The investigators pooled accelerometer data from 40\u2008327 adults aged at least 40 years across seven cohorts in Sweden, Norway, and the USA. Participants averaged 9\u00b76 h\/day in sedentary time, 4\u00b78 h\/day in light-intensity activity, and 28 min\/day in moderate-to-vigorous intensity activity. To estimate the potential impact fraction (percentage of deaths averted), the authors first estimated mortality risks for substituting time spent in various behaviours\u2014eg, the risk associated with a 30-min reduction in sedentary time and a corresponding increase in total activity (ie, light-intensity and moderate-to-vigorous intensity activity). Next, they estimated the cumulative proportion of deaths averted for different hypothetical changes in time spent in each behaviour across the range of the exposure. Additionally, the effect of two distinct public health intervention strategies focused on the high-risk segment of the population (least active 20%) or a more general population approach (least active 80%) was estimated, and sensitivity analyses were conducted.<\/p>\n\n\n\n
Both intervention strategies were found to have a substantial impact. For example, increasing moderate-to-vigorous intensity activity in the population by 10 min\/day using the high-risk approach was estimated to avert 8\u00b78% of deaths annually, whereas the population approach averted 14\u00b79%. Deaths prevented through reductions in sedentary time and increased light-intensity and total physical activity were also substantial. For example, with the population intervention strategy, reducing sedentary time by 30 min\/day was estimated to avert 7\u00b73% of deaths annually, and increasing light-intensity activity by the same amount was associated with 5\u00b74% fewer deaths and total physical activity with 6\u00b71% fewer deaths. Sensitivity results considering gender, body size, and chronic conditions appeared robust, but evidence for adults younger than 60 years was limited, and results from the UK Biobank were weaker.<\/p>\n\n\n\n
This report has many strengths, including the use of harmonised accelerometer data from geographically diverse, population-representative cohorts, which enabled greater precision in the behaviour\u2013mortality dose\u2013response curves and, for the first time, impact estimates for sedentary behaviour, light-intensity activity, and total physical activity. However, some uncertainty persists about the magnitude of the impact of moderate-to-vigorous intensity activity on the deaths averted. The previous study3<\/sup><\/a> estimated about half the impact from a 10-min increase in moderate-to-vigorous intensity activity of 6\u00b79%, compared with 14\u00b79% of deaths averted reported here. Two key differences between the studies might offer insights. First, Ekelund and colleagues4<\/sup><\/a> assumed that everyone in the population was physically capable of increasing their activity level, whereas the other study3<\/sup><\/a> assumed that disability and frailty would limit the ability of some adults to increase their activity (eg, by walking for exercise). When this assumption was tested in Ekelund and colleagues\u2019 study in two cohorts by excluding adults with mobility limitations, the proportion of deaths averted was reduced by 15\u201320%. This analytical decision could also affect the magnitude of impact for the other behaviours examined. Second, the methods used to estimate moderate-to-vigorous intensity activity differed between studies, and this variation also affected the results. When Ekelund and colleagues4<\/sup><\/a> used an alternate method in sensitivity analyses, the proportion of deaths averted was substantially lower for the same increase in moderate-to-vigorous intensity activity. We agree that there is no consensus on the best method for measuring moderate-to-vigorous intensity activity, and that the venerable Freedson method5<\/sup><\/a> and others developed more than 25 years ago have been informative within the earliest accelerometer-based mortality studies pooled in this report. However, there is a need for stronger evidence regarding the primary method used here, since several other viable methods have been shown to produce higher values.6<\/a>, 7<\/a> By contrast, empirical support exists for the accuracy of the method used to estimate sedentary time and, consequently, total activity, in comparison with strong reference measures.8<\/a>, 9<\/a> High-quality performance-evaluation studies are needed to strengthen our accelerometer-based measurement methods for public health applications.10<\/a>, 11<\/a>, 12<\/a> That said, the methodological differences noted here probably affect only the magnitude of the impact for moderate-to-vigorous intensity physical activity but do not alter the more comprehensive findings and importance of this report.<\/p>\n\n\n\n
![\"\"\/](\"https:\/\/ars.els-cdn.com\/content\/image\/1-s2.0-S0140673626000930-fx1_lrg.jpg\")
<\/figure>\n","protected":false},"excerpt":{"rendered":"