Neighborhood Activity Hotspots for Multi-Ethnic Youth in Copenhagen, Denmark – Using GPS, Accelerometry and GIS: The WCMC Study

Background Changes in physical activity (PA) levels can be observed among youth. The majority of 11 year olds in Denmark do meet the recommend level of PA; the majority of 15 year olds do not. The chances of not meeting the recommended level of PA as adults are correlated with not meeting them at 15. For that reason, reducing the number of 15 year olds that do not meet the recommended level of PA is of great interest. There are many hypotheses as to what causes these changes, but no consensus has been reached. One hypothesis currently explored in Denmark is that youth are more active if they live in an attractive neighborhood with abundant possibilities for them to be physically active. However, relatively little is known about where youth like to be active, the activity hotspots, and how these sites look. A large scale neighborhood renewal project, with focus on increasing the possibilities for youth to be active, is currently being executed by the City of Copenhagen. This neighborhood renewal project provided a unique opportunity to study the changes in activity levels in this neighborhood in a natural experiment that will include both baseline and follow-up measurements.

Objectives The overall objective of the When Cities Move Children (WCMC) study is to determine the effects of urban renewal on objectively measured physical activity levels and activity patterns among youth (10–16yrs old) living in a deprived area with 40% having a multi-ethnic background, in Copenhagen, Denmark. The specific objective for this paper is to identify and describe hotspots for physical activity of the study participants at baseline.

Methods For our baseline study, carried out in spring 2010, 551 youth enrolled at three public schools were asked to wear an accelerometer (ActiGraph GT3X) and a GPS (Qstarz BT-Q1000X) for 7 days (5 week days, 2 weekend days) to determine their level of activity and movement patterns. Their GPS position was recorded every 15 seconds and their activity level was recorded every 2 seconds. Accelerometer data were compiled using the computer software Propero, recently developed by researchers at the University of Southern Denmark. Data were examined for extreme accelerometer values (above 20.000 counts and negative values) and participants were included in further analysis if they had at least 4 valid days (including at least 1 weekend day) of at least 8 hours wear time between 6am and midnight. Non-wear was defined as 60 or more minutes of consecutive zeros, allowing for two activity epochs in each block of non-wear. All GPS and accelerometer data were compiled and joined using an internet based computer program, the Physical Activity Location Measurement System (PALMS), developed by researchers at the Center for Wireless & Population Health Systems at the University of California, San Diego. The outputs PALMS produces consist of filtered and cleaned GPS points linked with PA data for those points. All PALMS outputs were imported into ArcGIS, a Geographic Information Software package, which enabled inclusion of environmental data. ArcGIS served as platform for further analysis and identification of neighborhood activity hotspots. Neighborhood activity hotspots were defined as all locations with a least 5 different individual hotspots. For each participant individual hotspots consist of all locations where at least 4 consecutive GPS points that are part of an activity bout can be found within a 50 meter radius. An activity bout was defined as a period of at least 5 minutes of at least moderate activity (using Freedsons age adjusted cut-off points), allowing for maximum 2 minutes of activity below this threshold. Results Accelerometer data were downloaded for 454 participants, and 271 participants qualified for inclusion in further analysis. Based on data from these participants, neighborhood activity hotspots primarily consist of schoolyards, sports facilities and shared backyards between multistory social housing complexes. Our results also show age, gender and time differences with different areas being popular for different age groups, at different times of the day, and at different days of the week.

Conclusions Shared backyards to large social housing areas are crucial places for PA among many of our respondents, suggesting that these sites deserve more attention in future studies. Our study furthermore shows the potential of combining accelerometer, GPS and GIS data, but the large amount of data, and need for specialized software and advanced GIS tools to work with the data means that many methodological and technical challenges remain to be solved.

Link to Abstract: http://www.activelivingresearch.org/node/12640