Delayed Sleep Phase Disorder. Prevalence, sleep, circadian rhythm and treatment

Abstract

Adolescence is often characterized by delayed and irregular sleep patterns, with potential negative consequences in terms of school performance and daytime functioning. At the most extreme, a stably delayed sleep phase may reflect a circadian rhythm sleep disorder of the delayed sleep phase type (delayed sleep phase disorder, DSPD). DSPD is assumed to be common amongst adolescents and young adults, but little is known about its prevalence and aetiology, and no guidelines exist with respect to treatment. The aims of the theses were 1) to address the prevalence and correlates of a delayed sleep phase (DSP) in a large and representative sample of Norwegian high school students, 2) to investigate objective measures of sleep, circadian rhythm and phase angle relationship between the underlying circadian rhythm and sleep timing in adolescents and young adults with DSPD when allowed to sleep on a self-chosen schedule and 3) to investigate short- and long-term effects on sleep of bright light and melatonin administered alongside gradual advancement of rise time in adolescents and young adults with DSPD. The aims were addressed using two different study designs. Study 1 was an internet-based, cross sectional survey conducted on 1285 Norwegian high school students. The survey included items on background, demography, sleep and daytime functioning and the validated questionnaires the Hospital Anxiety and Depression Scale (HADS) and the Alcohol Use Disorder Identification Test (AUDIT). DSP was operationalized as difficulties falling asleep before 2 a.m. at least 3 nights per week together with much/very much difficulties awakening in the morning. The results showed a prevalence of DSP of 8.4% (paper 1). Students with DSP slept less during weekdays and had more often weekend rebound sleep than students without DSP. DSP was associated with negative outcomes in terms of poorer school grades, more smoking, more use of alcohol and increased symptoms of anxiety and depression. Study 2 combined a case control study and a clinical trial in a comprehensive design into which 40 adolescents and young adults with DSPD and 21 healthy controls were recruited to participate. In the case control study, sleep on a self-chosen schedule was assessed in patients with DSPD and controls by means of polysomnography (PSG). Circadian rhythm was assessed by measuring salivary dim light melatonin onset (DLMO). Results showed delayed timing of sleep and delayed DLMO in the patients with DSPD compared to the healthy controls (paper 2). Sleep, however, appeared to occur at a similar phase angle, and once sleep was initiated no differences in sleep duration or sleep architecture were observed between the groups. In the clinical trial, the DSPD patients were randomized to receive treatment for two weeks in one of 4 treatment conditions: dim light and placebo capsules, bright light and placebo capsules, dim light and melatonin capsules and bright light and melatonin capsules. In a follow-up study, participants were re-randomized to either receive treatment with the combination of bright light and melatonin or no treatment in an open label trial for approximately three months. Light and melatonin were always administered alongside gradual advancement of rise times. Sleep was assessed by sleep diaries, actigraphy recordings, the Pittsburgh Sleep Quality Index (PSQI) and the Bergen Insomnia Scale (BIS). Circadian rhythm was assessed by measuring DLMO. Results showed that the timing of sleep and DLMO were advanced in all groups after short-term treatment, with no additional effect of bright light and melatonin (paper 3). Termination of treatment produced a relapse to baseline levels whereas continued treatment using bright light and melatonin together with adjunct behavioural instructions allowed maintenance of the sleep rhythm. In conclusion, we found in study 1 a high prevalence of DSP in our sample of Norwegian high school students (paper 1). DSP was associated with lower school grades, more smoking, more alcohol use and higher scores on anxiety and depression. In study 2 we found that patients with DSPD had delayed timing of sleep and DLMO, whereas sleep architecture and phase angle relationship appeared to be normal when patients were allowed to sleep according to a self-chosen sleep schedule (paper 2). Short-term treatment of patients with DSPD involving bright light and melatonin alongside gradual advancement of rise time produced a phase advance irrespective of treatment condition (paper 3). Long-term treatment with bright light and melatonin alongside gradual advancement of rise time allowed maintenance of the sleep rhythm whereas termination of treatment caused relapse into delayed sleep times.