To determine the value of a detailed family history for the assessment
of the risk of recurrence of febrile seizures, 115 children who visit
ed the emergency room of an academic children's hospital were studied
prospectively. The recurrence risk of febrile seizures was analysed in
relation to the child's family history and the proportion of relative
s affected by febrile seizures using Kaplan-Meier estimates and Cox pr
oportional hazard models. A first degree family history positive for f
ebrile seizures (parents or siblings affected by febrile seizures) inc
reased a child's two year recurrence risk from 27 to 52%. No significa
nt increase of recurrence risk for febrile seizures was found in child
ren with second degree relatives (grandparents and uncles/aunts) or co
usins only affected by febrile seizures. Recurrence risk was significa
ntly correlated with the proportion of first degree relatives affected
by febrile seizures: risks were 27, 40, and 83% in children whose pro
portion was 0, 0-0.5, and greater than or equal to 0.5 respectively. A
nalysis of the recurrence risk in relation to a weighted proportion, a
djusted for the attained age and sex of first degree relatives, showed
similar results. It is concluded that the application of the proporti
on of first degree relatives affected by febrile seizures generates a
more differentiated assessment of the recurrence risk of febrile seizu
res.