Heritability and mortality risk of insomnia-related symptoms: a genetic epidemiologic study in a population-based twin cohort

Abstract

Study objectives: Our aim was to estimate heritability in phenotypic insomnia and the association between insomnia and mortality.

Design: Representative follow-up study.

Participants: 1990 survey of the Finnish Twin Cohort (N = 12502 adults; 1554 monozygotic and 2991 dizygotic twin pairs).

Measurements: Current insomnia-related symptoms (insomnia in general, difficulty in initiating sleep, sleep latency, nocturnal awakening, early morning awakening, and non-restorative sleep assessed in the morning and during the day) were asked. Latent class analysis was used to classify subjects into different sleep quality classes. Quantitative genetic modelling was used to estimate heritability. Mortality data was obtained from national registers until end of April 2009.

Results: The heritability estimates of each symptom were similar in both genders varying from 34% (early morning awakening) to 45% (nocturnal awakening). The most parsimonious latent class analysis produced 3 classes: good sleepers (48%), average sleepers (up to weekly symptoms, 40%), and poor sleepers (symptoms daily or almost daily, 12%). The heritability estimate for the cluster was 46% (95% confidence interval 41% to 50%). In a model adjusted for smoking, BMI, and depressive symptoms, the all-cause mortality of poor sleepers was elevated (excess mortality 55% in men and 51% in women). Further adjustment for sleep length, use of sleep promoting medications, and sleep apnea-related symptoms did not change the results.

Conclusions: Insomnia-related symptoms were common in both genders. The symptoms and their clusters showed moderate heritability estimates. A significant association was found between poor sleep and risk of mortality, especially in those with somatic disease.

Keywords: Insomnia; heritability; mortality; population; twins.

Figure 1

Proportions (95% confidence intervals; y-axis range 0–1) of insomnia-related symptoms of the three sleeper categories in cluster analysis. This figure shows the proportions of those with insomnia symptoms in each of the three clusters when the response alternatives are dichotomized according to the mean responses of good sleepers. Low values indicate absence or infrequent occurrence of each symptom (in sleep latency short time) and high values increasing frequency of the symptom (long sleep latency) N = 12,126. InsSymptom, insomnia-symptom; DIS, difficulty in initiating sleep; SleepLat, sleep latency; NoctAwak, nocturnal awakenings; EMA, early morning awakening; NRSMorn, non-restorative sleep (morning); NRSDay, non-restorative sleep (day).