Accelerometer-measured physical activity and life expectancy: a lifetable analysis
Abstract Overview
Background:
Low physical activity (PA) levels are associated with increased mortality. Improved measurement via accelerometery has shown a stronger association between PA and all-cause mortality compared with self-reported survey data, but this has not yet been translated to estimates of the population level disease burden attributable to low PA.
Purpose:
To estimate by how much low physical activity reduces life expectancy, and how much lifetime could be gained by increasing physical activity levels for both populations and individuals.
Methods:
We used device-measured physical activity risk estimates of all-cause mortality and a lifetable model analysis to construct a model of the 2019 American population aged 40+ years. Mortality data were from the National Centre for Health Statistics, 2017.
Results:
If all were as active as the top 25% of the population, Americans over the age 40 could expect to live an extra 5.3 years (95% uncertainty interval 3.7 to 6.8) on average. The greatest average gain in lifetime per single hour of walking was seen for individuals in the lowest activity quartile, where a one-hour walk could add 376.3 minutes (~6.3 hours) to life expectancy (95% uncertainty interval 321.5 to 428.5 minutes), on average.
Conclusions:
Physical activity provides substantially larger health benefits than previously thought.
Practical implications:
Increased investment in physical activity promotion and creating physical activity promoting living environments can yield large health gains.
Funding:
Previous funding from the New South Wales Ministry of Health, Sydney, Australia, is gratefully acknowledged.
Additional Authors
Name: Jakob Tarp
Affiliation: Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital
Presenting Author: yes
Name: Ruth Wijaya
Affiliation: School of Medicine and Dentistry, Griffith University, Gold Coast, Queensland, Australia
Presenting Author: no
Name: Mary Wanjau
Affiliation: Public Health & Economics Modelling Group, School of Medicine and Dentistry, Griffith University, Gold Coast, Queensland, Australia
Presenting Author: no
Name: Holger Möller
Affiliation: School of Population Health, University of New South Wales, Sydney, Australia
Presenting Author: no
Name: Fiona Haigh
Affiliation: Health Equity Research and Development Unit (HERDU), University of New South Wales, Sydney, Australia
Presenting Author: no
Name: Peta Lucas
Affiliation: Centre for Population Health, NSW Ministry of Health, Sydney, Australia
Presenting Author: no
Name: Andrew Milat
Affiliation: School of Public Health, University of Sydney, Camperdown, New South Wales, Australia
Presenting Author: no
Information
Study Type: Epidemiology
Subtheme: Other (please specify)
Subtheme other Title: Burden of Disease
Population Group: Adults
Setting: Whole System
Submitting Author
Lennert Veerman
Professor of Public Health
Griffith University
Australia
