Inconsistency between prospectively and retrospectively reported febrile seizures

Area of Science:

• Pediatrics
• Epidemiology
• Neurology

Background:

• Febrile seizures (FS) are common in childhood.
• Accurate incidence data are crucial for understanding FS and its long-term implications.
• Prospective and retrospective methods may yield different prevalence estimates.

Purpose of the Study:

• To compare prospective incidence data with retrospective parental recall of febrile seizures (FS).
• To evaluate the accuracy of parental recall in identifying FS cases up to age 12.
• To assess biases in retrospective data collection for FS.

Main Methods:

• A cohort of 807 children had prospective FS data collected up to age 5.
• At age 12, parental recall of FS was assessed using the same questionnaire.
• Statistical analysis compared prospective incidence with retrospective prevalence, calculating sensitivity, specificity, and predictive values.

Main Results:

• Prospective cumulative incidence was 7.1%, while retrospective prevalence was 5.6%.
• Retrospective data showed under-reporting (19 children) and over-reporting (8 children).
• Retrospective approach had 65% sensitivity, 99% specificity, 82% positive predictive value, and 97% negative predictive value.

Conclusions:

• Retrospective parental recall underestimates the frequency of febrile seizures (FS) despite high specificity.
• Prospective data may miss some FS cases, as suggested by recall data.
• Biases in recall methods must be considered when using FS data for predicting future health effects.