Electronic Health Record-Based Algorithm for Monitoring Respiratory Virus-Like Illness.

Viral respiratory illness surveillance has traditionally focused on single pathogens (e.g., influenza) and required fever to identify influenza-like illness (ILI). We developed an automated system applying both laboratory test and syndrome criteria to electronic health records from 3 practice groups in Massachusetts, USA, to monitor trends in respiratory viral-like illness (RAVIOLI) across multiple pathogens. We identified RAVIOLI syndrome using diagnosis codes associated with respiratory viral testing or positive respiratory viral assays or fever. After retrospectively applying RAVIOLI criteria to electronic health records, we observed annual winter peaks during 2015-2019, predominantly caused by influenza, followed by cyclic peaks corresponding to SARS-CoV-2 surges during 2020-2024, spikes in RSV in mid-2021 and late 2022, and recrudescent influenza in late 2022 and 2023. RAVIOLI rates were higher and fluctuations more pronounced compared with traditional ILI surveillance. RAVIOLI broadens the scope, granularity, sensitivity, and specificity of respiratory viral illness surveillance compared with traditional ILI surveillance.

SARS-CoV-2, influenza A/B and respiratory syncytial virus positivity and association with influenza-like illness and self-reported symptoms, over the 2022/23 winter season in the UK: a longitudinal surveillance cohort.

BACKGROUND: Syndromic surveillance often relies on patients presenting to healthcare. Community cohorts, although more challenging to recruit, could provide additional population-wide insights, particularly with SARS-CoV-2 co-circulating with other respiratory viruses. METHODS: We estimated the positivity and incidence of SARS-CoV-2, influenza A/B, and RSV, and trends in self-reported symptoms including influenza-like illness (ILI), over the 2022/23 winter season in a broadly representative UK community cohort (COVID-19 Infection Survey), using negative-binomial generalised additive models. We estimated associations between test positivity and each of the symptoms and influenza vaccination, using adjusted logistic and multinomial models. RESULTS: Swabs taken at 32,937/1,352,979 (2.4%) assessments tested positive for SARS-CoV-2, 181/14,939 (1.2%) for RSV and 130/14,939 (0.9%) for influenza A/B, varying by age over time. Positivity and incidence peaks were earliest for RSV, then influenza A/B, then SARS-CoV-2, and were highest for RSV in the youngest and for SARS-CoV-2 in the oldest age groups. Many test positives did not report key symptoms: middle-aged participants were generally more symptomatic than older or younger participants, but still, only ~ 25% reported ILI-WHO and ~ 60% ILI-ECDC. Most symptomatic participants did not test positive for any of the three viruses. Influenza A/B-positivity was lower in participants reporting influenza vaccination in the current and previous seasons (odds ratio = 0.55 (95% CI 0.32, 0.95)) versus neither season. CONCLUSIONS: Symptom profiles varied little by aetiology, making distinguishing SARS-CoV-2, influenza and RSV using symptoms challenging. Most symptoms were not explained by these viruses, indicating the importance of other pathogens in syndromic surveillance. Influenza vaccination was associated with lower rates of community influenza test positivity.

Enterovirus D68 disease burden and epidemiology in hospital-admitted influenza-like illness, Valencia region of Spain, 2014-2020 influenza seasons.

Enterovirus D68 (EV-D68) is an emerging agent for which data on the susceptible adult population is scarce. We performed a 6-year analysis of respiratory samples from influenza-like illness (ILI) admitted during 2014-2020 in 4-10 hospitals in the Valencia Region, Spain. EV-D68 was identified in 68 (3.1%) among 2210 Enterovirus (EV)/Rhinovirus (HRV) positive samples. Phylogeny of 59 VP1 sequences showed isolates from 2014 clustering in B2 (6/12), B1 (5/12), and A2/D1 (1/12) subclades; those from 2015 (n = 1) and 2016 (n = 1) in B3 and A2/D1, respectively; and isolates from 2018 in A2/D3 (42/45), and B3 (3/45). B1 and B2 viruses were mainly detected in children (80% and 67%, respectively); B3 were equally distributed between children and adults; whereas A2/D1 and A2/D3 were observed only in adults. B3 viruses showed up to 16 amino acid changes at predicted antigenic sites. In conclusion, two EV-D68 epidemics linked to ILI hospitalized cases occurred in the Valencia Region in 2014 and 2018, with three fatal outcomes and one ICU admission. A2/D3 strains from 2018 were associated with severe respiratory infection in adults. Because of the significant impact of non-polio enteroviruses in ILI and the potential neurotropism, year-round surveillance in respiratory samples should be pursued.

Evaluation of Machine Learning to Detect Influenza Using Wearable Sensor Data and Patient-Reported Symptoms: Cohort Study.

BACKGROUND: Machine learning offers quantitative pattern recognition analysis of wearable device data and has the potential to detect illness onset and monitor influenza-like illness (ILI) in patients who are infected. OBJECTIVE: This study aims to evaluate the ability of machine-learning algorithms to distinguish between participants who are influenza positive and influenza negative in a cohort of symptomatic patients with ILI using wearable sensor (activity) data and self-reported symptom data during the latent and early symptomatic periods of ILI. METHODS: This prospective observational cohort study used the extreme gradient boosting (XGBoost) classifier to determine whether a participant was influenza positive or negative based on 3 models using symptom-only data, activity-only data, and combined symptom and activity data. Data were collected from the Home Testing of Respiratory Illness (HTRI) study and FluStudy2020, both conducted between December 2019 and October 2020. The model was developed using the FluStudy2020 data and tested on the HTRI data. Analyses included participants in these studies with an at-home influenza diagnostic test result. Fitbit (Google LLC) devices were used to measure participants' steps, heart rate, and sleep parameters. Participants detailed their ILI symptoms, health care-seeking behaviors, and quality of life. Model performance was assessed by area under the curve (AUC), balanced accuracy, recall (sensitivity), specificity, precision (positive predictive value), negative predictive value, and weighted harmonic mean of precision and recall (F2) score. RESULTS: An influenza diagnostic test result was available for 953 and 925 participants in HTRI and FluStudy2020, respectively, of whom 848 (89%) and 840 (90.8%) had activity data. For the training and validation sets, the highest performing model was trained on the combined symptom and activity data (training AUC=0.77; validation AUC=0.74) versus symptom-only (training AUC=0.73; validation AUC=0.72) and activity-only (training AUC=0.68; validation AUC=0.65) data. For the FluStudy2020 test set, the performance of the model trained on combined symptom and activity data was closely aligned with that of the symptom-only model (combined symptom and activity test AUC=0.74; symptom-only test AUC=0.74). These results were validated using independent HTRI data (combined symptom and activity evaluation AUC=0.75; symptom-only evaluation AUC=0.74). The top features guiding influenza detection were cough; mean resting heart rate during main sleep; fever; total minutes in bed for the combined model; and fever, cough, and sore throat for the symptom-only model. CONCLUSIONS: Machine-learning algorithms had moderate accuracy in detecting influenza, suggesting that previous findings from research-grade sensors tested in highly controlled experimental settings may not easily translate to scalable commercial-grade sensors. In the future, more advanced wearable sensors may improve their performance in the early detection and discrimination of viral respiratory infections.

Application of the Time Derivative (TD) Method for Early Alert of Influenza Epidemics.

BACKGROUND: In order to minimize the spread of seasonal influenza epidemic to communities worldwide, the Korea Disease Control and Prevention Agency has issued an influenza epidemic alert using the influenza epidemic threshold formula based on the results of the influenza-like illness (ILI) rate. However, unusual changes have occurred in the pattern of respiratory infectious diseases, including seasonal influenza, after the coronavirus disease 2019 (COVID-19) pandemic. As a result, the importance of detecting the onset of an epidemic earlier than the existing epidemic alert system is increasing. Accordingly, in this study, the Time Derivative (TD) method was suggested as a supplementary approach to the existing influenza alert system for the early detection of seasonal influenza epidemics. METHODS: The usefulness of the TD method as an early epidemic alert system was evaluated by applying the ILI rate for each week during past seasons when seasonal influenza epidemics occurred, ranging from the 2013-2014 season to the 2022-2023 season to compare it with the issued time of the actual influenza epidemic alert. RESULTS: As a result of applying the TD method, except for the two seasons (2020-2021 season and 2021-2022 season) that had no influenza epidemic, an influenza early epidemic alert was suggested during the remaining seasons, excluding the 2017-2018 and 2022-2023 seasons. CONCLUSION: The TD method is a time series analysis that enables early epidemic alert in real-time without relying on past epidemic information. It can be considered as an alternative approach when it is challenging to set an epidemic threshold based on past period information. This situation may arise when there has been a change in the typical seasonal epidemic pattern of various respiratory viruses, including influenza, following the COVID-19 pandemic.

Forecasting severe respiratory disease hospitalizations using machine learning algorithms.

BACKGROUND: Forecasting models predicting trends in hospitalization rates have the potential to inform hospital management during seasonal epidemics of respiratory diseases and the associated surges caused by acute hospital admissions. Hospital bed requirements for elective surgery could be better planned if it were possible to foresee upcoming peaks in severe respiratory illness admissions. Forecasting models can also guide the use of intervention strategies to decrease the spread of respiratory pathogens and thus prevent local health system overload. In this study, we explore the capability of forecasting models to predict the number of hospital admissions in Auckland, New Zealand, within a three-week time horizon. Furthermore, we evaluate probabilistic forecasts and the impact on model performance when integrating laboratory data describing the circulation of respiratory viruses. METHODS: The dataset used for this exploration results from active hospital surveillance, in which the World Health Organization Severe Acute Respiratory Infection (SARI) case definition was consistently used. This research nurse-led surveillance has been implemented in two public hospitals in Auckland and provides a systematic laboratory testing of SARI patients for nine respiratory viruses, including influenza, respiratory syncytial virus, and rhinovirus. The forecasting strategies used comprise automatic machine learning, one of the most recent generative pre-trained transformers, and established artificial neural network algorithms capable of univariate and multivariate forecasting. RESULTS: We found that machine learning models compute more accurate forecasts in comparison to naïve seasonal models. Furthermore, we analyzed the impact of reducing the temporal resolution of forecasts, which decreased the model error of point forecasts and made probabilistic forecasting more reliable. An additional analysis that used the laboratory data revealed strong season-to-season variations in the incidence of respiratory viruses and how this correlates with total hospitalization cases. These variations could explain why it was not possible to improve forecasts by integrating this data. CONCLUSIONS: Active SARI surveillance and consistent data collection over time enable these data to be used to predict hospital bed utilization. These findings show the potential of machine learning as support for informing systems for proactive hospital management.

Enhanced laboratory surveillance of respiratory infection disclosed the rapid rise of enterovirus D68 cases, northern Italy, August to September 2024.

We report a considerable increase in enterovirus D68 (EV-D68) cases since July 2024, culminating in an ongoing outbreak of acute respiratory infections in northern Italy, accounting for nearly 90% of all enterovirus infections. The outbreak was identified by community- and hospital-based surveillance systems, detecting EV-D68 in individuals with mild-to-severe respiratory infections. These strains belonged to B3 and a divergent A2 lineage. An increase in adult cases was observed. Enhanced surveillance and molecular characterisation of EV-D68 across Europe are needed.

Real-time investigation of an influenza A(H3N2) virus outbreak in a refugee community, November 2022.

OBJECTIVES: To report epidemiological and virological results of an outbreak investigation of influenza-like illness (ILI) among refugees in Northern Italy. STUDY DESIGN: Outbreak investigation of ILI cases observed among nearly 100 refugees in Northern Italy unvaccinated for influenza. METHODS: An epidemiological investigation matched with a differential diagnosis was carried out for each sample collected from ILI cases to identify 10 viral pathogens (SARS-CoV-2, influenza virus type A and B, respiratory syncytial virus, metapneumovirus, parainfluenza viruses, rhinovirus, enterovirus, parechovirus, and adenovirus) by using specific real-time PCR assays according to the Centers for Disease Control and Prevention (CDC) protocols. In cases where the influenza virus type was identified, complete hemagglutinin (HA) gene sequencing and the related phylogenetic analysis were conducted. RESULTS: The outbreak was caused by influenza A(H3N2): the attack rate was 83.3% in children aged 9-14 years, 84.6% in those aged 15-24 years, and 28.6% in adults ≥25 years. Phylogenetic analyses uncovered that A(H3N2) strains were closely related since they segregated in the same cluster, showing both a high mean nucleotide identity (100%), all belonging to the genetic sub-group 3C.2a1b.2a.2, as those mainly circulating into the general population in the same period. CONCLUSIONS: The fact that influenza outbreak strains as well as the community strains were genetically related to the seasonal vaccine strain suggests that if an influenza prevention by vaccination strategy had been implemented, a lower attack rate of A(H3N2) and ILI cases might have been achieved.

Continuous evolution and emerging lineage of seasonal human coronaviruses: A multicenter surveillance study.

The seasonal human coronaviruses (HCoVs) have zoonotic origins, repeated infections, and global transmission. The objectives of this study are to elaborate the epidemiological and evolutionary characteristics of HCoVs from patients with acute respiratory illness. We conducted a multicenter surveillance at 36 sentinel hospitals of Beijing Metropolis, China, during 2016-2019. Patients with influenza-like illness (ILI) and severe acute respiratory infection (SARI) were included, and submitted respiratory samples for screening HCoVs by multiplex real-time reverse transcription-polymerase chain reaction assays. All the positive samples were used for metatranscriptomic sequencing to get whole genomes of HCoVs for genetical and evolutionary analyses. Totally, 321 of 15 677 patients with ILI or SARI were found to be positive for HCoVs, with an infection rate of 2.0% (95% confidence interval, 1.8%-2.3%). HCoV-229E, HCoV-NL63, HCoV-OC43, and HCoV-HKU1 infections accounted for 18.7%, 38.3%, 40.5%, and 2.5%, respectively. In comparison to ILI cases, SARI cases were significantly older, more likely caused by HCoV-229E and HCoV-OC43, and more often co-infected with other respiratory pathogens. A total of 179 full genome sequences of HCoVs were obtained from 321 positive patients. The phylogenetical analyses revealed that HCoV-229E, HCoV-NL63 and HCoV-OC43 continuously yielded novel lineages, respectively. The nonsynonymous to synonymous ratio of all key genes in each HCoV was less than one, indicating that all four HCoVs were under negative selection pressure. Multiple substitution modes were observed in spike glycoprotein among the four HCoVs. Our findings highlight the importance of enhancing surveillance on HCoVs, and imply that more variants might occur in the future.

Coupling an individual adaptive-decision model with a SIRV model of influenza vaccination reveals new insights for epidemic control.

Past seasonal influenza epidemics and vaccination experience may affect individuals' decisions on whether to be vaccinated or not, decisions that may be constantly reassessed in relation to recent influenza related experience. To understand the potentially complex interaction between experience and decisions and whether the vaccination rate is likely to reach a critical coverage level or not, we construct an adaptive-decision model. This model is then coupled with an influenza vaccination dynamics (SIRV) model to explore the interaction between individuals' decision-making and an influenza epidemic. Nonlinear least squares estimation is used to obtain the best-fit parameter values in the SIRV model based on data on new influenza-like illness (ILI) cases in Texas. Uncertainty and sensitivity analyses are then carried out to determine the impact of key parameters of the adaptive decision-making model on the ILI epidemic. The results showed that the necessary critical coverage rate of ILI vaccination could not be reached by voluntary vaccination. However, it could be reached in the fourth year if mass media reports improved individuals' memory of past vaccination experience. Individuals' memory of past vaccination experience, the proportion with histories of past vaccinations and the perceived cost of vaccination are important factors determining whether an ILI epidemic can be effectively controlled or not. Therefore, health authorities should guide people to improve their memory of past vaccination experience through media reports, publish timely data on annual vaccination proportions and adjust relevant measures to appropriately reduce vaccination perceived cost, in order to effectively control an ILI epidemic.

Comparison of Influenza-Like Illness (ILI) incidence data from the novel LeCellPHIA participatory surveillance system with COVID-19 case count data, Lesotho, July 2020 - July 2021.

BACKGROUND: While laboratory testing for infectious diseases such as COVID-19 is the surveillance gold standard, it is not always feasible, particularly in settings where resources are scarce. In the small country of Lesotho, located in sub-Saharan Africa, COVID-19 testing has been limited, thus surveillance data available to local authorities are limited. The goal of this study was to compare a participatory influenza-like illness (ILI) surveillance system in Lesotho with COVID-19 case count data, and ultimately to determine whether the participatory surveillance system adequately estimates the case count data. METHODS: A nationally-representative sample was called on their mobile phones weekly to create an estimate of incidence of ILI between July 2020 and July 2021. Case counts from the website Our World in Data (OWID) were used as the gold standard to which our participatory surveillance data were compared. We calculated Spearman's and Pearson's correlation coefficients to compare the weekly incidence of ILI reports to COVID-19 case count data. RESULTS: Over course of the study period, an ILI symptom was reported 1,085 times via participatory surveillance for an average annual cumulative incidence of 45.7 per 100 people (95% Confidence Interval [CI]: 40.7 - 51.4). The cumulative incidence of reports of ILI symptoms was similar among males (46.5, 95% CI: 39.6 - 54.4) and females (45.1, 95% CI: 39.8 - 51.1). There was a slightly higher annual cumulative incidence of ILI among persons living in peri-urban (49.5, 95% CI: 31.7 - 77.3) and urban settings compared to rural areas. The January peak of the participatory surveillance system ILI estimates correlated significantly with the January peak of the COVID-19 case count data (Spearman's correlation coefficient = 0.49; P < 0.001) (Pearson's correlation coefficient = 0.67; P < 0.0001). CONCLUSIONS: The ILI trends captured by the participatory surveillance system in Lesotho mirrored trends of the COVID-19 case count data from Our World in Data. Public health practitioners in geographies that lack the resources to conduct direct surveillance of infectious diseases may be able to use cell phone-based data collection to monitor trends.

Spatial distribution and driving factors of the associations between temperature and influenza-like illness in the United States: a time-stratified case-crossover study.

BACKGROUND: Several previous studies investigated the associations between temperature and influenza in a single city or region without a national picture. The attributable risk of influenza due to temperature and the corresponding driving factors were unclear. This study aimed to evaluate the spatial distribution characteristics of attributable risk of Influenza-like illness (ILI) caused by adverse temperatures and explore the related driving factors in the United States. METHODS: ILI, meteorological factors, and PM2.5 of 48 states in the United States were collected during 2011-2019. The time-stratified case-crossover design with a distributed lag non-linear model was carried out to evaluate the association between temperature and ILI at the state level. The multivariate meta-analysis was performed to obtain the combined effects at the national level. The attributable fraction (AF) was calculated to assess the ILI burden ascribed to adverse temperatures. The ordinary least square model (OLS), spatial lag model (SLM), and spatial error model (SEM) were utilized to identify driving factors. RESULTS: A total of 7,716,115 ILI cases were included in this study. Overall, the temperature was negatively associated with ILI risk, and lower temperature gave rise to a higher risk of ILI. AF ascribed to adverse temperatures differed across states, from 49.44% (95% eCI: 36.47% ~ 58.68%) in Montana to 6.51% (95% eCI: -6.49% ~ 16.46%) in Wisconsin. At the national level, 29.08% (95% eCI: 27.60% ~ 30.24%) of ILI was attributable to cold. Per 10,000 dollars increase in per-capita income was associated with the increment in AF (OLS: ß = -6.110, P = 0.021; SLM: ß = -5.496, P = 0.022; SEM: ß = -6.150, P = 0.022). CONCLUSION: The cold could enhance the risk of ILI and result in a considerable proportion of ILI disease burden. The ILI burden attributed to cold varied across states and was higher in those states with lower economic status. Targeted prevention programs should be considered to lower the burden of influenza.

Platelet count and risk of severe illness in hospitalised children with Influenza-Like illness.

AIM: To examine the clinical significance of thrombocytosis (platelets > 500 × 109 /L) in admitted children with an influenza-like illness. METHODS: We performed a database analysis consisting of patients evaluated at our medical centers with an influenza-like illness between 2009 and 2013. We included paediatric patients and examined the association between platelet count, respiratory viral infections, and admission outcomes (hospital length of stay and admission to the paediatric intensive care unit) using regression models adjusting for multiple variables. RESULTS: A total of 5171 children were included in the study cohort (median age 0.8 years; interquartile range, 0.2-1.8; 58% male). Younger age, and not the type of viral infection, was associated with a high platelet count (p < 0.001). Elevated platelet count independently predicted admission outcomes (p ≤ 0.05). The presence of thrombocytosis was associated with an increased risk for a prolonged length of stay (odds ratio = 1.2; 95% Confidence interval = 1.1 to 1.4; p = 0.003) and admission to the paediatric intensive care unit (odds ratio = 1.5; 95% Confidence interval = 1.1 to 2.0; p = 0.002). CONCLUSION: In children admitted with an influenza-like illness, a high platelet count is an independent predictor of admission outcomes. Platelet count may be used to improve risk assessment and management decisions in these paediatric patients.

Influenza Epidemic Trend Surveillance and Prediction Based on Search Engine Data: Deep Learning Model Study.

BACKGROUND: Influenza outbreaks pose a significant threat to global public health. Traditional surveillance systems and simple algorithms often struggle to predict influenza outbreaks in an accurate and timely manner. Big data and modern technology have offered new modalities for disease surveillance and prediction. Influenza-like illness can serve as a valuable surveillance tool for emerging respiratory infectious diseases like influenza and COVID-19, especially when reported case data may not fully reflect the actual epidemic curve. OBJECTIVE: This study aimed to develop a predictive model for influenza outbreaks by combining Baidu search query data with traditional virological surveillance data. The goal was to improve early detection and preparedness for influenza outbreaks in both northern and southern China, providing evidence for supplementing modern intelligence epidemic surveillance methods. METHODS: We collected virological data from the National Influenza Surveillance Network and Baidu search query data from January 2011 to July 2018, totaling 3,691,865 and 1,563,361 respective samples. Relevant search terms related to influenza were identified and analyzed for their correlation with influenza-positive rates using Pearson correlation analysis. A distributed lag nonlinear model was used to assess the lag correlation of the search terms with influenza activity. Subsequently, a predictive model based on the gated recurrent unit and multiple attention mechanisms was developed to forecast the influenza-positive trend. RESULTS: This study revealed a high correlation between specific Baidu search terms and influenza-positive rates in both northern and southern China, except for 1 term. The search terms were categorized into 4 groups: essential facts on influenza, influenza symptoms, influenza treatment and medicine, and influenza prevention, all of which showed correlation with the influenza-positive rate. The influenza prevention and influenza symptom groups had a lag correlation of 1.4-3.2 and 5.0-8.0 days, respectively. The Baidu search terms could help predict the influenza-positive rate 14-22 days in advance in southern China but interfered with influenza surveillance in northern China. CONCLUSIONS: Complementing traditional disease surveillance systems with information from web-based data sources can aid in detecting warning signs of influenza outbreaks earlier. However, supplementation of modern surveillance with search engine information should be approached cautiously. This approach provides valuable insights for digital epidemiology and has the potential for broader application in respiratory infectious disease surveillance. Further research should explore the optimization and customization of search terms for different regions and languages to improve the accuracy of influenza prediction models.

Characterization of Influenza-Like Illness Burden Using Commercial Wearable Sensor Data and Patient-Reported Outcomes: Mixed Methods Cohort Study.

BACKGROUND: The burden of influenza-like illness (ILI) is typically estimated via hospitalizations and deaths. However, ILI-associated morbidity that does not require hospitalization remains poorly characterized. OBJECTIVE: The main objective of this study was to characterize ILI burden using commercial wearable sensor data and investigate the extent to which these data correlate with self-reported illness severity and duration. Furthermore, we aimed to determine whether ILI-associated changes in wearable sensor data differed between care-seeking and non-care-seeking populations as well as between those with confirmed influenza infection and those with ILI symptoms only. METHODS: This study comprised participants enrolled in either the FluStudy2020 or the Home Testing of Respiratory Illness (HTRI) study; both studies were similar in design and conducted between December 2019 and October 2020 in the United States. The participants self-reported ILI-related symptoms and health care-seeking behaviors via daily, biweekly, and monthly surveys. Wearable sensor data were recorded for 120 and 150 days for FluStudy2020 and HTRI, respectively. The following features were assessed: total daily steps, active time (time spent with >50 steps per minute), sleep duration, sleep efficiency, and resting heart rate. ILI-related changes in wearable sensor data were compared between the participants who sought health care and those who did not and between the participants who tested positive for influenza and those with symptoms only. Correlative analyses were performed between wearable sensor data and patient-reported outcomes. RESULTS: After combining the FluStudy2020 and HTRI data sets, the final ILI population comprised 2435 participants. Compared with healthy days (baseline), the participants with ILI exhibited significantly reduced total daily steps, active time, and sleep efficiency as well as increased sleep duration and resting heart rate. Deviations from baseline typically began before symptom onset and were greater in the participants who sought health care than in those who did not and greater in the participants who tested positive for influenza than in those with symptoms only. During an ILI event, changes in wearable sensor data consistently varied with those in patient-reported outcomes. CONCLUSIONS: Our results underscore the potential of wearable sensors to discriminate not only between individuals with and without influenza infections but also between care-seeking and non-care-seeking populations, which may have future application in health care resource planning. TRIAL REGISTRATION: Clinicaltrials.gov NCT04245800; https://clinicaltrials.gov/ct2/show/NCT04245800.

Long Short-term Memory-Based Prediction of the Spread of Influenza-Like Illness Leveraging Surveillance, Weather, and Twitter Data: Model Development and Validation.

BACKGROUND: The potential to harness the plurality of available data in real time along with advanced data analytics for the accurate prediction of influenza-like illness (ILI) outbreaks has gained significant scientific interest. Different methodologies based on the use of machine learning techniques and traditional and alternative data sources, such as ILI surveillance reports, weather reports, search engine queries, and social media, have been explored with the ultimate goal of being used in the development of electronic surveillance systems that could complement existing monitoring resources. OBJECTIVE: The scope of this study was to investigate for the first time the combined use of ILI surveillance data, weather data, and Twitter data along with deep learning techniques toward the development of prediction models able to nowcast and forecast weekly ILI cases. By assessing the predictive power of both traditional and alternative data sources on the use case of ILI, this study aimed to provide a novel approach for corroborating evidence and enhancing accuracy and reliability in the surveillance of infectious diseases. METHODS: The model's input space consisted of information related to weekly ILI surveillance, web-based social (eg, Twitter) behavior, and weather conditions. For the design and development of the model, relevant data corresponding to the period of 2010 to 2019 and focusing on the Greek population and weather were collected. Long short-term memory (LSTM) neural networks were leveraged to efficiently handle the sequential and nonlinear nature of the multitude of collected data. The 3 data categories were first used separately for training 3 LSTM-based primary models. Subsequently, different transfer learning (TL) approaches were explored with the aim of creating various feature spaces combining the features extracted from the corresponding primary models' LSTM layers for the latter to feed a dense layer. RESULTS: The primary model that learned from weather data yielded better forecast accuracy (root mean square error [RMSE]=0.144; Pearson correlation coefficient [PCC]=0.801) than the model trained with ILI historical data (RMSE=0.159; PCC=0.794). The best performance was achieved by the TL-based model leveraging the combination of the 3 data categories (RMSE=0.128; PCC=0.822). CONCLUSIONS: The superiority of the TL-based model, which considers Twitter data, weather data, and ILI surveillance data, reflects the potential of alternative public sources to enhance accurate and reliable prediction of ILI spread. Despite its focus on the use case of Greece, the proposed approach can be generalized to other locations, populations, and social media platforms to support the surveillance of infectious diseases with the ultimate goal of reinforcing preparedness for future epidemics.

Healthcare-Seeking Behavior for Respiratory Illness Among Flu Near You Participants in the United States During the 2015-2016 Through 2018-2019 Influenza Seasons.

BACKGROUND: Flu Near You (FNY) is an online participatory syndromic surveillance system that collects health-related information. In this article, we summarized the healthcare-seeking behavior of FNY participants who reported influenza-like illness (ILI) symptoms. METHODS: We applied inverse probability weighting to calculate age-adjusted estimates of the percentage of FNY participants in the United States who sought health care for ILI symptoms during the 2015-2016 through 2018-2019 influenza season and compared seasonal trends across different demographic and regional subgroups, including age group, sex, census region, and place of care using adjusted χ 2 tests. RESULTS: The overall age-adjusted percentage of FNY participants who sought healthcare for ILI symptoms varied by season and ranged from 22.8% to 35.6%. Across all seasons, healthcare seeking was highest for the <18 and 65+ years age groups, women had a greater percentage compared with men, and the South census region had the largest percentage while the West census region had the smallest percentage. CONCLUSIONS: The percentage of FNY participants who sought healthcare for ILI symptoms varied by season, geographical region, age group, and sex. FNY compliments existing surveillance systems and informs estimates of influenza-associated illness by adding important real-time insights into healthcare-seeking behavior.

Circulation of Enterovirus D68 during Period of Increased Influenza-Like Illness, Maryland, USA, 2021.

We report enterovirus D68 circulation in Maryland, USA, during September-October 2021, which was associated with a spike in influenza-like illness. The characterized enterovirus D68 genomes clustered within the B3 subclade that circulated in 2018 in Europe and the United States.

Epidemiology of influenza under the coronavirus disease 2019 pandemic in Nanjing, China.

PURPOSE: Since the pandemic of coronavirus disease-19 (COVID-19), the incidence of influenza has decreased significantly, but there are still few reports in the short period before and after the pandemic period. This study aimed to explore influenza activity and dynamic changes before and during the pandemic. METHODS: A total of 1 324 357 influenza-like illness (ILI) cases were reported under the ILI surveillance network from January 1, 2018, to September 5, 2021, in Nanjing, of which 16 158 cases were detected in a laboratory. Differences in ILI and influenza were conducted with the χ2 test. RESULTS: The number of ILI cases accounted for 8.97% of outpatient and emergency department visits. The influenza-positive ratio (IPR) was 7.84% in ILI cases. During the COVID-19 pandemic, ILI% and IPR dropped by 6.03% and 11.83% on average, respectively. Besides this, IPR rose slightly in Week 30-35 of 2021. Not only differences in gender, age, and employment status, but also the circulating strains had changed from type A to B through the COVID-19 pandemic. CONCLUSION: The level of influenza activity was severely affected by COVID-19, but it seems that it is inevitable to be vigilant against the co-circulation in the future.

Dynamics of influenza-like illness under urbanization procedure and COVID-19 pandemic in the subcenter of Beijing during 2013-2021.

Influenza-like illness (ILI) varies in intensity year by year, generally keeping a stable pattern except for great changes of its epidemic pattern. Of the most impacting factors, urbanization has been suggested as shaping the intensity of influenza epidemics. Besides, growing evidence indicates the nonpharmaceutical interventions (NPIs) to severe acute respiratory syndrome coronavirus 2 offer great advantages in controlling infectious diseases. The present study aimed to evaluate the impact of urbanization and NPIs on the dynamic of ILI in Tongzhou, Beijing, during January 2013 to March 2021. ILI epidemiological surveillance data in Tongzhou district were obtained from Beijing Influenza Surveillance Network and separated into three periods of urbanization and four intervals of coronavirus disease 2019 pandemic. Standardized average incidence rates of ILI in each separate stages were calculated and compared by using Wilson method and time series model of seasonal ARIMA. Influenza seasonal outbreaks showed similar epidemic size and intensity before urbanization during 2013-2016. Increased ILI activity was found during the process of Tongzhou's urbanization during 2017-2019, with the rate difference of 2.48 (95% confidence interva [CI]: 2.44, 2.52) and the rate ratio of 1.75 (95% CI: 1.74, 1.76) of ILI incidence between preurbanization and urbanization periods. ILI activity abruptly decreased from the beginning of 2020 and kept at the bottom level almost in every epidemic interval. The top decrease in ILI activity by NPIs was shown in 5-14 years group in 2020-2021 influenza season, as 92.2% (95% CI: 78.3%, 95.2%). The results indicated that both urbanization and NPIs interrupted the epidemic pattern of ILI. We should pay more attention to public health when facing increasing population density, human contact, population mobility, and migration in the process of urbanization. NPIs and influenza vaccination should be implemented as necessary measures to protect people from common infectious diseases like ILI.