CEBM Response: “Does physical activity attenuate, or even eliminate, the detrimental association of sitting time with mortality?”
This new study by Ekelund and colleagues reports that the increased risk of early death associated with too much sitting is mitigated in adults who say they do 60 to 75 minutes of moderate vigorous intensity physical activity (MVPA)per day, suggesting that with enough of the right intensity of daily physical activity there is little risk of prolonged sitting for early death. The study does however suggest that the WHO recommendations of 30 minutes MVPA per day may not be sufficient to eliminate this risk.
However, there are a number of strengths and weakness associated with the study as well as further nuances that need to be taken into account when considering the wider adoption of the findings. For example, the authors found that specific types of sitting activity, such as TV viewing time, was still found to increase people’s risk of early death regardless of how much daily MVPA people said they performed, although this risk was attenuated with higher levels of activity.
Furthermore, paradoxically, the authors fail to point out that no amount of physical activity was found to provide a protective effect (reducing the risk of early death).
That said, one of the strengths was that the study performed was a systematic review of prospective cohort studies including over 1 million participants and follow-up of 2 to 18.1 years. Additional strengths include to re-analysis of original data, thus reducing the methodological and statistical heterogeneity between studies, and the use of official registries for ascertainment of mortality outcomes.
However, there are also weaknesses. For example, it is possible that the authors may have missed relevant studies by excluding non-English studies and using only 1 reviewer for screening titles.
The authors use a non-standardised method to assess included study quality. This method relied on a scoring system; and found, in general, that included studies scored “high”, thus indicating them to be of high quality by the authors. However, scoring systems have inherent problems, not least because they cannot distinguish between more important aspects of bias and lesser ones. For this reason, the Cochrane group recommend against their use, preferring a more qualitative assessment. The authors do not narratively discuss the quality of included studies, presumably relying on the reported high quality scores.
The authors do not use a GRADE or similar approach to determine the quality and certainty of the effect estimates where as such the nature of the observational data would automatically be considered low quality (i.e. “further research is likely to have an important impact on confidence in the estimate of effect and may change the estimate”).
Effect sizes referred to throughout are relative estimates and it is important to consider the absolute effects. For example, whilst the authors report a 59% relative increase in risk of death in people who state sitting for over 8 hours a day and doing less than 2 hours of MVPA compared with people sitting for less than 4 hours a day and doing 35.5 hours of MVPA a week, this confers an absolute increase in risk of 3.1%, meaning 32 people would need to be sedentary and inactive to this degree for there to be one additional death.
Forest plots and meta-analyses are presented in such a way that it is not possible to ascertain the contributing studies nor how much weight each included study adds to the pooled summary estimates.
Given the nature of included observational data, there is likely to be residual confounding, something that is acknowledged by the authors. For example, socioeconomic status has a big impact in cohort studies such as those included in this review; only 2 of the included studies appear to control for this factor in their analysis. This may be why for some analyses the dose-response relationship was not observed (e.g. >5 h/day of TV-viewing time and all-cause mortality [table 3]). As acknowledged by the authors, reverse causality is an additional confounding factor that was not controlled for. This would likely result in exaggerated effect estimates. Another potential confounding factor is the length of sitting at any one period. It may be that more active people are also less likely to sit for prolonged periods compared with less active people independent of total sitting time, potential exaggerating effect estimates based on total sitting time alone.
As the authors acknowledge, there is an issue of recall bias by the use of self-report to assess physical activity and sedentary time. Indeed, some of the surveys used asked people to recall their average daily sitting time from the previous year. It is known that using these methods people often underestimate sedentary time and overestimate physical activity level when compared with objective measures.
This study did not investigate associations of physical activity and sedentary time with risk of developing lifestyle related disease. It may be that physical activity plays an even greater role in the onset of diabetes and other metabolic conditions associated with high levels of sedentary time.
Finally, the study was unfunded and the authors declare no conflict of interest. This study was conducted as part of a Lancet Physical Activity Series of which it is worth noting that a number of authors are members of the steering group committee and have authored reports for various Departments of Health recommending promotion of daily physical activity as part of their public health strategies.
Given the above, any discussion of attenuating benefit of physical activity at the individual level must be tempered by the observational nature and quality of included data and the absolute benefit. However, in consideration of its strengths and weaknesses, it is encouraging that this study suggests the negative associations of too much sitting with early death due to all-causes, cardiovascular disease and cancer can be attenuated, and in some cases mitigated, by participation in daily physical activity of 30 minutes or greater. As the greatest gains were observed for highest physical activity and lowest sitting and TV-watching time, we should continue to support promotion of daily physical activity alongside strategies to reduce time in sedentary activities.
Dr David Nunan – Departmental Lecturer in Evidence Based Medicine
Centre for Evidence Based Medicine, Nuffield Department of Primary Care Health Sciences, University of Oxford.
This response will be posted on www.cebm.ox.ac.uk
Dr David Nunan is a member of the Royal College of General Practitioners (RCGP) steering committee to support the new Physical Activity and Lifestyle clinical priority. He has received funding for research from the NHS National Institute for Health Research School for Primary Care Research (NIHR SPCR) and the RCGP for independent research projects related to physical activity and dietary interventions. The views expressed are those of the author and not necessarily those of the NHS, the NIHR, the RCGP or the Department of Health.
He declares no other relevant conflicts of interest.