Field Project 3:

1
Field Project 3 Surveillance and control of
Rift Valley Fever in the Greater Horn of Africa
and the Middle East
www.tomuphoto.com/gallery/ Landscapes/rift_valley
2
What is Rift Valley Fever?

• Febrile disease that affects sheep, cattle,
  goats, humans, primates, camels
• Vector
• Most human cases are mild and involve fever,
  myaglia, vomiting, diarrhea, and/or hepatitis
• Can become severe in small of human cases
• Eye disease 0.5-2
• Meningoencephalitis lt1
• Hemorrhagic fever lt1

Aedes (Neomelaniconion) and Aedes (Stegomyia),
Culex, Mansonia, Anopheles and Eretmapodites,
have all been shown to transmit the virus.
Photo news.bbc.co.uk/.../ newsid_934000/934032.st
m
3
RVF Epizootics

• Humans infected from
• Mosquitoes
• Contact with blood/body fluids/organs of infected
  animals
• First isolated in 1930 among sheep on a farm in
  the Rift Valley of Kenya
• Recent Outbreaks
• Egypt 1993, 1997
• Kenya, Somalia 1997-98
• First cases outside African Continent not until
  2000 in Saudi Arabia Yemen

4
Economic Impacts of RVF
Pastoralists in the Horn of Africa
Markets in the Arabian Peninsula
BOYCOTT
Photos, clockwise from top left www.uni-mainz.de,
www.asergeev.com, www.fao.org
Sheep infected with RVF
5
RVF Management Group Decision Problem

• Monitoring and forecasting system depends on
  cooperation between producers and consumers
• If benefits of monitoring/forecasting are not
  spread across each group, essential players may
  not want to participate
• Groups span across production, export,
  consumption, continents, religions,
  nationalities, and ethnicities.

6
How to overcome trade constraints?

• Development of a model that could
• Identify areas that are RVF enzootic with
  epizootic potential
• Identify periods of high risk of RVF
• Determine lead time in which high risk areas
  periods can be identified
• Establish linkages between model outputs
  decision-making options at multiple levels

7
Developing a Monitoring Forecasting System

• Identify areas that are RVF enzootic with
  epizootic potential
• Using historical data from Kenya, assemble a set
  of environmental layers to identify the areas
  where RVF is enzootic with epizootic potential
• Apply them to the entire Horn of Africa Middle
  East

(Completed)
(In progress)
8
Developing a Monitoring Forecasting System

• Identify periods of high risk of RVF (almost
  completed)
• Determine lead time in which high risk areas
  periods can be identified
• Identify threshold values of rainfall, soil
  moisture, vegetation greenness, inundation
  associated with historical outbreaks
• Assess ability to predict exceedance of these
  thresholds at different times using global
  climate models

(In progress)
(In progress)
9
Developing a Monitoring Forecasting System

• 4. Establish linkages between model outputs
  decision-making options at multiple levels
• Barriers to group cooperation must be identified,
  understood, and addressed in development of the
  system if cooperation is important
• It may be necessary to design group cooperation
  mechanisms into reporting components of the
  monitoring and forecasting system

10
Reputation and Vulnerability

• Reporting information on RVF outbreaks may hurt a
  producers credibility and reputation
• A reputation of being a producer with high risk
  for RVF may make a vulnerable producer more
  vulnerable
• Are there ways to address these problems?

11
RVF Decision Making

• Potential areas to explore
• Incentives for voluntary monitoring
• Sharing risk through sales contracts
• Compensation schemes producers who provide
  sensitive information
• Designing systems that take advantage of
  inter/intra group credibility effects
• Incentives for vaccination of animals
• Temporary restriction of trade from affected areas

12
Mechanism Design

• Buyers in Middle East could compensate
  pastoralists for reporting sick sheep, making
  monitoring and auditing costs less prohibitive

• Questions
• Would buyers blacklist pastoralists because they
  shared information about the presence of the
  virus, can a contract be designed to overcome
  this?
• Do pastoralists report differently as
  individuals, as groups, across ethnic groups?

13
Groups and Contract Design

• When making contracts, does the cost of revealing
  private information differ between individuals
  and groups?
• We will build on existing research by developing
  tools to explore
• What constitutes a group and what are the
  premiums that group dynamics bring?
• Experimental game in lab (Summer 2005) possibly
  at conference in the field

14
QUESTIONS?
Photo www.freshtracks.ca/group_incentive.shtml