Introduction: Recent proposals have called for reform of the traditional medical education model and development of learner centered paradigms that prioritize conceptual mastery and enhance engagement. Blended learning models mediate a shift from lecture settings to collaborative active learning environments. Studies suggest that blended learning modalities may improve affective and objective outcomes compared to passive didactics. However, there is limited evidence regarding the implementation of blended learning in preclinical medical school courses. We developed an innovative blended learning curriculum for a preclinical second year medical student hematology course using a flipped classroom large-group active learning model and investigated its impact on student performance, satisfaction, and engagement. Methods: We used constructivist education theory and a mixed methods model to develop a blended learning curriculum. In 2015 we converted the didactic lecture component of the white blood cell dyscrasias unit into nine online video modules and two large-group active learning sessions. Blended learning content was structured to maintain time neutrality and allocate sufficient time for independent review. Large-group active learning sessions were planned for an Education Center active learning studio and designed to emphasize collaborative team-based discussion and included clinical problem solving, audience response questions, and gamification. Attendance at large-group sessions was optional in 2015 and mandatory in 2016 - 2017. The white blood cell dyscrasias question final examination subset scores were compared between the historical 2014 cohort and blended learning cohorts from 2015 - 2017 with one-way ANOVA with post-hoc comparisons via Dunnett's multiple comparisons test with a single pooled variance. The examination questions compared were identical. Course evaluations were compared between 2014 and 2015 via an independent samples t-test, but 2016 and 2017 evaluations were not directly compared due to an institution-wide change in evaluation scales. Qualitative student evaluations regarding the traditional and blended learning curriculum were collected between 2014 - 2017. Results: Overall satisfaction with the course (5, excellent) was similar (p=0.38) in 2014 (4.13±0.79) and 2015 (4.05±0.82). Satisfaction remained high (4, excellent) in 2016 (3.33±0.72) and 2017 (3.12±0.84). Student performance was marginally worse on the white blood cell dyscrasias question subset (17-point maximum) in 2015 (Mean Score, 14.78±2.01, p=0.0069), but not significantly different in 2016 (14.96±1.94, p=0.0856) or 2017 (15.10±1.82, p=0.3803) compared to 2014 (15.37±1.64). In 2015, on average, 34% (65/191) of students attended didactic lectures and active learning sessions were attended by 33% of students (60/183). Attendance was mandatory in 2016 (n=184) and 2017 (n=175). Between 2015 - 2017, 60% of students reported not completing video modules prior to corresponding active learning sessions. Qualitative analysis from 2015 - 2017 of student feedback revealed themes on active learning sessions focused on perceived lack of time neutrality, difficulty adapting to a novel learning style, appreciation of clinical case-based learning, and the value of real-time facilitator feedback. Conclusions: We found that conversion of conceptually challenging material in a preclinical hematology course to a blended learning curriculum did not improve performance. When mandated, attendance at active learning sessions did not incentivize student pre-session preparation. These findings carry implications for the use of blended learning in preclinical hematology education. Inadequate student preparation for sessions may have mitigated the benefit of active learning and detrimentally impacted performance. Students habituated to a lecture-based curriculum experienced difficulty adapting to a novel learning approach emphasizing collaborative engagement, weakening the utility of the active learning sessions. Our findings suggest the importance of accountability and preparation for active learning sessions in a blended learning curriculum. As blended learning becomes prevalent in medical education, further studies of pedagogic methods to overcome barriers to successful student performance, satisfaction, and engagement are necessary. Disclosures Murakhovskaya: Momenta: Membership on an entity's Board of Directors or advisory committees.