Dataset Information

Using web search queries to monitor influenza-like illness: an exploratory retrospective analysis, Netherlands, 2017/18 influenza season.

ABSTRACT: BackgroundDespite the early development of Google Flu Trends in 2009, standards for digital epidemiology methods have not been established and research from European countries is scarce.AimIn this article, we study the use of web search queries to monitor influenza-like illness (ILI) rates in the Netherlands in real time.MethodsIn this retrospective analysis, we simulated the weekly use of a prediction model for estimating the then-current ILI incidence across the 2017/18 influenza season solely based on Google search query data. We used weekly ILI data as reported to The European Surveillance System (TESSY) each week, and we removed the then-last 4 weeks from our dataset. We then fitted a prediction model based on the then-most-recent search query data from Google Trends to fill the 4-week gap ('Nowcasting'). Lasso regression, in combination with cross-validation, was applied to select predictors and to fit the 52 models, one for each week of the season.ResultsThe models provided accurate predictions with a mean and maximum absolute error of 1.40 (95% confidence interval: 1.09-1.75) and 6.36 per 10,000 population. The onset, peak and end of the epidemic were predicted with an error of 1, 3 and 2 weeks, respectively. The number of search terms retained as predictors ranged from three to five, with one keyword, 'griep' ('flu'), having the most weight in all models.DiscussionThis study demonstrates the feasibility of accurate, real-time ILI incidence predictions in the Netherlands using Google search query data.

SUBMITTER: Schneider PP

PROVIDER: S-EPMC7268271 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Using web search queries to monitor influenza-like illness: an exploratory retrospective analysis, Netherlands, 2017/18 influenza season.

Schneider Paul P PP van Gool Christel Jaw CJ Spreeuwenberg Peter P Hooiveld Mariëtte M Donker Gé A GA Barnett David J DJ Paget John J

Euro surveillance : bulletin Europeen sur les maladies transmissibles = European communicable disease bulletin 20200501 21

BackgroundDespite the early development of Google Flu Trends in 2009, standards for digital epidemiology methods have not been established and research from European countries is scarce.AimIn this article, we study the use of web search queries to monitor influenza-like illness (ILI) rates in the Netherlands in real time.MethodsIn this retrospective analysis, we simulated the weekly use of a prediction model for estimating the then-current ILI incidence across the 2017/18 influenza season solely ...[more]

PMID: 32489174

Similar Datasets

Project description:BackgroundSeasonal influenza causes a considerable burden to healthcare services every year. To better measure the impact of severe influenza cases in Romania, we analyzed active surveillance data collected during the 2017-2018 season from patients admitted for influenza-like illness (ILI) at a tertiary care hospital in Bucharest.MethodsPatients admitted for acute ILI were included if they were resident in the Bucharest-Ilfov region, had been hospitalized for at least 24 h, and had onset of symptoms within 7 days before admission. Patient demographics, healthcare use, vaccination status, and outcome data were collected by questionnaire or by searching clinical records. Respiratory swabs were also obtained from each patient to confirm influenza A (A/H1 and A/H3 subtypes) or influenza B (Yamagata and Victoria lineages) infection by real-time reverse-transcription polymerase chain reaction assay.ResultsThe study included 502 patients, many (45.2%) of whom were aged < 5 years. Overall, 108 patients (21.5%) had one or more comorbidities. Seventeen adults aged 18-64 years (3.4%) had been vaccinated against influenza. Patients were hospitalized for a median of 5 days and most (90.4%) were prescribed antiviral treatment. More than one-half of the patients (n = 259, 51.6%) were positive for influenza. Most influenza cases were caused by B viruses (172/259, 66.4%), which were mostly of the B/Yamagata lineage (85 of 94 characterized, 90.4%). Most of the subtyped A viruses were A/H1 (59/74, 79.7%). A/H1 viruses were frequently detected in influenza-positive admissions throughout the 2017-2018 season, whereas the predominant B/Yamagata viruses were detected around the middle of the season, with a peak in cases at week 7 of 2018. Eleven patients were admitted to an intensive care unit; of these, one patient with confirmed B/Yamagata infection died.ConclusionsThese results show that seasonal influenza results in considerable hospitalization in Bucharest-Ilfov, Romania and suggest vaccine coverage should be extended, especially to the youngest age groups. The data from this study should help inform and optimize national influenza healthcare policies.

Project description:BackgroundReal-time air pollution monitoring is a valuable tool for public health and environmental surveillance. In recent years, there has been a dramatic increase in air pollution forecasting and monitoring research using artificial neural networks. Most prior work relied on modeling pollutant concentrations collected from ground-based monitors and meteorological data for long-term forecasting of outdoor ozone (O3), oxides of nitrogen, and fine particulate matter (PM2.5). Given that traditional, highly sophisticated air quality monitors are expensive and not universally available, these models cannot adequately serve those not living near pollutant monitoring sites. Furthermore, because prior models were built based on physical measurement data collected from sensors, they may not be suitable for predicting the public health effects of pollution exposure.ObjectiveThis study aimed to develop and validate models to nowcast the observed pollution levels using web search data, which are publicly available in near real time from major search engines.MethodsWe developed novel machine learning-based models using both traditional supervised classification methods and state-of-the-art deep learning methods to detect elevated air pollution levels at the US city level by using generally available meteorological data and aggregate web-based search volume data derived from Google Trends. We validated the performance of these methods by predicting 3 critical air pollutants (O3, nitrogen dioxide, and PM2.5) across 10 major US metropolitan statistical areas in 2017 and 2018. We also explore different variations of the long short-term memory model and propose a novel search term dictionary learner-long short-term memory model to learn sequential patterns across multiple search terms for prediction.ResultsThe top-performing model was a deep neural sequence model long short-term memory, using meteorological and web search data, and reached an accuracy of 0.82 (F1-score 0.51) for O3, 0.74 (F1-score 0.41) for nitrogen dioxide, and 0.85 (F1-score 0.27) for PM2.5, when used for detecting elevated pollution levels. Compared with using only meteorological data, the proposed method achieved superior accuracy by incorporating web search data.ConclusionsThe results show that incorporating web search data with meteorological data improves the nowcasting performance for all 3 pollutants and suggest promising novel applications for tracking global physical phenomena using web search data.

Project description:BackgroundThe objective of this study is to investigate predictive utility of online social media and web search queries, particularly, Google search data, to forecast new cases of influenza-like-illness (ILI) in general outpatient clinics (GOPC) in Hong Kong. To mitigate the impact of sensitivity to self-excitement (i.e., fickle media interest) and other artifacts of online social media data, in our approach we fuse multiple offline and online data sources.MethodsFour individual models: generalized linear model (GLM), least absolute shrinkage and selection operator (LASSO), autoregressive integrated moving average (ARIMA), and deep learning (DL) with Feedforward Neural Networks (FNN) are employed to forecast ILI-GOPC both one week and two weeks in advance. The covariates include Google search queries, meteorological data, and previously recorded offline ILI. To our knowledge, this is the first study that introduces deep learning methodology into surveillance of infectious diseases and investigates its predictive utility. Furthermore, to exploit the strength from each individual forecasting models, we use statistical model fusion, using Bayesian model averaging (BMA), which allows a systematic integration of multiple forecast scenarios. For each model, an adaptive approach is used to capture the recent relationship between ILI and covariates.ResultsDL with FNN appears to deliver the most competitive predictive performance among the four considered individual models. Combing all four models in a comprehensive BMA framework allows to further improve such predictive evaluation metrics as root mean squared error (RMSE) and mean absolute predictive error (MAPE). Nevertheless, DL with FNN remains the preferred method for predicting locations of influenza peaks.ConclusionsThe proposed approach can be viewed a feasible alternative to forecast ILI in Hong Kong or other countries where ILI has no constant seasonal trend and influenza data resources are limited. The proposed methodology is easily tractable and computationally efficient.

Project description:BackgroundInternet search query trends have been shown to correlate with incidence trends for select infectious diseases and countries. Herein, the first use of Google search queries for malaria surveillance is investigated. The research focuses on Thailand where real-time malaria surveillance is crucial as malaria is re-emerging and developing resistance to pharmaceuticals in the region.MethodsOfficial Thai malaria case data was acquired from the World Health Organization (WHO) from 2005 to 2009. Using Google correlate, an openly available online tool, and by surveying Thai physicians, search queries potentially related to malaria prevalence were identified. Four linear regression models were built from different sub-sets of malaria-related queries to be used in future predictions. The models' accuracies were evaluated by their ability to predict the malaria outbreak in 2009, their correlation with the entire available malaria case data, and by Akaike information criterion (AIC).ResultsEach model captured the bulk of the variability in officially reported malaria incidence. Correlation in the validation set ranged from 0.75 to 0.92 and AIC values ranged from 808 to 586 for the models. While models using malaria-related and general health terms were successful, one model using only microscopy-related terms obtained equally high correlations to malaria case data trends. The model built strictly of queries provided by Thai physicians was the only one that consistently captured the well-documented second seasonal malaria peak in Thailand.ConclusionsModels built from Google search queries were able to adequately estimate malaria activity trends in Thailand, from 2005-2010, according to official malaria case counts reported by WHO. While presenting their own limitations, these search queries may be valid real-time indicators of malaria incidence in the population, as correlations were on par with those of related studies for other infectious diseases. Additionally, this methodology provides a cost-effective description of malaria prevalence that can act as a complement to traditional public health surveillance. This and future studies will continue to identify ways to leverage web-based data to improve public health.

Dataset Information

Using web search queries to monitor influenza-like illness: an exploratory retrospective analysis, Netherlands, 2017/18 influenza season.

Publications

Using web search queries to monitor influenza-like illness: an exploratory retrospective analysis, Netherlands, 2017/18 influenza season.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets