Retrospective Parameter Estimation and Forecast of Respiratory Syncytial Virus in the United States

Current state and challenges in respiratory syncytial virus drug discovery and development

Human respiratory syncytial virus (RSV) is a leading cause of lower respiratory tract infections (LRTI) in young children and elderly people worldwide. Recent significant progress in our understanding of the structure and function of RSV proteins has led to the discovery of several clinical candidates targeting RSV fusion and replication. These include both the development of novel small molecule interventions and the isolation of potent monoclonal antibodies. In this review, we summarize the state-of-the-art of RSV drug discovery, with a focus on the characteristics of the candidates that reached the clinical stage of development. We also discuss the lessons learned from failed and discontinued clinical developments and highlight the challenges that remain for development of RSV therapies.

The acute bronchiolitis rebound in children after COVID-19 restrictions: a retrospective, observational analysis

BACKGROUND AND AIM

Bronchiolitis represents the main cause of illness and hospitalization in infants and young children. The aim of this study was to compare the Pediatric Emergency Department (PED) admissions for bronchiolitis during the post-COVID (Coronavirus disease) period to those of previous seasons and to analyze their etiology during COVID and post-COVID period.

METHODS

We compared demographics, clinical and microbiological data of children admitted to PED with bronchiolitis between September 2021 and March 2022 (post-COVID period) to the previous seasons (COVID and pre-COVID period).

RESULTS

During the post-COVID period the bronchiolitis season started earlier than usual, with a peak reached in November 2021; a gradual reduction was subsequently observed between December 2021 and January 2022. Our data showed a prevalence of High Priority code in children admitted to the PED with bronchiolitis during the post-COVID period (61.4%) compared the pre-COVID period (34.8%) (p=0.00). Also regarding the hospitalization of these patients, we found a major rate of hospitalization during this epidemic season (p=0.035). In addition, only 4 (1.5%) of the tested children resulted positive for SARS-CoV-2 and all of them were admitted to PED during the post-COVID period. The search for the other respiratory viruses showed during the current season a prevalence of respiratory syncytial virus (RSV) (60.2%), followed by Human Rhinovirus (30.1%).

CONCLUSIONS

The post-COVID period was characterized by an early and short-term peak in acute bronchiolitis, with an increased rate of hospitalization. In addition, SARS-CoV-2 infection was rarely cause of bronchiolitis in children under 2 years old.

Impact of COVID-19 Countermeasures on Pediatric Infections

(1) Background: General infection control measures have been implemented at the societal level against COVID-19 since the middle of 2020, namely, hand hygiene, universal masking, and social distancing. The suppressive effect of the social implementation of general infection control measures on pediatric infections has not been systematically assessed. (2) Methods: We addressed this issue based on publicly available data on 11 pediatric infections reported weekly by sentinel sites in Osaka and Iwate prefectures in Japan since 2010. We obtained the 5-year average for 2015-2019 and compared it to the weekly report for 2020-2021. (3) Results: The rate of 6 of the 11 pediatric infections decreased significantly during 2020-2021, regardless of the magnitude of the prevalence of COVID-19 in both areas. However, only RSV infection, one of the six infections, was endemic in 2021. Exanthem subitum was not as affected by COVID-19 countermeasures as other diseases. (4) Conclusions: The social implementation of infectious disease control measures was effective in controling certain infectious diseases in younger age groups, where compliance with the countermeasures should not be as high as that of adults.

An Outbreak of Severe Neonatal Pneumonia Caused by Human Respiratory Syncytial Virus BA9 in a Postpartum Care Centre in Shenyang, China

Human respiratory syncytial virus (HRSV) is a major pathogen of lower respiratory tract infections in children (<5 years) and older individuals, with outbreaks mainly reported among infants in hospital pediatric departments and intensive care units (ICUs). An outbreak of severe neonatal pneumonia occurred in a postpartum center in Shenyang city, China, from January to February 2021. In total, 34 respiratory samples were collected from 21 neonates and 13 nursing staff. The samples were screened for 27 pathogens using a TaqMan low-density array, and 20 samples tested positive for HRSV, including 16 neonates and 4 nursing staff samples. Among the 16 hospitalized neonates, seven were admitted to an ICU and nine to general wards. Four of the nursing staff had asymptomatic infections. To investigate the genetic characteristics of the HRSV responsible for this outbreak, the second hypervariable region (HVR2) sequences of the G gene were obtained from six neonates and two nursing staff. Phylogenetic analyses revealed that all eight sequences (SY strains) were identical, belonging to the HRSV BA9 genotype. Our findings highlight the necessity for strict hygiene and disease control measures so as to prevent cross-infection and further avoid potential outbreaks of severe infectious respiratory diseases. IMPORTANCE Human respiratory syncytial virus (HRSV) is one of the leading causes of acute lower respiratory infections (ALRI) worldwide. In this study, we first reported an outbreak of severe neonatal pneumonia caused by HRSVB BA9 at a postpartum care center in mainland China. Among 20 confirmed cases, 16 were hospitalized neonates with 7 in the ICU ward, and the other four were nursing staff with asymptomatic infections. Our findings highlighted the importance of preventing cross-infection in such postpartum centers.

RSV disease in infants and young children: Can we see a brighter future?

Respiratory syncytial virus (RSV) is a highly contagious seasonal virus and the leading cause of Lower Respiratory Tract Infections (LRTI), including pneumonia and bronchiolitis in children. RSV-related LRTI cause approximately 3 million hospitalizations and 120,000 deaths annually among children <5 years of age. The majority of the burden of RSV occurs in previously healthy infants. Only a monoclonal antibody (mAb) has been approved against RSV infections in a restricted group, leaving an urgent unmet need for a large number of children potentially benefiting from preventive measures. Approaches under development include maternal vaccines to protect newborns, extended half-life monoclonal antibodies to provide rapid long-lasting protection, and pediatric vaccines. RSV has been identified as a major global priority but a solution to tackle this unmet need for all children has yet to be implemented. New technologies represent the avenue for effectively addressing the leading-cause of hospitalization in children <1 years old.

Comparison of the efficacy and safety of different immunization routes induced by human respiratory syncytial virus F protein with CpG adjuvant in mice

Human respiratory syncytial virus (HRSV) is a leading cause worldwide of severe respiratory illness in infants and the elderly. The ideal HRSV vaccine should induce a systemic immune response, especially mucosal immunity. In this study, mice were immunized twice with F protein combined with CpG adjuvant to compare the safety and efficacy of 4 immunization routes, including intranasal primed/intramuscular boosted immunization (CpG + F/in+im), intramuscular primed/intranasal boosted immunization (CpG + F/im+in), intramuscular primed/intramuscular boosted immunization (CpG + F/im + im) and intranasal primed/intranasal boosted immunization (CpG + F/in+in). Compared with the control group (CpG/in+im, CpG/im+in, CpG/im + im and CpG/in+in), all 4 immunization routes induced a high titer of neutralizing antibodies and a strong cellular immune response. Mice in the CpG + F/in+in group induced the highest antibody neutralization titer, and IgA antibody in bronchoalveolar lavage fluid (BALF) was the highest. The copy of HRSVs in the lung decreased by approximately 3 log10. As seen from the IgG1/IgG2a and IFN-γ/IL-4-secreting lymphocyte ratios, compared with the mice in the CpG + F/im + im group, mice in the CpG + F/in+in group induced Th1-baised humoral and cellular immune responses and significantly reduced lung pathological injury. In conclusion, among the 4 immunization routes, the safety and efficacy induced by the mice in the CpG + F/in+in group were the best. We can conclude that intranasal immunization is superior to intramuscular immunization using F protein with CpG adjuvant as vaccine candidates.

Distinguishing viruses responsible for influenza-like illness

The many respiratory viruses that cause influenza-like illness (ILI) are reported and tracked as one entity, defined by the CDC as a group of symptoms that include a fever of 100 degrees Fahrenheit, a cough, and/or a sore throat. In the United States alone, ILI impacts 9-49 million people every year. While tracking ILI as a single clinical syndrome is informative in many respects, the underlying viruses differ in parameters and outbreak properties. Most existing models treat either a single respiratory virus or ILI as a whole. However, there is a need for models capable of comparing several individual viruses that cause respiratory illness, including ILI. To address this need, here we present a flexible model and simulations of epidemics for influenza, RSV, rhinovirus, seasonal coronavirus, adenovirus, and SARS/MERS, parameterized by a systematic literature review and accompanied by a global sensitivity analysis. We find that for these biological causes of ILI, their parameter values, timing, prevalence, and proportional contributions differ substantially. These results demonstrate that distinguishing the viruses that cause ILI will be an important aspect of future work on diagnostics, mitigation, modeling, and preparation for future pandemics.

Use of mathematical modelling to assess respiratory syncytial virus epidemiology and interventions: a literature review

Respiratory syncytial virus (RSV) is a leading cause of acute lower respiratory tract infection worldwide, resulting in approximately sixty thousand annual hospitalizations of< 5-year-olds in the United States alone and three million annual hospitalizations globally. The development of over 40 vaccines and immunoprophylactic interventions targeting RSV has the potential to significantly reduce the disease burden from RSV infection in the near future. In the context of RSV, a highly contagious pathogen, dynamic transmission models (DTMs) are valuable tools in the evaluation and comparison of the effectiveness of different interventions. This review, the first of its kind for RSV DTMs, provides a valuable foundation for future modelling efforts and highlights important gaps in our understanding of RSV epidemics. Specifically, we have searched the literature using Web of Science, Scopus, Embase, and PubMed to identify all published manuscripts reporting the development of DTMs focused on the population transmission of RSV. We reviewed the resulting studies and summarized the structure, parameterization, and results of the models developed therein. We anticipate that future RSV DTMs, combined with cost-effectiveness evaluations, will play a significant role in shaping decision making in the development and implementation of intervention programs.

‘Exploring socioeconomic status as a global determinant of COVID-19 prevalence, using statistical, exploratory data analytic, and supervised machine learning techniques.’ (Preprint)

Background

The COVID-19 pandemic represents the most unprecedented global challenge in recent times. As the global community attempts to manage the pandemic in the long term, it is pivotal to understand what factors drive prevalence rates and to predict the future trajectory of the virus.

Objective

This study had 2 objectives. First, it tested the statistical relationship between socioeconomic status and COVID-19 prevalence. Second, it used machine learning techniques to predict cumulative COVID-19 cases in a multicountry sample of 182 countries. Taken together, these objectives will shed light on socioeconomic status as a global risk factor of the COVID-19 pandemic.

Methods

This research used exploratory data analysis and supervised machine learning methods. Exploratory analysis included variable distribution, variable correlations, and outlier detection. Following this, the following 3 supervised regression techniques were applied: linear regression, random forest, and adaptive boosting (AdaBoost). Results were evaluated using k-fold cross-validation and subsequently compared to analyze algorithmic suitability. The analysis involved 2 models. First, the algorithms were trained to predict 2021 COVID-19 prevalence using only 2020 reported case data. Following this, socioeconomic indicators were added as features and the algorithms were trained again. The Human Development Index (HDI) metrics of life expectancy, mean years of schooling, expected years of schooling, and gross national income were used to approximate socioeconomic status.

Results

All variables correlated positively with the 2021 COVID-19 prevalence, with R² values ranging from 0.55 to 0.85. Using socioeconomic indicators, COVID-19 prevalence was predicted with a reasonable degree of accuracy. Using 2020 reported case rates as a lone predictor to predict 2021 prevalence rates, the average predictive accuracy of the algorithms was low (R²=0.543). When socioeconomic indicators were added alongside 2020 prevalence rates as features, the average predictive performance improved considerably (R²=0.721) and all error statistics decreased. Thus, adding socioeconomic indicators alongside 2020 reported case data optimized the prediction of COVID-19 prevalence to a considerable degree. Linear regression was the strongest learner with R²=0.693 on the first model and R²=0.763 on the second model, followed by random forest (0.481 and 0.722) and AdaBoost (0.454 and 0.679). Following this, the second model was retrained using a selection of additional COVID-19 risk factors (population density, median age, and vaccination uptake) instead of the HDI metrics. However, average accuracy dropped to 0.649, which highlights the value of socioeconomic status as a predictor of COVID-19 cases in the chosen sample.

Conclusions

The results show that socioeconomic status is an important variable to consider in future epidemiological modeling, and highlights the reality of the COVID-19 pandemic as a social phenomenon and a health care phenomenon. This paper also puts forward new considerations about the application of statistical and machine learning techniques to understand and combat the COVID-19 pandemic.

No Change of Pneumocystis jirovecii Pneumonia after the COVID-19 Pandemic: Multicenter Time-Series Analyses

Consolidated infection control measures imposed by the government and hospitals during COVID-19 pandemic resulted in a sharp decline of respiratory viruses. Based on the issue of whether Pneumocystis jirovecii could be transmitted by airborne and acquired from the environment, we assessed changes in P. jirovecii pneumonia (PCP) cases in a hospital setting before and after COVID-19. We retrospectively collected data of PCP-confirmed inpatients aged ≥18 years (N = 2922) in four university-affiliated hospitals between January 2015 and June 2021. The index and intervention dates were defined as the first time of P. jirovecii diagnosis and January 2020, respectively. We predicted PCP cases for post-COVID-19 and obtained the difference (residuals) between forecasted and observed cases using the autoregressive integrated moving average (ARIMA) and the Bayesian structural time-series (BSTS) models. Overall, the average of observed PCP cases per month in each year were 36.1 and 47.3 for pre- and post-COVID-19, respectively. The estimate for residuals in the ARIMA model was not significantly different in the total PCP-confirmed inpatients (7.4%, p = 0.765). The forecasted PCP cases by the BSTS model were not significantly different from the observed cases in the post-COVID-19 (−0.6%, 95% credible interval; −9.6~9.1%, p = 0.450). The unprecedented strict non-pharmacological interventions did not affect PCP cases.

Early Impact of Social Distancing in Response to Coronavirus Disease 2019 on Hospitalizations for Acute Bronchiolitis in Infants in Brazil

Background

Interventions to tackle the COVID-19 pandemic may affect the burden of other respiratory diseases. Considering the repercussion of these unique social experiences in infant’s health, this study aims to assess the early impact of social distancing due to the COVID-19 pandemic in hospital admissions for acute bronchiolitis.

Methods

Data from hospitalizations of acute bronchiolitis in infants under one year were obtained from the Department of Informatics of the Brazilian Public Health database (DATASUS) for the period between 2016 and 2020. These data were also analyzed by macro-regions of Brazil (North, Northeast, Southeast, South and Midwest). To evaluate the effect of social distancing strategy on the incidence of acute bronchiolitis, the absolute and relative reduction was calculated by analyzing the yearly subsets of 2016vs2020, 2017vs2020, 2018vs2020, and 2019vs2020.

Results

There was a significant reduction in all comparisons, ranging from -78% [IRR 0.22 (0.20 to 0.24)] in 2016vs2020 at -85% [IRR 0.15 (0.13 to 0.16)] in 2019vs2020, for the data from Brazil. For analyzes by macro-regions, the reduction varied from -58% [IRR 0.41 (0.37 to 0.45)] in the Midwest in 2016vs2020 to -93% [IRR 0.07 (0.06 to 0.08)] in the South in 2019vs2020.

Conclusions

There was a significant reduction in hospitalization for acute bronchiolitis in children under one year old, in Brazil, of the order of more than 70% for most analysis. Our data suggest an important impact of social distance on reducing the transmission of viruses related to acute bronchiolitis. Such knowledge may guide strategies for prevention of viruses spread.

Pediatric Infectious Disease Group (GPIP) position paper on the immune debt of the COVID-19 pandemic in childhood, how can we fill the immunity gap?

Since the beginning of the COVID-19 pandemic, reduced incidence of many viral and bacterial infections has been reported in children: bronchiolitis, varicella, measles, pertussis, pneumococcal and meningococcal invasive diseases. The purpose of this opinion paper is to discuss various situations that could lead to larger epidemics when the non-pharmaceutical interventions (NPI) imposed by the SARS-CoV-2 epidemic will no longer be necessary. While NPIs limited the transmission of SARS-CoV-2, they also reduced the spread of other pathogens during and after lockdown periods, despite the re-opening of schools since June 2020 in France. This positive collateral effect in the short term is welcome as it prevents additional overload of the healthcare system. The lack of immune stimulation due to the reduced circulation of microbial agents and to the related reduced vaccine uptake induced an "immunity debt" which could have negative consequences when the pandemic is under control and NPIs are lifted. The longer these periods of "viral or bacterial low-exposure" are, the greater the likelihood of future epidemics. This is due to a growing proportion of "susceptible" people and a declined herd immunity in the population. The observed delay in vaccination program without effective catch-up and the decrease in viral and bacterial exposures lead to a rebound risk of vaccine-preventable diseases. With a vaccination schedule that does not include vaccines against rotavirus, varicella, and serogroup B and ACYW Neisseria meningitidis, France could become more vulnerable to some of these rebound effects.

A Spatiotemporal Tool to Project Hospital Critical Care Capacity and Mortality From COVID-19 in US Counties

Objectives. To create a tool to rapidly determine where pandemic demand for critical care overwhelms county-level surge capacity and to compare public health and medical responses.Methods. In March 2020, COVID-19 cases requiring critical care were estimated using an adaptive metapopulation SEIR (susceptible‒exposed‒infectious‒recovered) model for all 3142 US counties for future 21-day and 42-day periods from April 2, 2020, to May 13, 2020, in 4 reactive patterns of contact reduction-0%, 20%, 30%, and 40%-and 4 surge response scenarios-very low, low, medium, and high.Results. In areas with increased demand, surge response measures could avert 104 120 additional deaths-55% through high clearance of critical care beds and 45% through measures such as greater ventilator access. The percentages of lives saved from high levels of contact reduction were 1.9 to 4.2 times greater than high levels of hospital surge response. Differences in projected versus actual COVID-19 demands were reasonably small over time.Conclusions. Nonpharmaceutical public health interventions had greater impact in minimizing preventable deaths during the pandemic than did hospital critical care surge response. Ready-to-go spatiotemporal supply and demand data visualization and analytics tools should be advanced for future preparedness and all-hazards disaster response.

Differential effects of intervention timing on COVID-19 spread in the United States

Assessing the effects of early nonpharmaceutical interventions on coronavirus disease 2019 (COVID-19) spread is crucial for understanding and planning future control measures to combat the pandemic. We use observations of reported infections and deaths, human mobility data, and a metapopulation transmission model to quantify changes in disease transmission rates in U.S. counties from 15 March to 3 May 2020. We find that marked, asynchronous reductions of the basic reproductive number occurred throughout the United States in association with social distancing and other control measures. Counterfactual simulations indicate that, had these same measures been implemented 1 to 2 weeks earlier, substantial cases and deaths could have been averted and that delayed responses to future increased incidence will facilitate a stronger rebound of infections and death. Our findings underscore the importance of early intervention and aggressive control in combatting the COVID-19 pandemic.

Differential Effects of Intervention Timing on COVID-19 Spread in the United States

Assessing the effects of early non-pharmaceutical interventions^1–5 on COVID-19 spread in the United States is crucial for understanding and planning future control measures to combat the ongoing pandemic^6–10. Here we use county-level observations of reported infections and deaths¹¹, in conjunction with human mobility data¹² and a metapopulation transmission model^13,14, to quantify changes of disease transmission rates in US counties from March 15, 2020 to May 3, 2020. We find significant reductions of the basic reproductive numbers in major metropolitan areas in association with social distancing and other control measures. Counterfactual simulations indicate that, had these same control measures been implemented just 1–2 weeks earlier, a substantial number of cases and deaths could have been averted. Specifically, nationwide, 56.5% [95% Cl: 48.1%−65.9%] of reported infections and 54.0% [95% Cl: 43.6%−63.8%] of reported deaths as of May 3, 2020 could have been avoided if the same control measures had been implemented just one week earlier. We also examine the effects of delays in re-implementing social distancing following a relaxation of control measures. A longer response time results in a stronger rebound of infections and death. Our findings underscore the importance of early intervention and aggressive response in controlling the COVID-19 pandemic.

Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2)

Estimation of the prevalence and contagiousness of undocumented novel coronavirus [severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2)] infections is critical for understanding the overall prevalence and pandemic potential of this disease. Here, we use observations of reported infection within China, in conjunction with mobility data, a networked dynamic metapopulation model, and Bayesian inference, to infer critical epidemiological characteristics associated with SARS-CoV-2, including the fraction of undocumented infections and their contagiousness. We estimate that 86% of all infections were undocumented [95% credible interval (CI): 82-90%] before the 23 January 2020 travel restrictions. The transmission rate of undocumented infections per person was 55% the transmission rate of documented infections (95% CI: 46-62%), yet, because of their greater numbers, undocumented infections were the source of 79% of the documented cases. These findings explain the rapid geographic spread of SARS-CoV-2 and indicate that containment of this virus will be particularly challenging.

Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2)

Estimation of the prevalence and contagiousness of undocumented novel coronavirus [severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2)] infections is critical for understanding the overall prevalence and pandemic potential of this disease. Here, we use observations of reported infection within China, in conjunction with mobility data, a networked dynamic metapopulation model, and Bayesian inference, to infer critical epidemiological characteristics associated with SARS-CoV-2, including the fraction of undocumented infections and their contagiousness. We estimate that 86% of all infections were undocumented [95% credible interval (CI): 82-90%] before the 23 January 2020 travel restrictions. The transmission rate of undocumented infections per person was 55% the transmission rate of documented infections (95% CI: 46-62%), yet, because of their greater numbers, undocumented infections were the source of 79% of the documented cases. These findings explain the rapid geographic spread of SARS-CoV-2 and indicate that containment of this virus will be particularly challenging.

Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV-2)

Estimation of the prevalence and contagiousness of undocumented novel coronavirus [severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2)] infections is critical for understanding the overall prevalence and pandemic potential of this disease. Here, we use observations of reported infection within China, in conjunction with mobility data, a networked dynamic metapopulation model, and Bayesian inference, to infer critical epidemiological characteristics associated with SARS-CoV-2, including the fraction of undocumented infections and their contagiousness. We estimate that 86% of all infections were undocumented [95% credible interval (CI): 82-90%] before the 23 January 2020 travel restrictions. The transmission rate of undocumented infections per person was 55% the transmission rate of documented infections (95% CI: 46-62%), yet, because of their greater numbers, undocumented infections were the source of 79% of the documented cases. These findings explain the rapid geographic spread of SARS-CoV-2 and indicate that containment of this virus will be particularly challenging.

What is Machine Learning? A Primer for the Epidemiologist

Machine learning is a branch of computer science that has the potential to transform epidemiologic sciences. Amid a growing focus on "Big Data," it offers epidemiologists new tools to tackle problems for which classical methods are not well-suited. In order to critically evaluate the value of integrating machine learning algorithms and existing methods, however, it is essential to address language and technical barriers between the two fields that can make it difficult for epidemiologists to read and assess machine learning studies. Here, we provide an overview of the concepts and terminology used in machine learning literature, which encompasses a diverse set of tools with goals ranging from prediction to classification to clustering. We provide a brief introduction to 5 common machine learning algorithms and 4 ensemble-based approaches. We then summarize epidemiologic applications of machine learning techniques in the published literature. We recommend approaches to incorporate machine learning in epidemiologic research and discuss opportunities and challenges for integrating machine learning and existing epidemiologic research methods.

Incidence of respiratory syncytial virus infection in children with congenital heart disease undergoing immunoprophylaxis with palivizumab in Pará state, north region of Brazil

Background

Palivizumab prophylaxis for the human respiratory syncytial virus (HRSV) has been reported to reduce the risk of hospital admissions related to HRSV in children with congenital heart disease (CHD). These children are at high risk of developing severe lower respiratory tract infection (LRTI) due to HRSV infection. Our goal was to evaluate the incidence of HRSV infection in children with CHD after being submitted to immunoprophylaxis with palivizumab in Pará state, North region of Brazil.

Methods

A prospective and observational cohort study was performed in children ≤2 years of age with CHD who received palivizumab immunoprophylaxis between January 1 and June 31, 2016. A questionnaire about basic non-medical care measures was applied to parents/legal representatives. Data on patients’ demographic characteristics, household environment, and respiratory infections were evaluated. HRSV infection was determined by qPCR.

Results

There were 104 children enrolled in this investigation and the results showed a mean age of 10.6 months, an average weight of 7.3 kg and 3.5 doses of palivizumab per children during seasonality of HRSV. Respiratory infection was observed in 27.9% of cases, of which 9.6% were LRTI. No case of children who received palivizumab immunoprophylaxis and developed influenza-like symptoms tested positive for HRSV.

Conclusion

Although the lack of a control group doesn’t allow to affirm the effectiveness of HRSV passive immunization, the immunoprophylaxis with palivizumab appeared to be totally efficient in preventing respiratory infection by HRSV in children up to two years of age with CHD.

Context matters: using reinforcement learning to develop human-readable, state-dependent outbreak response policies

The number of all possible epidemics of a given infectious disease that could occur on a given landscape is large for systems of real-world complexity. Furthermore, there is no guarantee that the control actions that are optimal, on average, over all possible epidemics are also best for each possible epidemic. Reinforcement learning (RL) and Monte Carlo control have been used to develop machine-readable context-dependent solutions for complex problems with many possible realizations ranging from video-games to the game of Go. RL could be a valuable tool to generate context-dependent policies for outbreak response, though translating the resulting policies into simple rules that can be read and interpreted by human decision-makers remains a challenge. Here we illustrate the application of RL to the development of context-dependent outbreak response policies to minimize outbreaks of foot-and-mouth disease. We show that control based on the resulting context-dependent policies, which adapt interventions to the specific outbreak, result in smaller outbreaks than static policies. We further illustrate two approaches for translating the complex machine-readable policies into simple heuristics that can be evaluated by human decision-makers. This article is part of the theme issue 'Modelling infectious disease outbreaks in humans, animals and plants: epidemic forecasting and control'. This theme issue is linked with the earlier issue 'Modelling infectious disease outbreaks in humans, animals and plants: approaches and important themes'.

Predictability in process-based ensemble forecast of influenza

Process-based models have been used to simulate and forecast a number of nonlinear dynamical systems, including influenza and other infectious diseases. In this work, we evaluate the effects of model initial condition error and stochastic fluctuation on forecast accuracy in a compartmental model of influenza transmission. These two types of errors are found to have qualitatively similar growth patterns during model integration, indicating that dynamic error growth, regardless of source, is a dominant component of forecast inaccuracy. We therefore examine the nonlinear growth of model initial error and compute the fastest growing directions using singular vector analysis. Using this information, we generate perturbations in an ensemble forecast system of influenza to obtain more optimal ensemble spread. In retrospective forecasts of historical outbreaks for 95 US cities from 2003 to 2014, this approach improves short-term forecast of incidence over the next one to four weeks.

Superensemble forecast of respiratory syncytial virus outbreaks at national, regional, and state levels in the United States

Respiratory syncytial virus (RSV) infections peak during the winter months in the United States, yet the timing, intensity, and onset of these outbreaks vary each year. An RSV vaccine is on the cusp of being released; precise models and accurate forecasts of RSV epidemics may prove vital for planning where and when the vaccine should be deployed. Accurate forecasts with sufficient spatial and temporal resolution could also be used to support the prevention or treatment of RSV infections. Previously, we developed and validated an RSV forecast system at the regional scale in the United States. This model-inference system had considerable forecast skill, relative to the historical expectance, for outbreak peak intensity, total outbreak size, and onset, but only marginal skill for predicting the timing of the outbreak peak. Here, we use a superensemble approach to combine three forecasting methods for RSV prediction in the US at three different spatial resolutions: national, regional, and state. At the regional and state levels, we find a substantial improvement of forecast skill, relative to historical expectance, for peak intensity, timing, and onset outbreak up to two months in advance of the predicted outbreak peak. Moreover, due to the greater variability of RSV outbreaks at finer spatial scales, we find that improvement of forecast skill at the state level exceeds that at the regional and national levels. Such finer scale superensemble forecasts may be more relevant for effecting local-scale interventions, particularly in communities with a high burden of RSV infection.

Real-time decision-making during emergency disease outbreaks

In the event of a new infectious disease outbreak, mathematical and simulation models are commonly used to inform policy by evaluating which control strategies will minimize the impact of the epidemic. In the early stages of such outbreaks, substantial parameter uncertainty may limit the ability of models to provide accurate predictions, and policymakers do not have the luxury of waiting for data to alleviate this state of uncertainty. For policymakers, however, it is the selection of the optimal control intervention in the face of uncertainty, rather than accuracy of model predictions, that is the measure of success that counts. We simulate the process of real-time decision-making by fitting an epidemic model to observed, spatially-explicit, infection data at weekly intervals throughout two historical outbreaks of foot-and-mouth disease, UK in 2001 and Miyazaki, Japan in 2010, and compare forward simulations of the impact of switching to an alternative control intervention at the time point in question. These are compared to policy recommendations generated in hindsight using data from the entire outbreak, thereby comparing the best we could have done at the time with the best we could have done in retrospect. Our results show that the control policy that would have been chosen using all the data is also identified from an early stage in an outbreak using only the available data, despite high variability in projections of epidemic size. Critically, we find that it is an improved understanding of the locations of infected farms, rather than improved estimates of transmission parameters, that drives improved prediction of the relative performance of control interventions. However, the ability to estimate undetected infectious premises is a function of uncertainty in the transmission parameters. Here, we demonstrate the need for both real-time model fitting and generating projections to evaluate alternative control interventions throughout an outbreak. Our results highlight the use of using models at outbreak onset to inform policy and the importance of state-dependent interventions that adapt in response to additional information throughout an outbreak.

Respiratory syncytial virus tracking using internet search engine data

Background

Respiratory Syncytial Virus (RSV) is the leading cause of hospitalization in children less than 1 year of age in the United States. Internet search engine queries may provide high resolution temporal and spatial data to estimate and predict disease activity.

Methods

After filtering an initial list of 613 symptoms using high-resolution Bing search logs, we used Google Trends data between 2004 and 2016 for a smaller list of 50 terms to build predictive models of RSV incidence for five states where long-term surveillance data was available. We then used domain adaptation to model RSV incidence for the 45 remaining US states.

Results

Surveillance data sources (hospitalization and laboratory reports) were highly correlated, as were laboratory reports with search engine data. The four terms which were most often statistically significantly correlated as time series with the surveillance data in the five state models were RSV, flu, pneumonia, and bronchiolitis. Using our models, we tracked the spread of RSV by observing the time of peak use of the search term in different states. In general, the RSV peak moved from south-east (Florida) to the north-west US.

Conclusions

Our study represents the first time that RSV has been tracked using Internet data results and highlights successful use of search filters and domain adaptation techniques, using data at multiple resolutions. Our approach may assist in identifying spread of both local and more widespread RSV transmission and may be applicable to other seasonal conditions where comprehensive epidemiological data is difficult to collect or obtain.

Simulation of four respiratory viruses and inference of epidemiological parameters

While influenza has been simulated extensively to better understand its behavior and predict future outbreaks, most other respiratory viruses have seldom been simulated. In this study, we provide an overview of four common respiratory viral infections: respiratory syncytial virus (RSV), respiratory adenovirus, rhinovirus and parainfluenza, present specimen data collected 2004–2014, and simulate outbreaks in 19 overlapping regions in the United States. Pairing a compartmental model and data assimilation methods, we infer key epidemiological parameters governing transmission: the basic reproductive number R₀ and length of infection D. RSV had been previously simulated, and our mean estimate of D and R₀ of 5.2 days and 2.8, respectively, are within published clinical and modeling estimates. Among the four virus groupings, mean estimates of R₀ range from 2.3 to 3.0, with a lower and upper quartile range of 2.0–2.8 and 2.6–3.2, respectively. As rapid PCR testing becomes more common, estimates of the observed virulence and duration of infection for these viruses could inform decision making by clinicians and officials for managing patient treatment and response.

Forecasting the spatial transmission of influenza in the United States

Recurrent outbreaks of seasonal and pandemic influenza create a need for forecasts of the geographic spread of this pathogen. Although it is well established that the spatial progression of infection is largely attributable to human mobility, difficulty obtaining real-time information on human movement has limited its incorporation into existing infectious disease forecasting techniques. In this study, we develop and validate an ensemble forecast system for predicting the spatiotemporal spread of influenza that uses readily accessible human mobility data and a metapopulation model. In retrospective state-level forecasts for 35 US states, the system accurately predicts local influenza outbreak onset,-i.e., spatial spread, defined as the week that local incidence increases above a baseline threshold-up to 6 wk in advance of this event. In addition, the metapopulation prediction system forecasts influenza outbreak onset, peak timing, and peak intensity more accurately than isolated location-specific forecasts. The proposed framework could be applied to emergent respiratory viruses and, with appropriate modifications, other infectious diseases.

Counteracting structural errors in ensemble forecast of influenza outbreaks

For influenza forecasts generated using dynamical models, forecast inaccuracy is partly attributable to the nonlinear growth of error. As a consequence, quantification of the nonlinear error structure in current forecast models is needed so that this growth can be corrected and forecast skill improved. Here, we inspect the error growth of a compartmental influenza model and find that a robust error structure arises naturally from the nonlinear model dynamics. By counteracting these structural errors, diagnosed using error breeding, we develop a new forecast approach that combines dynamical error correction and statistical filtering techniques. In retrospective forecasts of historical influenza outbreaks for 95 US cities from 2003 to 2014, overall forecast accuracy for outbreak peak timing, peak intensity and attack rate, are substantially improved for predicted lead times up to 10 weeks. This error growth correction method can be generalized to improve the forecast accuracy of other infectious disease dynamical models.

Inaccuracy of influenza forecasts based on dynamical models is partly due to nonlinear error growth. Here the authors address the error structure of a compartmental influenza model, and develop a new improved forecast approach combining dynamical error correction and statistical filtering techniques.