Mortality in autism: a prospective longitudinal community-based study.

The purposes of the present study were to establish the mortality rate in a representative group of individuals (n = 120) born in the years 1962-1984, diagnosed with autism/atypical autism in childhood and followed up at young adult age (>or=18 years of age), and examine the risk factors and causes of death. The study group, which constituted a total population sample of children with these diagnoses, were followed up in Swedish registers. Nine (7.5%) of the 120 individuals with autism had died at the time of follow-up, a rate 5.6 times higher than expected. The mortality rate was significantly higher among the females. Associated medical disorders (including epilepsy with cognitive impairment) and accidents accounted for most of the deaths, and it was not possible to determine whether autism "per se" actually carries an increased mortality risk.