RESUMO
Although sleep disturbances are commonly reported among children exposed to violence, objective evidence of such disturbances is rare. This longitudinal, home-based study assessed the effects of a known community- or family-violence incident on both actigraphy-derived and subjectively reported sleep outcomes of an ethnically mixed, urban sample of children aged 8-16 years. We hypothesized that increased event severity (child physical assault, witnessed homicide) would be associated with lower sleep duration and poorer sleep quality both at baseline and at 3-month follow-up. Covariate-adjusted analyses based on a generalized estimating equations approach showed that children physically assaulted during the event showed lower sleep duration and sleep efficiency and greater wake after sleep onset than those not physically assaulted. Physically assaulted children were more likely to have a later bedtime than non-assaulted children, but this difference decreased at 3 months. Children witnessing a homicide showed greater wake after sleep onset at baseline and reported greater sleep problems than those witnessing a non-homicide event, but these differences decreased at 3 months. They were also somewhat more likely to have greater nightly variation in sleep duration. Collectively, results suggest that violence exposure influences children's sleep, but that specific dimensions of sleep may exhibit different susceptibility to different characteristics of violence, especially over time.