Forecasting state-level premature deaths from alcohol, drugs, and suicides using Google Trends data

Seasonal Trends in Suicide Attempts-Keywords Related Searches: A Google Trends Analysis

Suicide is a significant public health concern globally, with its varying rates influenced by numerous factors, including seasonal changes. Online search behaviors, particularly searches related to suicide and mental health, have been proposed as real-time indicators of suicidal ideation in populations. In this study, a cross-sectional time series analysis was conducted, utilizing data on suicide attempts from the Polish Police Headquarters and online search behavior from Google Trends over a decade. Suicide attempt data were analyzed alongside the frequency of Google searches for suicide-related keywords derived from the Polish Corpus of Suicide Notes. A total of 66 keywords were selected for analysis to identify seasonal trends and patterns in search behavior. The study employed linear regression, Seasonal Mann-Kendall tests, and TBATS models to analyze the data. Suicide rates show seasonal patterns, peaking in warmer months. However, keyword searches did not strongly correlate with peak suicide months. This study enhances our understanding of suicide-related search trends and their potential connection to suicide rates. It suggests avenues for more effective prevention efforts and the potential for future algorithms to predict suicide rates and identify at-risk groups.

Google Trends for health research: Its advantages, application, methodological considerations, and limitations in psychiatric and mental health infodemiology

The high utilization of infodemiological tools for psychiatric and mental health topics signals the emergence of a new discipline. Drawing on the definition of infodemiology by Eysenbach, this emerging field can be termed “psychiatric and mental health infodemiology,” defined as the science of distribution and determinants of information in an electronic medium, including the internet, or in a population to inform mental health services and policies. Since Google Trends is one of its popular tools, this minireview describes its advantages, application, methodological considerations, and limitations in psychiatric and mental health research. The advantage of Google Trends is the nature of its data, which may represent the actual behavior rather than their users' stated preferences in real-time through automatic anonymization. As such, it can provide readily available data about sensitive health topics like mental disorders. Therefore, Google Trends has been used to explore public concerns, interests, and behaviors about psychiatric and mental health phenomena, service providers, and specific disciplines. In this regard, several methodological can be considered by studies using Google Trends, including documenting their exact keywords, query category, time range, location, and date of retrieval. Likewise, its limitations should be accounted for in its interpretation, including restricted representation of people who use the Google search engine, limited validity in areas with low internet penetration or freedom of speech, does not provide absolute search volumes, unknown sampled queries, and limited transparency in its algorithm, especially the terms and idioms it subsumes under its “topic” keywords.

Trends in public interest in child abuse in the United States: An infodemiology study of Google Trends from 2004 to 2022

INTRODUCTION

More than 1 in 7 children in the United States experience abuse annually with rates remaining consistent over the past 2 decades. During this timeframe, several high-profile cases of child abuse and neglect were publicized in national media in addition to multiple investigations uncovering Indigenous children dying from abuse at Indian Boarding Schools. Increased media attention among other public health and medical topics has been linked to increased public interest, thus, our objective was to investigate trends in public interest from 2004 to 2022.

METHODS

To assess trends in public interest, we extracted monthly relative search interest in child abuse from Google Trends. We constructed linear regression to determine the long-term trajectory of interest, and also compared the slope of the trend to other topics, such as domestic violence. Further, we compared mean relative search interest (RSI) from Child Abuse Awareness Month (April) to other months via t-test. Lastly, we assess by-state correlations of RSI and number of children abused.

RESULTS

Since 2004, search interest in child abuse has significantly declined in the United States-more than other related search terms. Child Abuse Awareness Month showed spikes in RSI which were greater than other months. By-state correlations of RSI and abuse were moderate to weak.

CONCLUSION

Despite heavy media attention covering stories of child abuse during the past 2 decades, search interest in child abuse has significantly declined. This trend may be related to aversion to secondary traumatic stress as news broadcasts often include stories of violence-of which child abuse stories may be most provoking. Following journalism guidance from the Centers for Disease Control and Prevention, reporting with focus on resiliency and prevention, rather than the individuals who perpetrated the crime, may provide more community support and increased public interest.

Internet search patterns for psychoactive substance use prevention and treatment in Mexico: A cross-sectional study

Objectives

This study was aimed at describing the patterns of searches for information on the prevention and treatment of psychoactive drug use in Mexico, among both the general population and the personnel dedicated to the prevention and treatment of this type of substance use in Mexico.

Methods

An exploratory cross-sectional quantitative study was performed with a validated online questionnaire to collect sociodemographic information, background information and self-reported internet search patterns on psychoactive substance use prevention. A chi-square test was used to identify differences between groups, and a classification tree was used to analyze the search patterns. The combinations of the search criteria with the search topics were entered into Google Trends to validate the information.

Results

The participants (n = 544 adults) were mostly women (65%), 18-30 years of age and bachelor's degree holders (57%). A total of 32% were students, 59% searched the Internet for drug use prevention or treatment, and 12% professionally engaged in drug use prevention or treatment.

Conclusions

Statistically significant differences were found between the general population and professionals dedicated to drug dependency services. We identified six search patterns used in the decision-making process by people seeking information on drug prevention and treatment on the Internet. These patterns were graphically visualized with a classification tree, although, this method did not allow clear differentiation of patterns between groups. The search patterns were successfully validated with Google trends.

Digital Public Health Surveillance Tools for Alcohol Use and HIV Risk Behaviors

There is a need for real-time and predictive data on alcohol use both broadly and specific to HIV. However, substance use and HIV data often suffer from lag times in reporting as they are typically measured from surveys, clinical case visits and other methods requiring extensive time for collection and analysis. Social big data might help to address this problem and be used to provide near real-time assessments of people's alcohol use and/or alcohol. This manuscript describes three types of social data sources (i.e., social media data, internet search data, and wearable device data) that might be used in surveillance of alcohol and HIV, and then discusses the implications and potential of implementing them as additional tools for public health surveillance.

The Language of the Times: Using the COVID-19 Pandemic to Assess the Influence of News Affect on Online Mental Health-Related Search Behavior across the United States

Background

The digital era has ushered in an unprecedented volume of readily accessible information, including news coverage of current events. Research has shown that the sentiment of news articles can evoke emotional responses from readers on a daily basis with specific evidence for increased anxiety and depression in response to coverage of the recent COVID-19 pandemic. Given the primacy and relevance of such information exposure, its daily impact on the mental health of the general population within this modality warrants further nuanced investigation.

Objective

Using the COVID-19 pandemic as a subject-specific example, this work aimed to profile and examine associations between the dynamics of semantic affect in online local news headlines and same-day online mental health term search behavior over time across the United States.

Methods

Using COVID-19–related news headlines from a database of online news stories in conjunction with mental health–related online search data from Google Trends, this paper first explored the statistical and qualitative affective properties of state-specific COVID-19 news coverage across the United States from January 23, 2020, to October 22, 2020. The resultant operationalizations and findings from the joint application of dictionary-based sentiment analysis and the circumplex theory of affect informed the construction of subsequent hypothesis-driven mixed effects models. Daily state-specific counts of mental health search queries were regressed on circumplex-derived features of semantic affect, time, and state (as a random effect) to model the associations between the dynamics of news affect and search behavior throughout the pandemic. Search terms were also grouped into depression symptoms, anxiety symptoms, and nonspecific depression and anxiety symptoms to model the broad impact of news coverage on mental health.

Results

Exploratory efforts revealed patterns in day-to-day news headline affect variation across the first 9 months of the pandemic. In addition, circumplex mapping of the most frequently used words in state-specific headlines uncovered time-agnostic similarities and differences across the United States, including the ubiquitous use of negatively valenced and strongly arousing language. Subsequent mixed effects modeling implicated increased consistency in affective tone (Spin_VA β=–.207; P<.001) as predictive of increased depression-related search term activity, with emotional language patterns indicative of affective uncontrollability (Flux_A β=.221; P<.001) contributing generally to an increase in online mental health search term frequency.

Conclusions

This study demonstrated promise in applying the circumplex model of affect to written content and provided a practical example for how circumplex theory can be integrated with sentiment analysis techniques to interrogate mental health–related associations. The findings from pandemic-specific news headlines highlighted arousal, flux, and spin as potentially significant affect-based foci for further study. Future efforts may also benefit from more expansive sentiment analysis approaches to more broadly test the practical application and theoretical capabilities of the circumplex model of affect on text-based data.

Seasonal variations and public search interests in Toxoplasma: a 16-year retrospective analysis of big data on Google Trends

BACKGROUND

This study aims to understand whether there is a seasonal change in the internet search interest for Toxoplasma by using the data derived from Google Trends (GT).

METHODS

The present study searched for the relative search volume (RSV) for the search term 'Toxoplasma' in GT within six major English-speaking countries (Australia, New Zealand [Southern Hemisphere] and Canada, Ireland, the UK and the USA [Northern Hemisphere] from 1 January 2004 to 31 December 2019, utilizing the category of 'health'. Data regarding the RSV of Toxoplasma was obtained and further statistical analysis was performed in R software using the 'season' package.

RESULTS

There were significantly seasonal patterns for the RSV of the search term 'Toxoplasma' in five countries (all p<0.05), except for the UK. A peak in December-March and a trough in July-September (Canada, Ireland, the UK and the USA) were observed, while a peak in June/August and a trough in December/February (Australia, New Zealand) were also found. Moreover, the presence of seasonal patterns regarding RSV for 'Toxoplasma' between the Southern and Northern Hemispheres was also found (both p<0.05), with a reversed meteorological month.

CONCLUSIONS

Overall, our study revealed the seasonal variation for Toxoplasma in using internet search data from GT, providing additional evidence on seasonal patterns in Toxoplasma.

Association of the Timing of School Closings and Behavioral Changes With the Evolution of the Coronavirus Disease 2019 Pandemic in the US

Importance

The consequences of school closures for children's health are profound, but existing evidence on their effectiveness in limiting severe acute respiratory syndrome coronavirus 2 transmission is unsettled.

Objective

To determine the independent associations of voluntary behavioral change, school closures, and bans on large gatherings with the incidence and mortality due to coronavirus disease 2019 (COVID-19).

Design, Setting, and Participants

This population-based, interrupted-time-series analysis of lagged independent variables used publicly available observational data from US states during a 60-day period from March 8 to May 18, 2020. The behavioral measures were collected from anonymized cell phone or internet data for individuals in the US and compared with a baseline of January 3 to February 6, 2020. Estimates were also controlled for several state-level characteristics.

Exposures

Days since school closure, days since a ban on gatherings of 10 or more people, and days since residents voluntarily conducted a 15% or more decline in time spent at work via Google Mobility data.

Main Outcomes and Measures

The natural log of 7-day mean COVID-19 incidence and mortality.

Results

During the study period, the rate of restaurant dining declined from 1 year earlier by a mean (SD) of 98.3% (5.2%) during the study period. Time at work declined by a mean (SD) of 40.0% (7.9%); time at home increased by a mean (SD) of 15.4% (3.7%). In fully adjusted models, an advance of 1 day in implementing mandatory school closures was associated with a 3.5% reduction (incidence rate ratio [IRR], 0.965; 95% CI, 0.946-0.984) in incidence, whereas each day earlier that behavioral change occurred was associated with a 9.3% reduction (IRR, 0.907; 95% CI, 0.890-0.925) in incidence. For mortality, each day earlier that school closures occurred was associated with a subsequent 3.8% reduction (IRR, 0.962; 95% CI, 0.926-0.998), and each day of advance in behavioral change was associated with a 9.8% reduction (IRR, 0.902; 95% CI, 0.869-0.936). Simulations suggest that a 2-week delay in school closures alone would have been associated with an additional 23 000 (95% CI, 2000-62 000) deaths, whereas a 2-week delay in voluntary behavioral change with school closures remaining the same would have been associated with an additional 140 000 (95% CI, 65 000-294 000) deaths.

Conclusions and Relevance

In light of the harm to children of closing schools, these findings suggest that policy makers should consider better leveraging the public's willingness to protect itself through voluntary behavioral change.

Insights on the implications of COVID-19 mitigation measures for mental health

Given the unprecedented level and duration of mitigation policies during the 2020 COVID-19 pandemic, it is not surprising that the public and the media have raised important questions about the potential for negative mental health consequences of the measures. To answer them, natural variability in policy implementation across US states and over time was analyzed to determine if mitigation policies correlated with Google searches for terms associated with symptoms of depression and anxiety. Findings indicated that restaurant/bar limits and stay-at-home orders correlated with immediate increases in searches for isolation and worry but the effects tapered off two to four weeks after their respective peaks. Moreover, the policies correlated with a reduction in searches for antidepressants and suicide, thus revealing no evidence of increases in severe symptomatology. The policy implications of these findings are discussed.

Google searches for suicide and suicide risk factors in the early stages of the COVID-19 pandemic

A novel coronavirus (SARS-CoV-2), which causes the COVID-19 respiratory illness, emerged in December of 2019 and has since spread globally. The dramatic lifestyle changes and stressors associated with this pandemic pose a threat to mental health and have the potential to exacerbate risk factors for suicide. We used autoregressive integrated moving average (ARIMA) models to assess Google Trends data representing searches in the United States for 18 terms related to suicide and known suicide risk factors following the emergence of COVID-19. Although the relative proportion of Google searches for suicide-related queries was lower than predicted during the early pandemic period, searches for the following queries representative of financial difficulty were dramatically elevated: "I lost my job" (226%; 95%CI, 120%-333%), "laid off" (1164%; 95%CI, 395%-1932%), "unemployment" (1238%; 95%CI, 560%-1915%), and "furlough" (5717%; 95%CI, 2769%-8665%). Searches for the Disaster Distress Helpline, which was promoted as a source of help for those impacted by COVID-19, were also remarkably elevated (3021%; 95%CI, 873%-5169%). Google searches for other queries representative of help-seeking and general mental health concerns were moderately elevated. It appears that some indices of suicidality have fallen in the United States in this early stage of the pandemic, but that COVID-19 may have caused an increase in suicide risk factors that could yield long-term increases in suicidality and suicide rates.

Using Application Programming Interfaces to Access Google Data for Health Research: Protocol for a Methodological Framework

Background

Individuals are increasingly turning to search engines like Google to obtain health information and access resources. Analysis of Google search queries offers a novel approach, which is part of the methodological toolkit for infodemiology or infoveillance researchers, to understanding population health concerns and needs in real time or near-real time. While searches predominantly have been examined with the Google Trends website tool, newer application programming interfaces (APIs) are now available to academics to draw a richer landscape of searches. These APIs allow users to write code in languages like Python to retrieve sample data directly from Google servers.

Objective

The purpose of this paper is to describe a novel protocol to determine the top queries, volume of queries, and the top sites reached by a population searching on the web for a specific health term. The protocol retrieves Google search data obtained from three Google APIs: Google Trends, Google Health Trends (also referred to as Flu Trends), and Google Custom Search.

Methods

Our protocol consisted of four steps: (1) developing a master list of top search queries for an initial search term using Google Trends, (2) gathering information on relative search volume using Google Health Trends, (3) determining the most popular sites using Google Custom Search, and (4) calculating estimated total search volume. We tested the protocol following key procedures at each step and verified its usefulness by examining search traffic on birth control in 2017 in the United States. Two separate programmers working independently achieved similar results with insignificant variation due to sample variability.

Results

We successfully tested the methodology on the initial search term birth control. We identified top search queries for birth control, of which birth control pill was the most popular and obtained the relative and estimated total search volume for the top queries: relative search volume was 0.54 for the pill, corresponding to an estimated 9.3-10.7 million searches. We used the estimates of the proportion of search activity for the top queries to arrive at a generated list of the most popular websites: for the pill, the Planned Parenthood website was the top site.

Conclusions

The proposed methodological framework demonstrates how to retrieve Google query data from multiple Google APIs and provides thorough documentation required to systematically identify search queries and websites, as well as estimate relative and total search volume of queries in real time or near-real time in specific locations and time periods. Although the protocol needs further testing, it allows researchers to replicate the steps and shows promise in advancing our understanding of population-level health concerns.

International Registered Report Identifier (IRRID)

RR1-10.2196/16543

Using Twitter to Surveil the Opioid Epidemic in North Carolina: An Exploratory Study

BACKGROUND

Over the last two decades, deaths associated with opioids have escalated in number and geographic spread, impacting more and more individuals, families, and communities. Reflecting on the shifting nature of the opioid overdose crisis, Dasgupta, Beletsky, and Ciccarone offer a triphasic framework to explain that opioid overdose deaths (OODs) shifted from prescription opioids for pain (beginning in 2000), to heroin (2010 to 2015), and then to synthetic opioids (beginning in 2013). Given the rapidly shifting nature of OODs, timelier surveillance data are critical to inform strategies that combat the opioid crisis. Using easily accessible and near real-time social media data to improve public health surveillance efforts related to the opioid crisis is a promising area of research.

OBJECTIVE

This study explored the potential of using Twitter data to monitor the opioid epidemic. Specifically, this study investigated the extent to which the content of opioid-related tweets corresponds with the triphasic nature of the opioid crisis and correlates with OODs in North Carolina between 2009 and 2017.

METHODS

Opioid-related Twitter posts were obtained using Crimson Hexagon, and were classified as relating to prescription opioids, heroin, and synthetic opioids using natural language processing. This process resulted in a corpus of 100,777 posts consisting of tweets, retweets, mentions, and replies. Using a random sample of 10,000 posts from the corpus, we identified opioid-related terms by analyzing word frequency for each year. OODs were obtained from the Multiple Cause of Death database from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (CDC WONDER). Least squares regression and Granger tests compared patterns of opioid-related posts with OODs.

RESULTS

The pattern of tweets related to prescription opioids, heroin, and synthetic opioids resembled the triphasic nature of OODs. For prescription opioids, tweet counts and OODs were statistically unrelated. Tweets mentioning heroin and synthetic opioids were significantly associated with heroin OODs and synthetic OODs in the same year (P=.01 and P<.001, respectively), as well as in the following year (P=.03 and P=.01, respectively). Moreover, heroin tweets in a given year predicted heroin deaths better than lagged heroin OODs alone (P=.03).

CONCLUSIONS

Findings support using Twitter data as a timely indicator of opioid overdose mortality, especially for heroin.

Flattening the Mental Health Curve: COVID-19 Stay-at-Home Orders Are Associated With Alterations in Mental Health Search Behavior in the United States

BACKGROUND

The coronavirus disease (COVID-19) has led to dramatic changes worldwide in people's everyday lives. To combat the pandemic, many governments have implemented social distancing, quarantine, and stay-at-home orders. There is limited research on the impact of such extreme measures on mental health.

OBJECTIVE

The goal of this study was to examine whether stay-at-home orders produced differential changes in mental health symptoms using internet search queries on a national scale.

METHODS

In the United States, individual states vary in their adoption of measures to reduce the spread of COVID-19; as of March 23, 2020, 11 of the 50 states had issued stay-at-home orders. The staggered rollout of stay-at-home measures across the United States allows us to investigate whether these measures impact mental health by exploring variations in mental health search queries across the states. This paper examines the changes in mental health search queries on Google between March 16-23, 2020, across each state and Washington, DC. Specifically, this paper examines differential changes in mental health searches based on patterns of search activity following issuance of stay-at-home orders in these states compared to all other states. The participants were all the people who searched mental health terms in Google between March 16-23. Between March 16-23, 11 states underwent stay-at-home orders to prevent the transmission of COVID-19. Outcomes included search terms measuring anxiety, depression, obsessive-compulsive, negative thoughts, irritability, fatigue, anhedonia, concentration, insomnia, and suicidal ideation.

RESULTS

Analyzing over 10 million search queries using generalized additive mixed models, the results suggested that the implementation of stay-at-home orders are associated with a significant flattening of the curve for searches for suicidal ideation, anxiety, negative thoughts, and sleep disturbances, with the most prominent flattening associated with suicidal ideation and anxiety.

CONCLUSIONS

These results suggest that, despite decreased social contact, mental health search queries increased rapidly prior to the issuance of stay-at-home orders, and these changes dissipated following the announcement and enactment of these orders. Although more research is needed to examine sustained effects, these results suggest mental health symptoms were associated with an immediate leveling off following the issuance of stay-at-home orders.

Strengths and weaknesses of existing data sources to support research to address the opioids crisis

Better opioid prescribing practices, promoting effective opioid use disorder treatment, improving naloxone access, and enhancing public health surveillance are strategies central to reducing opioid-related morbidity and mortality. Successfully advancing and evaluating these strategies requires leveraging and linking existing secondary data sources. We conducted a scoping study in Fall 2017 at RAND, including a literature search (updated in December 2018) complemented by semi-structured interviews with policymakers and researchers, to identify data sources and linking strategies commonly used in opioid studies, describe data source strengths and limitations, and highlight opportunities to use data to address high-priority public health research questions. We identified 306 articles, published between 2005 and 2018, that conducted secondary analyses of existing data to examine one or more public health strategies. Multiple secondary data sources, available at national, state, and local levels, support such research, with substantial breadth in data availability, data contents, and the data's ability to support multi-level analyses over time. Interviewees identified opportunities to expand existing capabilities through systematic enhancements, including greater support to states for creating and facilitating data use, as well as key data challenges, such as data availability lags and difficulties matching individual-level data over time or across datasets. Multiple secondary data sources exist that can be used to examine the impact of public health approaches to addressing the opioid crisis. Greater data access, improved usability for research purposes, and data element standardization can enhance their value, as can improved data availability timeliness and better data comparability across jurisdictions.

Lifestyle Disease Surveillance Using Population Search Behavior: Feasibility Study

Background

As the process of producing official health statistics for lifestyle diseases is slow, researchers have explored using Web search data as a proxy for lifestyle disease surveillance. Existing studies, however, are prone to at least one of the following issues: ad-hoc keyword selection, overfitting, insufficient predictive evaluation, lack of generalization, and failure to compare against trivial baselines.

Objective

The aims of this study were to (1) employ a corrective approach improving previous methods; (2) study the key limitations in using Google Trends for lifestyle disease surveillance; and (3) test the generalizability of our methodology to other countries beyond the United States.

Methods

For each of the target variables (diabetes, obesity, and exercise), prevalence rates were collected. After a rigorous keyword selection process, data from Google Trends were collected. These data were denormalized to form spatio-temporal indices. L1-regularized regression models were trained to predict prevalence rates from denormalized Google Trends indices. Models were tested on a held-out set and compared against baselines from the literature as well as a trivial last year equals this year baseline. A similar analysis was done using a multivariate spatio-temporal model where the previous year’s prevalence was included as a covariate. This model was modified to create a time-lagged regression analysis framework. Finally, a hierarchical time-lagged multivariate spatio-temporal model was created to account for subnational trends in the data. The model trained on US data was, then, applied in a transfer learning framework to Canada.

Results

In the US context, our proposed models beat the performances of the prior work, as well as the trivial baselines. In terms of the mean absolute error (MAE), the best of our proposed models yields 24% improvement (0.72-0.55; P<.001) for diabetes; 18% improvement (1.20-0.99; P=.001) for obesity, and 34% improvement (2.89-1.95; P<.001) for exercise. Our proposed across-country transfer learning framework also shows promising results with an average Spearman and Pearson correlation of 0.70 for diabetes and 0.90 and 0.91 for obesity, respectively.

Conclusions

Although our proposed models beat the baselines, we find the modeling of lifestyle diseases to be a challenging problem, one that requires an abundance of data as well as creative modeling strategies. In doing so, this study shows a low-to-moderate validity of Google Trends in the context of lifestyle disease surveillance, even when applying novel corrective approaches, including a proposed denormalization scheme. We envision qualitative analyses to be a more practical use of Google Trends in the context of lifestyle disease surveillance. For the quantitative analyses, the highest utility of using Google Trends is in the context of transfer learning where low-resource countries could benefit from high-resource countries by using proxy models.

Online Searching and Social Media to Detect Alcohol Use Risk at Population Scale

INTRODUCTION

Harnessing engagement in online searching and social media may provide complementary information for monitoring alcohol use, informing prevention and policy evaluation, and extending knowledge available from national surveys.

METHODS

Relative search volumes for 7 alcohol-related keywords were estimated from Google Trends (data, 2014-2017), and the proportion of alcohol use-related Twitter posts (data, 2014-2015) was estimated using natural language processing. Searching/posting measures were created for all 50 U.S. states plus Washington, D.C. Survey reports of alcohol use and summaries of state alcohol policies were obtained from the Behavioral Risk Factor Surveillance System (data, 2014-2016) and the Alcohol Policy Scale. In 2018-2019, associations among searching/posting measures and same state/year Behavioral Risk Factor Surveillance System reports of recent (past-30-day) alcohol use and maximum number of drinks consumed on an occasion were estimated using logistic and linear regression, adjusting for sociodemographics and Internet use, with moderation tested in regressions that included interactions of select searching/posting measures and the Alcohol Policy Scale.

RESULTS

Recent alcohol use was reported by 52.93% of 1,297,168 Behavioral Risk Factor Surveillance System respondents, which was associated with all state-level searching/posting measures in unadjusted and adjusted models (p<0.0001). Among drinkers, most searching/posting measures were associated with maximum number of drinks consumed (p<0.0001). Associations varied with exposure to high versus low levels of state policy controls on alcohol.

CONCLUSIONS

Strong associations were found among individual alcohol use and state-level alcohol-related searching/posting measures, which were moderated by the strength of state alcohol policies. Findings support using novel personally generated data to monitor alcohol use and possibly evaluate effects of alcohol control policies.

Access to Primary Care and Internet Searches for Walk-In Clinics and Emergency Departments in Canada: Observational Study Using Google Trends and Population Health Survey Data

Background

Access to primary care is a challenge for many Canadians. Models of primary care vary widely among provinces, including arrangements for same-day and after-hours access. Use of walk-in clinics and emergency departments (EDs) may also vary, but data sources that allow comparison are limited.

Objective

We used Google Trends to examine the relative frequency of searches for walk-in clinics and EDs across provinces and over time in Canada. We correlated provincial relative search frequencies from Google Trends with survey responses about primary care access from the Commonwealth Fund’s 2016 International Health Policy Survey of Adults in 11 Countries and the 2016 Canadian Community Health Survey.

Methods

We developed search strategies to capture the range of terms used for walk-in clinics (eg, urgent care clinic and after-hours clinic) and EDs (eg, emergency room) across Canadian provinces. We used Google Trends to determine the frequencies of these terms relative to total search volume within each province from January 2011 to December 2018. We calculated correlation coefficients and 95% CIs between provincial Google Trends relative search frequencies and survey responses.

Results

Relative search frequency of walk-in clinic searches increased steadily, doubling in most provinces between 2011 and 2018. Relative frequency of walk-in clinic searches was highest in the western provinces of British Columbia, Alberta, Saskatchewan, and Manitoba. At the provincial level, higher walk-in clinic relative search frequency was strongly positively correlated with the percentage of survey respondents who reported being able to get same- or next-day appointments to see a doctor or a nurse and inversely correlated with the percentage of respondents who reported going to ED for a condition that they thought could have been treated by providers at usual place of care. Relative search frequency for walk-in clinics was also inversely correlated with the percentage of respondents who reported having a regular medical provider. ED relative search frequencies were more stable over time, and we did not observe statistically significant correlation with survey data.

Conclusions

Higher relative search frequency for walk-in clinics was positively correlated with the ability to get a same- or next-day appointment and inversely correlated with ED use for conditions treatable in the patient’s regular place of care and also with having a regular medical provider. Findings suggest that patient use of Web-based tools to search for more convenient or accessible care through walk-in clinics is increasing over time. Further research is needed to validate Google Trends data with administrative information on service use.

Seasonality and global public interest in psoriasis: an infodemiology study

OBJECTIVE

Although patients with psoriasis frequently report seasonal changes in their symptoms, the seasonality of psoriasis has rarely been explored. This study aims to investigate the seasonal pattern of and global public interest in psoriasis using Google search data.

METHODS

Internet search data were collected from Google Trends. Data on the relative search volume (RSV) from January 2004 to December 2018 were retrieved using the term psoriasis. Cosinor analyses were conducted to examine the seasonality of psoriasis using data from two southern hemisphere countries (Australia and New Zealand) and four northern hemisphere countries (USA, Canada, UK and Ireland).

RESULTS

Overall, searches for psoriasis steadily decreased between 2004 and 2010, and then rose from 2011 to 2018. On cosinor analyses, RSV of 'psoriasis' displayed a significant seasonal variation worldwide (p<0.025). Further analyses confirmed the seasonality of psoriasis-related RSV in Australia, New Zealand, USA, Canada, UK and Ireland (p<0.025 for all), with peaks in the late winter/early spring months and troughs in the late summer/early autumn months. The top 11 rising topics were calcipotriol/betamethasone dipropionate, ustekinumab, apremilast, shampoo, eczema, guttate psoriasis, seborrhoeic dermatitis, dermatitis, psoriatic arthritis, atopic dermatitis and arthritis.

CONCLUSION

There was a significant seasonal pattern for psoriasis, with peaks in the late winter/early spring and troughs in the late summer/early autumn. Further studies are warranted to confirm the seasonal pattern of psoriasis using clinical data and to explore the underlying mechanisms.

The Validity of Google Trends Search Volumes for Behavioral Forecasting of National Suicide Rates in Ireland

Annual suicide figures are critical in identifying trends and guiding research, yet challenges arising from significant lags in reporting can delay and complicate real-time interventions. In this paper, we utilized Google Trends search volumes for behavioral forecasting of national suicide rates in Ireland between 2004 and 2015. Official suicide rates are recorded by the Central Statistics Office in Ireland. While similar investigations using Google trends data have been carried out in other jurisdictions (e.g., United Kingdom, United Stated of America), such research had not yet been completed in Ireland. We compiled a collection of suicide- and depression-related search terms suggested by Google Trends and manually sourced from the literature. Monthly search rate terms at different lags were compared with suicide occurrences to determine the degree of correlation. Following two approaches based on vector autoregression and neural network autoregression, we achieved mean absolute error values between 4.14 and 9.61 when incorporating search query data, with the highest performance for the neural network approach. The application of this process to United Kingdom suicide and search query data showed similar results, supporting the benefit of Google Trends, neural network approach, and the applied search terms to forecast suicide risk increase. Overall, the combination of societal data and online behavior provide a good indication of societal risks; building on past research, our improvements led to robust models integrating search query and unemployment data for suicide risk forecasting in Ireland.

Developing an early warning system of suicide using Google Trends and media reporting

BACKGROUND

Conventional surveillance systems for suicides typically suffer from a substantial time lag of six months to two years. This study aims to develop an early warning system of possible suicide outbreaks in Hong Kong using Google Trends and suicide-related media reporting.

METHODS

Data on 3,534 suicides from 2011 to 2015 were obtained from Hong Kong Census and Statistics Department, and the Coroner's Court. Using data from Google Trends and features extracted from media reporting on suicide news, we fitted Poisson regression models to predict the number and estimate the intensity of suicides on a weekly basis, for six subgroups, defined by gender and age. We adopted the cumulative sum (CUSUM) control chart-based method to identify outbreaks of suicide.

RESULTS

The proposed model was able to predict the number of suicides with reasonably low normalized root mean squared errors, ranging from 15.6% for young females to 24.16% for old females. The suicide intensity curves were well captured by the proposed models for young males and females, but not for other groups. The Sensitivity, Precision and F1 Score of the CUSUM-based method were 50%, 100% and 67% for young females, and 93%, 54% and 68% for young males.

LIMITATIONS

This study focused only on predicting the number of suicides in the current week, not in the future weeks. The model did not include social media, socioeconomic and climate data.

CONCLUSIONS

Our results indicate that Google Trends search terms and media reporting data may be valuable data sources for predicting possible outbreak of suicides in Hong Kong. The proposed system could support effective and targeted interventions.

Assessing the Methods, Tools, and Statistical Approaches in Google Trends Research: Systematic Review

Background

In the era of information overload, are big data analytics the answer to access and better manage available knowledge? Over the last decade, the use of Web-based data in public health issues, that is, infodemiology, has been proven useful in assessing various aspects of human behavior. Google Trends is the most popular tool to gather such information, and it has been used in several topics up to this point, with health and medicine being the most focused subject. Web-based behavior is monitored and analyzed in order to examine actual human behavior so as to predict, better assess, and even prevent health-related issues that constantly arise in everyday life.

Objective

This systematic review aimed at reporting and further presenting and analyzing the methods, tools, and statistical approaches for Google Trends (infodemiology) studies in health-related topics from 2006 to 2016 to provide an overview of the usefulness of said tool and be a point of reference for future research on the subject.

Methods

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for selecting studies, we searched for the term “Google Trends” in the Scopus and PubMed databases from 2006 to 2016, applying specific criteria for types of publications and topics. A total of 109 published papers were extracted, excluding duplicates and those that did not fall inside the topics of health and medicine or the selected article types. We then further categorized the published papers according to their methodological approach, namely, visualization, seasonality, correlations, forecasting, and modeling.

Results

All the examined papers comprised, by definition, time series analysis, and all but two included data visualization. A total of 23.1% (24/104) studies used Google Trends data for examining seasonality, while 39.4% (41/104) and 32.7% (34/104) of the studies used correlations and modeling, respectively. Only 8.7% (9/104) of the studies used Google Trends data for predictions and forecasting in health-related topics; therefore, it is evident that a gap exists in forecasting using Google Trends data.

Conclusions

The monitoring of online queries can provide insight into human behavior, as this field is significantly and continuously growing and will be proven more than valuable in the future for assessing behavioral changes and providing ground for research using data that could not have been accessed otherwise.

What, where and when? Using Google Trends and Google to investigate patient needs and inform pharmacy practice

Objectives

The aim was to provide a comprehensive overview (using pertinent examples) of the various ways that Google Trends and Google data could inform pharmacy practice. The objectives were to: examine what type of information people search for in relation to a common class of medicines; ascertain where people are directed to (websites) following an initial search for a medicine or medical condition; and establish information about when they search.

Methods

The methodology differed depending on whether Google Trends or Google was being interrogated, but the search domain was always limited to the United Kingdom. Google Trends was queried, typically for a 5-year time frame, and data downloaded for many search inputs relating to medical conditions (self-treatable and non-self-treatable) and medicines (bought over-the-counter and prescribed). Google was queried and data collected for searches related to ‘antibiotics’.

Key findings

Google Trends revealed a previously unknown seasonality pattern for irritable bowel syndrome. Related searches for ‘antibiotics’ revealed a high level of interest in the appropriateness of concomitant alcohol consumption and queries about what antibiotics are. Largely, people were being directed to reputable websites following their initial search input about a prescription-only medicine. However, searches for over-the-counter medicines were more likely to lead to commercial domains.

Conclusions

This is one of the first studies to investigate use of Google Trends and Google in a pharmacy-specific context. It is relevant for practice as it could inform marketing strategies, public health policy and help tailor patient advice and counselling.

Facilitating Factors and Barriers to the Use of Emerging Technologies for Suicide Prevention in Europe: Multicountry Exploratory Study

BACKGROUND

This study provides an analysis on the use of emerging technologies for the prevention of suicide in 8 different European countries.

OBJECTIVE

The objective of this study was to analyze the potentiality of using emerging technologies in the area of suicide prevention based on the opinion of different professionals involved in suicide prevention.

METHODS

Opinions of 3 groups of stakeholders (ie, relevant professionals in suicide field) were gathered using a specifically designed questionnaire to explore dimensions underlying perceptions of facilitating factors and barriers in relation to the use of emerging technologies for suicide prevention.

RESULTS

Goal 1 involved facilitating factors for the use of emerging technologies in suicide prevention. Northern European countries, except for Belgium, attach greater relevance to those that optimize implementation and benefits. On the other hand, Southern European countries attach greater importance to professionally oriented and user-centered facilitating factors. According to different stakeholders, the analysis of these facilitating factors suggest that professionals in the field of social work attach greater relevance to those that optimize implementation and benefits. However, professionals involved in the area of mental health, policy makers, and political decision makers give greater importance to professionally oriented and user-centered facilitating factors. Goal 2 was related to barriers to the usability of emerging technologies for suicide prevention. Both countries and stakeholders attach greater importance to barriers associated with resource constraints than to those centered on personal limitations. There are no differences between countries or between stakeholders. Nevertheless, there is a certain stakeholders-countries interaction that indicates that the opinions on resource constraints expressed by different stakeholders do not follow a uniform pattern in different countries, but they differ depending on the country.

CONCLUSIONS

Although all countries and stakeholders agree in identifying resource constraints as the main barrier to the use of emerging technologies, factors facilitating their use in suicide prevention differ among countries and among stakeholders.

Ontology-Based Approach to Social Data Sentiment Analysis: Detection of Adolescent Depression Signals

BACKGROUND

Social networking services (SNSs) contain abundant information about the feelings, thoughts, interests, and patterns of behavior of adolescents that can be obtained by analyzing SNS postings. An ontology that expresses the shared concepts and their relationships in a specific field could be used as a semantic framework for social media data analytics.

OBJECTIVE

The aim of this study was to refine an adolescent depression ontology and terminology as a framework for analyzing social media data and to evaluate description logics between classes and the applicability of this ontology to sentiment analysis.

METHODS

The domain and scope of the ontology were defined using competency questions. The concepts constituting the ontology and terminology were collected from clinical practice guidelines, the literature, and social media postings on adolescent depression. Class concepts, their hierarchy, and the relationships among class concepts were defined. An internal structure of the ontology was designed using the entity-attribute-value (EAV) triplet data model, and superclasses of the ontology were aligned with the upper ontology. Description logics between classes were evaluated by mapping concepts extracted from the answers to frequently asked questions (FAQs) onto the ontology concepts derived from description logic queries. The applicability of the ontology was validated by examining the representability of 1358 sentiment phrases using the ontology EAV model and conducting sentiment analyses of social media data using ontology class concepts.

RESULTS

We developed an adolescent depression ontology that comprised 443 classes and 60 relationships among the classes; the terminology comprised 1682 synonyms of the 443 classes. In the description logics test, no error in relationships between classes was found, and about 89% (55/62) of the concepts cited in the answers to FAQs mapped onto the ontology class. Regarding applicability, the EAV triplet models of the ontology class represented about 91.4% of the sentiment phrases included in the sentiment dictionary. In the sentiment analyses, "academic stresses" and "suicide" contributed negatively to the sentiment of adolescent depression.

CONCLUSIONS

The ontology and terminology developed in this study provide a semantic foundation for analyzing social media data on adolescent depression. To be useful in social media data analysis, the ontology, especially the terminology, needs to be updated constantly to reflect rapidly changing terms used by adolescents in social media postings. In addition, more attributes and value sets reflecting depression-related sentiments should be added to the ontology.