Using Google Takeout Donations for Suicide Prevention Research: Study Design and Lessons Learned (Preprint)

BACKGROUND Social media and search engine tech companies, such as Google, use the wealth of data at their disposal to create algorithms to predict suicide risk, and provide interventions based on company policies and outputs from these algorithms. Neither the details of these algorithms or their accuracy in risk prediction are shared with the public, thus research is needed to verify whether algorithms applied to search engine data can accurately predict suicide risk in a platform user. The onus is on researchers to study the utility of these data for prediction, which requires donation of search data from people with lived experience of suicidal thoughts and behavior. There is little guidance, however, from the research community or tech companies as to how to best assist participants in sharing their data easily and securely. OBJECTIVE This paper describes study methods and experiences recruiting and retaining participants in a year long, fully remote prospective study of suicide prediction in people with variable risk for suicide, using Google Search data. METHODS We have recruited 485 United States dwelling and English-speaking adults who fall into one of 3 categories: participants with a suicide attempt in the past year, participants with a lifetime suicide attempt but not one in the past year, and participants with no lifetime suicide attempt. RESULTS We describe important data supporting the methods we used to recruit a remote sample of participants with risk for suicide, Google Search data quality, amount of fraudulent acting, and retention at each study phase and follow up assessment. CONCLUSIONS Despite past success with remote clinical research methods, we found that recruitment can be significantly hampered by fraudulent acting, despite the use of state-of-the-art methods to mitigate this problem. Further, the ability to acquire useful Google Search data is extremely challenging and necessitates the use of technical assistance and continuous monitoring of Google’s Takeout data structure to ensure accurate data downloads. The solutions we provide for this type of research should be helpful for future endeavors. However, research is still limited by the barriers to obtaining data from technology companies. CLINICALTRIAL Not applicable – not a clinical trial