Emerging ethical considerations for the use of geospatial data in research on the social determinants of health

There is a growing interest in using geospatial data to explore associations between place characteristics, behaviors, and health outcomes. However, using social-spatial data, which links a person to a geographic location (e.g., home, work), for research on the social determinants of health also jeopardizes participant confidentiality. The potential for social harms associated with breaches in confidentiality is particularly salient when exploring associations among place and stigmatizing behaviors or conditions, such as sexual risk behaviors and HIV serostatus.

To date, standard Good Clinical Practice and Human Subjects Protection certifications do not include a module on use of social-spatial data nor are there universally accepted guidelines for using linked social-spatial data for social and behavioral research. Recent software advances and the availability of primary and secondary datasets that link spatial information (e.g., participant home address or census tract) to participant study data have created new opportunities to explore these relationships and to disseminate results using maps. Public health professionals may be unaware of what types and units of geospatial data constitute protected health information, the risks associated with using linked social-spatial data (e.g., reverse identification) and approaches to mitigate these risks (e.g., geomasking, data sharing strategies) while ensuring the integrity of research findings.

Using HIV/AIDS as a case study, we will explore ethical considerations for using linked social-spatial data for research on the social determinants of health, with a focus on: 1) what types and units of geospatial data are considered protected health information; 2) how confidentiality considerations may vary depending on the user (e.g., primary data collection vs. secondary analysis) and the research stage (e.g., analysis vs. dissemination), and; 3) balancing the need to protect participant confidentiality while preserving scientific integrity.