Transcript Getting real data on BTB: The wildlife / livestock / human interface

Evaluating the potential burden of
zoonotic mycobacteria in Africa:
Can modelling disease in wildlife
populations help?
Claire Geoghegan & Wayne Getz
Mammal Research Institute, Department of Zoology and
Entomology, University of Pretoria, South Africa
&
Department of Environmental Science, Policy & Management,
University of California – Berkeley, USA
Introduction
Introduction
• Drivers of disease
• Tuberculosis and HIV
• The role of Bovine tuberculosis (BTB) in animal health
• Research to date
• Future work
• ‘Throughout the region
people are walking a thin
tightrope between life and
death. The combination of
widespread hunger, chronic
poverty and the HIV/AIDS
pandemic is devastating
and may soon lead to a
catastrophe. Policy failures
and mismanagement have
only exacerbated an
already serious situation.’
James Morris , World Food
Programme’s Executive Director, July
2002,
Disease in animals and humans – why should we care?
Of all human pathogens, 62% are zoonotic and attributed to animals
Livestock pathogens
that can infect
wildlife
Human pathogens
that can infect
wildlife
54%
44%
Cunningham et al.
If a pathogen can infect wildlife, >
2x likely to cause an emerging human disease
Pathogens in species
E. J Woolhouse et al, 2005
Number of zoonotic pathogen species associated
with different types of nonhuman host
Important to understand
the temporal and spatial
dynamics of pathogens
in human and animal
reservoirs and
populations
Emerging infections
• novel paths to infect naïve hosts
• drastic effect on host health and mortality
• infect multiple species, promoting residence of pathogen in the system
• affect population levels and fecundity rates
• impact on conservation management
• economic and social consequences (direct and indirect)
What drives disease?
M. E. J Woolhouse et al, 2005
It is imperative to understand the fundamental dynamics of
infectious diseases in order to mitigate the impacts on public
health, wildlife and livestock economies
Tuberculosis
1/3 of people are infected and
have latent or active
tuberculosis
Over 90% of people in Africa
have been exposed to the TB
bacilli
75%
86%
HIV
HIV and TB
TB is an ancient
contagious disease,
discovered in 5000 B.C
L. Blanc et al, 2002
Global distribution
8 million new cases / year
~3 million deaths / year
80% of global
case load in
developing
countries
The Real
World
MRC report, 2000 / Hosegood et al
Provinces
Eastern Cape
TB incidence per
100 000 people
504
Estimated TB Proportion of TB
cases
cases
34 371
20.4%
Free State
282
8 272
32.1%
Gauteng
375
26 378
25.2%
KwaZulu/Natal
381
34 178
45.0%
Mpumalanga
286
8 716
39.5%
Northern Cape
340
2 675
13.6%
Northern Province
260
13 927
16.7%
North West
271
9 557
25.9%
Western Cape
South Africa
559
362
20 615
158 689
12.0%
27.0%
Bovine Tuberuclosis
(BTB)
Reported BTB Disease Status in Africa
W. Y Ayele et al, 2004
Bovine TB – a hidden threat
Global distribution
Listed as a category B disease by the OIE
Chronic disease that has an
effect on animal populations and
productivity
Annual worldwide losses
~$3 billion (trade)
Wide host range, including;
ruminants, predators,
scavengers, small mammals
Difficult to eradicate due to the
large disease reservoir apparent
F Biet et al, 2005
in wildlife
Clinical Signs and Symptoms
Infected cattle may present with progressive emaciation, capricious appetite
and a fluctuating fever.
However, many infected animals do not show any clinical abnormalities.
Tuberculin Skin Test
Test uses comparative reaction to
M.bovis and M. avium
•
•
Sensitivity ranges from 68 – 95%
Specificty ranges from 96 – 99%
Results are affected by:
•
•
•
•
•
•
•
potency and dose of tuberculin
the interval of time post-infection
desensitisation
deliberate interference
post-partum immunosuppression
observer variation
exposure to M. avium,
M. paratuberculosis and
environmental mycobacteria and by
skin tuberculosis
Routes of Transmission
1 Oral; 2 Aerosol; 3 Passive; 4 Derivative Product; 5 Vertical; 6 Horizontal; 7 Predation
Why is zoonotic TB so serious?
•
Causes extra-pulmonary
manifestations (9.4% of global TB)
•
Slow to develop and infects
many organs, which makes
treatment difficult
•
Multi Drug Resistant to the top 10
frontline drugs. This increases the
duration and cost ( x 10) of
treatment
Why should we be concerned?
Thoen and Steele (1995)
•
In Africa, 80% of the population is
rural and depend solely on livestock
for food and wealth (AU 2002)
•
85% cattle, 82% people live where
BTB is only partially controlled
•
90% of the total milk produced in
Africa is consumed raw or soured
The story so far….
The Great Limpopo Transfrontier Park
Links South Africa, Mozambique and
Zimbabwe
Health challenges in the park
TSETSE FLIES
FMD STRAINS
TB
BRUCELLOSIS
FMD
RABIES
TSETSE FLY
TB
BRUCELLOSIS
FMD STRAINS
CANINE DIST.
MAJOR LOCAL COMMUNITIES WITH DOMESTIC ANIMALS IN AND AROUND PARK
2020
2006
Why was the prediction so wrong?
Bovine tuberculosis is an exotic disease introduced from Europe
No co-evolution of host and pathogen
BTB was first noted in the 1990’s but probably entered the park in
the South East in the 1960’s
Incorrect temporal scale used for prediction
Thought to only infect buffalo
Found in lions, kudu, warthog, baboons, small antelope
Not the top priority
Anthrax, rabies and FMD more threatening!
Study design
Collared 100+ buffalo in
Kruger National Park
Followed herds to get
visual data on individuals
Branded ~500 buffalo
(roughly 2% of population)
Mass captures to test for
BTB
Marked additional buffalo
with ID collars
Removed infected buffalo
for pathology analysis
How the network of connections between individuals and the
interactions of group size, movement and recovery affect the
probability of BTB infection in structured populations.
Why was this approach unique?
Traditional animal disease models
assume random mixing of individuals,
not the individual connections
Spatial disease models
assume limited dispersal
between fixed groups
BUT: individuals risk of infection depends on the global state of the population
Network perspective: individual risk of infection depends on
the number and frequency of connections with infected individuals
• Population structure
• Landscape topology
• Total number of infected individuals
• Speed of the disease spread
Important in determining the probability of disease infection and invasion
Monthly radio-tracking data used
to create social networks
Balls represent individual buffalo
and lines show all non-zero
association values. Individuals
are distributed vertically
according to herd membership
These were used to simulate
disease dynamics along with
other factors including scale and
behaviour (females move!)
Cluster analysis indicated that
buffalo were less tightly clustered
in 2003 compared to 2002
Thus, increased host mixing
during this time (dry year) would
help facilitate disease invasion
spread
Climate may play a role in herd
movements and in BTB spread
Cross et al. 2004
Five critical issues:
1. What defines a contact for
airborne diseases?
2. What are the appropriate time
and spatial scales to sample an
animal network?
3. How do you confidently scale up
a sample to represent an entire
population?
4. How to allow for birth and deaths
and changes of association
patterns while maintaining the
overall properties of the network?
5. Is there a difference in behaviour
between susceptible and infected
individuals?
Is variance in connection strengths
and frequency of contact in individuals
important?
How does the duration of
infectiousness affect the degree of
disease experienced by the
population structure?
Why are some hosts affected
more than others?
How does incorporation of non-random
association data affect predictions about the
speed and intensity of a disease outbreak?
How do we get more empirical data and projects to run that require that data?
Models are constructions of
knowledge and caricatures of reality
Beissinger and Westphal,1998
Complex web of socio-economic
factors pertinent to controlling
disease for feasible, affordable and
effective public health policies to be
devised and implemented
Host-pathogen interactions in
ecological and socio-economic
settings are complex, non-linear
systems which required detailed
maths and statistical analysis
Need experience of biological
systems and technical knowledge
Need improved health care systems
and information systems about
health in order to generate reliable
statistics that can be used to
monitor progress
The next steps…
Locate and Quantify Infection
Practical Risk Factors
Social, Cultural, Economic
Factors of Disease Dynamics
Model and Map for Predictions
The Way Forward…..
Policy
Integrated Science
‘One Medicine’
Stakeholder Involvement
Capacity Building / Retention of Ideas
“Knowing is not enough, we must apply.
Willing is not enough, we must do."
Goethe
Acknowledgements
The project is thankful for the support of the
DIMACS / SACEMA and AIMS
Mammal Research Institute, and the Department of Zoology and Entomology at the University of
Pretoria, South Africa
Division of Ecosystem Sciences, Department of Environmental Science, Policy and
Management at the University of California – Berkeley, USA.