SCoVMod - a spatially explicit mobility and deprivation adjusted model of first wave COVID-19 transmission dynamics.

Background: Mobility restrictions prevent the spread of infections to disease-free areas, and early in the coronavirus disease 2019 (COVID-19) pandemic, most countries imposed severe restrictions on mobility as soon as it was clear that containment of local outbreaks was insufficient to control spread. These restrictions have adverse impacts on the economy and other aspects of human health, and it is important to quantify their impact for evaluating their future value. Methods: Here we develop Scotland Coronavirus transmission Model (SCoVMod), a model for COVID-19 in Scotland, which presents unusual challenges because of its diverse geography and population conditions. Our fitted model captures spatio-temporal patterns of mortality in the first phase of the epidemic to a fine geographical scale. Results: We find that lockdown restrictions reduced transmission rates down to an estimated 12\% of its pre-lockdown rate. We show that, while the timing of COVID-19 restrictions influences the role of the transmission rate on the number of COVID-related deaths, early reduction in long distance movements does not. However, poor health associated with deprivation has a considerable association with mortality; the Council Area (CA) with the greatest health-related deprivation was found to have a mortality rate 2.45 times greater than the CA with the lowest health-related deprivation considering all deaths occurring outside of carehomes. Conclusions: We find that in even an early epidemic with poor case ascertainment, a useful spatially explicit model can be fit with meaningful parameters based on the spatio-temporal distribution of death counts. Our simple approach is useful to strategically examine trade-offs between travel related restrictions and physical distancing, and the effect of deprivation-related factors on outcomes.

A comparison of the value of two machine learning predictive models to support bovine tuberculosis disease control in England.

Nearly a decade into Defra's current eradication strategy, bovine tuberculosis (bTB) remains a serious animal health problem in England, with c.30,000 cattle slaughtered annually in the fight against this insidious disease. There is an urgent need to improve our understanding of bTB risk in order to enhance the current disease control policy. Machine learning approaches applied to big datasets offer a potential way to do this. Regularized regression and random forest machine learning methodologies were implemented using 2016 herd-level data to generate the best possible predictive models for a bTB incident in England and its three surveillance risk areas (High-risk area [HRA], Edge area [EA] and Low-risk area [LRA]). Their predictive performance was compared and the best models in each area were used to characterize herds according to risk. While all models provided excellent discrimination, random forest models achieved the highest balanced accuracy (i.e. average of sensitivity and specificity) in England, HRA and LRA, whereas the regularized regression LASSO model did so in the EA. The time since the last confirmed incident was resolved was the only variable in the top-ten ranking in all areas according to both types of models, which highlights the importance of bTB history as a predictor of a new incident. Risk categorisation based on Receiver Operating Characteristic (ROC) analysis was carried out using the best predictive models in each area setting a 99 % threshold value for sensitivity and specificity (97 % in the LRA). Thirteen percent of herds in the whole of England as well as in its HRA, 14 % in its EA and 31 % in its LRA were classified as high-risk. These could be selected for the deployment of additional disease control measures at national or area level. In this way, low-risk herds within the area considered would not be penalised unnecessarily by blanket control measures and limited resources be used more efficiently. The methodology presented in this paper demonstrates a way to accurately identify high-risk farms to inform a targeted disease control and prevention strategy in England that supplements existing population strategies.

Decision tree machine learning applied to bovine tuberculosis risk factors to aid disease control decision making.

Identifying and understanding the risk factors for endemic bovine tuberculosis (TB) in cattle herds is critical for the control of this disease. Exploratory machine learning techniques can uncover complex non-linear relationships and interactions within disease causation webs, and enhance our knowledge of TB risk factors and how they are interrelated. Classification tree analysis was used to reveal associations between predictors of TB in England and each of the three surveillance risk areas (High Risk, Edge, and Low Risk) in 2016, identifying the highest risk herds. The main classifying predictor for farms in England overall related to the TB prevalence in the 100 nearest cattle herds. In the High Risk and Edge areas it was the number of slaughterhouse destinations and in the Low Risk area it was the number of cattle tested in surveillance tests. How long ago the last confirmed incident was resolved was the most frequent classifier in trees; if within two years, leading to the highest risk group of herds in the High Risk and Low Risk areas. At least two different slaughterhouse destinations led to the highest risk group of herds in England, whereas in the Edge area it was a combination of no contiguous low-risk neighbours (i.e. in a 1 km radius) and a minimum proportion of 6-23 month-old cattle in November. A threshold value of prevalence in 100 nearest neighbours increased the risk in all areas, although the value was specific to each area. Having low-risk contiguous neighbours reduced the risk in the Edge and High Risk areas, whereas high-risk ones increased the risk in England overall and in the Edge area specifically. The best classification tree models informed multivariable binomial logistic regression models in each area, adding statistical inference outputs. These two approaches showed similar predictive performance although there were some disparities regarding what constituted high-risk predictors. Decision tree machine learning approaches can identify risk factors from webs of causation: information which may then be used to inform decision making for disease control purposes.

Using species distribution models to predict potential hot-spots for Rift Valley Fever establishment in the United Kingdom.

Vector borne diseases are a continuing global threat to both human and animal health. The ability of vectors such as mosquitos to cover large distances and cross country borders undetected provide an ever-present threat of pathogen spread. Many diseases can infect multiple vector species, such that even if the climate is not hospitable for an invasive species, indigenous species may be susceptible and capable of transmission such that one incursion event could lead to disease establishment in these species. Here we present a consensus modelling methodology to estimate the habitat suitability for presence of mosquito species in the UK deemed competent for Rift Valley fever virus (RVF) and demonstrate its application in an assessment of the relative risk of establishment of RVF virus in the UK livestock population. The consensus model utilises observed UK mosquito surveillance data, along with climatic and geographic prediction variables, to inform six independent species distribution models; the results of which are combined to produce a single prediction map. As a livestock host is needed to transmit RVF, we then combine the consensus model output with existing maps of sheep and cattle density to predict the areas of the UK where disease is most likely to establish in local mosquito populations. The model results suggest areas of high suitability for RVF competent mosquito species across the length and breadth of the UK. Notable areas of high suitability were the South West of England and coastal areas of Wales, the latter of which was subsequently predicted to be at higher risk for establishment of RVF due to higher livestock densities. This study demonstrates the applicability of outputs of species distribution models to help predict hot-spots for risk of disease establishment. While there is still uncertainty associated with the outputs we believe that the predictions are an improvement on just using the raw presence points from a database alone. The outputs can also be used as part of a multidisciplinary approach to inform risk based disease surveillance activities.

A new bovine tuberculosis model for England and Wales (BoTMEW) to simulate epidemiology, surveillance and control.

BACKGROUND: Bovine tuberculosis (bTB) is a zoonotic disease of cattle caused by Mycobacterium bovis, widespread in England and Wales. It has high incidence towards the South West of England and Wales, with much lower incidence to the East and North. A stochastic simulation model was developed to simulate M. bovis transmission among cattle, transfer by cattle movements and transmission from environmental reservoirs (often wildlife and especially badgers). It distinguishes five surveillance streams, including herd tests, pre-movement testing and slaughter surveillance. The model thereby simulates interventions in bTB surveillance and control, and generates outputs directly comparable to detailed disease records. An anonymized version of the executable model with its input data has been released. The model was fitted to cattle bTB records for 2008-2010 in a cross-sectional comparison, and its projection was compared with records from 2010 to 2016 for validation. RESULTS: The fitted model explained over 99% of the variation among numbers of breakdowns in four defined regions and surveillance streams in 2010. It classified 7800 (95% confidence interval c. 5500 - 14,000) holdings within high incidence regions as exposed to infectious environmental reservoirs, out of over 31,000 cattle holdings identified as potentially exposed to such sources. The model was consistent with previous estimates of low M. bovis transmission rate among cattle, but cattle to cattle transmission was clearly required to generate the number of cattle cases observed. When projected to 2016, the model as fitted to 2010 continued to match the distribution of bTB among counties, although it was notable that the actual distribution of bTB in 2010 was itself a close match for its distribution in 2016. CONCLUSIONS: The close model fit demonstrated that cattle movements could generate breakdowns as observed in low incidence regions, if persistent environmental reservoirs such as wildlife maintained infection levels in the high incidence regions. The model suggests that environmental reservoirs may be a challenge for control, because, although many holdings are exposed to infection from wildlife or the environment, they are a minority of holdings. Large impacts on disease in wildlife will be required to avoid each individual transmission event to cattle.

Epidemiology of extended spectrum beta-lactamase E. coli (CTX-M-15) on a commercial dairy farm.

The epidemiology of an extended spectrum beta-lactamase Escherichia coli (CTX-M-15) was observed and described on a commercial dairy farm located in the United Kingdom. During 2008 longitudinal sampling of faecal pat samples from different cattle groups comprising milking and non-milking cows, calving cows, calves, and the environment was carried out. The proportion of CTX-M-15 E. coli positive samples was significantly (p<0.0.01) higher in milking cows (30.3%, CI(95%) 26.8; 33.8) than in the herd as a whole (17.0%, CI(95%) 14.9; 19.0). In 2008 95.6% of sampled calves tested positive for CTX-M-15 E. coli at two days of age. A more detailed investigation in 2009 revealed that cows and heifers were approximately eight times more likely to test positive in the 10 days after calving than the 9 days before (OR 7.6, CI(95%) 2.32; 24.9). The CTX-M15 E. coli was also readily isolated from the immediate calving pen environment, including the water troughs. A cyclic pattern was apparent where cows immediately after calving and as high yielders were highly positive, but where the prevalence decreased during the dry period. The increased prevalence of the CTX-M-15 E. coli in certain cattle groups and farm environments including calving pens suggested that husbandry, antimicrobial usage and hygiene may play a significant role on a farm with regards to the epidemiology of CTX-M-15. This may offer a practical opportunity to reduce further dissemination through good practice and hygiene around calving.