60-70% of abortions annually. (Trussel 1992) EC is effective when taken 72 to 120 hours after intercourse, but is most effective the sooner it is taken. (Ellertson 2003) “Parent” Study Design & Setting A randomized, single-blind, controlled trial Conducted by investigators at UCSF July 2001 to June 2003 at four California clinics providing family planning services Three arms: advance provision, pharmacy access and clinic (control) Examined reproductive outcomes (pregnancy, STIs and contraceptive/sexual behaviors  Total N=2,117 Participants  Young women  15-24 years old  Spoke English or Spanish  Had sexual intercourse in the previous six months (“sexually active”)  Lived in the San Francisco Bay Area  Available for follow-up 6 months later Participants continued  Not pregnant or desiring pregnancy  Not using long-term hormonal contraception (injectables, IUD or implants)  Not requesting EC at the enrollment  Had not had unprotected intercourse in the previous three days  Not using “patch” or “ring” (not yet FDA approved) “Parent” Study Findings  No significant reduction in pregnancy rates in pharmacy access and advance provision groups  No increase in STIs  No differences by study group in: o contraceptive patterns o condom use o sexual behaviors EC use during the study N=571 (26.9%) “The public health impact may be negligible because of high rates of unprotected intercourse and relative underutilization of the method.” Raine TR, Harper CC, Rocca CH, Fischer R, Padian N, Klausner JD, Darney PD. Direct access to emergency contraception through pharmacies and effect on unintended pregnancy and STIs: a randomized, controlled trial. JAMA 2005; 293: 54–62. Study Hypotheses 1) Young women who have experienced pregnancy are more likely to use EC, than young women who have not experienced pregnancy. 2) Among young women who have experienced pregnancy, those who have experienced abortion are more likely to use EC, than young women who have not experienced abortion, who may or may not have experienced pregnancy. Measures  Outcome Measure (Follow-up visit): o EC use  Main Predictor Measures (Baseline visit): o Previous pregnancy (yes/no) o Previous abortion (yes/no-includes those who have been pregnant, but not had an abortion and those who have not been pregnant) Measures continued Predictor Measures: o Level of access to EC (study arm) o Clinic site (enrollment site) o Age (continuous) o Race/Ethnicity o Ever used EC (yes/no) o Frequency of intercourse o Unprotected intercourse past 6 mos (yes/no) o Contraceptive method Methods Chi-square tests: Differences between previous pregnancy/abortion and no previous pregnancy/abortion groups. Crude odds ratios: Association between previous pregnancy/abortion and characteristics of participants at enrollment (demographics, contraceptive/sexual behaviors and pregnancy attitudes). Multivariate logistic regression: Impact of previous pregnancy/abortion (and other predictors) on EC use during the study (adjusted). Results: Previous Pregnancy  Previous pregnancy N=686 (32.4%)  Used EC during study (outcome) N=200 (35.0%)  Chi-square tests showed significant differences between pregnancy experience groups in every characteristic examined except for level of access to EC (study arm) Multivariate Logistic Regression: Previous Pregnancy 32% higher odds of EC use among participants who had had a previous pregnancy compared to those who had not, adjusting for all the other variables Multivariate Logistic Regression: Previous Pregnancy Odds of EC use:  Unprotected intercourse  Ever used EC OC use and use of both OC and condoms *Adjusted for all the other variables Results: Previous Abortion  Of the 686 (32.4%) participants who reported pregnancy at baseline, 562 (81.9%) reported that they had had an abortion.  Used EC during study N=167 (29.2%)  Chi-square tests showed significant differences between abortion experience groups in every characteristic examined except for level of access to EC (study arm) and clinic site. Multivariate Logistic Regression: Previous Abortion 41% higher odds of EC use among participants who had had a previous abortion compared to those who had not, adjusting for all the other variables Multivariate Logistic Regression: Previous Abortion Odds of EC use:  Unprotected intercourse  Ever used EC OC use and use of both OC and condoms *Adjusted for all the other variables Limitations  Risk perception was not directly measured in the study  Self-reported pregnancy and abortion; likely under-reported, leading to underestimates of EC use  Data were not collected to test my hypotheses Conclusions  Increased rates of unprotected intercourse and increased EC use among young women with previous pregnancies and abortions (may cancel each other out)  A reduction in unintended pregnancies may not be observed in these subgroups of young women  Education about contraceptive risk in these subgroups is indicated Thank You!  Thesis Committee Members: Drs. Jane Rees (Chair), Cynthia Harper & Hendrika Meischke  Nate Mercaldo and the Dept. of Biostatistics consultants  Family and Friends (especially the “Friday Thesis Group” members)  Adam This work was supported in part by a grant from the US Department of Health and Human Services, Health Resources and Services Administration’s Maternal and Child Health Bureau (Title V, Social Security Act), grant # T76MC00011-21-00.