Ecohealth 2014 gianni lo iacono presentation on integrative modelling

Background Methods Predictions Conclusion References Appendix 1 / 17

Dynamic Drivers of Disease in AfricaIntegrating our understandings of zoonoses, ecosystems and wellbeing

A mechanistic model at the interface betweenepidemiology, ecology and environmental drivers.

EcoHealth 2014, Montreal, 11-15 August

Gianni Lo IaconoDepartment Veterinary MedicineUniversity of Cambridge, UK

August 13, 2014


Dynamic Drivers of Disease in AfricaIntegrating our understandings of zoonoses, ecosystems and wellbeing

The complex dynamics of Rift Valley FeverEcoHealth 2014, Montreal, 11-15 August

Gianni Lo IaconoDepartment Veterinary MedicineUniversity of Cambridge, UK

August 13, 2014


Background

Distribution Map of Rift Vally Fever. Source CDC

RVF is a viral diseaseaffecting livestock andhumans common in Africa.

Transmitted by differentspecies of mosquitoes.Associated with rainfall

and/orwater reservoir.

Associated with cultural andsocio-economic factors

Epidemiology

Ecology

Environmental, Economicand Political Issue

Anthropology and SocialScience

http://www.cdc.gov/


Background

Aedes Aegypti. Source Wikipedia.com

RVF is a viral diseaseaffecting livestock andhumans common in Africa.Transmitted by differentspecies of mosquitoes.

Associated with rainfall

and/orwater reservoir.


Epidemiology

Ecology



http://Wikipedia.com


Background

F. G. DAVIES ET AL.

RVF EPIZOOTICS

CC

: 21 000 - 1951-1953

c 15 000 -

s 75 9000,,3000-

X O-3000

Q C -90001951 195

C1991-1963 1997-1969

I..J..~~L~LL-...ri umI -LLJ.LIm . - JEd - A - L I I~1.~ 1. 1 ~L L -

I

1955

ri1969 1971 19731957 1959 1961 1963 1965 1967

Fig. 1. The relationship between Rift Valley fever (RVF) epizootics and rainfall in Kenya for the years 1951-82. Thegraph depicts a composite statistic based upon number of rainy days and rainfall for each month at five sites in Kenyawhere RVF epizootics occur. The zero line represents the 33-year mean rainfall for each month. Values above the zeroline represent periods of positive surplus rainfall.

Widespread rainfall in the region studied is causedby the changing characteristics of the intertropicalconvergence zone, the zone of confluence of aircurrents from north and south in the Africancontinent that determines the extent and persistenceof precipitation and cloud cover. The compositerainfall statistic in Fig. 1 is therefore a function of theactivity of this zone. Continued cloud cover may bean important determinant for the survival of adultmosquitos, a key factor for the generation ofmosquito-propagated virus epizootics (19).Convection and local air currents may be importantvehicles for the transport of infected mosquitos orother vectors of RVF during epizootics (15, 16), andthis could produce local or distant extension from theoriginal foci of epizootics. It should be noted, forexample, that the RVF outbreak in Egypt in 1977 wascoincident with that in East Africa.

Rift Valley fever is probably maintained ininterepizootic periods by transovarial transmission ofthe virus in mosquitos of the genus Aedes that breedin temporary ground pools (dambos). Periods ofpersistent rainfall raise the ground-water table to alevel where the breeding sites of Aedes mosquitosbecome flooded (8, 9). This flooding requires heavyand prolonged rainfall. Flooding of the damboformations induces the hatching of Aedes eggs andsubsequent emergence of very large numbers of theadult mosquitos, which feed preferentially uponcattle (10). If these mosquitos are infected with RVF,virus amplification occurs in the vertebrate hosts,leading to further infection of other mosquito speciesthat are capable of transmitting the virus (11). Many

of these opportunist mosquito species colonizedambos in the grasslands (2, 9). The humidconditions and cloud cover present during theprolonged rainy periods allow a greater proportion ofthe adult Aedes population to survive through morefeeding-oviposition cycles than in the hot, dryconditions usually prevailing in these areas.

It is interesting to note that the circumstances thatallow the generation of epizootics of RVF in Kenyaoften prevail simultaneously throughout a large partof the African continent. Epizootics of RVF havetherefore been recorded simultaneously in Kenya, theUnited Republic of Tanzania, and Zambia and oftenin the southern African countries (2, 14, 20).

This explanation of the mechanism of the effect ofrainfall upon RVF therefore explains the patterns ofRVF virus activity encountered in different parts ofAfrica. An annual emergence of infected Aedesmosquitos may occur in grassland areas with highannual rainfall (i.e., Zambia) or in the tropical forestbelt that traverses much of the continent, and this iscorroborated by serological studies of human andruminant populations in riverine and forest-edgesettlements in such areas (2, 5, 6, 12, 13). Emergenceof some RVF-infected Aedes mosquitos may beexpected to occur at 2-4-year intervals, coincidingwith the excessive rainfall that floods dambo breedingsites in the bushed and wooded savanna (with moder-ately high rainfall) that is contiguous with the forestbelt. Such areas are found in many parts of Africa inthe high plateau area east and south of the tropicalforest. Epizootics of RVF generally occur at 5-15-year intervals in the lower-rainfall areas of East and

1977-1979

6i r19791975 1977 1981

WHO 85952

, _,.r *, . o_- _, , , _ ,942

i I i'L A1r .11 .t9 v

Relationship between Rift Valley fever (RVF)

epizootics and rainfall in Kenya from Davies et al.

(1985)

RVF is a viral diseaseaffecting livestock andhumans common in Africa.Transmitted by differentspecies of mosquitoes.Associated with rainfall

and/orwater reservoir.Associated with cultural andsocio-economic factors

Epidemiology

Ecology




Background

Sand dam full of water. January 2013, Kenya.

Source thewaterproject.org

RVF is a viral diseaseaffecting livestock andhumans common in Africa.Transmitted by differentspecies of mosquitoes.Associated with rainfall and/orwater reservoir.


Epidemiology

Ecology



http://thewaterproject.org/community/projects/kenya/new-sand-dam-and-shallow-well-in-kenya-4025


Background

Livestock market in Kabul for Eid-al Adha festivities.

Source theguardian.com

RVF is a viral diseaseaffecting livestock andhumans common in Africa.Transmitted by differentspecies of mosquitoes.Associated with rainfall and/orwater reservoir.Associated with cultural andsocio-economic factors

Epidemiology

Ecology



http://theguardian.com


Background

Objective: To separate the complexity of RVF in its constituentparts to address questions like:

What make a RVF endemic site endemic?What happens if we change the ecosystem?What are the drivers and can we quantify these?What is the underlying mechanism?


Background


What make a RVF endemic site endemic?

What happens if we change the ecosystem?What are the drivers and can we quantify these?What is the underlying mechanism?


Background


What make a RVF endemic site endemic?What happens if we change the ecosystem?

What are the drivers and can we quantify these?What is the underlying mechanism?


Background


What make a RVF endemic site endemic?What happens if we change the ecosystem?What are the drivers and can we quantify these?

What is the underlying mechanism?


Background


What make a RVF endemic site endemic?What happens if we change the ecosystem?What are the drivers and can we quantify these?What is the underlying mechanism?


Methods: General Features of the Model

A deterministic model formulated in terms of a set of differentialequations:

track how the populations of infected and healthlivestock and infected and health mosquitoes change in time



A deterministic model formulated in terms of a set of differentialequations: track how the populations of infected and healthlivestock and infected and health mosquitoes change in time



One vertebrate host

Two mosquitoes species: Aedes and CulexAedes : subjected to vertical and (weak) horizontaltransmission Culex : strong horizontal transmission



One vertebrate hostTwo mosquitoes species: Aedes and Culex

Aedes : subjected to vertical and (weak) horizontaltransmission Culex : strong horizontal transmission



One vertebrate hostTwo mosquitoes species: Aedes and CulexAedes : subjected to vertical and (weak) horizontaltransmission Culex : strong horizontal transmission


Methods: Epidemiological Aspects

The host transit fromSusceptible, to Exposed,Infectious and Removedcategory (SEIR)

The mosquitoes transitfrom Susceptible toExposed and Infectiouscategory (SEI)


Methods: Epidemiological Aspects

The host transit fromSusceptible, to Exposed,Infectious and Removedcategory (SEIR)The mosquitoes transitfrom Susceptible toExposed and Infectiouscategory (SEI)


Methods: Ecological Aspects

Otero, Schweigmann, and Solari

Fig. 1 Populations and events of the stochastic model, where w(k) with k = 1,2, . . . ,14 are the transitionrates of Table 1 and Section 4.2.

5. Biological parameters

The rates of occurrence of the events described in Table 1 are specified for the case ofAedes aegypti. Consequently, the different parameters appearing in the stochastic processdescribed by Table 1 characterize Aedes aegypti biology.

5.1. Developmental and mortality rates

The five developmental rates that correspond to egg hatching, pupation, adult emergence,and gonotrophic cycles are evaluated using the results of the thermodynamic model devel-oped by Sharpe and DeMichele (1977) and were described in detail in the previous article(Otero et al., 2006). The different mortality rates as well as the emergence rate have beentaken from Focks et al. (1993), Christophers (1960) and were also described previously(Otero et al., 2006). The values of the developmental and mortality rates are available inAppendix A.

5.2. Breeding site dependent oviposition rate

Outside laboratory conditions, not every water container would be a breeding site. Theefficiency of the process of egg laying in the appropriate places (breeding sites) will beless than one. It just may happen that in a given homogeneous patch, there are no breedingsites and the eggs laid in the patch will not develop up to the adult stage.

Source: from Otero et al. (2008)

Stage-structured model forthe mosquitoes: eggs,larvae, pupae, naïvefemale, flyers, and femaleadults having laid eggs.


Methods: Environmental Aspects

The mosquitoes dynamics is driven by temperature theavailability of breeding sites (i.e. water ponds)

Effect of temperature on keyparametersAvailability of breeding sites(number and size) affects:

Number of laid eggsNumber of hatching eggs

Fluctuations on the availabilityof breeding sites (number andsize) affects the hatching rateof Aedes eggs




,I

I

; j

I

i:

,J

I " 7

Temperature-Dependent Development and Survival Rates ofCulex quinquefasciatus and Aedes aegypti (Diptera: Culicidae)

L. M. RUEDA, K. J. PATEL, R. C. AXTELL, AND R. E. STINNER

Department of Entomology, North Carolina State University,Raleigh, North Carolina 27695

J. Med. Entomol. 27(5):892-898(1990)ABSTRACT Development, growth, and survival of Culex quinquefasciatus Say and Aedesaegypti (L.) were determined at six constant temperatures (15, 20, 25, 27, 30, 34°C). TheSharpe & DeMichele four-parameter model with high-temperature inhibition described thetemperature-dependent median developmental rates of both mosquito species. In both species,body size generally decreased as temperature increased. Head capsule widths in all instarsin both species were significantly greater at 15 than at 30-34°C. Except for the thIrd instarof Ae. aegypti, the larval body lengths in both species were significantly greater at 15 thanat 34°C. All'instars and pupae of both species and the adults in Cx. quinquefasciatus weresignificantly heavier at 15 than at 27-34°C. In Cx. quinquefasciatus, survival from eclosionto adult emergence was highest in the range from 20 to 30°C(85-90%) and dropped drasticallyat 15 (38%) and 34°C (42%). In Ae. aegypti, survival to adult stage was high at 20 (92%)and 27°C (90%) and lowest at 15°C (3%).

KEY WORDS Insecta, temperature-dependent development, Culex quinquefasciatus, Aedesaegypti

INFORMATIONONTHE EFFECTSof temperature onthe rates of development and survival of the variousstages of mosquitoes are necessary in designingpopulation and control strategy models (Wagneret al. 1975, Moon 1976, Haile & Weidhaas 1977,Greever & Georghiou 1979). Data on the effects oftemperature on the development of Culex quin-quefasciatus Say (Shelton 1973, Madder et al. 1983,Rayah & Groun 1983, Kasule 1986, Service 1986)and Aedes aegypti (L.) (Bar-Zeev 1958, Kasule 1986,Southwood et al. 1972) have been reported. Un-fortunately, the reported data were not collectedin the appropriate form or published in sufficientdetail to provide the distributions of values re-quired for calculations in mathematical models oftemperature-dependent development. Because ofvariation among individuals, populations must besampled at several time intervals to document thedistribution of development times. Closer samplingintervals are required at high temperatures thanat low temperatures to obtain approximately equalnumbers of data points within the range of thedistribution.In the study reported here, the effect of constant

temperatures on the developmental rates, growth,and survival of the immature stages of Cx. quin-quefasciatus and Ae. aegypti was determined un-der laboratory conditions, and the temperature-dependent model of Sharpe & DeMichele (1977)was used to describe developmental rates. The dis-tribution of development was described using theapproach. of Stinner et al. (1975).

Materials and Methods

Mosquito eggs used in these studies were ob-tained from laboratory colonies whose stock par-ents originated from Raleigh, N.C., within the pre-vious 2 yr. Larvae hatching during the first hourafter the eggs were placed on water at 27°C wereused, and the developmental time was assumed tobe at the midpoint of this I-h period. One larvawas placed in each well (7 ml) of a covered 12-well tissue culture plate (Cat. 25815; Corning Com-pany, Corning, N.Y.) containing 3 mlliver solution(0.26 mg liver powder Iml deionized water). Thirtyplates were placed in each of five incubators (main-tained at 15, 20, 25, 30, and 34°C) and in a con-trolled-environment room (27°C). Every secondday, first and second instars were provided with 30J.LIof liver solution per well, and third and fourthinstars were given 50 J.Ll.Water was added as need-ed to maintain the volume. Based on our obser-vations of liver particles being present at all timesin the cells, this feeding regimen provided excessfood for the larvae at all temperatures. The incu-bators and the controlled room were illuminatedcontinuously with fluorescent light. In a prelimi-nary experiment, this procedure for rearing indi-vidual larvae in tissue culture plates under constantlight was compared with our routine rearing ingroups in pans under a 14:10 (L:D) photoperiod,and no differences in development times were ob-served.Mosquito larvae were examined at intervals of

0022-258~090/0892-0898$02.00/0 @ 1990 Entomological Society of America

~--"-~' '-'-- -" -'~---~-"'-T ~'---" . ,,,...

Effect of temperature on keyparameters

Availability of breeding sites(number and size) affects:







Number of laid eggs

Number of hatching eggs



Predictions:Mosquito population and mean size of breeding sites

Threshold in the mean size of breeding sitesSmall amount of water ! extinction of both speciesof mosquitoesLarge amount of water ! stable periodic oscillations



1 2 3 4 5year

Are

a of

the

bree

ding

site

s




0

10

20

30

1 2 3 4 5 6 7 8 9 10time

Eggs

Mos

quito

es

AedesCulex

0.0

2.5

5.0

7.5

10.0

1 2 3 4 5 6 7 8 9 10time

Adu

lt M

osqu

itoes

AedesCulex




0.0e+00

5.0e+07

1.0e+08

1.5e+08

2.0e+08

1 2 3 4 5 6 7 8 9 10time

Eggs

Mos

quito

es

AedesCulex

0e+00

2e+06

4e+06

6e+06

1 2 3 4 5 6 7 8 9 10time

Adu

lt M

osqu

itoes

AedesCulex



Predictions:Mosquito population and frequency of fluctuations of thebreeding sites

The amount of rainfall is not everythingIncrease in the frequency of fluctuations of thebreeding sites ! increase in the population ofAedes.Importance of the positive variation of rainfall insustaining endemicity of RVF



1 2 3 4 5year

Are

a of

the

bree

ding

site

s




0.0e+00

5.0e+07

1.0e+08

1.5e+08

2.0e+08

1 2 3 4 5 6 7 8 9 10time

Eggs

Mos

quito

es

AedesCulex

0e+00

2e+06

4e+06

6e+06

1 2 3 4 5 6 7 8 9 10time

Adu

lt M

osqu

itoes

AedesCulex



Predictions:Effect of rainfall on mosquito population and disease(in progress)

0.0e+00

5.0e+07

1.0e+08

1.5e+08

1 2 3 4 5time

Eggs

Mos

quito

es

AedesCulex

0e+00

1e+07

2e+07

3e+07

1 2 3 4 5time

Adu

lt M

osqu

itoes

AedesCulex

0

25000

50000

75000

100000

125000

1 2 3 4 5time

Pond

siz

e

Pond SizeRain


Conclusions and next steps

Unified eco-epidemiological model with inclusion of keybio-physical mechanisms

Importance of oscillations in the area of the breeding sitesCalibration with with fieldwork data and other (e.g. satellitedata)Further integration with outputs from ParticipatoryModellingInclusion of cultural and socio-economic effects



Unified eco-epidemiological model with inclusion of keybio-physical mechanismsImportance of oscillations in the area of the breeding sites

Calibration with with fieldwork data and other (e.g. satellitedata)Further integration with outputs from ParticipatoryModellingInclusion of cultural and socio-economic effects



Unified eco-epidemiological model with inclusion of keybio-physical mechanismsImportance of oscillations in the area of the breeding sitesCalibration with with fieldwork data and other (e.g. satellitedata)

Further integration with outputs from ParticipatoryModellingInclusion of cultural and socio-economic effects



Unified eco-epidemiological model with inclusion of keybio-physical mechanismsImportance of oscillations in the area of the breeding sitesCalibration with with fieldwork data and other (e.g. satellitedata)Further integration with outputs from ParticipatoryModelling

Inclusion of cultural and socio-economic effects



Unified eco-epidemiological model with inclusion of keybio-physical mechanismsImportance of oscillations in the area of the breeding sitesCalibration with with fieldwork data and other (e.g. satellitedata)Further integration with outputs from ParticipatoryModellingInclusion of cultural and socio-economic effects


A Language Issue

Biomathematics Conference Plain English

Transition of the sign of the realpart of Floquet exponents for thelinearized system of differentialequation..

What happens if you builda dam?


A Language Issue

Biomathematics Conference

Plain English




A Language Issue

Biomathematics Conference Plain English




Thank You

The Confusion of Tongues Gustave Dorè (1865)


Davies, F. G., Linthicum, K. J., & James, a. D. (1985). Rainfall and epizootic Rift Valley fever. Bulletin of the WorldHealth Organization, 63(5), 941–3.URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2536443&tool=

pmcentrez&rendertype=abstract

Otero, M., Schweigmann, N., & Solari, H. G. (2008). A stochastic spatial dynamical model for Aedes aegypti.Bulletin of mathematical biology , 70(5), 1297–325.

Ecohealth 2014 gianni lo iacono presentation on integrative modelling

Government & Nonprofit

Transcript of Ecohealth 2014 gianni lo iacono presentation on integrative modelling