Related Experiment Video
Updated: Nov 25, 2025
Visualization of Intensity Levels to Reduce the Gap Between Self-Reported and Directly Measured Physical Activity
Published on: March 7, 2019
Satellite Imaging-Based Residential Greenness and Accelerometry Measured Physical Activity at
Soile Puhakka1,2,3, Tiina Lankila1,3, Riitta Pyky1,4,5
1Department of Sports and Exercise Medicine, Oulu Deaconess Institute Foundation sr, PO Box 365, 90100 Oulu, Finland.
Background:
Recently, the importance of light physical activity (LPA) for health has been emphasized, and residential greenness has been positively linked to the level of LPA and a variety of positive health outcomes. However, people spend less time in green environments because of urbanization and modern sedentary leisure activities.
Aims:
In this population-based study, we investigated the association between objectively measured residential greenness and accelerometry measured physical activity (PA), with a special interest in LPA and gender differences.
Methods:
The study was based on the Northern Finland Birth Cohort 1966 (5433 members). Participants filled in a postal questionnaire and underwent clinical examinations and wore a continuous measurement of PA with wrist-worn Polar Active Activity Monitor accelerometers for two weeks. The volume of PA (metabolic equivalent of task or MET) was used to describe the participant's total daily activity (light: 2-3.49 MET; moderate: 3.5-4.99 MET; vigorous: 5-7.99 MET; very vigorous: ≥8 MET). A geographic information system (GIS) was used to assess the features of each individual's residential environment. The normalized difference vegetation index (NDVI) was used for the objective quantification of residential greenness. Multiple linear regression and a generalized additive model (GAM) were used to analyze the association between residential greenness and the amount of PA at different intensity levels.
Results:
Residential greenness (NDVI) was independently associated with LPA (unadjusted β = 174; CI = 140, 209) and moderate physical activity (MPA) (unadjusted β = 75; CI = 48, 101). In the adjusted model, residential greenness was positively and significantly associated with LPA (adjusted β = 70; CI = 26, 114). In men, residential greenness was positively and significantly associated with LPA (unadjusted β = 224; CI = 173, 275), MPA (unadjusted β = 75; CI = 48, 101), and moderate to vigorous physical activity (MVPA) (unadjusted β = 89; CI = 25, 152). In women, residential greenness was positively related to LPA (unadjusted β = 142; CI = 96, 188) and inversely associated with MPA (unadjusted β = -22; CI = -36, -8), vigorous/very vigorous physical activity (VPA/VVPA) (unadjusted β = -49; CI = -84, -14), and MVPA (unadjusted β = -71; CI = -113, -29). In the final adjusted models, residential greenness was significantly associated only with the amount of LPA in men (adjusted β = 140; CI = 75, 204).
Conclusions:
Residential greenness was positively associated with LPA in both genders, but the association remained significant after adjustments only in men. Residential greenness may provide a supportive environment for promoting LPA.
