Values are the observed rate (blue line) of all Google searches including the term suicide (excluding searches also mentioning squad representing searches for the movie Suicide Squad) and expected search rates (projection of historical trends; orange line) after the Trump Administration’s declaration of a national emergency in response to the COVID-19 pandemic (dashed line). The shaded area represents differences between the observed and expected search rates.
Values are the cumulative excess for the 20 suicide-related searches with corresponding 95% CIs from March 8, 2020, through July 5, 2020.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Ayers JW, Poliak A, Johnson DC, et al. Suicide-Related Internet Searches During the Early Stages of the COVID-19 Pandemic in the US. JAMA Netw Open. 2021;4(1):e2034261. doi:10.1001/jamanetworkopen.2020.34261
Experts anticipate that the societal fallout associated with the coronavirus disease 2019 (COVID-19) pandemic will increase suicidal behavior, and strategies to address this anticipated increase have been woven into policy decision-making without contemporaneous data.1,2 For instance, President Trump cited increased suicides as an argument against COVID-19 control measures during the first presidential debate on September 29, 2020.
Given the time delays inherent in traditional population mental health surveillance, it is important for decision-makers to seek other contemporaneous data to evaluate potential associations.3 To assess the value that free and public internet search query trends can provide to rapidly identify associations, we monitored suicide-related internet search rates during the early stages of the COVID-19 pandemic in the US.
In this cross-sectional study, we replicated the methods of a previous study on tracking suicidal ideation.4 Weekly Google search rates (per 10 million searches) for the term suicide after excluding searches mentioning squad (a reference to a popular film) that originating from the US between January 1, 2010, and July 5, 2020, were monitored using the Google Trends application programming interface (Alphabet Inc). We also monitored the top 20 unique queries related to suicide after unrelated terms (eg, suicide slide) had been excluded. This study was exempted from ethical review and certified as not qualifying as human participant research by the University of California San Diego Human Research Protections Program. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Changes in search rates were compared before and after the US declaration of a national emergency for the COVID-19 pandemic during the second week of March 2020. Search rates between January 1-7, 2010, and March 1-7, 2020, were used to forecast expected search rates between March 8-14, 2020, and July 5-12, 2020, using an autoregressive integrated moving average model. The ratio of observed and expected search rates with bootstrapped 95% CIs was computed weekly and cumulatively. Absolute volume of searches was estimated by multiplying the search rates by total search estimates from comScore.com (comScore Inc). Analyses were performed using R, version 3.6.1 (R Project for Statistical Computing)
All queries containing the term suicide cumulatively decreased by 22% (95% CI, 18%-26%) in the 18 weeks after President Trump declared a national emergency and never eclipsed their expected search rate for any week (Figure 1). In raw terms, this was approximately 7.8 million fewer searches than expected.
Moreover, searches for 15 of the 20 related terms significantly decreased, including suicide note (–47%; 95% CI, −52% to −43%), suicidal thoughts (−20%; 95% CI, −24% to −16%), and suicidal ideation (−22%; 95% CI, −28% to −15%), translating into approximately 245 000, 155 000, and 80 000 fewer searches than expected, respectively (Figure 2). The only search term that significantly increased was potentially associated with interest in suicide facts: how many people commit suicide (18%; 95% CI, 1%-36%; approximately 13 000 more searches than expected).
Internet searches for suicide, previously found to be associated with population changes in suicidal behavior,3,5 decreased during the early stages of the COVID-19 pandemic in the US. Although this study cannot independently confirm that changes in search rates were caused by changes in population-level suicide rates, it showed that COVID-19 may have been inversely associated with population suicide trends between March and July 2020. Counter to expectations, our early findings are supported by the literature on catastrophic events.6 In some cases, catastrophes are associated with increased social support and unify communities and are thereby associated with reduced suicidal outcomes (what experts call the “pulling together” phenomenon). These include acute events, such as the 1995 Great Hanshin-Awaji Earthquake, and long-lasting events, such as the response to the September 11 attacks on the US and the subsequent recovery efforts, and now potentially the COVID-19 pandemic. For instance, at the onset of the COVID-19 outbreak, business and community leaders adopted the United Nations’ “We are all in this together” campaign, a theme that has since permeated pop culture.
Still, search rates for information on suicide may change, even increase, especially given a prolonged pandemic, making continued monitoring crucial. Moreover, researchers can extend the approach that we used (including tracking online help-seeking searches and social media shares) to empirically assess complementary proxies for other population mental health outcomes. Decision-makers could track hundreds of mental health search queries, identify the subsets that have greater demand, and target resources to meet those needs. Timely, empirical evidence from contemporaneous digital data sources can help steer limited resources to align with the needs of the public and promote data-driven debate regarding the potential societal implications of the COVID-19 pandemic.
Accepted for Publication: November 30, 2020.
Published: January 21, 2021. doi:10.1001/jamanetworkopen.2020.34261
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Ayers JW et al. JAMA Network Open.
Corresponding Author: John W. Ayers, PhD, MA, Division of Infectious Diseases and Global Public Health, Department of Medicine, University of California San Diego, #333 CRSF 9500 Gilman Dr, La Jolla, CA 92093 (email@example.com).
Author Contributions: Drs Ayers and Nobles had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Ayers, Poliak, Leas, Dredze, Nobles.
Acquisition, analysis, or interpretation of data: Ayers, Johnson, Leas, Caputi, Nobles.
Drafting of the manuscript: Ayers, Leas, Caputi, Nobles.
Critical revision of the manuscript for important intellectual content: Poliak, Johnson, Leas, Dredze.
Statistical analysis: Leas, Caputi.
Obtained funding: Nobles.
Administrative, technical, or material support: Dredze.
Conflict of Interest Disclosures: Dr Ayers reported owning equity positions in Directing Medicine LLC, Health Watcher Inc, and Good Analytics Inc, outside the submitted work. Dr Poliak reported being a paid advisor for Bloomberg LP, MyFitnessPal, and Lincoln Labs. Dr Leas reported receiving personal fees from Good Analytics outside the submitted work. Dr Dredze reported holding equity in Good Analytics, receiving consulting fees from Directing Medicine, and receiving financial support from Sickweather and Bloomberg LP outside the submitted work. Mr Caputi reported receiving personal fees from Good Analytics and having an equity interest in Data Science Solutions outside the submitted work. No other disclosures were reported.
Funding/Support: This work was supported by grants 587873 and R00RG2545 from the University of California Office of the President Research Grants Program Office.
Role of the Funder/Sponsor: The University of California Office of the President Research Grants Program Office had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.