Background: Arousal parasomnias (night terrors, sleepwalking, and confusional arousals) have seldom been investigated in the adult general population. Clinical studies of parasomnias, however, show that these disorders may be indicators of underlying mental disorders and may have serious consequences.
Method: A representative sample of the United Kingdom population (N = 4972) was interviewed by telephone with the Sleep-EVAL system.
Results: Night terrors were reported by 2.2% (95% CI = 1.8% to 2.6%) of the sample, sleepwalking by 2.0% (1.6% to 2.4%), and confusional arousals by 4.2% (3.6% to 4.8%). The rate of these 3 parasomnias decreased significantly with age, but no gender difference was observed. Multivariate models identified the following independent factors as associated with confusional arousals (odds ratio [OR]): age of 15-24 years (OR = 4.1), shift work (OR = 2.1), hypnagogic hallucinations (OR = 3.3), deep sleep (OR = 1.6), daytime sleepiness (OR = 1.9), sleep talking (OR = 1.7), daily smoking (OR = 1.7), adjustment disorder (OR = 3.1), and bipolar disorder (OR = 13.0). Factors associated with night terrors were subjective sense of choking or blocked breathing at night (OR = 5.1), obstructive sleep apnea syndrome (OR = 4.1), alcohol consumption at bedtime (OR = 3.9), violent or injury-causing behaviors during sleep (OR = 3.2), hypnagogic hallucinations (OR = 2.2), and nightmares at least 1 night per month (OR = 4.0). Factors associated with sleepwalking were age of 15-24 years (OR = 5.2), subjective sense of choking or blocked breathing at night (OR = 5.1), sleep talking (OR = 5.0), and a road accident in the past year (OR = 3.9) after controlling for possible effects of sleep deprivation, life stress, and mental and sleep disorders.
Conclusion: Arousal parasomnias, especially night terrors and confusional arousals, are often the expression of a mental disorder. Other life or medical conditions, such as shift work or excessive need of sleep for confusional arousals and stressful events for sleepwalking, may also trigger parasomnias. Prevalence rates are based on self-reported data and, consequently, are likely underestimated.