- Big data that are relevant for oncology disparities research exist in a variety of formats; these include administrative claims, national cancer registries, electronic health records, and genomic data.
- Big data for oncology disparities research are united by their volume, complexity, and requirements for novel technological and analytic methods in order to extract useful and reliable information.
- Analyses based on big data can be compromised when researchers fail to understand the populations, biases, and limitations, as well as the strengths, of each unique data resource. However, analyses can be strengthened and made more representative of vulnerable populations if resources are linked together to offset these limitations.
- The next frontiers of big data research in oncology require increasing collaboration among interdisciplinary teams of clinical, laboratory, and information technology, as well as statistical experts willing to share methods and data between fields in secure, standardized ways.