Discussion
This is the first review to examine evidence regarding the association between sedentary behaviour and risk of anxiety. It is important to better understand this relationship as this information may help to inform the development of lifestyle change strategies for reducing the risk of anxiety in different population groups. It is clear from this review that the current body of evidence exploring the relationship between sedentary behaviour and risk of anxiety is limited, with only nine studies currently been published. On the balance, however, most studies (78 %) found at least one positive association between sedentary behaviour and anxiety risk. In other words, there is moderate evidence suggesting that engaging in overall sedentary behaviour was linked to an increased risk of anxiety. These findings are similar to those found in previous reviews that have assessed the relationship between sedentary behaviour and other specific mental health outcomes such as depression. However, in our review, based on the best-evidence synthesis and when considering the different types of sedentary behaviour separately, moderate evidence was found for the positive relationship between sitting time and anxiety risk, whilst inconsistent evidence was found for the relationship between screen time, television viewing time, computer use, and anxiety risk.
There is currently limited insight into the underlying mechanisms that may explain this positive relationship between sedentary behaviour and anxiety risk. As already discussed, plausible biological pathways may include central nervous system arousal, sleep disturbances or poor metabolic health resulting from engagement in sedentary behaviour. Furthermore, drawing from previously suggested hypotheses that have been used to explain the link between sedentary behaviour and other mental disorders (i.e. depression), it could be the displacement of physical activity when engaging in sedentary behaviour that explains the relationship with increased anxiety risk, since physical activity has been shown to be beneficial in reducing anxiety in both children/adolescents and adults. Alternatively, the link may be explained by a social withdrawal theory which posits that engaging in prolonged sedentary behaviours, such as television viewing, may lead to social solitude and withdrawing from interpersonal relationships which has been linked to increased feelings of social anxiety. On the other hand, it may be that those suffering anxiety symptoms are more inclined to engage in sedentary behaviours as a means of coping with anxiety, as has been suggested in previous research amongst adolescence with social physique anxiety.
In contrast, one cross-sectional study included in our review showed that sedentary behaviour was inversely associated with risk of anxiety in girls (i.e. those who spent less than 2-hours in screen-based entertainment were more likely to suffer symptoms related to anxiety). These findings may suggest that screen-based entertainment could be beneficial for relieving/managing anxiety symptoms in children. Alternatively, due to the cross-sectional nature of the study, they may suggest that children with higher levels of anxiety may be less likely to engage in screen-based entertainment and perhaps more likely to engage in other non-screen based sedentary activities such as reading/studying. There is a small body of literature that suggests some forms of sedentary behaviour may in fact have a positive impact on mental health, specifically depressive symptoms, however, further intervention and prospective studies are required to determine the direction of the relationship between sedentary behaviour and anxiety symptoms.
All studies included in this review were limited by several methodological weaknesses. For example, most studies (7/9) employed a cross-sectional study design and therefore causality and/or direction of relationships were unable to be determined. Secondly, self, or proxy-report measures of sedentary behaviour were utilised in most (n = 8) studies with such measures increasing the likelihood of recall problems and provision of socially desirable responses and thus to overcome these limitations further research involving objective measures of sedentary behaviour (e.g. accelerometers, posture monitors [i.e. activPALs]) is recommended. However, since the relationship between sedentary behaviour and anxiety risk may be dependent on the domain/type of sedentary behaviour (an aspect of sedentary behaviour that is not able to be measured using such monitors), a combination of both objective and subjective (e.g. self-report surveys) methods is warranted. This further highlights the need for the development of valid objective measures of sedentary behaviour which assess not only the dose (e.g. frequency, duration) but also the domain (e.g. leisure, work, transport), and context (TV viewing, computer use, tablet/smart phone use) in which these behaviours occur. Thirdly, only two studies compared different types of sedentary behaviours and their relationship with anxiety risk and therefore we were unable to clearly determine which specific sedentary behaviours may be linked to anxiety. Further, since few studies used the same method to define/assess sedentary behaviour, and every study included a different measure of anxiety symptoms, clear-cut conclusions were difficult to determine. Finally, given that anxiety, and depression are often co-morbid disorders, it is difficult to disentangle the relationship sedentary behaviour has with anxiety alone compared to comorbidity associations.
This review enhances the understanding of where the field is at in terms of sedentary behaviour and anxiety research. Although only limited evidence is currently available on the association between sedentary behaviour and risk of anxiety, on the balance this evidence suggests a positive association may exist between overall sedentary behaviour (e.g. sitting time specifically) and anxiety risk, whilst inconsistent evidence remains for other types of sedentary behaviours (e.g. computer use, television viewing, screen time), and their link with anxiety risk. This review further highlights the need for more high-quality longitudinal and intervention research to confirm and disentangle cross-sectional research findings.