Agreement between medical record and parent report for evaluation of childhood febrile seizures

Background: The monitoring of vaccine safety is critical to maintaining the public acceptance of vaccines required to ensure their continued success. Methods used to assess adverse events following immunization (AEFI) must accurately reflect their occurrence. Assessment of AEFI is often done via medical record review (MR) or via patient report (PR). However, these sources of data have not previously been compared for the analysis of AEFI. The objective of this study was to evaluate the concordance between MR and PR for young children identified as having had a febrile seizure (FS), an important AEFI, in an integrated health care system. The variables chosen for analysis were those recommended by the Brighton Collaboration Seizure Working Group for the evaluation of generalized seizure as an AEFI [1].

Methods: Parent report from phone interviews and mailed questionnaires was compared to abstracted medical records of 110 children with FS between ages 3 and 60 months. Concordance between PR and MR for characteristics and predisposing factors of FS was assessed by percent total agreement and kappa statistic.

Results: Percent total agreement between PR and MR was between 43.6 and 100% for variables studied, with 62.5% of items having >70% agreement. However, kappa was poor to fair for all measures (-0.04 to 0.33). While some variables, such as history of seizures in a sibling or parent and several seizure characteristics, were reported more often by PR, other items, such as maximum fever and several concurrent conditions, were reported more often by MR.

Conclusion: These findings demonstrate the limitations of using MR or PR alone to assess febrile seizures in children. This analysis supports the practice of collecting data from both MR and PR to most accurately portray the spectrum of predisposing factors and seizure characteristics when evaluating FS in children whenever feasible.