Emergence and resurgence of zoonotic and wildlife diseases

1
Emergence and resurgence of zoonotic and wildlife
diseases

• Approaches for research and control

Gideon Wasserberg (Ph.D., MPH) Walter-Reed
Institute (NRC fellow)
2
(No Transcript)
3
(No Transcript)
4
Emerging Diseases definition An emerging
disease is one that has appeared in a population
for the first time, or that may have existed
previously but is rapidly increasing in incidence
or geographic range (WHO 2008).
5
Zoonosis An infectious disease naturally cycling
within an animal system (reservoir host) that
could be transmitted to humans.
6
Environment
Agent
Host
Disease niche
Vector
7
(No Transcript)
8
Disease ecology
9
Todays talk
Approaches for the study of disease emergence and
resurgence

• Empirical and modeling research of cutaneous
  leishmaniasis in southern Israel.
• Modeling study of Chronic Wasting Disease in
  White-tailed deer in south-central Wisconsin.

10
General approach

• Observational studies pattern detection.
• Experimentation hypothesis testing.
• Modeling data integration, prediction.

11
Talk outline

• Cutaneous Leishmaniasis
• The effect of anthropogenic disturbance
• Epidemiological implications
• Modeling and implications for control
• On-going studies
• Chronic Wasting Disease (CWD)
• Overview
• Model
• Implications for control

12
Leishmaniases
Visceral (Kala-azar)
Cutaneous
Mucocutaneous
Diffuse cutaneous
13
World distribution of Leishmaniasis
14
Cutaneous Leishmaniasis in Israel
Ph. papatasi
L. major
P. obesus
L. major
15
Study question
Does anthropogenic disturbance enhance the
occurrence of Cutaneous Leishmaniasis in arid
environments, and if so how?
Ph. papatasi
P. obesus
16
Anthropogenic disturbance definition Any
human-induced modification of the original state
of the natural landscape.
17
Anthropogenic disturbance definition Any human
induced modification of the original state of the
natural landscape.
Alt. (m)
Precip (mm)
Land-use
Site
NIZ
286
87
Military, Ruins
SB
500
92
Old field
HAZ
-105
42
Settlement Edge
EY
-68
40
Ag., Edge
NHK
-352
30
Ag., Edge
18
Study approach

• Environmental aspect disease niche
• Temporal aspects
• Demographic aspects
• Spatial aspects
• Epidemiological aspects

19
Methods
20
The survey
Neot-Hakikar
21
Results
22
Pathogen identification
23
Reservoir host identification
24
Sand-fly vector identification 99.7 Phlebotomus
papatasi
Female
Male
25
Does anthropogenic disturbance affect infection
prevalence in the reservoir host?
26

• Log-linear model
• Infection prevalence is higher in the disturbed
  habitat (G 31.9, Plt.0001)
• Infection prevalence is higher in the less arid
  sites (G 42.0, Plt.0001)

Disturbed
.50
Undisturbed
.40
Infection Prevalence
.30
.20
.10
NIZ
SB
HAZ
EY
NHK
Site
27
Does anthropogenic disturbance affect vector
abundance?
28
1. Sand-flies are more abundant in disturbed
habitats (MW-U 625.5, c27.62, P0.006). 2.
Sand-flies are more abundant in the less arid
sites (KW20.16, P0.0005)
Site
29

• What is the mechanism?
• Alternative hypotheses
• Water addition
• Organic matter addition

30
3
Habitat effect P 0.009
2.5
A
Soil moisture
v
2
A
B

1.5
Soil moisture ()
B
C

1
C
0.5
C
0
NIZ
SB
HAZ
EY
NHK
Habitat effect P 0.18
0.5
A
Organic matter

0.4
B

A
B

0.3
Soil organic matter ()
0.2
C
0.1
0
NIZ
SB
HAZ
NHK
31
Does soil moisture affect sand-fly abundance?
32
Yes! Sandfly abundance increases with soil
moisture



Sand-fly abundance


Soil moisture
33
Does sand-fly abundance affect prevalence in the
sand-rat host?
34
YES! Prevalence increases with sandfly abundance
Dependent variable infection prevalence in P.
obesus (arcsine transformed)
Coefficient
SE
t
-test
-value
P
Intercept
0.18
0.10
1.72
0.10
Sand-fly abundance
0.46
0.11
4.27
gt0.001
P. obesus abundance
0.01
0.01
0.36
0.72
2
Note
Model
R
0.53, n 19.
35
How does disturbance affect the hosts food
source?
36
Plants are more lush in the disturbed habitats




Plant characteristic
Disturbed
Undisturbed
P-value
Percent cover

15.4

13.7

n.s.

Bush volume m3

16.8

12.3

n.s.

Bush transparency

0.65

0.61

n.s.

Lushness

1.82

1.49

0.009

Proportion cover

0.62

0.51

0.049


37
Which of these (lushness or chenopod cover)
affect host abundance?
38
Host abundance is positively correlated to plant
lushness
Dependent variable P. obesus abundance
(log-transformed)

Coefficient

SE

t
-
test

P
-
value

Intercept

0.56

0.88

0.63

0.53

Lushness

0.91

0.42

2.17

0.03

Prop. Chn.

0.61

0.86

0.70

0.49


39
Anthropogenic disturbance
Undisturbed
Disturbed
habitat
habitat
Dry
Moist
soil
soil
Sand-flies
Sand-flies
t 1
t
Sand-rats
Sand-rats
Susceptible
Infectious
Sand-flies
Sand-flies
Infectious
Susceptible
Host
Host
t
t 1
Chenopod plant
Chenopod plant
Dry
Lush
Human
Human
Susceptible
Infected
40
Spatial and epidemiologic patterns
41
Most endemic area Nizzana
Nizzana
soil moisture
42
P. obesus infection prevalence patterns
0
62
43
Sand-fly activity patterns
0
Sand-fly activity patterns
62
44
CL prevalence patterns among infantry soldiers
45

• Recruitment and training facts
• Three recruitment schedules August, November,
  March.
• Soldiers spend two month of basic-training in
  the base-camp at Nizzana (loess) and then move to
  Shunra sand-dunes for advanced training.

Hypothesis Soldiers get infected at their base
camp in Nizzana, but the lesions appear 3-4 month
later when soldiers are deployed at Shunra
sand-dunes
46
Alternative predictions Ranking of infection
prevalence by recruitment schedule 1. If most
exposure occurs at the base-camp August gt March
gt November 2. If most exposure occurs at the
camps in the sandy areas March gt August gt
November
Sand-fly activity
April
May
July
Sept.
Oct.
April
May
July
Sept.
Oct.
1
0.8
0.6
Sand-rat infection prevalence
0.4
0.2
0
Feb
Jun
Aug
Nov
Mar
47
The distribution of CL prevalence among infantry
soldiers in Nizzana region, by draft schedule
(1996 2000)
48
Supports base-camp exposure hypothesis
49

• Recommendations for control
• Focus control efforts to the loess area.
• Shift soldiers training area into the sandy
  habitat during the high risk period (May
  September).

50
Simulation modelCan we control CL by culling
sand-rats?
51
System initialization
Burrow
Juveniles
- Determine juvenile number
- Juvenile dispersal
Sand rat
Adults
- Reproduction (seasonal)
- Aging
- Mortality
t1
- Dispersal
If sand fly
activity period
(April - October)
- Disease transmission
Sand fly swarm
- Swarm dispersal
- Swarm aging
52
The sand-rat and sand-fly world

• Grid based
• Grid cell 2020m
• Time step week
• Grid size 100100 cells
• Object-based
• Sandrat burrows
• Sandrats
• Sandfly swarm
• Platform
• Borland C builder

2 KM
53
Summary of other findings used as model inputs

• Sand-flies tightly coupled to active host
  burrows
• Sand-flies effective dispersers effective
  transmission scale at least 500 m
• Spatial dynamics sand-rats determine the
  distribution whereas sand-flies determine the
  dynamics.
• Transmission mode density-dependent
• Annual sand-rat survival rare 17
• Differential disease-induced mortality F gt M

54
1
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
0
20
40
60
80
100
120
140
160
180
200
Infection prevalence in adult sand-rats
Week
1
0,9
0,8
0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
0
20
40
60
80
100
120
140
160
180
200
Week
55
Main results

• Infection dynamics is sensitive to vector
  seasonal dynamics.
• Infection dynamics is spatially in-sensitive.
  Cause strong vector-host coupling.
• Control measures host culling. Requires culling
  at least 85 of the host population to eradicate
  the disease

56
Conclusion

• Controlling the disease by host culling is not
  practical and could have severe ecological
  consequences on the natural-system.
• More practical - behavioral intervention
  including application of repellent, modification
  of training program to less risky habitats,
  increased awareness to times and places of
  increased risk.

57
Current research

• Soil moisture enhancement experiment the effect
  of soil moisture enhancement on sand-fly
  abundance, burrow residence time, and host
  infection probability (B. Kotler, R. Berger, BGU,
  Israel).
• Remote-sensing model Using Remote Sensing to
  predict Cutaneous Leishmaniasis infection risk in
  space and time (G. Glass, J. Clennon, JHSPH)

58
Pause for reflection
59
Problem-solving paradigm in applied ecology

• Define the problem
• Measure its magnitude
• Understand key determinants

Risk assessment
4. Develop intervention 5. Set policy/priorities
6. Implement and evaluate
Risk management
60
Problem-solving paradigm in applied ecology

• Define the problem
• Measure its magnitude
• Understand key determinants

Risk assessment
Model
4. Develop intervention 5. Set policy/priorities
6. Implement and evaluate
Risk management
61
Chronic Wasting Disease

• Disease agent Prion protein, identified in the
  1960s in CO and WY.
• The only TSE acting as infectious disease in
  free-ranging animals
• Origin unknown
• Occurs in North-America in farmed and
  free-ranging Deer, Elk, Moose.
• Highly resistant to environmental degradation
• Transmission poorly understood direct and/or
  environmental
• Potential serious impacts on deer-herd,
  recreational hunting, and concerns regarding
  potential transmission to cattle or humans
• Identified in Wisconsin in 2002

62
Concerns potential implications

• Cultural deer hunting as a cultural value
• Economical 1 billion dollar per year industry
• Conservational
• Public health
• Dairy industry

63
Spatial analysis and WI control policy

• Study Area high prevalence zone 120 mi2

• Goal Disease eradication
• Approach
• Extreme local reduction of deer density (DEZ).
• Creation of buffer area of increased hunting
  around the DEZ (HRZ).
• Enhanced surveillance

64
UW-Madison, Wildlife ecology group (Dr. M.
Samuel) - Research approach

• Spatial and temporal patterns of spread.
• Transmission mechanisms
• Population genetics
• Demographic patterns
• Human dimension (hunter behavior)
• Diagnostics
• Modeling (dynamic and statistical) my main task

65

• Modeling goals
• Understand the system
• Map-out and evaluate control strategy options

66
Transmission modes
Density-dependent
Contact rate
Frequency-dependent
Host density
DD Incidence rate function of host density FD
Incidence rate function of disease frequency
(i.e., prevalence)
67
Age and sex prevalence profile (2002-3)
68
Implications of sex-specific transmission and
sex-specific culling for disease control
69
Sex-specific contact models
Differential susceptibility
Male-based transmission
70
Competing models

• FD1 FD transmission, differential
  susceptibility
• FD2 FD transmission, male-based contact
• DD1 DD transmission, differential
  susceptibility
• DD2 DD transmission, male-based contact
• MFD1 FD transmission, no differential
  susceptibility
• MFD2 FD transmission, no male-based contact
• MDD1 DD transmission, no differential
  susceptibility
• MDD2 DD transmission, no male-based contact

71

• Study questions
• Is CWD transmission density- or frequency
  dependent?
• The importance of sex-specific transmission and
  culling for the dynamics and control of the
  disease.
• How long has CWD been present in WI?
• Can CWD coexist with white-tailed deer
  populations?
• Can we manage CWD using generalized deer culling,
  and if so how long would it take?
• What are the consequences of the mode of
  transmission on CWD and deer dynamics?

72
(No Transcript)
73

• Deterministic, multi-state population matrix
  model (e.g., Caswell 2001)
• Age
• Stage
• Sex
• Seasonality

74
Multi-state matrix model

• Non spatial
• Semi-annual time steps
• 4 states age, sex, infection stage, season

S
I
O
C
S
I
O
C
F
0
0
M
75
f
(i-a)s
(i-f)s
(i-g)s
(i-p)s
ps
gs
fs
O
S
I
C
i-s
i-s
i-s
i-s
a
Dead
Infection stages
Demographic and infection sub-matrices
p transmission sub-matrix
g incubation sub-matrix
f disease progression sub-matrix
s survival sub-matrix
a disease induced mortality sub-matrix
i identity sub-matrix
f fecundity sub-matrix
76

• Summer matrix Ms (Ls D) Hs
• With reproduction, no culling

77

• Summer matrix Ms (Ls D) Hs
• With reproduction, no culling
• Winter matrix Mw (Lw D ) Hw
• No reproduction, with culling

78

• Summer matrix Ms (Ls D) Hs
• With reproduction, no culling
• Winter matrix Mw (Lw D ) Hw
• No reproduction, with culling

Yearly matrix projection N(t) MwMsN(t-1)
79
Results
80
The effect of incidence rate on the finite
population growth rate (L) and equilibrium
prevalence
1.5
1


1.4
0.9
1.3
0.8
1.2
0.7
1.1
0.6
L
Equilibrium prevalence
1
0.5
0.9
0.4
0.8
0.3
0.7
0.2
0.6
0.1
0.5
0


0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
Incidence rate
81
The effect of incidence rate (p) on the finite
population growth rate (L) and equilibrium
prevalence
1.5
1


1.4
0.9
1.3
0.8
1.2
0.7
1.1
0.6
L
Equilibrium prevalence
1
0.5
0.9
0.4
0.8
0.3
0.7
0.2
0.6
0.1
0.5
0


0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
Incidence rate
82
The effect of incidence rate (p) on the finite
population growth rate (L) and equilibrium
prevalence
1.5
1


1.4
0.9
1.3
0.8
1.2
0.7
1.1
0.6
L
Equilibrium prevalence
1
0.5
0.9
0.4
0.8
0.3
0.7
0.2
0.6
0.1
0.5
0


0
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
Incidence rate
83
Likelihood profile analysis

• Iterating through a time since disease
  introduction (TDI), bm, bf parameter space.
• Each TDI corresponds to a given initial deer
  population size for that year (based on fitting
  our demographic model to historical deer harvest
  data).
• Fitted to age- and sex-specific prevalence
  distribution (2002-2006).

2002-3
Prevalence
84
Likelihood profile analysis Comparison between
DD and FD transmission
Estimation of the transmission coefficient, time
since disease introduction (TDI), annual
infection rate, and relative fit
85
Disease dynamics the effect of transmission
model and harvest pressure
86
DD transmission
FD transmission
Sustainable culling M48, F26
Sustainable culling M48, F26
0.1
0.5
Sustainable culling
0.08
0.4
0.3
0.06
0.04
0.2
0.02
0.1
0
0
Prevalence
Control culling M28, F39
Control culling M28, F39
0.1
0.5
0.08
0.4
0.06
0.3
0.04
0.2
0.02
0.1
0
0
0
5
10
15
20
0
10
20
30
Time (years)
87
Host dynamics the effect of transmission model
and harvest pressure
88
DD transmission
FD transmission
Sustainable culling
Sustainable culling
Deer density
Control culling
Control culling
30
30
20
20
10
10
0
0
0
5
10
15
20
0
10
20
30
Time (years)
89
The management implications of the sex-specific
transmission eradication versus containment
90
M25, F35
MF30
M35, F25
91

• Study questions
• CWD appears to be more consistent with FD
  transmission but this is far from being
  conclusive.
• Sex-specific transmission consistent with the
  data
• Disease introduction time FD 59 (38 100lt)
  years, DD 30 (24 34) years.
• Deer population can sustain a prevalence of up to
  62 if not hunted. However, if uncontrolled CWD
  can render hunting un-sustainable.

92
Study questions (cont.) 5. Implication of
transmission mode on control If DD CWD could
be eradicated, it would take ca. 15 yrs, results
in reduced deer population size If FD CWD could
be controlled only via host extermination
93
Study questions (cont.) 6. Recommendations
regarding sex-specific transmission depends on
the mode of transmission and management goal
DD female-biased culling. FD if
goaleradication, female-biased culling. If
goalcontainment, male-biased culling.
94
On-going and future studies

• Development of a spatial model (A. McClung, M.
  Samuel, UW-Madison)
• Study the effect of environmental reservoir (M.
  Samuel, UW-Madison)
• Experimental assessment of transmission mode
  effect of host reduction on disease transmission
  (M. Samuel, WDNR)
• Functional relations of deer density and hunter
  harvest rate (M. Samuel, T. Van-Deelan,
  UW-Wisconsin)

95
CONCLUSIONS
96
Problem-solving paradigm in applied ecology

• Define the problem
• Measure its magnitude
• Understand key determinants

Risk assessment
Model
4. Develop intervention 5. Set policy/priorities
6. Implement and evaluate
Risk management
97
Learn to manage by managing to learn
Control experiments
Model development
Analyze results
Re-evaluate
Empirical research
Problem definition Management goal
98
Acknowledgements

• Supervisors
• Z. Abramsky, B.P. Kotler, BGU, Israel, A.
  Warburg, Hebrew U.
• H. Thulke, F. Hansen, UFZ, Leipzig.
• M.D. Samuel, USGS-Wisconsin Cooperative Wildlife
  Research Unit. UW-Madison.
• Colleagues E. Osnas, A. McClung, R. Rolley, D.
  Grear, J. Blanchong, Greg Glass, Julie Clennon
• Funding sources
• Israel Ministry of Science, EU-Marie Curie Fund,
  Minerva fund.
• USGS-NWHC
• Data sources
• Wisconsin Department of Natural Resources

99
..Walked out this morning, dont believe what I
saw Hundred billion bottles washed up on the
shore. Seems Im not alone at being alone Hundred
billion castaways, looking for a home Ill send
an s.o.s. to the world Gordon Sumner (sting),
message in a bottle.
100
(No Transcript)
101
Development of alternative control plans
Model development
Analyze results
Re-evaluate
Empirical research
Problem definition Management goal
102
Development of alternative control plans
Model development
Analyze results
Re-evaluate
Empirical research
Problem definition Management goal
103
Control experiments
Model development
Analyze results
Re-evaluate
Empirical research
Problem definition Management goal
104
Control experiments
Model development
Analyze results
Re-evaluate
Empirical research
Problem definition Management goal
105
No. of sand flies/trapping station/night
No. of sand flies/trapping station/night
106
Can we predict Cutaneous Leishmaniasis Risk in
space and time using RS
Collaboration with Greg Glass, JHSPH
107
HYPOTHESIS

• The following environmental factors are important
  predictors of CL occurrence
• soil type
• plant-community type
• plant lushness
• soil moisture
• land-use
• climatic factors

108
METHODS

• soil type Digital soil map from the Israeli
  Center for Mapping
• plant-community type Plant community maps of
  the area are available from the literature (Danin
  2004).
• plant lushness NDVI, MSI
• soil moisture NDVI, MSI
• land-use range of RS methods
• climatic factors LANDSAT-7 and other weather
  satellites

109
NDVI Normalized Difference Vegetation Index
MSI moisture stress index
110

• Proximate goal creation of local-scale
  spatio-temporal CL risk map.
• Ultimate goal real-time, interactive, regional
  CL transmission-risk prediction system

111
Study area Nizzana (true colors)
112

• Study protocol
• Confirmation
• Projection
• Validation

113

• Further developments
• Sandfly population-dynamics model
• Reservoir host population dynamics model

114
Habitat effects

• Cutaneous Leishmaniasis in Israel.
• Chronic Wasting Disease in white-tailed deer,
  Wisconsin.

115
The epidemiological triad
Environment e.g., encroachment, antibiotics-use
Host
Agent Novel pathogens, drug resistance
116
The epidemiological triad
Environment
Host
Agent
117
..Walked out this morning, dont believe what I
saw Hundred billion bottles washed up on the
shore. Seems Im not alone at being alone Hundred
billion castaways, looking for a home Ill send
an s.o.s. to the world Gordon Sumner (sting),
message in a bottle)
118
The epidemiological triad
Environment
Host
Pathogen Pollutant Genetic
119
The epidemiological triad
Environment
Agent
Age, sex, genetics physiology behavior
120
The epidemiological triad
Climate, land-use Urban
environment Access to healthcare
Host
Agent
121
The epidemiological triad
Environment
Host
Agent
122
The epidemiological triad
Environment
Host
Agent
123
The epidemiological triad
Environment
Host
Agent
124
The epidemiological triad
Environment
Host
Agent
125
Aldo Leopold, 1949. The Land Ethics A land
ethic changes the role of Homo sapiens from
conqueror of the land community to plain member
and citizen of it. It implies for his fellow
members and also respect for the community as
such.
Van Rensselaer Potter, 1988. Global Bioethics
Bioethics as a profession needs to shift its
gaze to relationships in community, intuition,
and public and environmental health So my
colleagues, I care what you think and more what
you feel but, most importantly, what you do to
ensure a world for generations to come and for
all life on the planet.
126
Figure 1. The host-parasite ecological continuum
(here parasites include viruses and parasitic
prokaryotes). Most emerging diseases exist within
a host and parasite continuum between wildlife,
domestic animal, and human populations. Few
diseases affect exclusively any one group, and
the complex relations between host populations
set the scene for disease emergence. Examples of
EIDs that overlap these categories are canine
distemper (domestic animals to wildlife), Lyme
disease (wildlife to humans), cat scratch fever
(domestic animals to humans) and rabies (all
three categories). Arrows denote some of the key
factors driving disease emergence.
127
The epidemiological triad
Environment e.g., encroachment, antibiotics-use
Host
Agent
128
The epidemiological triad
Environment
Host
Agent
129
The epidemiological triad
Environment
Biodiversity and Emerging Diseases JEAN-CHARLES
MAILLARDaaCirad-Emvt/PRISE, Hanoi, Vietnam AND
JEAN-PAUL GONZALEZbbIRD/UR178, RCEVD, IST,
Mahidol University, Nakhonpathom,
ThailandaCirad-Emvt/PRISE, Hanoi, Vietnam
bIRD/UR178, RCEVD, IST, Mahidol University,
Nakhonpathom, Thailand Address for
correspondence Dr. Maillard Jean-Charles,
CIRAD-EMVT/PRISE c/o NIAH, Thuy Phuong, Tu Liem,
Hanoi, Vietnam. Voice 84-4-838-8068 fax
84-4-757-2177. e-mail maillard_at_fpt.vn Abstract
Abstract First we remind general considerations
concerning biodiversity on earth and particularly
the loss of genetic biodiversity that seems
irreversible whether its origin is directly or
indirectly linked to human activities. Urgent and
considerable efforts must be made from now on to
cataloge, understand, preserve, and enhance the
value of biodiversity while ensuring food safety
and human and animal health. Ambitious integrated
and multifield research programs must be
implemented in order to understand the causes and
anticipate the consequences of loss of
biodiversity. Such losses are a serious threat to
sustainable development and to the quality of
life of future generations. They have an
influence on the natural balance of global
biodiversity in particularly in reducing the
capability of species to adapt rapidly by genetic
mutations to survive in modified ecosystems.
Usually, the natural immune systems of mammals
(both human and animal), are highly polymorphic
and able to adapt rapidly to new situations. We
more specifically discuss the fact that if the
genetic diversity of the affected populations is
low the invading microorganisms, will suddenly
expand and create epidemic outbreaks with risks
of pandemic. So biodiversity appears to function
as an important barrier (buffer), especially
against disease-causing organisms, which can
function in different ways. Finally, we discuss
the importance of preserving biodiversity mainly
in the wildlife ecosystems as an integrated and
sustainable approach among others in order to
prevent and control the emergence or reemergence
of diseases in animals and humans (zoonosis).
Although plants are also part of
Host
Agent