scholarly journals Estimation of asthma symptom onset using Internet search queries: A lag-time series analysis (Preprint)

2020 ◽  
Author(s):  
Yulin Hswen ◽  
Amanda Zhang ◽  
Bruno Ventelou
Keyword(s):  
Time Series ◽  
Hospital Admission ◽  
Symptom Onset ◽  
Hospital Admissions ◽  
Pearson Correlation ◽  
Internet Search ◽  
Search Queries ◽  
Risk Of Death ◽  
Search Volume ◽  
Google Search

BACKGROUND Asthma affects over 330 million people worldwide. Timing of the asthma event is extremely important and lack of identification of asthma increases the risk of death. A major challenge for health systems is the length of time between symptom onset and care seeking, which could result in delayed treatment initiation and worsening of symptoms. OBJECTIVE This study evaluates the utility of the Internet search query data for the identification the onset of asthma symptoms. METHODS Pearson correlation coefficients between the time series of hospital admissions and Google searches were computed at lag times from 4 weeks prior to hospital admission to 4 weeks after hospital admission. RESULTS Google search volume for asthma had the highest correlation at 2 weeks before hospital admission. CONCLUSIONS Our findings demonstration Internet search queries can earlier predict asthma events and may be a better use for classifying the measurement of timing of symptom onset.

scholarly journals Population-Level Interest and Telehealth Capacity of US Hospitals in Response to COVID-19: Cross-Sectional Analysis of Google Search and National Hospital Survey Data (Preprint)

2020 ◽  
Author(s):  
Young-Rock Hong ◽  
John Lawrence ◽  
Dunc Williams Jr ◽  
Arch Mainous III
Keyword(s):  
United States ◽  
Health Care ◽  
Population Level ◽  
The United States ◽  
Health Care Facilities ◽  
Care Providers ◽  
Internet Search ◽  
American Hospital ◽  
Search Volume ◽  
Google Search

BACKGROUND As the novel coronavirus disease (COVID-19) is widely spreading across the United States, there is a concern about the overloading of the nation’s health care capacity. The expansion of telehealth services is expected to deliver timely care for the initial screening of symptomatic patients while minimizing exposure in health care facilities, to protect health care providers and other patients. However, it is currently unknown whether US hospitals have the telehealth capacity to meet the increasing demand and needs of patients during this pandemic. OBJECTIVE We investigated the population-level internet search volume for telehealth (as a proxy of population interest and demand) with the number of new COVID-19 cases and the proportion of hospitals that adopted a telehealth system in all US states. METHODS We used internet search volume data from Google Trends to measure population-level interest in telehealth and telemedicine between January 21, 2020 (when the first COVID-19 case was reported), and March 18, 2020. Data on COVID-19 cases in the United States were obtained from the Johns Hopkins Coronavirus Resources Center. We also used data from the 2018 American Hospital Association Annual Survey to estimate the proportion of hospitals that adopted telehealth (including telemedicine and electronic visits) and those with the capability of telemedicine intensive care unit (tele-ICU). Pearson correlation was used to examine the relations of population search volume for telehealth and telemedicine (composite score) with the cumulative numbers of COVID-19 cases in the United States during the study period and the proportion of hospitals with telehealth and tele-ICU capabilities. RESULTS We found that US population–level interest in telehealth increased as the number of COVID-19 cases increased, with a strong correlation (<i>r</i>=0.948, <i>P</i>&lt;.001). We observed a higher population-level interest in telehealth in the Northeast and West census region, whereas the proportion of hospitals that adopted telehealth was higher in the Midwest region. There was no significant association between population interest and the proportion of hospitals that adopted telehealth (<i>r</i>=0.055, <i>P</i>=.70) nor hospitals having tele-ICU capability (<i>r</i>=–0.073, <i>P</i>=.61). CONCLUSIONS As the number of COVID-19 cases increases, so does the US population’s interest in telehealth. However, the level of population interest did not correlate with the proportion of hospitals providing telehealth services in the United States, suggesting that increased population demand may not be met with the current telehealth capacity. Telecommunication infrastructures in US hospitals may lack the capability to address the ongoing health care needs of patients with other health conditions. More practical investment is needed to deploy the telehealth system rapidly against the impending patient surge.

scholarly journals Internet search query data improve forecasts of daily emergency department volume

2019 ◽  
Vol 26 (12) ◽  
pp. 1574-1583 ◽  
Author(s):  
Sam Tideman ◽  
Mauricio Santillana ◽  
Jonathan Bickel ◽  
Ben Reis
Keyword(s):  
Predictive Power ◽  
Linear Models ◽  
Data Sources ◽  
Weather Data ◽  
Volume Data ◽  
Internet Search ◽  
Search Activity ◽  
Search Volume ◽  
Day Of The Week ◽  
Google Search

Abstract Objective Emergency departments (EDs) are increasingly overcrowded. Forecasting patient visit volume is challenging. Reliable and accurate forecasting strategies may help improve resource allocation and mitigate the effects of overcrowding. Patterns related to weather, day of the week, season, and holidays have been previously used to forecast ED visits. Internet search activity has proven useful for predicting disease trends and offers a new opportunity to improve ED visit forecasting. This study tests whether Google search data and relevant statistical methods can improve the accuracy of ED volume forecasting compared with traditional data sources. Materials and Methods Seven years of historical daily ED arrivals were collected from Boston Children’s Hospital. We used data from the public school calendar, National Oceanic and Atmospheric Administration, and Google Trends. Multiple linear models using LASSO (least absolute shrinkage and selection operator) for variable selection were created. The models were trained on 5 years of data and out-of-sample accuracy was judged using multiple error metrics on the final 2 years. Results All data sources added complementary predictive power. Our baseline day-of-the-week model recorded average percent errors of 10.99%. Autoregressive terms, calendar and weather data reduced errors to 7.71%. Search volume data reduced errors to 7.58% theoretically preventing 4 improperly staffed days. Discussion The predictive power provided by the search volume data may stem from the ability to capture population-level interaction with events, such as winter storms and infectious diseases, that traditional data sources alone miss. Conclusions This study demonstrates that search volume data can meaningfully improve forecasting of ED visit volume and could help improve quality and reduce cost.

scholarly journals Using Application Programming Interfaces to Access Google Data for Health Research: Protocol for a Methodological Framework (Preprint)

2019 ◽  
Author(s):  
Anne Zepecki ◽  
Sylvia Guendelman ◽  
John DeNero ◽  
Ndola Prata
Keyword(s):  
Real Time ◽  
Birth Control ◽  
Google Trends ◽  
Search Term ◽  
Health Trends ◽  
Search Queries ◽  
Search Volume ◽  
Application Programming ◽  
Relative Search Volume ◽  
Google Search

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 <i>birth control</i> 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 <i>birth control</i>. We identified top search queries for <i>birth control</i>, of which <i>birth control pill</i> 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 RR1-10.2196/16543

Gauging public awareness and interest related to pancreatic cancer: An analysis of Google search volumes.

2018 ◽  
Vol 36 (4_suppl) ◽  
pp. 250-250
Author(s):  
Dhruvika Mukhija ◽  
Alok A. Khorana ◽  
Davendra Sohal
Keyword(s):  
Pancreatic Cancer ◽  
Colon Cancer ◽  
Public Awareness ◽  
P Value ◽  
Internet Search ◽  
Awareness Campaigns ◽  
Search Terms ◽  
Search Volume ◽  
News Events ◽  
Google Search

250 Background: Over the last 2 decades, the internet has become a major source of medical information. Infoveillance, i.e., public health surveillance using online content analysis has become a powerful tool and internet search activity has been used as a surrogate to gauge public awareness and interest for particular diseases. We aimed to evaluate the search volume for pancreatic cancer (PC), using colon cancer (CC), as a comparator, using data from a popular search engine. Methods: Using Google Trends, a public web facility of Google Inc., based on Google Search, we compared the relative frequency of search terms ‘pancreatic cancer’ and ‘colon cancer’ between 1st January 2004 and 31st August 2017 (n = 164 months). The program assigns a reference value of 100 for the point of maximum popularity from among all the search terms during the search period and provides comparative monthly scores, which we termed relative interest scores (RIS). The RIS for each cancer was then adjusted for incidence (i.e., 53,070 for PC and 95,270 for CC, based on 2016 data), calculated per 10,000 patients and termed ‘i-RIS’. A p-value of < 0.05 was considered significant. Results: For the entire duration, the maximum popularity (RIS = 100) corresponded to a point in March 2008 for PC, likely related to the diagnosis of a famous celebrity during that month. Similar but smaller surges in RIS were observed for other significant news events related to PC during other months (January 2009, October 2009 and October 2011). Overall, the mean (±S.D) RIS for PC and CC were 32.52±8.98 and 50.18±6.44, respectively (p < 0.001). However, the i-RIS was somewhat higher for PC (6.12±1.69) as compared with CC (5.26±0.67) (p < 0.001). Conclusions: Internet search data can provide estimates of public awareness and interest related to cancer. For PC, incidence-adjusted search volumes show spikes in search volumes related to major news events, providing internal validation of these results. Generating news items and promotion by celebrities may play a significant role in the success of cancer awareness campaigns.

scholarly journals Influence of Pharmaceutical Copayment on Emergency Hospital Admissions: A 1978–2018 Time Series Analysis in Spain

2021 ◽  
Vol 18 (15) ◽  
pp. 8009
Author(s):  
Antonio Palazón-Bru ◽  
Miriam Calvo-Pérez ◽  
Pilar Rico-Ferreira ◽  
María Anunciación Freire-Ballesta ◽  
Vicente Francisco Gil-Guillén ◽  
...  
Keyword(s):  
Time Series ◽  
Life Expectancy ◽  
Time Series Analysis ◽  
Hospital Admission ◽  
Hospital Admissions ◽  
Moving Average ◽  
Emergency Hospital ◽  
Admission Rates ◽  
Series Analysis ◽  
Emergency Hospital Admissions

No studies have evaluated the influence of pharmaceutical copayment on hospital admission rates using time series analysis. Therefore, we aimed to analyze the relationship between hospital admission rates and the influence of the introduction of a pharmaceutical copayment system (PCS). In July 2012, a PCS was implemented in Spain, and we designed a time series analysis (1978–2018) to assess its impact on emergency hospital admissions. Hospital admission rates were estimated between 1978 and 2018 each month using the Hospital Morbidity Survey in Spain (the number of urgent hospital admissions per 100,000 inhabitants). This was conducted for men, women and both and for all-cause, cardiovascular and respiratory hospital discharges. Life expectancy was obtained from the National Institute of Statistics. The copayment variable took a value of 0 before its implementation (pre-PCS: January 1978–June 2012) and 1 after that (post-PCS: July 2012–December 2018). ARIMA (Autoregressive Integrated Moving Average) (2,0,0)(1,0,0) models were estimated with two predictors (life expectancy and copayment implementation). Pharmaceutical copayment did not influence hospital admission rates (with p-values between 0.448 and 0.925) and there was even a reduction in the rates for most of the analyses performed. In conclusion, the PCS did not influence hospital admission rates. More studies are needed to design health policies that strike a balance between the amount contributed by the taxpayer and hospital admission rates.

scholarly journals 282Social isolation and incident coronary heart disease in two large UK prospective studies

2021 ◽  
Vol 50 (Supplement_1) ◽  
Author(s):  
Robert Smith ◽  
Isobel Barnes ◽  
Jane Green ◽  
Gillian Reeves ◽  
Valerie Beral ◽  
...  
Keyword(s):  
Coronary Heart Disease ◽  
Heart Disease ◽  
Hospital Admission ◽  
Social Isolation ◽  
Hospital Admissions ◽  
Cox Regression ◽  
Coronary Event ◽  
Living Alone ◽  
Risk Of Death ◽  
Increased Risk

Abstract Background Social isolation is associated with CHD mortality but evidence of association with incident CHD is mixed. We prospectively examined this association in the Million Women Study (MWS) and UK Biobank (UKB). Methods 481,946 MWS and 456,612 UKB participants reported on social isolation (living alone, little contact with family/friends/groups). Excluding those reporting previous CHD or stroke, participants were followed for incident CHD using linkage to hospital admission and death records. Cox regression yielded relative risks (RR) by 3 levels of social isolation, adjusted for relevant confounders. Results During 7 years follow-up in the MWS and UKB, there were 42,402 first coronary heart disease events in total (of which 1,834 were fatal without an associated hospital admission). After adjustment, social isolation was not associated with hospital admission for first CHD events (combined RR for both studies: RR = 1.01, 95% CI: 0.98–1.04). However, the risk of fatal first CHD events without an associated hospital admission was substantially higher in the most isolated group than the least isolated group (1.86 [1.63–2.12]) This association with fatal first CHD events was driven by the association with living alone. Conclusions Social isolation was not associated with increased risk of first CHD hospital admissions but was associated with increased risk of death from CHD. Key messages Social isolation is likely not a risk factor for developing CHD, but people living alone may be at greater risk of dying from a coronary event than those not living alone.

Forecasting Short-Term Housing Transaction Volumes using Time-Series and Internet Search Queries

2019 ◽  
Vol 23 (6) ◽  
pp. 2409-2416 ◽  
Author(s):  
Kanghyeok Lee ◽  
Hanbeen Kim ◽  
Do Hyoung Shin
Keyword(s):  
Time Series ◽  
Internet Search ◽  
Short Term ◽  
Search Queries
Sign in / Sign up

Export Citation Format

Share Document