Long-term survival of people with unprovoked seizures: a population-based study

Epilepsia. 1998 Jan;39(1):89-92. doi: 10.1111/j.1528-1157.1998.tb01279.x.


Purpose: Few population-based studies of long-term survival in people with seizures or epilepsy have been made.

Methods: Between January 1, 1960 and December 31, 1964, we identified 224 incidence cases of unprovoked seizures in Iceland and determined survivorship status and date of death for the cases as of January 1, 1996. We compared survivorship with that expected based on data from age-/sex-specific life tables from the country for 1961-1990 and calculated the standardized mortality ratio (SMR).

Results: By 30 years after diagnosis, there were 45 deaths among patients with unprovoked seizures as compared with an expected 28 deaths [standardized mortality ratio (SMR) 1.6; 95% confidence interval (CI) 1.2-2.2]. Patients with unprovoked seizures of unknown etiology did not have a significant increase in mortality overall (SMR 1.3, 95% CI 0.8-1.9) or in any time interval. For patients with remote symptomatic unprovoked seizures, mortality was increased (SMR 2.3, 95% CI 1.4-3.5). This increase was attributable to excess mortality for the first 15 years after diagnosis (SMR 4.1, 95% CI 2.4-6.6), and SMR was not different after that time.

Conclusions: Survivorship was decreased for the population of patients with unprovoked seizures. The increased mortality was primarily due to excess mortality in patients with remote symptomatic seizures, occurring in the first 15 years after diagnosis. Overall mortality for idiopathic unprovoked seizures was not significantly increased.

Publication types

  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Adult
  • Age Factors
  • Confidence Intervals
  • Epilepsy / epidemiology
  • Epilepsy / mortality
  • Female
  • Follow-Up Studies
  • Humans
  • Iceland / epidemiology
  • Incidence
  • Life Tables
  • Male
  • Middle Aged
  • Seizures / epidemiology*
  • Seizures / mortality*
  • Sex Factors
  • Survival Analysis