Sleep is a critical developmental need for healthy children and adolescents, and when disturbed, has been associated with problems in school attendance, performance, and mood, and impairment of daily functioning1
. In a recent study, adolescents with chronic pain and disturbed sleep were found to have reduced physical, social and emotional well-being compared to adolescents with chronic pain but fewer sleep disturbances2
Epidemiological studies indicate that about 50% of children and adolescents with chronic pain have sleep problems3
. Sleep disturbances have been documented in children with juvenile rheumatoid arthritis (JRA)4
, sickle cell disease5; 6
, and complex regional pain syndrome9
. Most commonly, these children complain of difficulties falling asleep, frequent night and early morning awakening, and excessive daytime sleepiness2
. On polysomnography (PSG), children with JRA and children with fibromyalgia have also evidenced sleep fragmentation10–12
Previous research has relied primarily on subjective methods to assess children’s sleep13
. With the exception of a few studies in pediatric rheumatology, there has been no other application of PSG to the assessment of sleep in children with chronic pain. Two published studies14; 15
have utilized another objective methodology for sleep assessment, actigraphy, which uses activity or motion counts as a proxy measure of sleep. It provides unobtrusive measurement of activity over extended periods, demonstrating excellent validity for total sleep duration compared with concomitant PSG, with agreement of up to 95%16
Bruni and colleagues14
compared sleep of 17 healthy controls and 18 children with migraines (ages 8 to 12 years) using 14 days of actigraphy. Although nocturnal motor activity was reduced preceding migraines, sleep was similar between the groups during the interictal period. Haim and colleagues15
compared sleep patterns in 25 children with recurrent abdominal pain and 15 healthy controls (ages 10 to 17 years) using actigraphy over 7 days. Although children with recurrent abdominal pain complained about disturbed sleep on self-report, actigraphy measures of sleep were similar between the groups. These findings highlight the importance of combining subjective and objective measures of sleep, which may provide different information about sleep in children and adolescents with chronic pain.
Although these two small studies are strengthened by the use of an objective sleep assessment, the sample and methods were limited in several ways. First, the samples were heterogeneous in regards to the frequency and intensity of experienced pain and may not accurately reflect the sleep experience of children with more severe chronic pain symptoms. Second, the methods were limited entirely to description of actigraphy findings, and the potential relationship between subjective assessment of sleep using well validated measures and objective assessment of sleep was not possible within the study design. Third, predictors of sleep disturbances were not described and thus, the identification of pain characteristics or psychological factors that may be related to poor sleep could not be appreciated.
In fact, few studies have identified behavioral or psychological correlates of sleep disturbances in children or adolescents with chronic pain. The pain itself has been found to be correlated with reports of sleep disturbances in children with migraine headaches8
and in children with polyarticular JRA10
. Pre-sleep arousal, which refers to worry at bedtime, has been studied in adults with chronic pain, with one study reporting the level of cognitive arousal experienced at bedtime, rather than pain severity, to be the primary predictor of sleep quality17
. Another potential correlate of disturbed sleep is psychiatric comorbidity. In an epidemiological study of the prevalence of insomnia in adolescents, 52.8% of those with insomnia had a comorbid psychiatric disorder, most often a mood disorder18
. Depressive symptoms have been found to be an important correlate of sleep disturbances in adolescents2
and adults19; 20
with chronic pain, predicting severity of sleep disturbances after controlling for demographic and pain-related variables. Anxiety, on the other hand, has not been found to be correlated with sleep quality19
To identify potential correlates of sleep, we used a framework described by Lewin and Dahl21
, examining the links between the regulation of sleep and pediatric pain. The primary tenet of the framework is that there are bi-directional effects between pain and sleep. Pain can directly affect sleep by prolonging sleep onset and interfering with the depth and continuity of sleep states, and the psychological and physiological sequelae of insufficient sleep (e.g., worry, negative thoughts, decrements in behavioral control) may have deleterious effects on pain management. Transient sleep problems may become chronic, independent of pain severity, due to pre-sleep arousal/worry and negative mood. Smith and colleagues22
describe a similar hypothesis in adults with chronic pain, that pre-sleep cognitive arousal may lead to insomnia secondary to chronic pain. For example, strong and problematic associations may develop between bedtime fears of separation and worry about physical sensations of pain leading to increased vigilance at bedtime that is incompatible with sleep. Clinical research supports the link between pain sensations and disturbed sleep4; 7; 23
, and experimental research supports the sleep-pain connection demonstrating that sleep deprivation produces hyperalgesia (i.e., enhanced pain sensitivity) in animal models24
and in otherwise healthy adults25; 26
The goal of our study was to extend previous research by conducting a comprehensive assessment of sleep using objective and subjective assessments in a sample of adolescents with chronic pain. We chose to focus on an adolescent sample because adolescents have been described as having a higher risk for sleep disturbances compared to younger children1
. Moreover, adolescents with similar frequency and intensity of chronic pain were recruited in order to obtain estimates of sleep within a treatment-seeking population that reflects those adolescents with more severe pain complaints. Our assessment plan was comprehensive, allowing description of objective sleep, sleep quality, sleep hygiene, and insomnia symptoms, which have not yet been described in children or adolescents with chronic pain.
Thus although our sample size for this study was small, the measurement plan would allow for estimation of effect sizes and identification of possible predictor variables that may inform future research in this area. We hypothesized that adolescents with chronic pain would demonstrate on actigraphy reduced total sleep time, more time awake after initial sleep onset, more wake bouts, and reduced sleep efficiency; and on subjective measures poorer sleep hygiene and sleep quality, and more insomnia symptoms, compared to an age and sex matched group of healthy peers. Increased depressive symptoms and higher levels of pre-sleep arousal/worry were hypothesized to be associated with poor sleep. Last, we anticipated moderate associations between subjective and objective sleep measures.