A session at AcademyHealth's Annual Research Meeting (ARM), "Harnessing Internet Search Data as a Potential Tool for Medical Diagnosis," brought together a panel of researchers and experts to discuss using internet search data for early diagnosis and health research. This session provided a comprehensive overview of the project, shared pioneering work, introduced pilot projects, and concluded with perspectives from the panelists.
The session featured:
- Greg Downing, D.O., Ph.D., Innovation Horizons, Inc.
- Henry Kautz, Ph.D., University of Virginia
- Michael Gluck, Ph.D., M.P.P., AcademyHealth
- Anna McCollister, Patient Advocate and Health Technology Innovator
- Jane Hamilton, Ph.D., M.P.H., LCSW, University of Texas Health Science Center Houston, McGovern Medical School
- John Torous, M.D., M.B.I. Beth Israel Deaconess Medical Center
Why This Session?
The ARM panel builds on work currently underway at AcademyHealth with funding from the Gordon and Betty Moore Foundation and in collaboration with colleagues at Innovation Horizons, which supports the integration of innovative technologies and services into value-based health care. The initiative addresses the scarcity of peer-reviewed studies leveraging search data for health research as well as tackles questions raised about the appropriateness of using such data.
AcademyHealth’s project assesses the utility of internet search data in aiding diagnoses, identify barriers to research, identifying best practices, and encouraging methodologies that integrate internet search data with health research data to identify diagnoses and conditions effectively.
The ARM session aimed to broaden the conversation by raising awareness among the larger health services research community about the potential applications of internet search data in medical research. Panelists discussed challenges, shared successes, and explored opportunities for collaboration, emphasizing overcoming technical, ethical, and methodological challenges. Two experts with extensive knowledge and experience in research methodologies for large data modeling and uses of data to support medical applications provided the following overview of this endeavor:
Henry Kautz provided an overview of the research landscape, highlighting the potential of internet search data to revolutionize early diagnosis and patient care, including the findings from a literature review conducted through this project. He discussed innovative methodologies and the importance of robust frameworks for integrating search data with clinical health data.
Anna McCollister shared a patient’s perspective, expressing cautious optimism about the research implications of patients sharing their search history with researchers and emphasizing the necessity of collaborating with patients. She highlighted the balance between maintaining patient privacy and leveraging data for health insights, reminding researchers of the human element central to their work.
Pilot Projects
The session introduced four pilot projects AcademyHealth has funded as part of this initiative:
- Jane Hamilton, Ph.D., M.P.H., LCSW described her project at the University of Texas Health Science Center Houston, McGovern Medical School, investigating the potential of search data to aid in early detection of chronic diseases, particularly alcohol use disorder. Her research aims to develop scalable methodologies for various health conditions.
- John Torous, M.D., M.B.I. discussed his project at Beth Israel Deaconess Medical Center, focusing on using internet search data to identify early signs of mental health conditions. His team explores ways to integrate search data with clinical records to enhance diagnostic accuracy and patient care.
- Dr. Elizabeth O'Donnell, M.D., DipABLM at Dana-Farber Cancer Institute leads a project on lung cancer, exploring whether analyzing internet search histories can facilitate early detection. This approach aims to address the late-stage diagnosis of lung cancer, the leading cause of cancer deaths in the U.S.
- Dr. Munmun De Choudhury, Ph.D. at Georgia Tech’s School of Interactive Computing leads a project on mental health disparities in minority communities. This project uses internet search data to uncover and address these disparities, aiming to develop culturally sensitive mental health interventions.
Work Groups
A critical component of this project is the establishment of two work groups comprised of subject matter experts who are developing tools that assist the health research community tackle technical challenges and address the human subjects and ethical concerns for this research. These groups act as a resource for researchers to seek advice and ask clarifying questions.
Conclusion
The ARM session on Harnessing Internet Search Data as a Potential Tool for Medical Diagnosis highlighted innovative work in this field and underscored the importance of collaboration among researchers, patients, and stakeholders. As the project progresses, the insights and methodologies developed will pave the way for new research avenues, enhancing our ability to diagnose and treat conditions earlier and more accurately. A video of the session can be viewed here.