Respiratory syncytial virus (RSV) is a leading cause of infant mortality, causing approximately 700 infant deaths per day . In this paper we investigate the drivers of seasonal RSV epidemics, across both temperate and tropical locations, with the aim of informing future interventions and control efforts. We find that humidity and precipitation affect transmission and determine epidemics dynamics, and that climate change will alter future patterns of disease spread.
While important work has been done in forecasting the effects of climate change on vector-borne diseases, relatively less work has considered infectious diseases that spread person-to-person i.e. directly-transmitted diseases, such as RSV. Much of our understanding of climate and directly-transmitted diseases comes from influenza. Comprehensive research, based on observational studies and laboratory experiments, has determined that cold, low humidity conditions increase transmission of influenza in temperate climates. This research is grounded in experiments on guinea pigs: guinea pigs can contract and transmit influenza much like humans do (i.e. sneezing!) and therefore experiments can be conducted on guinea pig populations while varying ambient climate conditions.
Unfortunately for RSV research (but fortunately for the guinea pigs), there is no similar animal, guinea pigs or otherwise, that can be used to study RSV. Therefore, we rely on observational data to develop our core understanding of climate drivers. In our study we used hospitalization data to answer two questions: what are the climate drivers of RSV, and how might climate change affect future RSV epidemics?
To answer these questions, we took county-level RSV hospitalization data from the USA and state-level hospitalization data from Mexico. Combining these datasets allowed us to look at locations with a wide range of climate conditions: from counties in Minnesota which experience a large seasonal change in temperature, to states in southern Mexico where climate conditions fluctuate little throughout the year. We found that RSV cases tend to track these climate patterns: northern counties see intense outbreaks in the winter months and very few cases in the summer months, while southern locations tend to see persistent RSV cases throughout the year.
One interesting feature of RSV case dynamics is that in some locations we see biennial patterns of outbreaks. This means an intense wintertime outbreak is followed by a less intense outbreak in the following year (Fig. 2). It was very important that our disease model be able to capture these types of epidemic patterns, as well as capturing patterns from locations where the epidemic is consistent year-to-year. By using a mechanistic disease model, grounded in core theoretical understanding of how these types of dynamics are created, we were able to model these current RSV epidemic dynamics as well as consider how these dynamics might change with global warming.
This work would not have been possible without an unusual collaboration between researchers in the geosciences department and researchers in disease ecology who came together as part of a new climate-disease group to tackle problems at the intersection of climate change and health. By working with climate scientists, we were able to dive into questions such as the extent to which uncertainty across climate models might affect our results. We found spatial heterogeneity in the importance of uncertainty, with uncertain precipitation projections leading to diverging results in tropical locations. Reducing uncertainty in climate model projections of precipitation may be an important next step to accurately forecasting future changes in disease dynamics.
Read the full paper here. https://www.nature.com/articles/s41467-019-13562-y