Drey E University of California, San Francisco, San Francisco, CA, USA Sokoloff A, Oderman L, Goldschlager N, Benowitz N, Steinauer J Objectives: The purpose of this study was to determine the safety of two routes and doses of digoxin before dilation and evacuation (D&E) at 19 to 23 weeks' gestation through an assessment of maternal systemic digoxin absorption, cardiac rhythm and coagulation parameters. Methods: Healthy women received either 1.5 mg digoxin through intra- amniotic injection (IAI) or 1.0 mg through intra-fetal injection (IFI). We measured serum digoxin levels serially for 48 h and assessed Holter monitoring and coagulation indices for 24 h. Digozin serum levels' reference range was 0.5 to 2.0 mcg/L. Results: Ten patients in each group completed the study. The mean serum digoxin peak concentration with IAI was 1.5±0.72 mcg/L (range 0.7 to 2.9 mcg/L) and with IFI was 1.4±0.74 mcg/L (range 0.5 to 2.4 mcg/L). Two subjects in the IAI and three in the IFI groups had supratherapeutic serum digoxin levels. The mean time to peak digoxin concentration was 10.7±8.5 h (range 1 to 24 h) following IAI and 1.5±0.97 h (range 1 to 4 h) after IFI. Maternal cardiac rhythm monitoring showed no digoxin-associated rhythm or conduction abnormalities. Coagulation assessments did not change significantly. Conclusions: The maximum peak serum digoxin concentration achieved after 1.5 mg IAI and 1 mg IFI generally was in the low therapeutic range. No cardiac rhythm or conduction abnormalities associated with digoxin were noted and coagulation parameters were unaffected. Five of 20 patients had supratherapeutic serum digoxin levels; the clinical importance of these serum levels remains to be determined. P25 ROLE OF SEMI-QUANTITATIVE URINE PREGNANCY TESTS IN MEDICAL ABORTION PROVISION Lynd K Stanford University Medical Center, Stanford, CA, USA Blum J, Lichtenberg S, Fischer D, Ricci R, Arnesen M, Ali R, Winikoff B, Blumenthal P Objectives: Standard medical abortion practice often involves in-clinic follow-up with serial serum hCG tests. Home use of a semi-quantitative pregnancy test, such as the dBest semi-quantitative test, could offer an alternative. Negative results on a validated pregnancy test with high predictive value could reduce the number of follow-up visits and tests, simplifying clinical practice worldwide. This study assessed the feasibility and acceptability for determining complete abortion status at home with a semi-quantitative test in lieu of clinic-based follow-up. Methods: Approximately 600 women from four sites in the US participated. Participants had a baseline semi-quantitative pregnancy test prior to mifepristone administration. They performed a test at home the morning of their clinic follow-up visit. At this visit, a clinician reviewed each woman's results. Results: Over 90% of participants found the test easy to use. At follow- up, 26% of participants thought they were still pregnant, but there were only two ongoing pregnancies. The test correctly identified both ongoing pregnancies as requiring additional clinic-based care: one as ongoing and one as inconclusive. Both participants with ongoing pregnancies reported that they thought they were still pregnant at the time of follow-up. Conclusions: These data show that the semi-quantitative pregnancy test may be an effective medical abortion follow-up tool. Preliminary results show that 25% of participants have been lost to follow-up, supporting the need to provide a feasible alternative follow-up with technologies such as the semi-quantitative pregnancy test. P26 SECOND-TRIMESTER ABORTION PRACTICE PATTERNS AND BARRIERS TO PROVISION Kerns J University of California, San Francisco, San Francisco, CA, USA Steinauer J, Landy U, Turk J Objectives: Family planning (FP) is an emerging specialty in obstetrics and gynecology. We sought to describe second-trimester abortion practices and barriers to provision among FP physicians. Methods: A web-based survey was sent by email to all past and present FP fellows and faculty affiliated with FP fellowships. Physicians reported on their second-trimester training and practices, as well as barriers to providing dilation and evacuation (D&E). Results: Of the 170 FP physicians invited to participate, 102 (60%) responded. The majority are female (86%), less than 40 years old (67%), work with trainees (95%) and practice in an academic setting (89%) in the west or northeast region (65%). All are trained in D&E, most up to 24 weeks' gestation (70%). While some received training in residency (35%), most were trained during fellowship (60%). FP physicians perform or supervise an average of 17 D&Es and 1 induction termination each month. Most D&Es are done in the operating room (42%) or a freestanding clinic (27%). Many FP physicians report barriers to providing D&Es and induction terminations (65% and 73%, respectively). The main barrier cited most frequently for both methods was unsupportive nurses. D&E barriers also included logistical difficulties (e.g., scheduling conflicts or lack of operating time) and restrictions (e.g., federal, state, hospital or insurance). Conclusions: Although most FP physicians are located in areas with fewer restrictions, they still face significant barriers to second-trimester abortion provision. Further study of perceived barriers among all members of the health care team may help direct efforts to alleviate some of these barriers. P27 BARRIERS AND SUPPORTS TO MEDICATION ABORTION PROVISION BY ADOLESCENT MEDICINE PROVIDERS Coles M University of Rochester Medical Center, Rochester, NY, USA Makino K, Phelps R Objectives: To understand the relative importance of barriers and supports to medication abortion provision by adolescent medicine providers. Methods: Using an online questionnaire, we surveyed US SAHM providers (n=797) on medication abortion counseling and provision. The survey had a 54% response rate (n=430). Clinicians who did not provide medication abortion (n=405) were asked to rate the importance of 18 potential barriers to offering the service. Those who were somewhat/very interested in providing medication abortion (n=124) were asked to rate the importance of seven potential supports to their interest. We performed descriptive analyses of provider responses; items were further grouped into four categories based upon the theory of planned behavior (TBP). Results: The most important barriers included lack of medication abortion training, lack of ultrasound training, inability to provide surgical abortion and lack of medication abortion knowledge. With the TPB framework, barriers that inhibited providers' perceived behavioral control appeared to be most strongly associated with their decision not to offer medication abortion. The most important supports were perceptions that medication abortion is noninvasive and that it increases patient access to abortion services. Conclusions: The top four barriers to offering medication abortion can be addressed by education and training, and two (lack of training in ultrasound and surgical abortion) reflect widespread misperceptions about skills 315 Abstracts / Contraception 84 (2011) 302–336