Evaluating the inverse, non-linear dose-response relationship between physical activity and the risk of disease and early death
A recent article published in the BMJ has attempted to further define the dose-response relationship between physical activity and the risk of all-cause, CVD-related and Cancer-related mortality and incidence.
The new research incorporates a sample size far greater than any previous publication in this space, utilising extensive methods to harmonise ‘exposure data’ and draw conclusions from an evidence base consisting of over 160 million person-years and 810,000 deaths. These figures are 17- and 7-fold larger, respectively, than those included in the previously largest physical activity dose–response analysis.
The top-line outcomes are unsurprising. In general, higher physical activity levels were associated with a lower risk of all negative health outcomes, and the differences in risk were greatest between 0 and 8.75 marginal metabolic equivalent of task-hours per week (mMET-hours/week), roughly equivalent to 150 mins/week of moderate-to-vigorous intensity physical activity (MVPA). This fits well with the existing research base, and provides further evidence to validate the inverse, non-linear dose-response relationship between physical activity and disease risk.
A particularly interesting outcome from the study relates to the benefits of doing ‘some’ physical activity versus doing nothing. The meta-analysis found that 1 in 10 premature deaths could potentially be avoided by doing the equivalent of 75 mins/week of MVPA. A number of media outlets have picked up on this, reducing their headlines to simple soundbites along the lines of: “Just 11 mins of activity per day can reduce risk of early death.” Whilst the findings can indeed be interpreted in this manner, we think there are two important points to consider.
Firstly, the risk reduction the authors are referring to here relates specifically to people who are completely inactive to begin with. If a previously inactive individual were to increase their activity level to achieve 75 mins/week of MVPA, 1 in 10 premature deaths would be avoided. This does not mean those individuals already doing more than this should reduce their activity level to match, as this would be detrimental to their health. Secondly, we must emphasise that the researchers assigned standardised mMET-hr values for different physical activity intensities, which unfortunately cannot account for the different types of daily activities performed at that specific intensity. As such, the activity intensities, and the subsequent public health-based messaging, are often boiled down to single activities (i.e., walking is good for you), with little recognition for everyday activities of a similar intensity (i.e., shopping, vacuuming can also be good for you). A simple tweak to the language and messaging would make the benefits of physical activity far more accessible and appealing to all.
Due to the sheer number of individual studies included in the meta-analysis to form the substantial sample size, many different methods for assessing physical activity are used across the different studies. Interestingly, the researchers acknowledged stronger associations between physical activity and health outcomes in studies which used objective physical activity monitoring, as opposed to self-report measures. This perhaps reflects the inconsistencies and inaccuracies associated with self-reported physical activity, which relies on the accuracy of an individual’s understanding and recall. Another small compromise made by the researchers, likely as a result of the large sample size, is the exclusion of occupational physical activity. Whilst there are a number of difficulties in accurately assessing occupational physical activity, particularly when using self-reported measures, it does make up a significant proportion of a typical adult’s daily lifestyle. As such, any analysis which excludes this specific component of physical activity must be mindful of this as a limitation, and we hope to see future studies build on this gap in the evidence base.
In summary, despite a number of limitations, many of which are highlighted by the authors, this new research adds significant weight to the inverse, non-linear dose-response relationship between physical activity and disease risk. Whilst it represents a step forward in the consolidation of the evidence base, concerns remain around how the findings are reported in the wider press. We understand the need to simplify the outcomes for a lay audience, however, the ‘one-size fits all’ messaging risks portraying physical activity as unidimensional, which is simply not true. From our experience, an alternative approach, which rethinks physical activity as far more than a set of simplistic, structured activities will be far more appealing and accessible to the wider population. Finally, now that the basic inverse, non-linear dose-response theory has been appropriately consolidated, we hope to see future studies adopt new, objective methods of PA assessment to enable wider analysis of ‘all’ daily physical activity, not just that which is non-occupational. This will enable the identification of the independent benefits and risks of different daily/weekly physical activity profiles, and support a personalised approach to physical activity and its potential impact on individual lives.
