Australasian Science: Australia's authority on science since 1938

Now Even Sitting Researchers Are Sitting on the Fence

By Tim Olds

To sit or stand has become an uncomfortable question for health researchers.

On 4 October this year, A/Prof Emmanuel Stamatakis of Sydney University released a study claiming that swapping just 1 hour of daily sitting with standing is linked to a 5% reduction in the risk of premature death.

This is not so surprising. Stamatakis is an excellent and well-respected epidemiologist, and we’ve heard the sitting message before. Fore example, a large American study found that people who sit more, and who sit for prolonged periods, are more likely to have wider waists, higher blood fats, higher levels of inflammation and greater overall risk of death.

What’s a little more surprising is that 1 month later Stamatakis was a co-author of a study which found that in a famous British cohort (Whitehall II), sitting was not associated with mortality risk after 16 years of follow-up. Neither total sitting time, time spent watching TV, leisure time sitting nor sitting at work had any relationship with the risk of death. Stamatakis’ group concluded that policy-makers should be cautious about recommending reductions in sitting time as a stand-alone public health intervention.

What’s a person to believe these days? One day, milk is good for you and your bones, the next it increases your risk of death and fractures. Saturated fat is a killer one day, the next day it’s almost a health food. Now we’re getting mixed messages about sitting.

So what’s at the bottom of it all, so to speak? I think there are two critical issues behind this uncertainty. The first has to do with why sitting appears to have such bad health outcomes, and the second has to do with the size of the effect.

Since the second issue is simpler, let’s start there. It’s easy to recommend sitting less, but how much less do we need to sit?

One way of quantifying this is to ask what the trade-off is between sitting and physical activity. How many minutes less sitting do I need to do to get the same benefit as doing 10 minutes more exercise each day? There are several studies which allow us to get a rough idea.

A recent analysis of a big American dataset found that in terms of reducing cardiovascular risk in adults, 10 minutes of physical activity was worth as much as 90 minutes less sitting. Sir David Spiegelhalter, a biostatistician at Cambridge University, estimates that in terms of reducing all-cause mortality, 10 minutes more exercise is equivalent to about 2 hours less sitting.

A 12-nation study looking at the link between physical activity, diet, sleep, sitting and obesity in kids found that kids would need to sit for about 6 hours less to get the same obesity-busting benefit as 10 more minutes of physical activity.

These figures are averages across the population, but the benefits from sitting are far greater among people who are very inactive. In fact, Stamatakis suggests that in the Whitehall study, the failure to find an association between sitting and all-cause mortality may be due in part to the fact that this cohort was abnormally active.

This is what we also found in our analysis of sitting and cardiovascular risk: people who get about 2 hours of physical activity per day get virtually no extra benefit from sitting less, a conclusion confirmed in a recent meta-analysis.

The more interesting and complex issue is that of the mechanisms behind the sitting–health link. Is it just because sitting displaces other activities that are healthy (e.g. standing, light physical activity, exercise, sleep)? And does it make any difference whether sitting displaces standing, light physical activity or exercise? Or is it because there is something special about sitting, such as the lack of muscular activity in the big muscles of the legs?

A paper by the Franco-Scottish sedentary behaviour expert Sébastien Chastin gives us some clues. Chastin used a radical new statistical paradigm called compositional analysis, which takes account of the fact that if we do less of one thing (sitting) we must be doing more of something else (such as physical activity).

Compositional analysis looks not at individual behaviours but at mixes of behaviours called compositions. Using this procedure, Chastin looked at the relationship between how people spend their time and cardiometabolic risk in American adults. A striking finding is that sitting, as part of the daily mix of activities, has at best very weak associations with almost all health outcomes (fatness, blood pressure, blood fats, inflammatory markers, blood sugar control). In particular, providing you get a reasonable amount of physical activity (e.g. 30–50 minutes per day), it makes no difference whether you are sitting, standing or doing light physical activity – for all outcomes except blood insulin levels. Sitting is mainly harmful because it displaces physical activity.

So here’s the bottom line, as I read it: if you’re reasonably active you can forget about the standing desk, and needn’t feel guilty about staying seated at meetings. If you’re relatively inactive you’ll need to do an awful lot of standing for it to make much difference. Standing, and light physical activity, may still beat sitting if you have diabetes or pre-diabetes, or if you’re very, very inactive.

All this means that it’s an uncomfortable time for sitting researchers – they’re either sitting on the fence or walking both sides of the street.

Professor Tim Olds leads the Health and Use of Time Group at the Sansom Institute for Health Research, University of South Australia.