Sir

Paul Torgerson and David Torgerson argue in Correspondence that research and surveillance for tuberculosis (TB) in cattle can only be justified on animal-health grounds because bovine TB poses a negligible threat to human health ('Does risk to humans justify high cost of fighting bovine TB?' Nature 455, 1029; 2008). But the relationship between surveillance of cattle and the risk of humans contracting the disease is not as simple as they imply.

The current surveillance to remove infected cattle at an early stage in the disease process reduces the prevalence and, more important, curtails the course of the disease in individual cattle. This is particularly relevant in areas of higher incidence, where testing is an annual event. For example, three UK counties breached 10% herd incidence of confirmed cases in 2007.

Without a test-and-cull policy and abattoir surveillance, bovine TB would advance to its more infectious stages and lead to an increased risk of transmission of this primarily airborne disease to humans. Removing animals at an early stage and controlling the prevalence of the disease in the national herd is therefore likely to reduce the probability of transmission to humans.

We would argue that active surveillance for bovine TB followed by the removal of infected cattle is part of the reason why this disease is not seen more often in humans. When using a cat to control mice, you shouldn't get rid of the cat just because you don't see any more mice.

We agree that protecting animal health and reducing losses in animal productivity are important justifications for surveillance and research of bovine TB. But keeping the risk of human exposure as low as possible must also be factored into the equation.

See also Bovine TB: stopping disease control would block all live exports