Sleep in pediatric oncology: measuring, identifying and intervening

The main aim of this thesis was to examine sleep in adolescents and young adults during the first 10 years after diagnosis and treatment for childhood cancer: what has been the impact of being critically ill, undergoing extensive treatment and the type of cancer on sleep and how do we measure, identify and intervene in sleep problems and disorders in this patient group. In chapter 2, we evaluated the psychometric properties of the Dutch versions of the Pediatric PROMIS item banks for sleep disturbance (SD), sleep-related impairment (SRI) and fatigue in Dutch children and adolescents from the general population, since they have not been validated yet in this translation. The SD item bank did not meet all assumptions for item response theory (IRT) modeling (not unidimensional). The SRI and fatigue item banks met the assumptions and measured reliably, in which computerized adaptive testing (CAT) outperformed the full and short forms. Overall, this study shows that the Dutch pediatric PROMIS item banks for SRI and Fatigue are efficient and reliable in measuring sleep-related impairment and fatigue in Dutch children and adolescents, but further research is needed for the SD item bank. In chapter 3, we systematically reviewed the literature on risk factors associated with steroid-induced adverse psychological reactions (APRs) and sleep problems in children diagnosed with Acute Lymphoblastic Leukemia (ALL). The findings reveal that there were few high-quality prospective studies and sample sizes were small, necessitating caution in interpreting the results. We therefore formulated recommendations regarding risk factors associated with steroid-induced APRs and sleep problems and in addition recommend to invest in more high-quality studies to draw more substantiated conclusions. We recommend using standardized prospective registration of steroid-induced APRs and sleep problems, alongside the identification of risk factors in forthcoming studies involving children with ALL. In chapter 4, we assessed the prevalence of six sleep disorders based on the International Classification of Sleep Disorders, second edition (ICSD-2) in 576 adolescent and young adult childhood cancer patients in follow-up after treatment (aged 12-26; at least six months post-treatment). We found insomnia (9.6%) to be most prevalent disorder followed by circadian rhythm sleep disorder (CRSD; 8.1%), restless legs syndrome (7.6%), parasomnia (3.5%), hypersomnia (3.5%), and sleep-related breathing disorders (1.8%). A considerable proportion of participants expressed a need for sleep treatment (ranging from 42% to 72% depending on the sleep disorder), but only a small percentage received such treatment (0% to 5.6%). These findings emphasize the treatment access gap for sleep treatment among childhood cancer patients during their follow-up care. In chapter 5, we assessed the prevalence of insomnia-only, daytime fatigue-only, the coexistence of insomnia and daytime fatigue symptoms, and associated risk factors among 565 childhood cancer patients (aged 12-26; at least six months post-treatment). In total, 61.8% of the patients reported no insomnia or daytime fatigue symptoms, while 38.2% experienced symptoms. Among those with symptoms, 48.1% reported both insomnia and daytime fatigue, 34.7% had insomnia only, and 17.1% experienced daytime fatigue only. Our results indicate that approximately 40% of patients suffer from either insomnia or fatigue, with half of them experiencing both simultaneously. In chapter 6, we describe the study protocol of our randomized-controlled trial evaluating the effectiveness of online cognitive behavioral therapy for insomnia (online CBT-i) i-Sleep youth on insomnia symptoms in adolescents and young adults who finished childhood cancer treatment. We hypothesize that the intervention will improve sleep efficiency (actigraphy) and insomnia severity (self-report) on the short-term compared to the control group, which is maintained on the long-term up to a year after the intervention. We also aim to study the acceptability of the online intervention in the adolescent target group.