Conjunction of factors triggering waves of seasonal influenza

• Published in: eLife Vol. 7
• Main Authors: Chattopadhyay, Ishanu, Kiciman, Emre, Elliott, Joshua W, Shaman, Jeffrey L, Rzhetsky, Andrey
• Format: Journal Article
• Language: English
• Published: England eLife Sciences Publications Ltd 27.02.2018; eLife Sciences Publications, Ltd
• Subjects: Accounting; Air travel; Antigenic drift; Behavior; Causality; Disease Transmission, Infectious; Epidemiology and Global Health; Granger causality; Gravitational waves; Gravity; Humans; Humidity; Incidence; Infections; Infectious diseases; Influenza; Influenza, Human - epidemiology; Influenza, Human - transmission; Insurance claims; Longitudinal Studies; matching analysis; Mortality; Orthomyxoviridae - genetics; Orthomyxoviridae - immunology; Phase transitions; poisson regression; Public health; Seasons; Solar radiation; Spatio-Temporal Analysis; Travel; US influenza epidemic; Weather; United Kingdom > UK; United States > US; Chicago Illinois; global health; US influenza epidemic; epidemiology; poisson regression; human; virus; Granger causality; matching analysis
• ISSN: 2050-084X; 2050-084X
• DOI: 10.7554/eLife.30756

Using several longitudinal datasets describing putative factors affecting influenza incidence and clinical data on the disease and health status of over 150 million human subjects observed over a decade, we investigated the source and the mechanistic triggers of influenza epidemics. We conclude that the initiation of a pan-continental influenza wave emerges from the simultaneous realization of a complex set of conditions. The strongest predictor groups are as follows, ranked by importance: (1) the host population’s socio- and ethno-demographic properties; (2) weather variables pertaining to specific humidity, temperature, and solar radiation; (3) the virus’ antigenic drift over time; (4) the host population’€™s land-based travel habits, and; (5) recent spatio-temporal dynamics, as reflected in the influenza wave auto-correlation. The models we infer are demonstrably predictive (area under the Receiver Operating Characteristic curve 80%) when tested with out-of-sample data, opening the door to the potential formulation of new population-level intervention and mitigation policies. Influenza – or ‘the flu’ – is a contagious disease which sweeps across the globe like clockwork, claiming tens of thousands of lives. This is known as ‘seasonal flu’. Many scientists have tried to identify the factors that spark these yearly outbreaks. Some past studies have found that seasonal flu occurs when air that is normally humid turns dry, suggesting weather patterns play an important part. Other research has shown that air travel contributes to the flu spreading across the world. However, these studies typically focus on just one or two factors on their own. It is still not clear how exactly these factors combine to drive outbreaks, and then sustain the wave of infection. To address this, Chattopadhyay et al. analyze the medical histories of 150 million American people over a decade, combining this information with large datasets about the different factors that trigger flu outbreaks. This includes detailed data about air travel and weather patterns, as well as census data that describe features of the population. Patterns of movement are also examined, for example by processing billions of Twitter messages “tagged” with a location. Chattopadhyay et al. used all of these datasets to model outbreaks of the flu in the United States, and see which factors play the biggest role. It turns out that yearly outbreaks of seasonal flu are a result of a combination of elements. Some factors interact to help trigger the start of the wave, like humid weather in a highly populated area with nearby airports. Other factors, such how people move, encourage the spread of the infection. Finally, certain features of the population, for example how closely knitted a community is, make specific areas of the country more susceptible to the arrival of the disease. Overall, some of the most important elements of the model relate to the characteristics of the populations, the weather, the type of virus, and the number of short-distance journeys (rather than air travel). Understanding how and why outbreaks occur can help policy-makers design strategies that reduce the spread and impact of seasonal flu, which could potentially save thousands of lives. Ultimately, the model developed by Chattopadhyay et al. could be used to test whether these policies would work before they are implemented in the real world.