Why are adults sedentary?

Today’s article comes from Dr Stephanie Prince Ware.  More information on Dr Prince Ware can be found at the bottom of this post.  Her recent paper on correlates of sedentary behaviour in adults can be found at the following link:

Prince SA, Reed JL, McFetridge C, Tremblay MS, Reid RD. Correlates of sedentary behaviour in adults: a systematic review. Obes Rev 2017;18(8):915-935. https://www.ncbi.nlm.nih.gov/pubmed/28524615

Background

As adults, we spend far too much time being sedentary; in fact, we spend most of our waking time sitting or lying down (1,2).  Sedentary time is an accumulation of our time spent sitting/reclining/lying engaged in different behaviours at a low energy expenditure (e.g., using a computer, watching TV, sitting in a car) across multiple domains of our life (work, home, play, transportation) (3).  We are beginning to understand that all of this sedentary time is not great for our health; having been positively associated with obesity, chronic disease and mortality (4-5).

Understanding the “why” behind our sedentary habits is important for the development of interventions targeting these behaviours and in understanding the barriers and enablers of behaviour change.  Many individual studies have looked at various reasons for why individuals engage in specific sedentary behaviours.  Previous reviews that have synthesized evidence around the correlates of sedentary behaviour have largely focused on factors that operate at different levels of influence (e.g., intrapersonal, home environment, neighbourhood).  Examining these levels separately does not provide a complete picture of all the known modifiable factors associated with sedentary behaviour and the relative trend in the strength of association across these factors.  This information is ultimately needed to inform intervention design by identifying good targets. Further, it has become increasingly apparent that the reasons behind why adults engage in sedentary time are dependent upon the individual behaviours themselves, as well as the domains in which they take place.

Therefore, the objective of our research was to systematically review the literature and identify and describe the intrapersonal, social environment, physical environmental and policy correlates of domain-specific sedentary behaviours and total sedentary time among adults.

What did we do?

To examine all available literature we conducted a systematic review which searched six different bibliographic databases (you can see more details on this in the full article) looking for any study which described a correlate (related factor) of sedentary behaviour (self-reported or objectively measured) among adults (18+ years).  We were very fortunate to obtain several unpublished results provided to us upon request from authors of original studies where these relationships were not reported upon in previous articles.

We assessed the risk of bias of every individual study.  We summarized the results qualitatively to describe overall trends (and strengths) in the associations between the correlates and sedentary behaviours across domains.  We also did our best to look at differences between men and women, age groups, populations, self-reported vs. objectively measured sedentary behaviour, published vs. unpublished, and looked at trends over time.

What did we find?

After we removed duplicate papers, we ended up retrieving 22,726 relevant papers.  That was a lot of papers to screen!  After many hours, days and months looking at the abstracts and then the full articles, 257 studies were kept in the review having met our inclusion criteria.  See Supplemental figure 1 in the article for the PRISMA flow diagram to get a sense of how we screened out papers.

Most studies relied on self-reported sedentary behaviour measures; only 2% used both a self-report and objective measure.  In addition, many of these self-report measures were not validated.  Almost all (90%) of the studies used cross-sectional data.  The most studied domain of sedentary behaviour was leisure time; TV time being the most examined.  Most of the studies reported on individual-level correlates of sedentary behaviour; education, employment, income/SES and marital status were the most common.  Fewer studies examined factors in the built and social environments and only a couple reported on policy-level correlates.

Not surprisingly, our review found that being employed full time was consistently associated with lower leisure, but higher transportation-related sedentary time.  Also in the intuitive direction, ownership of a TV was consistently associated with greater leisure sedentary behaviour.  Good news for parents, having children was fairly consistently associated with lower total sitting.  Living in more urban areas trended toward greater sitting and total sedentary time.  Surprisingly, most social environmental factors (e.g., social participation, crime, social support, social trust/cohesion, norms) were not associated with leisure sedentary time – though these factors were less studied than individual-level correlates.  This provides a glimpse into some of our findings; see the full article for the exhaustive list.

Interestingly, we found differences in the direction of association across correlates between weekday and weekend sedentary time.  We also found that the direction of association with several of the correlates differed between self-reported vs. objectively measured total sedentary time.  For example, greater income/SES was positively associated with greater self-reported sedentary time, but not associated with objectively measured sedentary time.  Unfortunately, because almost all of the evidence came from cross-sectional studies, we are unable to infer causation and thus, cannot definitively conclude that any of the factors cause adults to spend more time being sedentary.

Take-home messages

The results of our review provide a comprehensive summary of the correlates of sedentary behaviour in adults using a socio-ecological approach.  Results provide guidance for the design of future studies by identifying gaps in the evidence, as well as identifying possible areas of promise for interventions.  There is a need for more research using prospective and experimental designs to establish causality, and for a greater understanding of the role of factors in the social, built and policy environments.  When possible, future research should consider looking at both total and domain-specific sedentary behaviour using a combination of self-report and objective measures, and look at weekday and weekend sedentary time separately.

Ultimately, understanding the strongest and consistent correlates of sedentary behaviour provides an opportunity for targeted and evidence-based intervention development.

About the author:

Dr. Stephanie Prince Ware is a CIHR Health Systems Impact Fellow working at the Public Health Agency of Canada and the University of Ottawa Heart Institute.  You can follow her on Twitter @SPrinceWare.

References:

1. Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS. Physical activity of Canadian adults: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep2011; 22: 7–14.
2. Matthews CE, Chen KY, Freedson PSet al. Amount of time spent in sedentary behaviors in the United States, 2003–2004. Am J Epidemiol 2008; 167: 875–881.
3. Tremblay MS, Aubert S, Barnes JD et al. Sedentary Behavior Research Network (SBRN) – Terminology Consensus Project process and outcome. Int J Behav Nutr Phys Act 2017;14(1):75.