Increasing an Understanding of Antibiotic Prescribing Practices in the Neonatal Intensive Care Unit (NICU) by Using Clinical Case Vignettes

Background: Surveys using clinical vignettes can be useful predictors of antimicrobial prescribing practices.

Objective: To use vignettes to assess diversity in antimicrobial prescribing practices by neonatal practitioners as well as agreement with recommendations for appropriate antibiotic use as recommended by the Centers for Disease Control and Prevention.  

Methods: An anonymous survey was administered to neonatal practitioners from 4 tertiary care NICUs.  Respondents reviewed 13 vignettes derived from past cases in the study NICUs and rated the described antibiotic management as ‘appropriate’, ‘inappropriate’, or ‘cannot determine’. Vignettes described use of empiric antibiotics; targeted antibiotics for a known pathogen; prophylaxis; and duration of therapy. Five of the vignettes described appropriate antibiotic use, 5 described inappropriate use, and 3 described ambiguous use. Vignettes were validated by pilot testing with infectious diseases physicians and neonatologists. Individual question and composite scores for correct assessment of appropriate, inappropriate or ambiguous use were analyzed using t tests, Fisher’s exact tests, and ANOVA.

Results: Of 271 potential respondents, 161 (59%) completed the survey (51 attendings, 29 fellows, 48 residents, 21 nurse practitioners, and 7 hospitalists) of whom 37% had worked in NICUs for > 7 years.  Most (82%) considered a short duration (24 hours) of post-surgical prophylaxis to be appropriate and most (75%) considered prolonged antibiotic prophylaxis for chest tubes to be inappropriate. Most (97%) thought narrowing vancomycin to oxacillin to treat methicillin-susceptible Staphylococcus aureus infection was appropriate, but fewer (53%) thought continuing use of a broad-spectrum cephalosporin to treat cefazolin-susceptible E. coli inappropriate. Attending status and > 7 years of NICU experience significantly predicted the highest composite scores (p=.04 and .02, respectively).  While composite scores were similar among the 4 sites, site differences were noted for specific vignettes describing the duration of post-surgical prophylaxis (p< .001) and treatment of necrotizing enterocolitis (p=.03).  

Conclusions: The survey identified substantial differences in antibiotic prescribing practices. While narrowing therapy for Gram positive cocci appeared to be standard practice, narrowing therapy for Gram negative bacilli was less common.  Clinical experience predicted appropriate prescribing. Such data can be used to provide prescribers with feedback and improve antimicrobial prescribing for the NICU population.