Assessments

Height and weight were measured onsite at baseline (T0) and upon discharge from treatment (T1) using standardized techniques. Participants' height and weight were converted to BMI, which were applied to the appropriate sex-specific BMI reference charts to determine BMI percentile and z-score at each time point (28). Percentage body fat was determined using a Lipometer (Möller Messtechnik, Graz, Austria) at 15 prespecified anatomical points, providing noninvasive, reliable, and valid measures (29, 30).

Standardized and age-validated questionnaires were completed by the children and parents in both a cross-sectional (completed by OBE and NW and their parents at T0) and longitudinal design (completed at T0 and T1 by OBE). These questionnaires covered aspects of sleep duration and sleep behaviors, and psychological factors such as anxiety, stress and depression, self-esteem, and quality of life. The specific details of the questionnaires are summarized in Table 1.

Questionnaires to assess sleep and psychological factors.

The specific details of the questionnaires are summarized below.

Sleep Self-Report. This established, validated self-report questionnaire is a screening tool for assessing overall sleep problems. The development of the SSR was based on the Children's Sleep Habits Questionnaire (CSHQ). The SSR consists of 26 items with three answer options ranging from “common (5–7×/week)” to “uncommon (0–1×/week)”. The items are about sleep duration, bedtime resistance, night awakenings, and daytime sleepiness. A score of 23 items is calculated. For the German version, the internal consistency was acceptable with a Chronbach's α = 0.73 and a retest reliability of r = 0.58 (31). In this study population, Chronbach's α = 0.75.

Pediatric Daytime Sleepiness Scale. This validated questionnaire assesses daytime sleepiness and consists of eight items (32). The five answer options range from “never” to “always.” Internal consistency is good with Chronbach's α = 0.8. In this study population, Chronbach's α = 0.75.

Inventory for Recording the Quality of life in Children and Adolescents Version for Children. This validated tool assesses quality of life. It consists of 10 items and covers the thematic areas school, family, social contacts among peers, interests and free-time activities, physical health, and psychological health. A sum score is calculated for overall quality of life. The five answer options range from “very good” to “very bad.” Chronbach's α was between 0.55 and 0.76 and the test–retest reliability between r = 0.6 and 0.8 (33). In this study population, Chronbach's α = 0.72.

Depression Inventory for Children and Adolescents. This validated questionnaire is used to assess the severity of depressive symptoms according to Diagnostic and Statistical Manual of Mental Disorders (DSM−5) criteria. It consists of 26 items and three predefined answer alternatives, allowing for calculation of a sum score. Internal consistency of the German version was reported with Chronbach's α = 0.87 to 0.92 (36). In this study population, Chronbach's α = 0.88.

Somatization Inventory for Children and Adolescents. This validated questionnaire provides information about intensity, frequency, and types of somatic complaints in the areas pseudoneurological symptoms, cardiovascular symptoms, gastrointestinal symptoms, and pain (34). It consists of 35 items and has five answer options ranging from “not at all” to “much”. The internal consistency was high with Chronbach's α = 0.93 (43). In this study population, Chronbach's α = 0.90.

State-Trait Anxiety Inventory for Children. The established, validated questionnaire measures state anxiety (anxiety about an event) and trait anxiety (anxiety as a personal characteristic). In this study, trait anxiety was assessed, consisting of 20 items and allowing for calculation of a final sum score. The three answer options range from “almost never” to “often.” Internal consistency of the German version was reported with Chronbach's α = 0.81 and test–retest reliability with r = 0.64 (35). In this study population, Chronbach's α = 0.90.

Stress and Stress Coping Questionnaire for Children and Adolescents. This validated tool assesses the degree of self-reported stress and coping with stress. It consists of 84 items covering the subscales stress vulnerability, stress cooping, and stress symptoms. The five answer options range from “never” to “always.” Values for internal consistency (Chronbach's α) are between 0.67 and 0.89 and test–retest reliability between r = 0.56 and 0.82 (37). In this study population Chronbach's α ranges between the subscales from 0.69 to 0.87.

Self-esteem Scale by Rosenberg. This validated questionnaire assesses self-esteem using 10 items with subsequent calculation of a sum score. The four answer options range from “strongly disagree” to “strongly agree.” In the German version, the internal consistency Chronbach's α ranges between 0.72 and 0.85 (38). In this study population, Chronbach's α = 0.90.

CSHQ. This questionnaire is an established, validated parents' questionnaire for assessing sleep problems of their children. It consists of 33 items with three answer options ranging from “common (5–7×/week)” to “uncommon (0–1×/week).” A total score of 33 items and 8 subscores comprising the areas bedtime resistance, sleep-onset delay, sleep duration, sleep anxiety, night wakings, parasomnias, sleep-disordered breathing, and daytime sleepiness are calculated. In the German version, the internal consistency Chronbach's α = 0.68, and test–retest reliability of r = 0.76 (39, 44). In this study population, Chronbach's α = 0.78.

Epworth Sleepiness Scale. This validated parents' questionnaire is used to assess daytime sleepiness in their children (45). The development was based on the ESS version for adults with Chronbach's α = 0.88 and test–retest reliability of r = 0.82 (41, 46). It consists of eight items and four answer options ranging from “would never doze” to “dozing high probability”. Finally, a sum score is calculated. In this study population, Chronbach's α = 0.73.

Pediatric Sleep Habits Questionnaire. This validated parents' questionnaire is used to assess sleep-disordered breathing, snoring, sleepiness, and behavioral problems of their children. It consists of 22 items and the three subscales snoring, sleepiness, and behavior. In addition, a total sum score is calculated. The answer options are “yes”, “no”, and “don't know”. For items 17 to 22, the four answer options range from “does not apply” to “meets most of time”. The internal consistency (Chronbach's α = 0.89) and test–retest reliability were considered as good (40). In the German study population, Chronbach's α = 0.72 (47). In this study, Chronbach's α = 0.73.

Inventory for Recording the Quality of life in Children and Adolescents Version for Parents. This validated parent's questionnaire assesses quality of life of their children and is a similar construct to the children/adolescents' version as described above but consists of 11 items (33). In this study, Chronbach's α = 0.62.

Strength and Difficulties Questionnaire. This validated parent's questionnaire is used to measure pro-social behavior, emotional symptoms, conduct problems, hyperactivity, and peer problems of their children. It consists of 25 items which allow calculation of the five subscales emotional symptoms, conduct problems, hyperactivity–inattention, peer relationship problems, and prosocial behaviors. A sum problem score is calculated including all items. The three answer options range from “does not apply” to “clearly applicable.” Internal consistence (Chronbach's α = 0.82) was considered as good for the German version (42, 48). In this study, Chronbach's α = 0.73.

In addition to the questionnaires, sleep duration was self-reported by the children at T0 with the question “How many hours per night do you sleep on average?” and at T1 with the question “How many hours per night have you been sleeping on average while here in Wangen?” The answers were recorded using a five-point scale (5 = 9–11 h, 4 = 8–9 h, 3 = 7–8 h, 2 = 5–7 h, 1 = <5 h). Sleep duration was also reported by the parents at baseline as a component of the CSHQ and was recorded to one decimal place (e.g., 8.5 h).

Sleep quality was self-reported by the children at T0 with the question “How well do you sleep at home on average?” and at T1 with the question “How well have you slept on average during your inpatient stay?” The answers were collected using a visual 10-point scale (0 = very good to 10 = very bad).