Potential use of personal protection online search during COVID-19 pandemic for predicting and monitoring public response

COVID-19 has become a global public health emergency in almost all over the world, including in Indonesia. Effective risk communication becomes an emergency response to increase awareness and determine appropriate interventions. The study aimed to assess the success of risk communication monitoring using Google Trends during the COVID-19 pandemic in Indonesia. Quantitative and qualitative research uses time-series data (31 December 2019-2 May 2020). The relative search volume (RSV) of keyword „masker‟ (mask) and „cuci tangan‟ (handwashing) from Google Trends (GT) and the number of COVID-19 daily cases were collected. Analyzed qualitatively. RSV search data and daily case comparisons were performed based on Pearson correlation analysis and time lag correlation with significance <0.05. The keyword „masker‟ has four peaks and „cuci tangan‟ has three peaks with fluctuations due to the increase in mask prices, government policies, news, and official WHO recommendations. Validation using time-lag correlation shows the significant results between RSV keywords related to personal protection and the number of COVID-19 cases. The highest correlation was achieved by the keyword „masker‟ three days before the number of COVID-19 cases. Google Trends can potentially be used as a complement and support for early warning systems in the surveillance system and improve public health responses in Indonesia.

Download Full-text

Analysis of public search interest towards immune system improvement during the COVID-19 pandemic using Google Trends

International Journal of Public Health Science (IJPHS) ◽

10.11591/ijphs.v9i4.20518 ◽

2020 ◽

Vol 9 (4) ◽

pp. 414

Author(s):

Linda Amelia ◽

Rizma Adlia Syakurah

Keyword(s):

Immune System ◽

Risk Communication ◽

Public Awareness ◽

Time Lag ◽

Google Trends ◽

Series Data ◽

Search Term ◽

Search Volume ◽

System Improvement ◽

Time Lag Correlation

In combating COVID-19, maintaining the immune system is important. Providing this information to the general population will increase public awareness towards improving their immune system. The use of Google Trends for exploring web behavior related to a topic or search term also considered as a tool for monitoring public awareness to help risk communication during the COVID-19 pandemic. Therefore, this study was conducted to assess the use of Google Trends to monitor public awareness to immune system improvement during the COVID-19 pandemic in Indonesia. This quantitative and qualitative research used time-series data from 31 December 2019 to 2 May 2020. The time-lag correlation analysis was performed to compare between relative search volume (RSV) of “Vitamin C”, “Vaksin” (Vaccine), “Berjemur” (Sunbathing) from Google Trends (GT), and the number of reported COVID-19 new cases. Validation using time-lag correlation shows the significant correlation between RSV keywords related to public awareness towards immune system improvement and the number of COVID-19 cases in Indonesia in 1-3 days before an increase in the number of cases occurs. Google Trends has the potential to become an early warning system and a tool for monitoring risk communication towards immune system improvement during the COVID-19 pandemic by Indonesia Government.

Download Full-text

Increase in public interest concerning alternative medicine during the COVID-19 pandemic in Indonesia: a Google Trends study

F1000Research ◽

10.12688/f1000research.25525.2 ◽

2021 ◽

Vol 9 ◽

pp. 1201

Author(s):

Dewi Rokhmah ◽

Khaidar Ali ◽

Serius Miliyani Dwi Putri ◽

Khoiron Khoiron

Keyword(s):

Alternative Medicine ◽

Public Interest ◽

Time Lag ◽

Google Trends ◽

Rank Test ◽

Search Term ◽

Alternative Medicines ◽

Search Volume ◽

Relative Search Volume ◽

Time Lag Correlation

Background: The COVID-19 pandemic has triggered individuals to increase their healthy behaviour in order to prevent transmission, including improving their immunity potentially through the use of alternative medicines. This study aimed to examine public interest on alternative medicine during the COVID-19 pandemic using Google Trends in Indonesia. Methods: Employing a quantitative study, the Spearman rank test was used to analyze the correlation between Google Relative Search Volume (RSV) of various search terms, within the categories of alternative medicine, herbal medicine and practical activity, with COVID-19 cases. In addition, time lag correlation was also investigated. Results: Public interest toward alternative medicine during COVID-19 pandemic in Indonesia is dramatically escalating. All search term categories (alternative medicine, medical herbal, and alternative medicine activities) were positively associated with COVID-19 cases (p<0.05). The terms ‘ginger’ (r=0.6376), ‘curcumin’ (r=0.6550) and ‘planting ginger’ (0.6713) had the strongest correlation. Furthermore, time lag correlation between COVID-19 and Google RSV was also positively significant (p<0.05). Conclusion: Public interest concerning alternative medicine related terms dramatically increased after the first COVID-19 confirmed case was reported in Indonesia. Time lag correlation showed good performance using weekly data. The Indonesian Government will play an important role to provide and monitor information related to alternative medicine in order for the population to receive the maximum benefit.

Download Full-text

Increase in public interest concerning alternative medicine during the COVID-19 pandemic in Indonesia: a Google Trends study

F1000Research ◽

10.12688/f1000research.25525.1 ◽

2020 ◽

Vol 9 ◽

pp. 1201

Author(s):

Dewi Rokhmah ◽

Khaidar Ali ◽

Serius Miliyani Dwi Putri ◽

Khoiron Khoiron

Keyword(s):

Alternative Medicine ◽

Public Interest ◽

Time Lag ◽

Google Trends ◽

Rank Test ◽

Search Term ◽

Alternative Medicines ◽

Search Volume ◽

Relative Search Volume ◽

Time Lag Correlation

Download Full-text

Suitability of Google Trends™ for digital surveillance during ongoing COVID-19 epidemic: a case study from India

Disaster Medicine and Public Health Preparedness ◽

10.1017/dmp.2021.249 ◽

2021 ◽

pp. 1-28

Author(s):

Parmeshwar Satpathy ◽

Sanjeev Kumar ◽

Pankaj Prasad

Keyword(s):

Information Seeking ◽

Media Coverage ◽

Time Lag ◽

National Level ◽

Google Trends ◽

Health Seeking ◽

High Burden ◽

Search Volume ◽

Digital Surveillance ◽

Time Lag Correlation

Abstract Objective: Digital surveillance has shown mixed results as supplement to traditional surveillance. Google Trends™ (GT) has been used for digital surveillance of H1N1, Ebola and MERS. We used GT to correlate the information seeking on COVID-19 with number of tests and cases in India. Methods: We obtained data on daily tests and cases from WHO, ECDC and covid19india.org. We used a comprehensive search strategy to retrieve GT data on COVID-19 related information-seeking behaviour in India between 1st January and 31st May 2020 in the form of relative search volume (RSV). We used time-lag correlation analysis to assess the temporal relationships between RSV and daily new COVID-19 cases and tests. Results: GT RSV showed high time-lag correlation with both daily reported tests and cases for the terms “COVID 19”, “COVID”, “social distancing”, “soap” and “lockdown” at national level. In five high-burden states, high correlation was observed for these five terms along with “Corona”. Peaks in RSV both at national level and high-burden states corresponded with media coverage or government declarations on the ongoing pandemic. Conclusion: The correlation observed between GT data and COVID-19 tests/cases in India may be either due to media-coverage induced curiosity or health-seeking.

Download Full-text

Analysis of public search interest regarding government containment policy on COVID-19 new cases in Indonesia, Malaysia and Singapore

Health Promotion Perspectives ◽

10.34172/hpp.2021.46 ◽

2021 ◽

Vol 11 (3) ◽

pp. 360-368

Author(s):

Muhammad Farid Rizqullah ◽

Rizma Adlia Syakurah

Keyword(s):

Risk Communication ◽

Large Scale ◽

Pearson Correlation ◽

Time Lag ◽

Circuit Breaker ◽

Search Volume ◽

Containment Policy ◽

The Government ◽

Relative Search Volume ◽

The Relationship

Background: As preventive measures to curb coronavirus disease 2019 (COVID-19)transmission, Indonesia, Malaysia and Singapore had imposed web-accessible regulations where the popularity of relative internet search volume can be obtained from Google Trends(GT). This research aimed to seek the relationship between public search interest and countries policies, furthermore to observe whether the GT data could be utilized as a tool to make a risk communication during this pandemic. Methods: This retrospective study used GT to analyze the relative search volume (RSV) of keywords large-scale social restrictions (Pembatasan Sosial Berskala Besar – PSBB ), MovementControl Order (MCO) or Perintah Kawalan Pergerakan (PKP) and Circuit Breaker (CB) for Indonesia, Malaysia and Singapore respectively. Daily number of COVID-19 confirmed cases were collected and analyzed using Pearson correlation and time-lag with P<0.05. Every search interest peak and mobility trends changes were qualitatively analyzed. Results: The results exhibited the relationship between the government containment policy, the peaks of analyzed RSV keywords and the mobility trends. The containment policy has significant relationships with COVID-19 daily cases (P<0.05). Conclusion: These results indicated that the government could use GT RSV as a strategy of crisis and risk communication to intervene public behavior towards the pandemic.

Download Full-text

An Epidemiological Model Considering Isolation to Predict COVID-19 Trends in Tokyo, Japan: Numerical Analysis (Preprint)

10.2196/preprints.23624 ◽

2020 ◽

Author(s):

Motoaki Utamura ◽

Makoto Koizumi ◽

Seiichi Kirikami

Keyword(s):

Public Health ◽

Transmission Rate ◽

Time Lag ◽

Sir Model ◽

Epidemiological Model ◽

Basic Reproductive Number ◽

Polymerase Chain Reaction Test ◽

Series Data ◽

Reproductive Number ◽

Alternative Formulation

BACKGROUND COVID-19 currently poses a global public health threat. Although Tokyo, Japan, is no exception to this, it was initially affected by only a small-level epidemic. Nevertheless, medical collapse nearly happened since no predictive methods were available to assess infection counts. A standard susceptible-infectious-removed (SIR) epidemiological model has been widely used, but its applicability is limited often to the early phase of an epidemic in the case of a large collective population. A full numerical simulation of the entire period from beginning until end would be helpful for understanding COVID-19 trends in (separate) counts of inpatient and infectious cases and can also aid the preparation of hospital beds and development of quarantine strategies. OBJECTIVE This study aimed to develop an epidemiological model that considers the isolation period to simulate a comprehensive trend of the initial epidemic in Tokyo that yields separate counts of inpatient and infectious cases. It was also intended to induce important corollaries of governing equations (ie, effective reproductive number) and equations for the final count. METHODS Time-series data related to SARS-CoV-2 from February 28 to May 23, 2020, from Tokyo and antibody testing conducted by the Japanese government were adopted for this study. A novel epidemiological model based on a discrete delay differential equation (apparent time-lag model [ATLM]) was introduced. The model can predict trends in inpatient and infectious cases in the field. Various data such as daily new confirmed cases, cumulative infections, inpatients, and PCR (polymerase chain reaction) test positivity ratios were used to verify the model. This approach also derived an alternative formulation equivalent to the standard SIR model. RESULTS In a typical parameter setting, the present ATLM provided 20% less infectious cases in the field compared to the standard SIR model prediction owing to isolation. The basic reproductive number was inferred as 2.30 under the condition that the time lag <i>T</i> from infection to detection and isolation is 14 days. Based on this, an adequate vaccine ratio to avoid an outbreak was evaluated for 57% of the population. We assessed the date (May 23) that the government declared a rescission of the state of emergency. Taking into consideration the number of infectious cases in the field, a date of 1 week later (May 30) would have been most effective. Furthermore, simulation results with a shorter time lag of <i>T</i>=7 and a larger transmission rate of α=1.43α0 suggest that infections at large should reduce by half and inpatient numbers should be similar to those of the first wave of COVID-19. CONCLUSIONS A novel mathematical model was proposed and examined using SARS-CoV-2 data for Tokyo. The simulation agreed with data from the beginning of the pandemic. Shortening the period from infection to hospitalization is effective against outbreaks without rigorous public health interventions and control.

Download Full-text

ANALISIS PENELUSURAN MASKER SEBAGAI PROTOKOL KESEHATAN SAAT PANDEMI COVID-19 DI INDONESIA: STUDI GOOGLE TRENDS

VISIKES: Jurnal Kesehatan Masyarakat ◽

10.33633/visikes.v20i1.4191 ◽

2021 ◽

Vol 20 (1) ◽

Author(s):

Ully Febra Kusuma ◽

Nurunnisa Arsyad ◽

Melissa Shalimar Lavinia ◽

Selvia Rahayu ◽

Muhammad Khairul Kahfi Pasaribu ◽

...

Keyword(s):

Public Interest ◽

Keyword Search ◽

Time Series Data ◽

Preventive Measures ◽

Time Lag ◽

Google Trends ◽

Series Data ◽

Qualitative And Quantitative ◽

Search Results ◽

Qualitative And Quantitative Study

AbstrakMedical masks are also used by people at risk who are indicated to need them. The supply of medical masks is limited, the general public is encouraged to use non-medical masks or cloth masks. This article will discuss the comparison of search results for sensi masks, cloth masks and N-95 masks using google trend analysis. This research method is a qualitative and quantitative study using time series data with quantitative analysis, time-lag correlation is used to assess whether an increase in GT data is correlated with an increase in COVID-19 cases. Data from google trends regarding keywords related to one of the preventive measures for COVID-19, namely masks such as "sensi masks", "cloth masks" and "N-95 masks". Each search interest usually reaches a peak depending on the situation and conditions that occur at that time. The keyword search for "N-95 masks" experienced a peak when 2 Indonesians were confirmed positive for COVID-19, namely on March 2, 2020 and the day after that the keyword "sensi mask" also experienced the highest peak of searches. The keyword search for "cloth masks" peaked on March 6, 2020, when the price of sensi masks began to rise. The results of the keyword correlation test for “sensi mask”, “cloth mask” and “N-95 mask” show that the keyword search results on Google trended a decline in line with the increase in COVID-19 cases in Indonesia. Public interest in tracing increased at the beginning of COVID-19 entering Indonesia. However, the interest in this search continues to decline and is inversely proportional to the increase in the incidence of COVID-19 cases in Indonesia. Keywords: COVID-19, sensi masks, medical masks, cloth masks, N-95 masks, Google trends

Download Full-text

Real-time Prediction of the Daily Incidence of COVID-19 in 215 Countries and Territories Using Machine Learning: Model Development and Validation (Preprint)

10.2196/preprints.24285 ◽

2020 ◽

Author(s):

Yuanyuan Peng ◽

Xinjian Chen ◽

Yibiao Rong ◽

Chi Pui Pang ◽

Xinjian Chen ◽

...

Keyword(s):

Machine Learning ◽

Real Time ◽

Correlation Coefficient ◽

Pearson Correlation ◽

Google Trends ◽

Pearson Correlation Coefficient ◽

Spearman Correlation ◽

Time Prediction ◽

Search Volume ◽

The Mean

BACKGROUND Advanced prediction of the daily incidence of COVID-19 can aid policy making on the prevention of disease spread, which can profoundly affect people's livelihood. In previous studies, predictions were investigated for single or several countries and territories. OBJECTIVE We aimed to develop models that can be applied for real-time prediction of COVID-19 activity in all individual countries and territories worldwide. METHODS Data of the previous daily incidence and infoveillance data (search volume data via Google Trends) from 215 individual countries and territories were collected. A random forest regression algorithm was used to train models to predict the daily new confirmed cases 7 days ahead. Several methods were used to optimize the models, including clustering the countries and territories, selecting features according to the importance scores, performing multiple-step forecasting, and upgrading the models at regular intervals. The performance of the models was assessed using the mean absolute error (MAE), root mean square error (RMSE), Pearson correlation coefficient, and Spearman correlation coefficient. RESULTS Our models can accurately predict the daily new confirmed cases of COVID-19 in most countries and territories. Of the 215 countries and territories under study, 198 (92.1%) had MAEs <10 and 187 (87.0%) had Pearson correlation coefficients >0.8. For the 215 countries and territories, the mean MAE was 5.42 (range 0.26-15.32), the mean RMSE was 9.27 (range 1.81-24.40), the mean Pearson correlation coefficient was 0.89 (range 0.08-0.99), and the mean Spearman correlation coefficient was 0.84 (range 0.2-1.00). CONCLUSIONS By integrating previous incidence and Google Trends data, our machine learning algorithm was able to predict the incidence of COVID-19 in most individual countries and territories accurately 7 days ahead.

Download Full-text

An Epidemiological Model Considering Isolation to Predict COVID-19 Trends in Tokyo, Japan: Numerical Analysis

JMIR Public Health and Surveillance ◽

10.2196/23624 ◽

2020 ◽

Vol 6 (4) ◽

pp. e23624

Author(s):

Motoaki Utamura ◽

Makoto Koizumi ◽

Seiichi Kirikami

Keyword(s):

Public Health ◽

Transmission Rate ◽

Time Lag ◽

Sir Model ◽

Epidemiological Model ◽

Basic Reproductive Number ◽

Polymerase Chain Reaction Test ◽

Series Data ◽

Reproductive Number ◽

Alternative Formulation

Background COVID-19 currently poses a global public health threat. Although Tokyo, Japan, is no exception to this, it was initially affected by only a small-level epidemic. Nevertheless, medical collapse nearly happened since no predictive methods were available to assess infection counts. A standard susceptible-infectious-removed (SIR) epidemiological model has been widely used, but its applicability is limited often to the early phase of an epidemic in the case of a large collective population. A full numerical simulation of the entire period from beginning until end would be helpful for understanding COVID-19 trends in (separate) counts of inpatient and infectious cases and can also aid the preparation of hospital beds and development of quarantine strategies. Objective This study aimed to develop an epidemiological model that considers the isolation period to simulate a comprehensive trend of the initial epidemic in Tokyo that yields separate counts of inpatient and infectious cases. It was also intended to induce important corollaries of governing equations (ie, effective reproductive number) and equations for the final count. Methods Time-series data related to SARS-CoV-2 from February 28 to May 23, 2020, from Tokyo and antibody testing conducted by the Japanese government were adopted for this study. A novel epidemiological model based on a discrete delay differential equation (apparent time-lag model [ATLM]) was introduced. The model can predict trends in inpatient and infectious cases in the field. Various data such as daily new confirmed cases, cumulative infections, inpatients, and PCR (polymerase chain reaction) test positivity ratios were used to verify the model. This approach also derived an alternative formulation equivalent to the standard SIR model. Results In a typical parameter setting, the present ATLM provided 20% less infectious cases in the field compared to the standard SIR model prediction owing to isolation. The basic reproductive number was inferred as 2.30 under the condition that the time lag T from infection to detection and isolation is 14 days. Based on this, an adequate vaccine ratio to avoid an outbreak was evaluated for 57% of the population. We assessed the date (May 23) that the government declared a rescission of the state of emergency. Taking into consideration the number of infectious cases in the field, a date of 1 week later (May 30) would have been most effective. Furthermore, simulation results with a shorter time lag of T=7 and a larger transmission rate of α=1.43α0 suggest that infections at large should reduce by half and inpatient numbers should be similar to those of the first wave of COVID-19. Conclusions A novel mathematical model was proposed and examined using SARS-CoV-2 data for Tokyo. The simulation agreed with data from the beginning of the pandemic. Shortening the period from infection to hospitalization is effective against outbreaks without rigorous public health interventions and control.

Download Full-text

Suitability of Google Trends ™ for digital surveillance during ongoing COVID-19 epidemic: a case study from India

10.1101/2020.08.24.20176321 ◽

2020 ◽

Author(s):

Parmeshwar D Satpathy ◽

Sanjeev Kumar ◽

Pankaj Prasad

Keyword(s):

Media Coverage ◽

Time Lag ◽

National Level ◽

High Time ◽

Google Trends ◽

High Burden ◽

Social Distancing ◽

Digital Surveillance ◽

Traditional Surveillance ◽

Time Lag Correlation

Background: India went into the largest population-level lockdown on 25th March 2020 in response to the declaration of COVID-19 pandemic by World Health Organization (WHO). Digital surveillance has been shown to be useful to supplement the traditional surveillance. Google Trends ™ (GT) is one such platform reported to be useful during pandemics of H1N1, Ebola and MERS. Objective: We used GT to correlate the information seeking behaviour regarding COVID-19 of Indians with curiosity and apprehensiveness generated through media coverage as well as status of the epidemic both at national and state levels. Methods: We retrieved GT data between 1st January 2020 to 31st May 2020 for India using a comprehensive search strategy. We obtained data on daily tests and cases from WHO, ECDC and covid19india.org websites. We explored the trends of COVID-19 in the form of relative search volume (RSV) from GT platform and correlated them with media reports. We used time-lag correlation analysis to assess the temporal relationships between Google search terms and daily new COVID-19 cases and daily tests for 14 days. Results: Peaks in RSV correlated with media coverage or government declarations suggestive of curiosity and apprehensiveness both at national level and high-burden states. High time-lag correlation was observed between both the daily reported number of tests and cases and RSV for the terms ″COVID 19″, ″COVID″, ″social distancing″, ″soap″ and ″lockdown″ at national level. Similar high time-lag correlation was observed for the terms ″COVID 19″, ″COVID″, ″Corona″, ″social distancing″, ″soap″, ″lockdown″ in five high-burden states. Conclusion: This study reveals the advantages of infodemiology using GT to monitor an emerging infectious disease like COVID-19 in India. Google searches in India during the ongoing COVID-19 pandemic reflects mostly curiosity and apprehension of citizens. GT can also complement traditional surveillance in India as well as high burden states.

Download Full-text