Cosmos Q&A: Taking the exercise path

Were there any surprises in the findings?

Researchers this week released what they say this is the first study to demonstrate that the intensity of physical activity is important over and above the total volume of activity in lowering the risk of an early death. The findings show, for example, that adding two minutes of brisk walking to a 35-minute daily stroll could cut that risk by 21%.

Cosmos spoke to the Australia-based member of the team, Dr Paddy Dempsey from the MRC Epidemiology Unit at the University of Cambridge (UK), the Leicester Diabetes Centre (UK), and the Baker Heart and Diabetes Institute in Melbourne.

You collected data from 100,000 people. What story did this tell?

The role of intensity over and above activity volume was ambiguous previously, but the size of this study and the availability of a week’s objectively measured physical activity data for each participant allowed us for the first time to address this question in an integrated way. The key finding from this study was that doing more physical activity of any intensity is beneficial, but that expending those calories in more intense activity is better still.

Moreover, the results also showed that activity volumes accumulated almost exclusively through light-intensity activity could still halve the mortality risk. The key take-home message therefore is that there are several different pathways or options to maintaining good health (e.g. through progressively increasing activity volume or intensity, but ideally a combination of both) and people can choose the path that works best for them.

Were there any surprises in the findings?

For physical activity researchers at least, the key findings of this study will not be all that surprising. There is a wealth of epidemiological evidence showing that higher levels of physical activity are associated with a lower risk of mortality and other adverse health outcomes.

However, this previous evidence, and current physical activity guidelines, are based largely on epidemiological studies in which physical activity was assessed by self-report questionnaires. Self-report measures of physical activity, while still important, are prone to reporting bias and measurement error, and are unable to robustly quantify light-intensity activities of daily living (a key contributor to total activity volume).

What will perhaps be more surprising for many is that the benefits of physical activity shown in this study using wearable devices may be greater than previously considered, and that the levels of activity at which they occur may be lower. This could have important implications for public-health guidelines in the future.

At first read, the studies suggest you actually don’t have to do a lot to achieve significant gains. Is that a correct reading?

Yes, to a degree. The caveat is that our inactive reference group are not completely bed-rested – they do 1-2 hours of light activity and 5-10 minutes of moderate activity each day, which is still not a lot because we count every five seconds that register into these categories.

Expending more energy at any physical activity intensity was strongly associated with a lower risk of death over the following three years. For example, participants who accumulated 20 kJ/kg/day through physical activity were a third less likely to die compared to those who accumulated 15 kJ/kg/day, when the proportion from at least moderate intensity activity was 10% in both cases.

The additional activity is the equivalent to a 35-minute stroll, with an extra two minutes at a brisker pace. When considering the simpler associations of physical activity energy expenditure (PAEE) – that is, adjusting for confounders but not activity intensity – undertaking just an additional 5 kJ/kg/day (relative to a low baseline of 15 kJ/kg/day) was associated with a 37% lower risk of all-cause mortality.

The results from this study also suggest that the greatest potential for a reduction in mortality risk was for those with the lowest levels of PAEE (i.e. the most inactive), whereas those who were already achieving higher levels of PAEE (i.e. greater than the median amount in the population) may achieve less additional gains from doing more. This is in line with the classic curvilinear dose-response curve we often see for the association between physical activity and health – where you get the “most bang for your public health buck” so to speak when you shift more inactive people to becoming more active.

These findings support the general public-health message that “any activity is better than none, and more is better still”, and benefits are still achieved at levels below the guideline target.

Is exercising to be healthy the same as exercising to be fit?

I think to best answer this requires a good understanding of the key concepts around fitness and activity. Both “physical activity” (any general bodily movement that results in energy expenditure) and “exercise” (a behavioural subset of physical activity that is more planned, structured or and repetitive) are behaviours that help improve and maintain various components of “physical fitness” (a set of attributes that people have or achieve) – which can be more ‘health-related’ or ‘performance-related’.

So physical activity and fitness are separate concepts that are both important for overall health but are distinct in how they are measured. Different types of activity promote different types of physiological changes and different health outcomes, therefore our conceptual “starting point” for baseline activity or fitness, and ultimate goals, likely also matter. From a public health perspective, it is likely more important to understand the relationships among type, volume and intensity for sedentary, light-intensity, and moderate-intensity activities, whereas understanding more vigorous intensity activities might be more relevant to athletes focused on performance-related goals.

Is being sedentary the same as not being active?

Sedentary behaviour is defined as any waking behaviour characterised by an energy expenditure ≤1.5 metabolic equivalents (METs), while in a sitting, reclining, or lying posture. It is considered conceptually distinct from physical inactivity, and it is certainly possible to meet or exceed the public health guidelines for moderate-to-vigorous physical activity (e.g. 150 min/week), and yet also spend most waking hours in sedentary behaviours (picture your typical desk based worker who walks or cycles into work).

This is a different question to what was addressed directly in our study, but relevant for public health, since behaviourally when we look to replace our typically high volumes of sedentary time throughout the day, it is most likely to be with light-intensity activities of daily living, and that would drive overall activity volume up, everything else being equal.

There are a lot of myths around exercise. Does research need to take account of that?

That’s a pretty open question, but I think a key myth relevant to the general context of this study is that you must hurt and sweat to gain any “real” or “meaningful” health benefits (i.e. the “no pain no gain” mantra). But evidence increasingly shows that even light-intensity activity appears to provide benefits and is likely preferable to sitting still for large portions of the day.

When increasing physical activity toward a desired level, small and well-spaced increments will reduce the incidence of adverse health events and likely improve adherence. These key concepts are relatively unambiguous and detailed in current guidelines (though sometimes misunderstood). Nevertheless, we still have a lot more to understand about the optimal “balance” of behaviours to maximise risk reduction.

Exercise science is an increasingly sophisticated and multi-disciplinary field. What have been the major advances?

A key factor in advancing the field of physical activity and exercise science has been the advances in tools for measuring activities of all intensities, and even differentiating between some types of activity (e.g. cycling, walking). This study used wearable devices, backed up by a series of validation studies, to estimate physical activity energy expenditure more robustly and on a larger scale than ever before, providing a proof-of-concept for wearable device use at scale in the population and new insights into the dose-response relationships of physical activity volume and intensity with mortality risk.

The use of wearable devices that monitor physical activity is projected to increase dramatically in the future, which could help provide people with more personalised information and feedback to improve their health. However, there is a lot more research to be done in this space on multiple fronts. For example, how can and might different people use wearable devices in everyday living, what are the specific health benefits for different amounts and types of activities, and how might the data gathered compare to what is typically obtained in research studies (be that device-based or through self-report)?

Do we understand how genes and other biological factors influence physical activity and sedentary behaviour?

All human behavioural traits are usually determined by both environmental/social and biological factors, and there is growing evidence in both humans and animals which suggests that intrinsic biological factors may also play an important role in the regulation of daily physical activity. These may include complex influences on brain circuitry related to personality, affect regulation, relative reinforcing, and reward processing, or via influencing cardiorespiratory and muscle capacity/function to regularly engage in physical activity.

These elements are likely to also interact with self-efficacy and other core components of social–behavioural models for both physical activity and sedentary behaviours. Since successful efforts to improve levels of physical activity in the population are likely contingent upon an accurate understanding of both biological and societal determinants of habitual activity, it will be important to better integrate such evidence into our understanding of the overall scientific picture.

Seems to me what we ultimately will do with specific biological or genetic information is debatable, as regardless of their relative importance, we all need to engage in physical activity to maintain our health. It could perhaps help with more precise and effective targeting of physical activity interventions to those who may benefit from it most, but it is equally possible that we will get more “bang for our buck” by improving societal and environmental infrastructure to support active living for everyone.