Foot-and-mouth disease virus (FMD) is endemic amongst the approximately 650,000 cattle that populate the Lake Chad Basin, Cameroon. FMD is highly contagious, and though rarely lethal by itself, can cause debilitating symptoms in its hosts including blisters on the mouth and feet which can cause difficulty eating and lameness. These clinical signs may prevent or delay the necessary weather driven nomadic patterns of the cattle movement in Cameroon, and prevent animals from obtaining the daily food and water that are critical to their survival and reproduction. FMD is therefore detrimental to herds' survival and the economics of the cattle industry in Cameroon, which supports a majority of households in the Lake Chad Basin.
Though there are far-reaching effects of FMD in Cameroon, very little research has been conducted on the role of the carrier class in the endemicity of the disease. This project seeks to understand this role by using an SIR-type ordinary differential equations system (SEICR - susceptible, exposed, infectious, carrier, and recovered) to model FMD under an endemic setting.
The Ohio State University's field team in Cameroon gathered statistics on the number of carrier and recovered cattle. This information is used in conjunction with the model and previous work to investigate associated parameters of the system, parameterize the model, and calculate the basic reproductive number (Ro) using the next generation matrix technique. Sensitivity analysis is performed using Latin hypercube sampling to investigate the properties of Ro and how the attributes related to carriers affect Ro, while assuming endemic equilibrium.
Results suggest that, when matching data from this endemic situation, the relative infectivity of the carrier class has no effect on the magnitude of Ro. However, the carrier cattle do still contribute to the spread of the disease. A carrier class with relative infectivity of only 6.55% is shown to be sufficient to drive the disease to be endemic from the carriers alone.
The parameterization of the model elucidates mathematical means for reducing Ro, and how to calculate either an upper or lower bound on the length of time spent in incubation, infectious, and carrier phases, and also the percentage of infectious cattle that become carriers. Sensitivity analysis shows that over full biologically plausible parameter ranges, Ro varies widely with an extremely long tail to its distribution. However, over reduced parameter ranges, where Ro is less sensitive to parameters, Ro ranges from only 6 to 14.