Bovine tuberculosis

48 A.J.K. Conlan and J.L.N. Wood

transmission was not observed on pasture con-
taminated by artificially inoculated animals
(Maddock, 1934). However, animals in this sec-
ond study were only exposed to pasture for
3 weeks after the infected animals were removed
to eliminate the risk of direct transmission. This
is a very short period of exposure given the low
rates of direct transmission that have been
observed between susceptible and test-positive
animals held in direct contact over far longer
exposure times of 12 months (Khatri et al.,
2012). Environmentally mediated exposure
cannot therefore be ruled out completely, par-
ticularly for extensively managed farms in devel-
oping countries (Ameni et al., 2007), but also as
a mechanism for within-farm persistence of
infection. The lower risk of recurrence of bovine
TB incidence in Scotland and low-risk areas of
England compared to high-risk areas in England
is most often attributed to a reduced prevalence
of disease within wildlife ( Karolemeas et al.,
2011). However, the reduced viability of envi-
ronmental M. bovis in Scotland due to climactic
factors could also be a contributory factor. Cli-
mate has been reported as a risk factor for TB
incidence (Wint et al., 2002); however, several
confounding factors could explain this associa-
tion including the density of cattle, badgers and
liver fluke which is known to interfere with the
action of the diagnostic skin test (Flynn et al.,
2007).

4.2 Patterns of Transmission Within

Unmanaged Herds

Regardless of the route of exposure, what is
clear is that bovine tuberculosis is a relatively
poorly transmissible pathogen that nonetheless
has a considerable potential for transmission
within herds due to the chronic and progressive
nature of infection. The potential for transmis-
sion of an infectious disease can be character-
ized by the basic reproduction number (R 0 ),
defined as the expected number of new infec-
tions when a single infectious individual is intro-
duced into a fully susceptible population.
Heuristically, the basic reproduction ratio can be
thought of as depending on two factors: the
duration of time they are infectious and the
probability of transmission per unit time (β′),

itself a function of the rate of contact between

susceptible and infected animals and the relative

infectiousness and susceptibility of different

hosts. For a chronic infection such as bovine

tuberculosis, the duration of infection is, to a

first approximation, the life expectancy of the

animal. Thus:

R 0 = βL

In an unmanaged population R 0 will be closely

related to the proportion of the population free

from disease (s) when the population is infected

endemically (Keeling and Rohani, 2008):

R 0 ≈ 1/s

Setting aside the likely presence of hidden infec-

tion, the fraction of endemically infected herds

unreactive to tuberculin can therefore provide a

‘first-estimate’ of the basic reproductive ratio of

bovine tuberculosis. Early prevalence studies in

Europe, collated by Francis (1947), reported

herd level prevalence from Denmark (1896) and

Great Britain (1945) of between 4% and 60%

corresponding to a range for the within herd R 0

of between 1.04 and 1.67.

Variations between regions, animal hus-

bandry, breed and format of the tuberculin test is

responsible for some of this variability, but two

systematic aspects were highlighted by Francis

in his seminal book that have important implica-

tions for how epidemiological modellers would

come to describe cattle-to-cattle transmission.

4.2.1 Density-dependent transmission

Francis noted that the proportion of cattle test-

ing positive to tuberculin increased with herd

size (Fig. 4.2a): ‘For every addition of ten cattle

to a herd the incidence of tuberculosis increases

by about 4 per cent’.

This observation, together with a more gen-

eral association of the risk of tuberculosis infec-

tion or breakdowns with herd size, is perhaps the

most consistently reported aspect of bovine

tuberculosis epidemiology in cattle (Skuce et al.,

2012). The implication of this observation is

that the potential for transmission of bovine

tuberculosis, as measured by R 0 , is increasing

along with the size of the herd (Fig. 4.2b).

Such a so-called density-dependent, but

really a herd size-dependent, relationship for