Physical inactivity, low physical fitness, and perceived stress during childhood and adolescence are presumed to be risk factors for various chronic conditions including cardiovascular diseases (Petersen et al., 2012; Bergh et al., 2015), depressive symptoms (Motl et al., 2004; Brunet et al., 2013), pain and subjective health complaints (Sundblad et al., 2008). In turn, physical fitness and physical activity are presumed to moderate the negative effects of stress and protect against stress-related health complaints in children and adolescents (Haugland et al., 2003; Hegberg and Tone, 2015; Gerber et al., 2017a). Gerber et al. (2017a) recently revealed that children (6- to 8-year-old) with higher physical fitness levels, in terms of cardiorespiratory fitness (measured by 20 m shuttle run test), experienced higher levels of psychological well-being relative to their less fit peers when exposed to elevated stress levels (parental questionnaire), and higher levels of physical wellbeing when stress levels were low. Furthermore, children with higher physical activity, as reported by their parents, demonstrated better overall health-related quality of life (parental questionnaire), but only if they experienced low stress (Gerber et al., 2017a). Recently, Cowley et al. (2017) confirmed an inverse relationship between total self-perceived stress and physical activity during spare time in adolescents.

Although strongly related, apparently (Caspersen et al., 1985; Tittlbach et al., 2017), physical activity and physical fitness are terms that describe different concepts, and which are defined and distinguished for health-related research: Physical activity is any bodily movement produced by skeletal muscles that results in energy expenditure. Physical fitness is a set of attributes that people have or achieve, and the health-related components of physical fitness are (a) cardiorespiratory endurance, (b) muscular endurance, (c) muscular strength, (d) body composition, and (e) flexibility (Caspersen et al., 1985). Of note, physical activity during childhood and adolescence seems only weakly related to cardiorespiratory fitness (a) in terms of peak oxygen uptake (Armstrong, 2013). According to a transactional model of stress, stress is experienced when a person perceives that the demands overload or exceed the personal and social resources the individual is able to mobilize (Lazarus and Folkman, 1984).

In particular, cardiorespiratory fitness is proposed to protect against health-threatening reactions to acute psychosocial stress (Wyss et al., 2016; Gerber et al., 2017a), and stress-related symptoms of burnout and depression (Gerber et al., 2013a,b; Elliot et al., 2015). Recently, authors have recommended vigorous physical activity in adolescents to prevent burnout symptoms in this group, based on the demonstrated negative correlations between these variables (Gerber et al., 2013b; Elliot et al., 2015). Cardiorespiratory fitness is also proposed to protect against overweight and adiposity in children, adolescents, and young adults (Gerber et al., 2017b; Hingorjo et al., 2017). Gerber et al. (2017b) found that children (7–8 years) experiencing elevated school-related stress had lower body mass index (BMI), body fat, and waist circumferences if they had high physical fitness (measured by 20 m shuttle-run test) and physical activity levels.

In general, physical activity and physical fitness are considered beneficial in adolescents, resulting in consensus statements and activity guidelines for this group (Hallal et al., 2006; Tremblay et al., 2011). Several of the health beneficial effects, with respect to various subjective health complaints including pain and psychological symptoms, apparently relate to reduced stress or stress resilience (Haugland et al., 2003; Sundblad et al., 2008; Moljord et al., 2014; Bergh et al., 2015; Hegberg and Tone, 2015; Gerber et al., 2017a). Mindfulness qualities also seem related to reduced stress in adolescents (Bluth et al., 2015; Galla, 2016). Moreover, studies have shown that physically active lifestyles are related to better mindfulness skills (Kangasniemi et al., 2014), and that mindfulness possibly mediates the beneficial (stress resilient) effects of physical fitness (Demarzo et al., 2014). Mindfulness is typically described as a form of nonjudgmental, nonreactive attention to present moment experiences, both internal and external (Brown and Ryan, 2003; Kabat-Zinn, 2003).

Hence, it might be reasonable to assume that physical fitness will relate negatively to perceived stress and pain and positively to mindfulness in adolescents. However, this is not previously investigated in Norwegian adolescents. Previous studies have demonstrated that resilience, which seems relevant for the positive effects of physical activity and physical fitness in adolescents, associates negatively with perceived stress (Montero-Marin et al., 2014) and positively with mindfulness qualities (Montero-Marin et al., 2015). Furthermore, previous studies have suggested plausible gender differences in the relationship between physical activity/physical fitness and psychological variables (Moksnes et al., 2014; Moljord et al., 2014), which are relevant to investigate further.

The main objective in this study was to evaluate how physical fitness, along with mindfulness qualities, pain, and BMI, relate to perceived stress in adolescents. Secondary objectives were to explore the relationship between physical fitness, mindfulness, and BMI more explicitly in the study sample, as well as to evaluate possible gender differences.

Table 2 summarizes the descriptive analyses of dichotomous variables, overall and separately for gender. 52.5% of the males and 45.2% of the females reported pain (Table 2). The prevalence of head pain/headache was highest among the females (Table 2). A similar portion of females and males related their pain to former injuries (25.8% of the females and 27.5% of the males) or to a known disease or disorder (9.7% of the females and 7.5% of the males). The prevalence of moderate to severe stress (PSQ ≥ 0.45) seemed slightly higher among the females, i.e., 27.4% of the females against 20% of the males reported PSQ ≥ 0.45. Descriptive analyses of the continuous variables are presented in means and standard deviation (SD) in Table 3, both overall and separately for gender. Higher scores on the physical fitness measures (except from sprint), indicate higher physical fitness. Higher scores on the PSQ, including factors of PSQ, indicate higher stress. Higher scores on the MAAS-A reflect more mindfulness-capacity with respect to the addressed mindfulness-quality. The pain variables (particularly pain intensity and pain sites) were not normally distributed; hence, pain descriptions in the subsamples of females and males reporting pain are also included in Table 3.

Spearman's rank correlation coefficient (r_s) was used to assess the associations between the outcome variables, since not all (pain) variables were normally distributed (Table 3). Tables 4, 5 summarize the results of the correlation analyses with respect to stress and physical fitness, respectively, both overall and separately for gender.

Table 4 shows significant (p < 0.001) and strong negative associations between stress and mindfulness, both overall and separately for gender. Among the females, demands appeared to be strongest related (r_s = −0.57) to mindfulness, while tension (r_s = −0.51) appeared strongest related to mindfulness in males.

Joy (lack of) seemed negatively associated (p < 0.05) to physical fitness in terms of pushing a medicine ball (strength) and reduced Cooper test (endurance) in the overall sample (Table 4). Also in the overall sample, worries seemed positively related (p < 0.05) to BMI.

Among the females, tension seemed positively associated (p < 0.05) to pain intensity (measured by VAS). Among the males in the study sample, several of the associations appeared reversed: Both total stress (total PSQ) and tension seemed positively associated (p < 0.05) to physical fitness in terms of pushing a medicine ball (strength). Additionally, joy (lack of) appeared negatively associated to pain sites and pain duration, while total stress (total PSQ), worries and joy (lack of) seemed negatively associated to pain intensity (VAS) (Table 4).

Table 5 presents more of the physical fitness associations. In the overall sample, physical fitness, measured by running fast 20 m, was negatively associated (p < 0.05) with BMI. Among the females, physical fitness, measured by the reduced Cooper test (endurance), was negatively associated (p < 0.005) to number of pain sites. Among the males, physical fitness in terms of running 20 m, standing broad jump and reduced Cooper, were negatively associated (p < 0.05) with BMI. There was no associations between physical fitness and mindfulness qualities measured by the MAAS-A (Table 5).

The linear regression analysis with stress (lack of joy) as dependent variable, revealed that mindfulness measured by the MAAS-A (β = −0.40, p < 0.001) and gender (PSQ, β = −0.33, p < 0.05) explained some of the variation in the stress factor (Table 6). Physical fitness (total test score), pain duration and BMI did not explain any variation in the stress factor, according to the regression analyses. The regression model explained 26.3 % of the variation in the stress factor (lack of joy). The R² (0.263) and significant F-value (5.36, p < 0.001) supported acceptable fit of the model. The adjusted R² (0.214) indicated very sparse loss of power. A Durbin-Watson test value close to two (1.70) indicated independence of the residuals.

In this study, physical fitness along with mindfulness qualities, pain and BMI were evaluated in relation to stress in adolescents. Other physical fitness-associations were also explored in addition to gender differences. According to the main findings, physical fitness cannot explain any variation in stress among the adolescents (Table 6). Nevertheless, there were some negative associations between one stress factor (lack of joy) and components of physical fitness (strength and endurance) at a group level (Table 5), possibly influenced by conditions not measured in this study. As opposed to physical fitness, mindfulness qualities seemed to explain variation in stress among the adolescents (Table 6). Physical fitness seemed entirely unrelated to mindfulness (MAAS-A) but somewhat (negatively) related to BMI, particularly among the males (Table 5). Among the females, higher physical fitness (in terms of endurance) seemed related to reduced number of pain sites (Table 5).

Regression analyses revealed that physical fitness did not explain any of the variation in stress (lack of joy) (Table 6). Hence, the results do not support that physical fitness protects against stress in adolescents. Rather, the results might correspond with the findings by Klaperski et al. (2013, 2014). They revealed unexpected psychological and physiological stress responses to exercises, including higher mood decrease in young women (Klaperski et al., 2013). Mindfulness qualities (MAAS-A), on the contrary, seemed to explain some of the variation in stress (Table 6). This is in compliance with previous findings, showing improved mindfulness qualities to be accompanied by reduced stress in adolescents (Bluth et al., 2015; Galla, 2016). Furthermore, physical fitness seemed unrelated to mindfulness qualities (MAAS-A) among the adolescents (Table 5). In sum, this might imply that the stress-preventive effect of physical fitness rely on mediating factors/variables such as mindfulness qualities, which is also previously suggested (Demarzo et al., 2014), or variables not measured in this study such as coping strategies.

The weak association (p < 0.05) between higher physical fitness (strength and endurance) and lower stress (increased joy) at a group level (Table 5) might be influenced by conditions not measured in this study. The finding might also be comparable to the results in Moljord et al. (2011). They found that adolescents who reported physical activity-participation two to three times per week or more scored significantly lower on stress (Norwegian version of the Adolescent Stress Questionnaire) and higher on happiness (Fordyce Happiness Scale) than those who participated in physical activity 1 day per week or less. However, they found no significant difference on stress and happiness between those being physically active two or three times a week and those being active almost every day. This might correspond to the findings by Lutz et al. (2010), revealing a positive relationship between stress and exercise frequency, intensity, and duration in students exercising three times a week or more (for 20 min or more each time and for more than 6 months). Put together, previous and present findings suggest that the health beneficial effects of physical activity and physical fitness probably relay on several factors. Among others, activity dose (volume, frequency, intensity) and the factors that initiates/regulates the activity probably influence the outcome (Moljord et al., 2011, 2014; Sibley et al., 2013; Gerber et al., 2015). Sibley et al. (2013) showed that higher intrinsic motivation (more intrinsic motives and intrinsic forms of motivation) predicted healthier levels of physical fitness in students in contrast to stronger introjected regulation and appearance motives, which predicted worse fitness.

At a group level (overall sample), there seemed to be a weak association (p < 0.05) between higher stress in terms of worries and higher BMI (Table 4). This might be considered in compliance with previous studies (Sethi et al., 2011; Gerber et al., 2017b), finding positive associations between stress and BMI. However, this was not evident in the analyses separately for gender. Moreover, higher physical fitness seemed associated (p < 0.05) with lower BMI, both in the overall sample and among the males (Table 5). The negative physical fitness-BMI association might be comparable to the findings by Gerber et al. (2017b). They found that children (7–8 years) experiencing elevated school-related stress had lower body mass index, body fat, and waist circumferences if they had high physical fitness (measured by 20 m shuttle-run test) and physical activity levels. Besides, the physical fitness tasks which were associated with reduced BMI in this study, i.e., running 20 m, standing broad jump, and reduced Cooper (Table 5), commonly involve cardiorespiratory training/fitness which associate with lower BMI in adolescents and young adults (Hingorjo et al., 2017). However, the negative physical fitness-BMI association was not possible to identify among the females (Table 5).

Among the females in this study, higher stress in terms of tension seemed associated with more pain in terms of higher pain intensity (Table 4). This corresponds to previous findings in Norwegian adolescents (Østerås et al., 2015, 2016). However, this finding was not evident in the overall sample, nor among the males, possibly due to the non-normal distribution of pain variables in the sample (Table 3). Also among the females, there appeared a negative physical fitness-pain association (p < 0.05) (Table 5). More specific, increased endurance (higher scores on the reduced Cooper test) seemed related to reduced number of pain sites (Table 5). At first, this might resemble activity-induced pain modulation or hypoalgesia (Naugle et al., 2013; Jones et al., 2014; Saanijoki et al., 2017). However, the study design does not allow such an evaluation. It is also likely that the females with fewer pain sites just performed better at the test-day. There were no other associations between physical fitness and pain in this study (Table 5). This is in contrast to previous studies, suggesting pain-moderating effects of physical activity and physical fitness in children and adolescents (Haugland et al., 2003; Sundblad et al., 2008; Gerber et al., 2017a). However, these prior results might have been mediated by reduced stress (Haugland et al., 2003; Sundblad et al., 2008; Gerber et al., 2017a).

The demonstrated gender differences in this study, with respect to stress (Table 3) and its' relation to other health variables (Tables 4, 5), might be considered in compliance with previous studies. Several authors have suggested gender differences in stress appraisals, reactions and responses, both on a behavioral and a neurobiological level (Wang et al., 2007; Mayor, 2015; Marrocco and McEwen, 2016; McEwen and Milner, 2017). Furthermore, in the study by Moljord et al. (2014) in Norwegian adolescents (13–18 years old), physical activity associated with psychological symptoms in the females but not in the males. In the present study also, some of the associations seemed reversed among the males. For instance, higher physical fitness (in terms of strength) seemed associated (p < 0.05) with higher stress among the males, in terms of total PSQ and tension (Table 4). This positive physical fitness-stress association might be comparable with the findings by Lutz et al. (2010) and Gerber et al. (2015). Gerber et al. (2015) found that exercise involvement was associated with increased stress in adolescents with high exercise self-regulation. The adolescents who accepted to participate in this study, which involved willingness to perform a physical fitness test, probably were relatively high in exercise self-regulation. However, this was not measured. Furthermore, among the males in the study sample there appeared negative associations between some of the stress factors and the pain variables (Table 4). This is in contrast to previous findings in Norwegian adolescents (Østerås et al., 2015, 2016). Probably, the findings in this study sample were influenced by the non-normal distribution of pain variables (Table 3). However, the findings might also imply that perceived stress associate differently with pain in a sample of fit adolescent males (see the “Strengths and limitations”-paragraph) than in the general adolescent population.

The lack of support in this study concerning the presumed stress and pain preventing effect of physical fitness in adolescents, might also be understood in light of modern societal trends and cultures. For instance, the promotion of exercise and physical fitness for appearance-motivated reasons are common in our modern society, extensively conveyed by “fitspiration” websites, which often reinforce an over-valuation of physical appearance, eating concerns, and excessive exercise (Boepple et al., 2016). This might contribute to more unhealthy physical fitness trends and habits, probably with less beneficial health outcomes as a result (Sibley et al., 2013; Cunningham et al., 2016). A recent population-based longitudinal study from Sweden revealed a shift in the covariates of perceived health, from physical fitness (cardiorespiratory) and chronic illness in 1990 to age, BMI, and educational level in 2015, i.e., a diminished impact of physical fitness on perceived health (Olsson et al., 2017).

This study evaluated physical fitness, mindfulness (MAAS-A), pain and BMI in relation to stress (PSQ) in Norwegian adolescents. Additional physical fitness associations were also explored as well as gender differences. Analyses revealed that physical fitness did not explain any variation in stress among the adolescents. However, there were some negative associations between one of the stress factors (lack of joy) and physical fitness components at a group level, which might have been influenced by conditions not measured in this study. As opposed to physical fitness, mindfulness qualities (MAAS-A), and to some degree gender, seemed to explain variation in stress among the adolescents. Physical fitness was unrelated to mindfulness (MAAS-A) but seemed somewhat related to BMI, particularly in males. Among the females, higher physical fitness (in terms of endurance) seemed related to fewer pain sites. The cross-sectional design did not allow us to determine any causal direction among the variables. Further studies might be needed to confirm these findings.

BØ has made substantial contributions to the conception and design of the work, the acquisition, analysis, and interpretation of data for the work. BØ has also drafted the work, revised it critically for important intellectual content, and approved the final version to be published. BØ agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. HS and MH has made contributions to the conception and design of the work, and has contributed in revising it, and given final approval of the version to be published. HS agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.