Mark S. Allen, Emma E. Walter, Christian Swann

This research synthesis sought to determine the magnitude of the association between sedentary behaviour (sitting time) and anxiety. A comprehensive literature search of eight electronic databases (and a manual search) identified 13 observational studies that met inclusion criteria (22 effect sizes; total n = 70,425). Pooled mean effects were computed using inverse-variance weighted random effects meta-analysis and moderation by study and population characteristics were tested using random effects meta-regression. Sedentary behaviour was associated with an increased risk of anxiety for non-adjusted effect sizes (k = 7, OR = 1.33 [95% CI: 1.14, 1.55]) and effect sizes adjusted for sociodemographic and health-related factors (k = 11, OR = 1.48 [95% CI: 1.25, 1.75]). There was no evidence of publication bias in the results. The regression models showed that effect sizes were not moderated by age or gender. However, there was some evidence of moderation by study quality and measurement of sedentary behaviour and anxiety. Measures of sitting time showed larger associations than measures of screen time, and measures of anxiety symptoms showed larger associations than measures of anxiety disorders. The research synthesis provides evidence that sedentary behaviour has a small positive association with anxiety, after controlling for sociodemographic and other health-related factors. Study limitations include low statistical power in meta-regression models and heterogeneity in measures of anxiety and sedentary behaviour. Findings might be of interest to health care professionals developing health care initiatives to reduce risk of anxiety.