s 687 – 838 687 The relationship of type of feedback to learning during simulation: a randomized clinical trial Sumit Saraf, Jyothshna Bayya, Howard Minkoff, Jaimie Glick, Nelli Fisher Maimonides Medical Center, OB Gyn, Brooklyn, NY, Maimonides Medical Center, Obstetrics and Gynecology, Brooklyn, NY, Maimonides Medical Center, Ob/Gyn, Brooklyn, NY, Albert Einstein College of Medicine/Montefiore Medical Center, Obstetrics & Gynecology and Women’s Health, Bronx, NY OBJECTIVE: To compare the effect of negative vs. positive feedback on students cognitive and technical performance, and confidence level during obstetrical simulation. STUDY DESIGN: Medical and nursing students without prior experience in performing normal vaginal delivery (NVD) were randomized to positive feedback (PF) or negative feedback (NF) groups. All students (N 59) participated in a teaching demonstration and practiced NVD using a simulator (Noelle 575, Gaumard Scientific). A baseline skills assessment in NVD was followed by a second intervention simulation, during which positive or negative feedback was provided using standardized scripts at predetermined stages of the simulation. Previously validated checklists assessing cognitive and technical performance as well as confidence measures were scored. Mean baseline and intervention scores were compared between groups. For each of the three measures (cognitive, technical, confidence) as dependent variable, a mixed linear model was constructed, with profession (medical, nursing), assessment (baseline, intervention), feedback arm (positive, negative) as factors. RESULTS: Considering both medical and nursing students, all participants in the PF arm demonstrated significant performance improvement after the intervention in all three assessment scores while participants in the NF arm registered a non-significant decline. Specifically, cognitive and technical performance scores, and the confidence scores in the PF group improved significantly with p values of 0.012, 0.026 and 0.036 respectively; while those of the NF group did not (p not statistically significant). CONCLUSION: A positive feedback environment strengthened students cognitive and technical performances and confidence while negative feedback worsened them. Further studies are needed to see if these differences persist over time, and whether they are replicable in a real world environment. 688 Comparison of perinatal outcomes in twins conceived by artificial reproductive techniques and spontaneous conception–a secondary analysis of ESPRiT study data Akila Anbazhagan, Alyson Hunter, Fionnuala Breathnach, Fionnuala McAuliffe, Mike Geary, Sean Daly, John Higgins, John Morrison, Gerry Burke, Shane Higgins, Pat Dicker, Fiona Manning, Steve Carroll, Fergal Malone Royal Jubilee Maternity Hospital, Obstetrics and Gynaecology, Belfast, United Kingdom, Royal College of Surgeons in Ireland, Obstetrics and Gynaecology, Dublin, Ireland, UCD Obstetrics and Gynaecology, School of Medicine and Medical Science, National Maternity Hospital, Holles St, Dublin, Ireland, Rotunda Hospital, Obstetrics and Gynaecology, Dublin, Ireland, The Coombe Women and Infants University Hospital, UCD School of Medicine and Medical Science, UCD, Dublin, Ireland, Cork University Maternity Hospital, Obstetrics, Cork, Ireland, National University of Ireland, Obstetrics & Gynaecology, Galway, Ireland, Regional Maternity Hospital, Obstetrics and Gynaecology, Limerick, Ireland, Our Lady of Lourdes Hospital, Obstetrics and Gynaecology, Drogheda, Ireland, Royal College of Surgeons in Ireland, Epidemiology and Public Health, Dublin, Ireland, National Maternity Hospital, Holles St, Dublin, Ireland OBJECTIVE: To compare perinatal outcomes of twin pregnancies conceived by assisted conception with those of natural conception. STUDY DESIGN: Evaluation of Sonographic Predictors of Restricted growth in Twins (ESPRiT) is a multicentric prospective observational study conducted in eight maternity centres in Ireland, by the Perinatal Ireland Research consortium. Of 1001 twin pregnancies completing the study over a 24 month period, method of conception was available for 874 twins. Statistical comparisons were made between methods of conception and types of artificial reproductive techniques (ARTs) for demographics, baseline characteristics and perinatal outcomes. RESULTS: There were 636 (73%) spontaneously conceived twins and 238 (27%) conceived by ART in the study. Nearly all (97%) of the ART twins were dichorionic as compared to 74% dichorionic twins of those conceived naturally. The ART group was older with a maternal age of 35 years as compared to 31 years for the spontaneous conception group (p 0.0001). The ART group was also more likely to be nulliparous (70% vs 40%, p 0.0001) and were less likely to be smokers (4% versus 15%, p 0.0001). There were no significant differences between the obstetric features such as preeclampsia, gestational diabetes, antepartum haemorrhage or preterm delivery. There were 16(1%) neonatal deaths in the naturally conceived group and 3(0.6%) in the assisted conception group (p 0.124) We found no difference (p-value 0.05) in gestational age at delivery for the ART group Poster Session V www.AJOG.org Saturday, February 11, 2012 • 10:00 am – 12:00 pm • Marsalis Hall, Dallas Hyatt Regency Supplement to JANUARY 2012 American Journal of Obstetrics & Gynecology S307