Skip to main content

Table 3 Assessment of causality using assessments of strength of association, consistency, specificity, temporality, coherence and biological plausibility, dose-response, and experimental evidence

From: Sedentary behaviour and adiposity in youth: a systematic review of reviews and analysis of causality

  Definition Summary of evidence
Strength of association How strong is the association between sedentary behaviour and adiposity in young people? Weak Consistently low strength of association values from cross-sectional evidence for self-reported screen time and objectively assessed sedentary time (e.g., r < 0.01); values close to zero for BMI per additional hour/day of screen time in prospective studies; small significant and non-significant effects from interventions.
Consistency How consistent is the evidence across different populations and in different settings? Moderate-to-weak Evidence on sex differences in inconclusive. Stronger evidence exists for an association in children than adolescents but this could be a function of the volume of research favouring younger age groups, as well as the issue of maturation confounding measures of adiposity. Evidence does not differ by country.
Consistency across measures of sedentary behaviour and markers of adiposity is weak.
Specificity Is adiposity mainly limited to the existence of sedentary behaviour? No It is clear that many factors can be listed that are associated with weight gain or higher levels of adiposity in young people. The factor of specificity, therefore, cannot be supported. However, Hill states that we must not overemphasise this issue because diseases may have more than one cause and that “one-to-one relationships are not frequent” (p. 297).
Temporality Does sedentary behaviour precede the development of adiposity? Weak Reviews addressing prospective studies show a mixed pattern of results. Data on self-reported screen time have shown ‘strong’ evidence for an association with BMI, but ‘insufficient’ for other measures of adiposity. Evidence concerning objective measures of total sedentary behaviour is largely null.
Coherence and biological plausibility Any interpretation of the data should not seriously conflict with what is known about weight status and adiposity in young people. Biological plausibility provides further support for causation. Moderate While it is plausible and coherent with current knowledge that low energy expenditure in the form of sitting could be obesogenic, often these behaviours (e.g., TV viewing) co-exist with other behaviours. These might include excessive dietary intake and prompts from TV advertising for unhealthy foods.
Individual sedentary behaviours are usually correlated in only a small way with moderate-to-vigorous physical activity, thus it cannot be claimed that one sedentary behaviour automatically precludes being physically active over time.
Dose-response Do higher levels of sedentary behaviour show higher levels of adiposity? Yes Two reviewers provide support for a dose-response relationship. Carson et al. showed that more or less regardless of how TV viewing categories were compared, higher viewing was associated with greater adiposity. Zhang et al. calculated an odds ratio per 1 h/day increment in TV watching as 1.13 (95% CI 1.03–1.19). Graphical data suggested a linear relationship.
Experimental evidence Is there evidence from interventions using experimental methods for changes in adiposity to result from changes in sedentary behaviour? Weak The analysis we have undertaken in this review of reviews summarises the evidence concerning effectiveness from interventions as ‘modest’, although some groups (e.g., obese) may gain more benefit. Effect sizes from meta-analyses, however expressed, are mostly small and both significant and non-significant.