A new R package to derive activity space metrics: Illustration with map-based questionnaire and well-being measures from Canadian COHESION cohort data in Canada and Montreal

Vortrag
Sitzungstermin
Mittwoch (20. September 2023), 16:30–18:00
Sitzungsraum
HZ 9
Autor*innen
Alexandre Naud (Université de Montréal)
Gregory Moullec (Hôpital du Sacré-Coeur de Montréal)
Benoit Thierry (Université de Montréal)
Camille Perchoux (Luxembourg Institute of Socio-Economic Research (LISER))
Philippe Gerber (Luxembourg Institute of Socio-Economic Research (LISER))
Basile Chaix (INSERM)
Yan Kestens (Université de Montréal)
Kurz­be­schreib­ung
We present a novel open-source R package with documented algorithms to facilitate the calculation of various activity space indicators. An empirical application of the package will be demonstrated using activity space and mental health data collected in Canada in the COHESION longitudinal survey.

Abstract

Availability of daily mobility data is increasing, as map-based questionnaires, GPS tracking, or cellphone location data become ubiquitous. To fully harnest their potential, such activity space data needs to be transformed into meaningful indicators. While the use activity space indicators have been increasing in health research in the last decade, there is currently no shared tool that provides robust ways to compile activity space indicators. As a consequence, indicators used by researcher can take many different forms, which limits comparability between studies. Some lack of information about how these indicators were processed does also limit reproductibility.

To address this issue, our team has developed an open-source R package with documented algorithms to facilitate the calculation of various activity space indicators. The package, meant to be updated with contributions from the research community, currently includes algorithms to derive spatial distribution indicators from activity locations, wether these are obtained from map-based surveys, GPS tracks, or cellphone location data. It also includes a visualization interface designed as a web application for rapid data exploration, and is designed to integrate seamlessly with other available analysis tools. For example, it can be used to describe distributions of exposure to environmental factors obtained from geographic information systems.

We will present the new activity space package, and demonstrate its applicability using activity location data obtained from Phase 2 of the COHESION project. COHESION (www.cohesionstudy.ca) is a Canadian cohort survey launched in Spring 2020, collecting data among Canadian adults to generate evidence about environmental and social determinants of health inequities. We will present spatial indicators using data collected through the VERITAS questionnaire in Spring 2022 among some 1,866 participants residing in metropolitan areas across Canada. Links between activity space indicators, social network indicators and measures of well-being will be discussed, along with recommendations for future activity space and health research.