The Epidemiology of Mycobacterium bovis Infection in Cattle 51
cattle testing positive to tuberculin peaks with
middle-age cattle.
4.3 Patterns of Transmission
in Managed Populations: The
Confounding Impact of ControlControl programmes for the management of
bovine tuberculosis in industrial countries pro-
vide perhaps the richest sources of epidemiologi-
cal data on bovine TB. However, the dynamic
and context-sensitive nature of control pro-
grammes for bovine TB confounds the patterns
of disease within herds that may be expected
from studying the disease in unmanaged popu-
lations. The frequency of testing and removal of
animals occurs on a timescale far shorter than
the natural timescales of transmission which is
reflected in the epidemiological patterns we see
in these populations.
In this section, we concentrate on the epide-
miology of bovine tuberculosis in the UK and the
Republic of Ireland. The relatively high preva-
lence of infection in the face of national control
programmes (Abernethy et al., 2013) combined
with detailed demographic and cattle movement
data (Mitchell et al., 2005) has facilitated highly
sophisticated analyses and modelling of trans-
mission. These analyses attempt to unpick the
mechanisms of transmission, local persistence
and efficiency of tuberculin testing that are
responsible for the successes, and failures, of
control in these populations.
4.3.1 Patterns of within-herd incidenceControl of bovine tuberculosis, particularly in
Europe, is targeted at the herd level. Herds are
tuberculin tested to demonstrate their freedom
from infection at an interval that depends on
their risk of infection. Detection of test-positive
animals triggers a so-called herd ‘breakdown’,
the lifting of officially TB-free status and the
imposition of movement restrictions onto the
affected herd. After the initial successes of the
attestation era (1935–1960) (Pritchard, 1988)
and before the introduction of annual testing for
the whole of Wales in 2008, the frequency of
routine testing in Great Britain was determined
by the historical incidence within a herd’s parish
(the Parish Testing Interval, or PTI). This trans-
lated to herds in high-risk areas being expected
to be tested annually, while lower-risk areas were
required to test at progressively longer intervals
of 2, 3 and 4 years.
Taken in itself, this temporal structuring of
testing in Great Britain might have been expected
to result in a greater number of reactor animals
being found in breakdowns that started in low-
risk areas. PTI 4 herds potentially had up to four
times longer for disease to transmit before it
could be disclosed by testing. However, the
distribution of reactor animals removed from
herds at their disclosing test was in fact remark-
ably consistent between testing areas – with the
most frequent result being a single reactor ani-
mal triggering a breakdown (Fig. 4.4). This
apparent incongruity is resolved when we note
that the second most frequent situation was for a
breakdown to start with no reactor animals –
the consequence of an animal with lesions con-
sistent with M. bovis infection being detected at
slaughter and triggering a whole herd test ear-
lier than the routine schedule would have
required.
This single statistic is a powerful illustra-
tion of the dynamic link between the frequency
of testing and potential for transmission within
managed populations. The consistency of the
within-herd distribution of reactors is a mea-
sure of the degree to which the second principle
arm of surveillance, meat inspection or slaugh-
terhouse surveillance, has worked to limit the
potential transmission of bTB within herds.
The proportion of breakdowns triggered at the
slaughterhouse was considerably higher in
low-risk areas (Conlan et al., 2012), effectively
limiting the duration of time that infection
could remain hidden within herds in low-risk
areas before herd-level measures could be
imposed. The power of slaughterhouse surveil-
lance comes from the sheer number of animals
that are inspected, such that even with a rela-
tively poor sensitivity at the individual animal
level, which can vary considerably between
slaughterhouses (Frankena et al., 2007; Olea-
Popelka et al., 2012; Shittu et al., 2013), there
can be a significant benefit at the population
level.