P3.T381: Improving Knowledge and Confidence in Neonatal Lumbar Punctures Among Pediatric Residents Through Simulation Training, a Quality Improvement Initiative

Purpose/Objectives: Performing a lumbar puncture (LP) is an essential procedural skill that must be attained by all pediatric residents during training. An unsuccessful LP can result in prolonged antimicrobial exposure, the need for repeat attempts, and increased stress for patients and caregivers. Variable opportunities to perform LPs on infants can lead to fluctuating competency among trainees. This may cause anxiety and discomfort surrounding the procedure for those with insufficient opportunities to practice. We implemented a quality improvement initiative to increase knowledge and confidence in performing LPs on infants among pediatric trainees.

Design/Methods: Medical students and pediatric residents of all training years participated in the LP training simulation, which included a standardized educational video created specifically for our training institution followed by staff-supervised, small group skill simulation sessions. This training was repeated six months later for all available learners. Trainees completed pre- and post-simulation questionnaires containing five knowledge questions (binomial, max composite score of 5) and three confidence questions (Likert scale 1-5), all related to preparation and performance of LPs on infants. Statistical analysis of questionnaire responses used the paired t-test method comparing mean scores of pre- and post-simulation questionnaires.

Results: There were a total of 29 trainees who completed one or both LP simulation sessions, which included 3 medical students, 10 PGY1, 7 PGY2 and 3 PGY3 residents between the two sessions (Figure 1). Mean knowledge composite scores nearly doubled for both sessions (initial 2.4 vs. 4.4, p< 0.01; follow-up 2.8 vs. 4.4, p< 0.01; Figure 2). Similarly, for both sessions confidence in LP setup increased (initial 2.1 vs. 3.3, p< 0.01; follow-up 2.3 vs. 3.4, p< 0.01), confidence to independently perform LPs on infants increased (initial 1.8 vs. 3.1, p< 0.01; follow-up 2.3 vs. 3.4, p< 0.01) and confidence in overall infant LP procedural skills increased (initial 1.9 vs. 3.1, p 0.01; follow-up 2.3 vs. 3.4, p< 0.01).

Seven trainees (3 PGY1 and 4 PGY2) attended both sessions. Comparing their baseline responses to initial post-session and follow-up post-session showed sustained increase in mean composite knowledge score: 1.9 vs. 4.6 (p < 0.01) then 1.9 vs. 4.4 (p < 0.01). Similarly, their confidence scores also increased for LP set-up (2.6 vs. 3.7, p< 0.01 then 2.6 vs. 3.8 p< 0.01) and independent performance (2.4 vs. 3.6, p< 0.01 then 2.4 vs. 4, p< 0.01). Confidence in infant LP procedural skills only increased from baseline after the follow-up post-session (2.7 vs. 4 p=0.01).

Conclusion/Discussion: Variation in LP training opportunities and subsequent deficits in knowledge and confidence can be mitigated by instituting a standardized, interactive simulation. Our quality improvement simulation increased a sustained knowledge base as well as confidence level in preparing for and performing infant LPs, both immediately and 6-months post-simulation for pediatric trainees and medical students.