December 8, 2022

expert reaction to study looking at short bursts of vigorous physical activity and risk of death

A study published in Nature Medicine looks at wearable device-measured vigorous physical activity and mortality.

Prof David Stensel, Professor of Exercise Metabolism, Loughborough University, said:

“This is a very nice paper emphasising the potential benefits of short (1 to 2 minute) bouts of vigorous exercise in people who do not consider themselves to be exercisers, and who report not taking part in exercise in their leisure time.  This study suggests there is potential for short bouts of vigorous physical activity completed during routine daily activities e.g., walking briskly uphill, climbing stairs, carrying shopping, even for a total of only 4 to 5 minutes per day, to have major benefits for reducing the risk of death due to cardiovascular disease, cancer and all-causes.  I use the word ‘potential’ because we can’t know cause and effect from this study, because it’s an observational study not a trial.

“This is good quality research backed up by solid data – notably the use of accelerometry to quantify vigorous intermittent lifestyle physical activity.  It fits with existing evidence that participating in vigorous physical activity is associated with a lower risk of all-cause, cardiovascular disease and cancer mortality and it moves the field forward by showing that this applies to unstructured forms of vigorous physical activity in people who report being non-exercisers.  The authors have accounted for confounding variables but this is still an observational study so there is always the possibility of reverse causation where people who have naturally high fitness levels also exhibit other constitutional factors which protect them from disease and the association is not necessarily cause and effect.  This is a limitation of all observational studies.  If these findings do represent cause and effect, then the implication in the real world is that people who squeeze in just a few (4 or 5) minutes of vigorous exercise each day into their daily activities can substantially lower their risk of mortality from chronic diseases.  This focus on the potential benefits of routine daily activities completed with vigour in ‘non-exercise settings’ is important.

“It’s important to note that the authors haven’t shown that vigorous intermittent lifestyle physical activity increases cardiorespiratory fitness (a potential mechanism to explain the findings), and they haven’t shown ‘reductions’ in mortality risk – they have shown a lower mortality risk in those reporting vigorous intermittent lifestyle physical activity than in those who do not report such activity.

“Nevertheless, the findings are important and provocative and should stimulate further research yielding greater insights into the potential health benefits of vigorous intermittent lifestyle physical activity.”

Prof Paul Leeson, Professor of Cardiovascular Medicine, University of Oxford, said:

“The findings are important because they are based on very robust wearable technology.  This was developed specifically for research purposes to capture precise information about activity in a large sample of the UK population.  Furthermore the results have been linked with data on the future health of participants, captured through the NHS, which is a unique strength of research performed in the UK.

“What is different about this research is that the investigators have studied how patterns of exercise link with future health, whether or not the participant intended to do the exercise.  They find the association between short bursts of vigorous activity and better health in later life is the same, whether the exercise was done purposefully as part of an exercise programme, or just because you happened to do the activity as part of your other daily activity, for example, by running for a bus.

“The major limitation is that this data is observational.  Therefore it does not tell us whether adding short bursts of vigorous activity into your lifestyle, if you are not doing this already, lowers your risk.  The findings just show us that the type of people who have lifestyles that include short bursts of vigorous activity, for whatever reason, tend to be the same people who live longer and avoid heart attacks.”

Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“This is an interesting and generally careful study, and one that couldn’t have been done until fairly recently because it depends on data from wearable accelerometers.  These days they are available in smart watches, and similar, and indeed in smartphones, though the data in this study come from about 25,000 participants in the UK Biobank study who were asked to wear accelerometers on their wrists for a week during 2013-15, and reported that they did not engage in leisure-time physical activity beyond one recreational walk a week.

“Some caution is needed in interpreting the findings.  The study is observational.  That is, nobody told the participants what to do in the way of exercise.  The researchers measured their movement using the wearable accelerometer.  Then they were followed up for about 7 years on average, during which time the researchers recorded deaths that occurred.  They did find an association – a correlation – between the amount of vigorous activity (outside leisure-time exercise) that the participants did and their chance of dying in a given period of time.  But it’s not possible to be certain that this association is one where the vigorous activity is the cause of the difference in death rates.  It certainly could be the cause, at least in part, but we can’t be certain.

“The reason is that, as you might expect, there were many differences between the people who took different amounts of this kind of exercise, apart from the amount of exercise they took.  For example, those who had the largest amounts of vigorous physical activity as part of their normal non-leisure lifestyle were, on average, younger than those who took no such exercise, and were less likely to be current smokers.  Those other differences between people could be the underlying cause of the differences in death rates.  Of course, the researchers were aware of this issue.  It’s possible to make statistical adjustments to allow for these other differences, provided data on the other differences are available, and the researchers did that for a wide range of factors, including age, sex, smoking, alcohol use, education level, consumption of fruit and vegetables, and a great deal more.  But one can never be certain that everything relevant has been adjusted for, and obviously you can’t adjust for factors that you didn’t observe and so have no data for.

“The researchers calculated so-called E-values, that measure how strong associations involving factors that weren’t adjusted for would need to be, to make the observed associations disappear if they could have been adjusted for.  The E-values that they found do give some reassurance that other factors may not have been too great an issue – but one still can’t be sure because E-values don’t allow for there being more than one other unobserved factor that jointly explain the findings.

“A particular issue here is what’s known as reverse causation.  This would mean that, instead of the differences in exercise levels being the cause (or partial cause) of the differences in death rates, the differences in death rates are (in a sense) the cause of the differences in exercise levels.  That sounds crazy, in that one can’t exercise after death.  But what’s meant is that people may have health issues that haven’t been diagnosed yet, or indeed may be latent and hidden so that they couldn’t be diagnosed, and that those issues first make it less likely that they will have short bursts of vigorous activity during their everyday lives, and later, independently, increase their chance of dying in a given period of time.  The researchers allowed for this possibility in several ways, for example by excluding people who already had cardiovascular disease or cancer from some of their analyses, and by adjusting for some medication use that could possibly have been related to pre-existing disease.  But again one can’t be sure that this has dealt with all possibilities of reverse causation.

“The researchers do, rightly, mention the possibility that some reverse causation remains, or that some relevant unmeasured factors couldn’t be adjusted for, in their paper.  But because of those issues, and possibly other issues to do with the way things were measured, some doubt does remain about cause and effect in the findings.  However, personally I’m not going to let that deter me from running for buses when I have to, as I’m pretty confident that does me more good than harm.”

‘Association of wearable device-measured vigorous intermittent lifestyle physical activity with mortality’ by Emmanuel Stamatakis et al. was published in Nature Medicine at 16:00 UK time on Thursday 8 December 2022.

DOI: 10.1038/s41591-022-02100-x

Declared interests

Prof David Stensel: “I am co-author / co-editor of a book with one of the authors of the article – Jason Gill.”

Prof Paul Leeson: “Member of the Imaging Advisory Board of UK Biobank.”

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  My quote above is in my capacity as an independent professional statistician.”