AcademyHealth, with support from The Gordon and Betty Moore Foundation, is funding health services research projects to understand the potential and challenges of using internet search data to improve diagnosis. The goal of this study is to address questions related to the feasibility and acceptability of leveraging private internet search data in an oncology population. The internet serves as a readily accessible and vast source of information, making it a common platform for individuals seeking medical information. Whether information on internet search histories can be harnessed to identify individuals requiring clinical surveillance is unclear, but the potential value would be considerable: lung cancer is the leading cause of cancer death in the US, and the vast majority of diagnoses occur at late stages. This study will gather preliminary evidence of specific search terms that are most common prior to a lung cancer diagnosis and evaluate the correlation with the stage of the cancer. In particular, the researchers will ascertain the feasibility of collecting internet search histories via Google Takeout from patients recently diagnosed with lung cancer. They will identify 50 patients at the Dana-Farber Cancer Institute (25 early stage (I-III), 25 late stage (IV)). They will consent patients for permission to request their internet search history prior to diagnosis. They will analyze their data using Natural Language Processing (NLP) to look for differences in search terms and other search query factors that may correlate with an early versus a late-stage diagnosis. Deliverables will include a project work plan, narrative and financial reports, and a range of products to reach researchers, funders, clinicians and other audiences with study findings.

Principal Investigator(s)


Elizabeth O’Donnell, M.D.

Director of Early Detection and Prevention - Dana-Farber Cancer Institute

Dr. O’Donnell received her medical degree from Vanderbilt University. Read Bio