Reader Comments

Post a new comment on this article

Response to "Reconsidering the sedentary behaviour paradigm"

Posted by BMLynch on 12 Feb 2014 at 06:10 GMT

Maher and colleagues report findings from a secondary analysis of accelerometer data collected as part of the US National Health and Nutrition Examination Survey (NHANES). They found no associations of sedentary time with cardiometabolic risk biomarkers, when controlling for a measure of “total physical activity” that was composed of time spent in both moderate- to vigorous-intensity and light-intensity physical activity (that is, all non-sedentary time), weighted by intensity (accelerometer counts). On this basis, they concluded that sedentary time may not have health effects independent of physical activity. Leaving aside legitimate concerns with interpreting null results in this fashion, a broader conclusion that might be drawn is that there is limited (or no) merit in searching for “independent” effects of behaviours that are unavoidably “interdependent”.

From our program of sedentary behaviour research – including evidence from cross-sectional and prospective observational studies, field experiments and laboratory studies – we conclude that energy expenditure and posture are the two key pathways by which sedentary behaviour may causally impact cardiometabolic health. Neither of these pathways would be detectable in the study by Maher and colleagues.

We postulate that sedentary behaviour (which involves very low energy expenditure) has adverse effects because it reduces total energy expenditure by displacing time that could be spent in alternative activities at a higher energy expenditure (predominantly light-intensity physical activity, because most adults engage in so little moderate-intensity activity and even smaller amounts of vigorous activity). In the case of the Maher et al study, independent effects examined by mutual statistical adjustment (i.e., each category is adjusted for all other categories) ignore effects that may occur because of such displacement.(1)

The potential importance of posture, supported by experimental studies from our group and others, suggests specific pathophysiological responses occur when seated and large skeletal muscle groups are not engaged.(2,3) The authors’ use of the NHANES accelerometer data classifies sedentary time based on intensity of vertical hip movement alone – a crude indicator of energy expenditure from ambulation that is not directly indicative of posture. Hence, these data are inherently limited in their capacity to detect effects that might be attributable to posture.

We agree with Maher and colleagues that previous approaches to sedentary behaviour analyses have limitations. However, their approach to gauging independent effects does not improve, nor does it discount in any material way, the current understanding of the health consequences of sedentary time. We have employed isotemporal substitution analysis to estimate the potential collective impacts of displacing time spent sedentary with time spent in light-, moderate- and vigorous physical activity.(4) These collective effects provide more relevant information than independent effects as, in practice, reduction in sedentary behaviour will naturally occur secondary to an increase in other activities (mostly light intensity physical activity).

Lastly, based on substantial scientific evidence, sedentary behaviour researchers have been meticulous in concluding that reduction of sedentary behaviour (i.e., increase in predominantly light-intensity activity) complements the promotion of moderate- to vigorous-intensity physical activity. This combined approach is acknowledged in physical activity recommendations by leading agencies such as the American College of Sports Medicine.(5) Unfortunately, Maher and colleagues have put forward an equivocal perspective that adds an unnecessary element of confusion to public health thinking.

Contributors to this response: David W Dunstan, Genevieve N Healy, Brigid M Lynch, N Owen, Nicola D Ridgers, Jo Salmon, Phillip B Sparling, Elisabeth A Winkler.

1. Mekary RA, Willett WC, Hu FB, Ding EL. Isotemporal substitution paradigm for physical activity epidemiology and weight change. Am J Epidemiol. Aug 15 2009;170(4):519-27.
2. Dunstan DW, Kingwell BA, Larsen R, et al. Breaking up prolonged sitting reduces postprandial glucose and insulin responses. Diabetes Care. 2012;35:976-83.
3. Peddie MC, Bone JL, Rehrer NJ, et al. Breaking prolonged sitting reduces postprandial glycemia in healthy, normal-weight adults: a randomized crossover trial. Am J Clin Nutr. Aug 2013;98(2):358-66.
4. Buman MP, Winkler EA, Kurka JM, et al. Reallocating Time to Sleep, Sedentary Behaviors, or Active Behaviors: Associations With Cardiovascular Disease Risk Biomarkers, NHANES 2005-2006. Am J Epidemiol. Feb 1 2014;179(3):323-34.
5. Garber CE, Blissmer B, Deschenes MR, et al. American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Medicine & Science in Sports & Exercise. 2011;43(7):1334-59.

No competing interests declared.

RE: Response to Dunstan and colleagues' comment

Carol_Maher replied to BMLynch on 17 Mar 2014 at 19:50 GMT

We thank Dunstan, Healy, Lynch, Owen, Ridgers, Salmon, Sparling and Winkler for their comments on our recently published paper. We can agree about the limitations of many previous studies on this topic. Our paper attempted to add information to, and advance, the growing body of evidence related to the health effects of sedentary behaviour. Given the issues related to multicollinearity across different intensities of human movement, we agree that other novel analytical approaches in the field, including isotemporal substitution analysis, will also offer additional evidence regarding the observed associations.

In the course of a person’s day, the key activity types are (1) sleep, (2) sedentary behaviour, (3) light physical activity, and (4) moderate to vigorous physical activity (MVPA). For most people, almost the entire non-sleeping day is made up of sedentary and light activities (since MVPA comprises a small portion of the day for most people). Given this, sedentary behaviour and light physical activity are inevitably highly inversely correlated. Therefore, any study that finds negative associations between sedentary behaviour and health outcomes will necessarily find positive associations between light physical activity and health outcomes. To date, epidemiologists studying sitting time have shied away from adjusting for light physical activity due, in part, to issues of multicollinearity, and have largely controlled for leisure-time MVPA. To address such limitations, we adjusted for total physical activity based on daily accelerometry counts accrued in non-sedentary minutes, and demonstrated that health associations with sedentary behaviour receded. These findings have two important implications. Firstly, we hope that our approach will expand thinking about multi-variable methods used to understand the independent health effects of sedentary behaviour. Secondly, we think our findings shed light on the potential mechanisms underpinning sedentary behaviour’s health associations, and suggest that reduced energy expenditure, rather than posture, appears to be a likely mechanism.

The commentators suggest that our paper may cause unnecessary public health confusion because we have brought a new perspective to the argument. On this issue we are in disagreement. We believe that, as scientists, we must be willing to challenge existing paradigms and rigorously test the validity of different perspectives. The implication of our scientific findings for public health messaging, like the results of any other single study, must be viewed in concert with the existing literature and should be considered in future scientific consensus processes.

The commentators also assert that “sedentary behaviour researchers have been meticulous in concluding that reduction of sedentary behaviour….complements the promotion of moderate- to vigorous-intensity physical activity.” However, we are not confident that the message “sit less” would necessarily be more effective than the message “keep moving.” We have several decades of interventional research demonstrating that increases in physical activity result in health improvements in the majority of people; however, the existing evidence related to the health effects of sedentary behaviour are largely observational in nature, and carefully designed intervention studies that reduce sedentary behaviour are just beginning to appear. As scientists, our fundamental duty is to seek the truth. We hope our study will open scientific debate, which will progress the field, and ultimately, improve health.

Contributors to this response: Carol Maher, Tim Olds, Emily Mire & Peter Katzmarzyk

No competing interests declared.