Introduction: Pediatric CNS infections may present with a variety of signs and symptoms, which may be subtle or inconsistent. Clinical diagnoses may vary depending on the assessor. For syndromic surveillance and reliable public health data, standardized case definitions should be implemented at the point-of-care (POC).

Methods: An automated case ascertainment tool (ChAT) was programmed based on published case definitions for aseptic meningitis (ASM), encephalitis (ENC), myelitis (MYE), and acute disseminated encephalomyelitis (ADEM). From 2010-2013, a total of 521 consecutive cases of suspected CNS infection (51.2% male, median age: 6.4 years, range 0.03;17.9 years) were assessed prospectively. An independent quality management team performed structured neurologic assessments using ChAT in real-time. Case classifications based on ChAT analysis were compared to: chart review applying the same standardized case definitions retroactively, versus screening of ICD-10-coded discharge diagnoses.

Results: With ChAT analysis, 200 cases fulfilled at least one case definition, compared to 170 cases by retrospective chart review. Review of ICD-10-coded discharge diagnoses revealed 83 cases. Positive and negative percent agreement between ChAT and retrospective analysis were 58.7% (95%CI =52.8-64.4%) and 97.9% (95%CI =97.1-98.5%; p<0.0001). Cohen’s kappa coefficients comparing ChAT to retrospective analysis were 0.92 (ASM), 0.56 (ENC), 0.3 (MYE) and 0.65 (ADEM), respectively.

Conclusions: Retrospective chart review and ICD-10-based screenings may be missing cases due to insufficient data. Taking case definitions to the POC will provide sufficient data for immediate and well-standardized case classifications eliminating observer bias and inter-rater variability.