TABLE 2.

Overcoming challenges to the use of big data in obesity research¹

Challenge	Potential solutions
Privacy	Automatic local/offline prechecks built into apps for identity-violating content Process location locally to geohashes or other aggregated values Assign tiered data access to investigators Involve domain experts/researchers, educational and citizens’ groups (end users) in privacy co-design
User bias	Align with city-lab and health equity and health promotion initiatives Specific targeting of communities of lower socioeconomic position via schools, clinics, and community organizations Control for access to health care
Veracity	Domain-specific standards for specific types of noisy, error-prone data and volume reduction Open-access, user-friendly tools to facilitate implementation of data standards
Legacy and FAIR data	Integrate big data in nutrition and public health with emerging research infrastructures Align emerging data and projects to the European Open Science Cloud and surveillance efforts [e.g., INFORMAS (81, 103)]
Industry vs. research interest	Ethical forums, conferences, and guidelines for big data industry–academic partnerships Research training in big data techniques and communication with technical experts

¹FAIR, findable, accessible, interoperable, reusable.