Evidence from big data in obesity research: international case studies

Area of Science:

• Public Health
• Epidemiology
• Data Science

Background:

• Obesity is a complex condition influenced by over 100 interacting factors.
• Traditional data collection methods for obesity research are costly and time-consuming.
• Big data, characterized by digital, large-sample, and multi-variable datasets, offers a potential solution.

Purpose of the Study:

• To showcase international research case studies on the use of big data in obesity research.
• To provide an in-depth view of the benefits, limitations, and challenges of big data in this field.
• To explore how big data analytics can enhance our understanding of obesity drivers.

Main Methods:

• Presentation of three international case studies from an Economic and Social Research Council (ESRC) seminar series.
• Case study 1: Spatial and individual-level data to assess the built environment's influence on physical activity.
• Case study 2: Linked electronic health datasets to investigate obesity surgery and cancer risk.
• Case study 3: Tax parcel values and survey data to examine sociodemographic determinants of obesity.

Main Results:

• Big data can augment traditional data, capturing a wider range of variables in the obesity system.
• Big data offers improvements in data size, coverage, temporality, and objectivity compared to traditional methods.
• Case studies highlighted challenges such as hidden biases and lack of contextual information.

Conclusions:

• Big data presents a valuable, yet underutilized, resource for obesity research.
• Despite limitations, big data holds significant promise for advancing our understanding of obesity drivers.
• Continued research and methodological development are crucial for maximizing big data's potential in public health.