A novel method to quantify the impact of weather on Lyme disease

Add to your list(s) Download to your calendar using vCal

• Gianni Loiacono (PHE)
• Friday 10 March 2017, 16:00-17:00
• Meeting room 4, Centre for Mathematical Sciences.

If you have a question about this talk, please contact Prof. Julia Gog.

Many infectious diseases exhibit seasonal changes driven by the weather, socio-economic, behavioral and other environmental factors. Separating and quantifying the contribution of each driver is extremely difficult as often some of the explanatory variables are related (collinearity).

Lyme Borreliosis is a tick-borne bacterial infection, particularly relevant for public health. The incidence is strongly influenced by the weather and other environmental variables such as the presence of vertebrate host and land use. The 11-year study dataset of human cases reported to PHE in England and Wales was linked with weather data (air temperature, rainfall, relative humidity and wind speed) at diagnostic laboratory postcode and specimen date. Then we statistically analyzed the subsets of epidemiological cases when all variables, except one, were within the same narrow range; in this way, we could detect the relevant explanatory variables, and remove the problem of collinearity. We also developed a novel method to define and quantify the time-lag between relevant changes in the weather variables and the risk of Lyme disease.

We provide an estimation of the probability of acquiring a disease conditional on the relevant weather variables. The risk of an infection increases non-linearly with the temperature, but decreases with relative humidity, probably due to vector and human behavioral factors. Other variables, such as daylight duration which affects diapause in ticks, might play important roles in the risk of disease.

We developed a robust, yet conceptually simple, method to identify the relevant weather variables involved in the disease transmission, and disentangled their relative contribution. By allowing the memory of past events in the model, this method could be generalized to a wider class of diseases for which human-to-human transmission and depletion of susceptible population are significant, such as respiratory diseases.

This talk is part of the Worms and Bugs series.

This talk is included in these lists:

• All CMS events
• All Talks (aka the CURE list)
• Cambridge Immunology
• Cambridge Infectious Diseases
• Meeting room 4, Centre for Mathematical Sciences
• Worms and Bugs
• bld31

Note that ex-directory lists are not shown.