Since 2013, the Centers for Disease Control and Prevention (CDC) has hosted an annual influenza season forecasting challenge. The 2015–2016 challenge consisted of weekly probabilistic forecasts of multiple targets, including fourteen models submitted by eleven teams. Forecast skill was evaluated using a modified logarithmic score. We averaged submitted forecasts into a mean ensemble model and compared them against predictions based on historical trends. Forecast skill was highest for seasonal peak intensity and short-term forecasts, while forecast skill for timing of season onset and peak week was generally low. Higher forecast skill was associated with team participation in previous influenza forecasting challenges and utilization of ensemble forecasting techniques. The mean ensemble consistently performed well and outperformed historical trend predictions. CDC and contributing teams will continue to advance influenza forecasting and work to improve the accuracy and reliability of forecasts to facilitate increased incorporation into public health response efforts.

No abstract

Background Respiratory viral infections are a major cause of morbidity and mortality worldwide. However, their characterization is incomplete because prevalence estimates are based on syndromic surveillance data. Here, we address this shortcoming through the analysis of infection rates among individuals tested regularly for respiratory viral infections, irrespective of their symptoms. Methods We carried out longitudinal sampling and analysis among 214 individuals enrolled at multiple New York City locations from fall 2016 to spring 2018. We combined personal information with weekly nasal swab collection to investigate the prevalence of 18 respiratory viruses among different age groups and to assess risk factors associated with infection susceptibility. Results 17.5% of samples were positive for respiratory viruses. Some viruses circulated predominantly during winter, whereas others were found year round. Rhinovirus and coronavirus were most frequently detected. Children registered the highest positivity rates, and adults with daily contacts with children experienced significantly more infections than their counterparts without children. Conclusion Respiratory viral infections are widespread among the general population with the majority of individuals presenting multiple infections per year. The observations identify children as the principal source of respiratory infections. These findings motivate further active surveillance and analysis of differences in pathogenicity among respiratory viruses.

Respiratory viral infections are a leading cause of disease worldwide. A variety of respiratory viruses produce infections in humans with effects ranging from asymptomatic to life-treathening. Standard surveillance systems typically only target severe infections (ED outpatients, hospitalisations, deaths) and fail to track asymptomatic or mild infections. Here we performed a large-scale community study across multiple age groups to assess the pathogenicity of 18 respiratory viruses. We enrolled 214 individuals at multiple New York City locations and tested weekly for respiratory viral pathogens, irrespective of symptom status, from fall 2016 to spring 2018. We combined these test results with participant-provided daily records of cold and flu symptoms and used this information to characterise symptom severity by virus and age category. Asymptomatic infection rates exceeded 70% for most viruses, excepting influenza and human metapneumovirus, which produced significantly more severe outcomes. Symptoms were negatively associated with infection frequency, with children displaying the lowest score among age groups. Upper respiratory manifestations were most common for all viruses, whereas systemic effects were less typical. These findings indicate a high burden of asymptomatic respiratory virus infection exists in the general population.

A variety of mechanistic and statistical methods to forecast seasonal influenza have been proposed and are in use; however, the effects of various data issues and design choices (statistical versus mechanistic methods, for example) on the accuracy of these approaches have not been thoroughly assessed. Here, we compare the accuracy of three forecasting approaches—a mechanistic method, a weighted average of two statistical methods and a super-ensemble of eight statistical and mechanistic models—in predicting seven outbreak characteristics of seasonal influenza during the 2016–2017 season at the national and 10 regional levels in the USA. For each of these approaches, we report the effects of real time under- and over-reporting in surveillance systems, use of non-surveillance proxies of influenza activity and manual override of model predictions on forecast quality. Our results suggest that a meta-ensemble of statistical and mechanistic methods has better overall accuracy than the individual methods. Supplementing surveillance data with proxy estimates generally improves the quality of forecasts and transient reporting errors degrade the performance of all three approaches considerably. The improvement in quality from ad hoc and post-forecast changes suggests that domain experts continue to possess information that is not being sufficiently captured by current forecasting approaches.