Increased self-reported and objectively assessed physical activity predict sleep quality among adolescents

Highlights

• Increased subjective and objective PA predict better subjective and objective sleep.

• Self-reported PA levels are a better predictor for self-reported sleep.

• Objective PA levels have a higher influence on objective sleep.

• Vigorous PA levels tend to be a better predictor for good sleep than moderate PA.

• Males reported fewer sleep complaints than females.

Abstract

Both scientists and the general public assume that physical activity (PA) is an effective, non-pharmacological approach to improvement in sleep quality. However, objective and reliable data on this relationship are scarce, particularly for adolescents. Therefore, the aims of the present study were to test the relationship by assessing both PA and sleep subjectively and objectively. A total of 56 adolescent vocational school students (Mean age = 17.98, SD = 1.36; 28 males, 28 females) participated in the study. Sleep and PA were subjectively assessed via questionnaires. Accelerometers objectively assessed PA, while sleep-EEG devices objectively assessed sleep. The data supported our prediction that adolescents with high PA levels would have longer TST, fewer wakening at night (WASO), fewer symptoms of insomnia, and higher sleep quality. However, gender influenced this pattern of results in that significant findings were only found between high self-reported PA levels and shorter perceived sleep onset latency (SOL). Though self-reported PA levels were a better predictor of good sleep than objectively assessed PA levels, gender was associated with sleep complaints; females reported more sleep complaints. Results indicate that among a non-clinical sample of adolescents increased PA is favorably associated with restoring sleep. Therefore, PA seems beneficial not only for physical and mental health, but also for sleep restoration.

Introduction

Adolescence is defined as a vulnerable period characterized by several changes in physiological, psychiatric, socio-cultural and psychological factors that affect health and behavior, including sleep quality and quantity [1], [2], [3], [4]. For vocational students, additional responsibilities arise in the transition to a vocational school and balancing academic and job requirements [5]. Neurobiological sleep-regulatory mechanisms also play a role; there are underlying changes in both circadian and homeostatic components of sleep which influence the sleep–wake cycle, as well as sleep timing, duration, and architecture [2], [6], [7]. The most striking change in sleep architecture is the decrease in slow wave sleep (SWS) [8], [9], [10].

Whereas poor sleep is associated with maladaptive functioning and a number of psychiatric and physical diseases [1], there is a dearth of knowledge on the sleep quality of adolescents attending vocational education and training. Pharmacological and psychotherapeutic treatments of insomnia have proved successful but entail negative side effects or require professional intervention [11], [12]. In contrast, physical activity (PA) is an alternative favored by many sleep experts [13], [14], [15]. There are two main hypotheses about the mechanism through which PA positively influences sleep. First, PA produces physiological changes favorable to sleep regulation. For example, Dworak et al. [16] were able to show that among young adolescents (12.6 years old) high-intensity exercise resulted in a significant rise in the proportion of SWS, less sleep in stage 2, greater sleep efficiency (SE %), and shorter sleep onset latency (SOL). Kalak et al. [17] assessed the sleep of 19 year olds via sleep-EEGs after three consecutive weeks of morning running, compared to a control condition. Objective sleep improved (SWS increased; SOL decreased) in the running group compared to the control group. The second hypothesis is that the sleep-promoting effects of PA are mediated via psychological functioning [18], [19], [20], by decreased symptoms of depression [21], [22], anxiety [23], [24], and stress [25], [26]. Thus, in previous studies of adolescents, we have found that self-reported exercise was positively related to adolescents' sleep and psychological functioning: Adolescent exercisers reported better sleep patterns, while males with low exercise levels were at increased risk for poor sleep and psychological functioning [12]. In a second study [22], objectively assessed but not self-reported sleep was significantly related to exercise level, higher positive coping and curiosity scores, and lower depressive symptom scores.

However, many adolescents reduce their PA levels during this period of life [27], [28], [29]. Vocational students in particular report lower levels of exercise than peers attending regular high schools [30], [31]. However, the relationship between PA and the functional importance of sleep among vocational students has received little attention thus far. This is surprising, given that 74% of all secondary school diplomas in Switzerland are granted to students with vocational education and training (VET) [32]. In Switzerland, VET serves as the primary gateway to numerous occupations. Most students start at the age of 16 or 17. Swiss VET combines apprenticeships in a company and vocational training in a school; students work three to four days a week in a company and attend a vocational school on one to two days a week. While apprentices receive a limited salary for their work in the company, they often have equal workloads in terms of working hours as well as responsibilities similar to adult employees. Overall, compared to academic high school students, vocational students face a dual burden of academic and work requirements: Depending on vocation, students have to manage this challenging condition for three to four years. With respect to the relationship between sleep and work, in one study the number of working hours was negatively correlated with hours of sleep, and decreased sleep quality was correlated with increased strain, depression and anxiety [33]. Additionally, employees who report poor sleep or symptoms of insomnia tend to consume more medical resources [34] and at work they exhibit lower self-esteem [35], lower work efficiency and higher absenteeism rates [36], [37]. Collectively, these findings suggest that adolescents attending vocational schools are at greater risk for developing sleep disturbances or sleep-related psychological dysfunction.

Despite the evidence from two studies [16], [17] that PA has beneficial effects on sleep, findings from cross-sectional studies among adolescents remain inconclusive [15], [20], [38], [39], [40], [41]. Whereas epidemiological data from self-reports have consistently revealed a positive relationship between PA and sleep [42], [43], [44], findings are difficult to interpret and compare given the vast array of different measurement methods in this field of research. To the best of our knowledge, no previous study has assessed PA and sleep via both subjective and objective methods in the adolescent population. Therefore, the present study aims to establish findings for the adolescent population, and overcome methodological limitations, by assessing PA and sleep both subjectively and objectively.

Five hypotheses were tested (see Fig. 1). (1) PA levels as assessed subjectively by questionnaire and objectively by accelerometer will correlate positively [45], [46]. (2) Sleep quality as assessed by questionnaires and by sleep-EEG device will positively intercorrelate [47]. Subjective sleep quality will be positively associated with both (3a) subjective PA level [12], [22], and (3b) objective PA level [48]. We further expected that more favorable objectively assessed sleep will be related to higher (4a) subjective, and (4b) objective PA levels [49], [50]. (5) Following cognitive models of insomnia [51], our final hypothesis was that subjective PA levels would better predict sleep quality than objective PA levels.