e4 ABSTRACTS ACADEMIC PEDIATRICS

Critical incident (CI) essays are a reflective narrative in whichlearners are asked to describe a specific event that has greatlyinfluenced their practice. As critical incidents often inspire novelinsights and learning goals, these essays provide unique insightinto the experiences and metacognitive processes of trainees.OBJECTIVES: (1) To describe the types of experiences depicted

in pediatric resident CI essays. (2) To describe the insights andlearning goals that emerge from personal reflection on criticalincidents.METHODS:We performed a qualitative, thematic analysis of 90

de-identified CI essays written by pediatric residents at a singleinstitution between 2002 and 2012. Essays were separated byyear of training and were randomly sampled until thematic satu-ration was achieved. Four reviewers coded all essays by an itera-tive process. Atlas.ti 7 software was used to assist with analysis.RESULTS: Experiences reported generally described one or

more of the following scenarios a near miss, a medical error, a pa-tient decompensation, the death of a patient, a communicationbreakdown, or a difficult social situation. Novel insights andlearning goals fit into 5 major themes: (1) development of self-trust and sense of personal responsibility; (2) maturation ofclinical skills, including critical thinking and comfort with uncer-tainty; (3) appreciation of humanism in patient care and commit-ment to social advocacy; (4) recognition of the importance ofgood communication and teamwork; and (5) professional identitydevelopment, including a commitment to lifelong learning.DISCUSSION: Pediatric resident CI essays emerge from a

diverse set of experiences and stimulate reflection and goal settingin many areas, including personal and professional development,communication, humanism and patient care. Many of these iden-tified learning goals overlap with pediatric ACGME core compe-tencies and the pediatric milestones. These findings help toilluminate the complex challenges faced by residents whileproviding patient care, and can inform our design of curriculumand support services.

10. EARLY CAREER EXPERIENCESOF PEDIATRICIANSPURSUING OR NOT PURSUING SUBSPECIALTYFELLOWSHIPSBobbi J. Byrne, MD, Indiana University, Indianapolis, IN,Shesha K. Katakam,MD,MPH, IndianaUniversity Health LaPorte, La Porte, IN, Mary Pat Frintner, MSPH,William L. Cull, PhD, American Academy of Pediatrics, ElkGrove Village, IL

BACKGROUND: Choosing career paths can be difficult decisionsfor residents who contemplate whether or not to pursue fellowshiptraining. There is not a significant body of literature for residencyand fellowship directors to draw from in counseling residents inthis area.OBJECTIVE: Compare the experiences of early career pediatri-

cians who did and did not pursue fellowships.METHODS: We analyzed nationally-representative, weighted

data from the AAP PLACES study of pediatricians 1e3 yrs(recent grad cohort) and 8e10 yrs (early career cohort) post-res-idency (n ¼ 1725). Work environment, work-life balance, finan-cial compensation, and career satisfaction were compared forpediatricians in or who had completed fellowship training (FT)and those that did not pursue fellowship training (NFT). Multivar-iable logistic regression examined the independent effects offellowship training while controlling for demographic differences(gender, IMG, race, and children) in each cohort.RESULTS: 44% of the recent grad cohort (384/880) and 39% of

the early career cohort (331/845) were FT. FT were more likely

than NFT to report learning opportunities in their work environ-ment in both the recent grad (38 vs 18%, p < .001) and earlycareer (25 vs 12%, p< .001) cohorts. FT in the early career cohortwere more likely to report an income of $125, 000 or higher (84 vs67%, p< .001). NFT in both cohorts were more likely than FT towork less than 50 hours per week (recent grad: 76 vs 27%,p < .001 and early career: 77 vs 39%, p < .001), have flexibilitywith their schedules (66 vs 51%, p < .001 and 64 vs 54%,p < .01), and be satisfied with time spent with their children(59 vs 37%, p < .001 and 59 vs 47%, p < .01). Both FT andNFT were generally satisfied with their life on whole and foundtheir work to be rewarding. (Table 1).CONCLUSION: Directors can reassure pediatric residents that

although important life and career differences need to be consid-ered when contemplating fellowship training, either choice usu-ally results in overall life and career satisfaction.

11. VALIDATION OF GLOBAL RATING SCALE ANDCHECKLIST INSTRUMENTS FOR THE INFANT LUMBARPUNCTURE PROCEDURE AMONG RESIDENT RATERSRenuka Mehta, MBBS, Medical College of Georgia atGeorgia Regents University, Augusta, GA,Colleen Braun, DO, Saint Louis University School ofMedicine, St Louis, MO, David O. Kessler, MD, ColumbiaUniversity Medical Center, New York, NY,Marc Auerbach, MD, Yale University School of Medicine,New Haven, CT, Anthony J. Scalzo, MD,James M. Gerard, MD, Saint Louis University School ofMedicine, St Louis, MO

INTRODUCTION: The International Network for Simulation-based Pediatric Innovation, Research and Education (INSPIRE)has developed tools for the assessment of competency to performthe infant lumbar puncture (ILP) procedure. We previously re-ported evidence to support the validity and reliability of these in-struments when used by expert raters in a simulated setting. Theobjective of this study was to evaluate the validity and reliabilityof these tools when used by resident raters to assess simulated ILPperformances.METHODS: Video recordings of 60 subjects performing an un-

supervised LP on an infant bench top simulator were collectedprospectively; 20 performed by subjects in each of three cate-gories (beginner, intermediate experienced, and expert). Sixblinded resident raters independently scored each subjects’ videorecording [3 via a global rating scale (GRS), 3 via a checklistinstrument].RESULTS: For the resident raters, across all subject groups,

higher GRS scores were found with advancing level of experience(P< 0.01). Total checklist scores were similar between the expert(80.0%) and intermediate experienced (76.6%) groups (P¼ 0.68).Both groups scored higher than the beginner group (50.4%) on thechecklist instrument (P < 0.01). Cronbach’s alpha coefficient forthe checklist was 0.77. The intraclass correlation coefficientsamong raters for the GRS and total checklist scores were 0.49and 0.47 respectively.CONCLUSIONS: Similar to that found for expert raters: 1)

acceptable internal consistency was found for the checklist instru-ment, and 2) The GRS instrument outperformed the checklist inits discriminant ability. Only moderate agreement among raterswas found for each of the scoring tools. Compared to expertraters, we found the level of agreement among resident raters tobe poorer for both of the scoring instruments. Reference: 1.Gerard JM, Kessler DO, Braun C, Mehta R, Scalzo AJ, AuerbachM. Validation of global rating scale and checklist instruments for