ORIGINAL ARTICLE

Benefit-Cost Analysis of Foot and Mouth Disease Control inLarge Ruminants in CambodiaJ. R. Young1, S. Suon2, L. Rast3, S. Nampanya1, P. A. Windsor1 and R. D. Bush1

1 Faculty of Veterinary Science, University of Sydney, Camden, NSW, Australia2 Department of Animal Health and Production, Ministry of Agriculture, Forestry and Fisheries, Phnom Penh, Cambodia3 Charles Sturt University, Wagga Wagga, NSW, Australia

Keywords:

public health; trade; cattle; buffalo;

smallholder farmers; value chain

Correspondence:

J. R. Young. Faculty of Veterinary Science,

University of Sydney, Camden, NSW,

Australia. Tel.: +64211427340;

Fax: +61293511693;

E-mail: jyou1412@uni.sydney.edu.au

Received for publication August 9, 2014

doi:10.1111/tbed.12292

Summary

Foot and mouth disease (FMD) is endemic in Cambodia and throughout the

Greater Mekong Subregion and causes significant losses to rural smallholders

owning the majority of the national large ruminant population. However, due to

underreporting, paucity of knowledge of FMD impacts, limited veterinary capac-

ity and deficits of data available for analysis, the quantifiable benefits of a national

FMD control programme are unknown. To address this deficit, existing literature

and research data from the ‘Best practice health and husbandry of cattle, Cambo-

dia’ project conducted between 2007 and 2012, were used to develop a three-

phase analysis framework to: assess the impacts of the recent widespread FMD

epizootic in Cambodia in 2010, conduct a value chain analysis of the large rumi-

nant market and estimate the costs and benefits for a national large ruminant

biannual FMD vaccination programme. A trader survey conducted in 2010–2011provided cattle and buffalo value chain information and was matched to village

herd structure data to calculate a total large ruminant farm-gate value of USD

1.271 billion in 2010. Monte Carlo simulation modelling that implemented a 5-

year biannual vaccination programme at a cost of USD 6.3 an animal per year

identified a benefit-cost ratio of 1.40 (95% CI 0.96–2.20) when accounting for

recent prices of cattle and buffalo in Cambodia and based on an expected annual

incidence of 0.2 (assuming one major epizootic in the 5-year vaccination pro-

gramme). Given that the majority of the large ruminants are owned by rural

smallholders, and mostly the poor are involved in agricultural employment, the

successful implementation of an FMD control programme in Cambodia would be

expected to avoid estimated losses of USD 135 million; equivalent to 10.6% of the

2010 farm-gate value and contributing to important reductions in rural poverty

and food insecurity.

Introduction

Cambodia together with Myanmar, Lao PDR, Thailand,

Vietnam and the Yunnan Province of China make up the

six states of the Greater Mekong Subregion (GMS), a natu-

ral economic area bound by the Mekong River basin com-

prising 2.6 million km2 and inhabited by ~326 million

people (ADB, 2013). Cambodia has a population of ~14.3million people, with an estimated 70% reliant on agricul-

ture for employment (MAFF, 2012; The World Bank,

2013). Cambodia has a total land area of 181 035 km2, with

75% consisting of the Tonle Sap Basin and Mekong Low-

lands. The relative contribution of agriculture to gross

domestic product (GDP) in Cambodia was 36.0% in 2010

(Trading Economics, 2014) of a total GDP of USD 14 bil-

lion (The World Bank, 2013). Cambodia is one of the poor-

est countries in South-East Asia, ranked 136th on the

Human Development Index slightly ahead of its neighbour

Lao PDR at 139th (UNPD, 2014). Although significant

improvements have been made in the last two decades,

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 1

Transboundary and Emerging Diseases

rural poverty remains an important issue with an estimated

30% of people living below the poverty line (MAFF (Minis-

try of Agriculture Forestry and Fisheries), 2012). While rice

production remains the major contributor to the agricul-

tural sector, livestock play an important role in smallholder

farm systems. Livestock often serve multiple purposes

including a source of draught power for transport and till-

age, manure for biodigester fuel and fertilizer, sale for beef,

and importantly as a cash asset store which can be sold

when money is needed (Shankar et al., 2012; Young et al.,

2013a).

Improving cattle and buffalo (large ruminant) produc-

tion in Cambodia is a recognized pathway to alleviate rural

poverty and improve food security in the GMS (Windsor,

2011). However, the national large ruminant herd declined

3.1% between 2009 and 2010 and a further 2.0% between

2010 and 2011 (Suon et al., 2013). In 2011, the national

herd comprised 3 406 972 cattle and 692 611 buffalo with

over 99% of large ruminants owned by smallholder farm-

ers, typically owning 3 or fewer animals (Young et al.,

2013a). Production levels are low, with mean average daily

weight gains of ~50 g per day, poor body condition year

round and poor reproductive performance (Young et al.,

2013a). Foot and mouth disease (FMD) and haemorrhagic

septicaemia (HS) have been identified as the most signifi-

cant transboundary animal diseases (TADs) affecting large

ruminants in the GMS (Young et al., 2013c).

In 2010–2011, a major FMD epizootic highlighted seri-

ous failures in international biosecurity in the GMS, with

regional epizootic peaks in December 2010 until February

2011 and sporadic outbreaks in the following months (OIE

Sub-Regional Representation for South-East Asia, 2011).

The Department of Animal Health and Production

(DAHP) in Cambodia reported 61 449 FMD cases in large

ruminants in 2010 and a further 11 335 in 2011. Further-

more, major FMD outbreaks that occurred in 2010 in both

Japan and South Korea (both previously FMD free without

vaccination) were of virus serotypes shown to have origi-

nated in South-East Asia (Knowles et al., 2012). The global

annual impact of FMD in terms of visible production losses

and vaccination costs in endemic regions has been esti-

mated to be between US$6.5 and US$21 billion (Knight-

Jones and Rushton, 2013).

To justify and direct allocation of resources in FMD con-

trol and eradication programmes, evaluation of the finan-

cial impacts at the national level is required. Several recent

studies have investigated the financial impacts of FMD in

Cambodia and within the GMS (Shankar et al., 2012; Ver-

gne et al., 2012; Young et al., 2013b). However, evaluation

of the national impact of FMD is limited by underreporting

of cases; absence of data on impacts on a per unit basis (i.e.

per animal, household or village), the inability to confi-

dently confirm the diagnosis of FMD and differentiate it

from HS and other diseases in the field, undetermined total

value of the large ruminant industry and inadequate esti-

mates of the quantity of international trade in livestock.

Between 2007 and 2012, the ‘Best practice health and hus-

bandry of cattle, Cambodia’ (BPHH) research project

sought to identify and evaluate a series of health and hus-

bandry interventions implemented at the smallholder

farmer and village level, using a collaborative adult educa-

tion-based extension and farming systems approach. This

project proposed that by increasing farmer knowledge of

large ruminant health and husbandry techniques improved

health and productivity would in turn lead to improved

rural smallholder livelihoods. This study aimed to develop

a framework to enable: (i) further understanding of the

large ruminant value chain; (ii) estimation of the impact of

the 2010 FMD epizootic on smallholder farmers; and (iii) a

benefit-cost analysis (BCA) for improved FMD control

through a national biannual vaccination programme to be

conducted.

FMD and associated impacts

The movement of live animals is considered the most

important method of transmission of FMD virus (Rosen-

berg et al., 1980; Rweyemamu, 1984; Fonnan, 1991; Ferris

et al., 1992; Perry et al., 2002; Windsor et al., 2011). How-

ever, there is a paucity of information describing the large

ruminant value chain in Cambodia, and indeed, significant

challenges are faced when attempting to capture this infor-

mation. Much of the trade is informal (and in some cases

illegal) with smallholders selling their livestock for a num-

ber of reasons other than for beef, such as to liberate an

asset in times of cash need. The National Bank of Cambo-

dia (central bank) reported total depositors reached

1 266 412 by the end of 2011 (National Bank of Cambodia,

2012). Assuming 95% of these are personal accounts and

5% business accounts, then the estimated proportion of the

population with a bank account is only ~8.4% (ANZ Ben

Smith, Personal communication), indicating the use of

large ruminants as a storage of wealth may be highly signifi-

cant. To date, the majority of research on FMD impacts has

focused at the producer level of the value chain.

The first study to report the financial and social impacts

of FMD in Cambodia utilized data from 59 households

affected by FMD in 2006–07 from a total of 117 households

(Shankar et al., 2012). The mean aggregate cost of FMD in

cattle per infected household was USD 45, which accounted

for treatment (USD 23), loss of current sales due to mortal-

ity (USD 15), lost draught work (USD 7) and other (USD

0.06), equating to 7.6% of annual household income or

10% for lower income households (Shankar et al., 2012). A

2010 study of 62 farmers impacted by FMD included the

change in value of sick animals (representing a mean

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.2

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

immediate opportunity cost of USD 200.94), mean draught

replacement costs (USD 31.22), treatment and manage-

ment costs (USD 15.13), and the salvage value of dead ani-

mals (USD 80.50) (Young et al., 2013c). The financial

impact of FMD had increased significantly (when com-

pared to Shankar et al., 2012), likely reflecting a rapid

increase in prices between 2006–2007 and 2010, plus the

inclusion of the significant opportunity cost due to weight

loss resulting from clinical FMD (O’Connell et al., 2013;

Young et al., 2013c). Of note, neither Cambodian study

investigated losses due to chronic FMD, despite an assess-

ment of FMD in South Sudan suggesting the chronic form

of FMD accounted for 28.2% of total FMD losses (Barasa

et al., 2008).

FMD reporting

Cambodia’s administrative structure contains 24 provinces,

185 districts, 1621 communes and 14 073 villages (MoP

(Ministry of Planning), 2009). Government veterinary ser-

vices have offices and staff in the central (Phnom Penh),

provincial and district levels. Although there are no animal

health resources dedicated at the commune level, 12,474

private village animal health workers (VAHWs) were

reported active in 2010, covering ~89% of the villages

(MAFF, 2011).

FMD was reported to occur nearly every year by 41.5%

of VAHW’s and every few years by 27.4% of VAHW’s

(Stratton, 2013). During village FMD outbreaks, 45.4%

VAHW’s reported that greater than 60% of the village herd

was affected (Stratton, 2013). FMD was the disease most

likely to be reported by VAHW’s at ~83%, with HS close

behind at ~81% (Stratton, 2013). However, underreporting

of FMD and HS cases is known to occur in Cambodia

(Shankar et al., 2012; Vergne et al., 2012; Kawasaki et al.,

2013; Young et al., 2013b) and is also likely throughout the

GMS. A two-source capture–recapture analysis for estimat-

ing the true number of villages experiencing clinical FMD

in 2009 was conducted in Svay Rieng province in Cambo-

dia (Vergne et al., 2012), evaluating the reporting rate to

provincial authorities at 0.05 (95% CI 0.03–0.13). In a large

study, 445 VAHW’s conducted in 2010, approximately

72% of VAHW’s claimed to report FMD immediately

(Stratton, 2013).

In a recent study in Laos investigating the financial

impact of FMD, 310 smallholder farmers were surveyed of

which 163 (~53%) responded that they reported FMD

(Nampanya et al., 2013), with a range between 25 and 65%

between the three northern provinces investigated. Of

interest was that farmers indicated they report FMD to the

village chief or village veterinary worker (equivalent to

Cambodia’s VAHW) 69–92% of the time and to the district

officer 8–31% of the time, generally 2–3 days after seeing

the disease (Nampanya et al., 2013). These survey results

confirm that non-reporting of disease occurs from initial

detection and subsequent transfers of information as was

experienced with highly pathogenic avian influenza in

South-East Asia (FAO, 2013). Further filtering can occur at

the next step (usually district level) if the outbreak is to be

judged to be a common endemic problem rather than a

major disease incident requiring immediate investigation

(FAO, 2013) or an emergency response intervention

(Windsor et al., 2011). As capturing disease incident infor-

mation may impose financial burdens from use of time and

communication costs, and these are often borne by those

volunteering information, there is likely to be a decline in

enthusiasm for the reporting process over time (FAO,

2013). Furthermore, this private–public partnership relies

on the efforts of ‘upstream’ officials and the quality of their

engagement with the private village-level workers (FAO,

2013). Surveillance is the foundation of infectious disease

control, and poor reporting sensitivity and specificity is

problematic (Coker et al., 2011), particularly when

attempting to estimate financial burdens and undertake

economic modelling for supporting control decisions.

Large ruminant trade in Cambodia

Large ruminant trade pathways have recently been

described (Cocks et al., 2009; Kerr et al., 2012) and suggest

export of the highest quality Cambodian animals to major

Vietnamese cities, supply of lower quality animals to the

domestic market in Phnom Penh and processing of poor

quality animals at local slaughter points for sale in local dis-

trict and provincial markets. Cambodia was also identified

as a major transit pathway for up to 150 000 large rumi-

nants moving through Cambodia from Thailand to Viet-

nam in 2009 (Kerr et al., 2012). These transit journeys are

usually rapid, crossing Cambodia in as little as 1 day, with

livestock often held in depots (along with cattle sourced

from Cambodia and Myanmar) for up to 4 days prior to

(often illegal) border crossings into Vietnam (Kerr et al.,

2013). It was ascertained that there was limited regulation

with FMD occurring within depots. Disease risks were asso-

ciated with direct contact of livestock from different con-

signments, indirect contact through uncleaned water

troughs and trucks moving through pens, plus limited bio-

security knowledge by depot staff (Kerr et al., 2013), identi-

fying depots as an unmanaged critical control point and

potential amplification point for TAD transmission. Trans-

boundary trade was also shown to be dynamic with the

Thai–Vietnam transit trade greatly decreasing by 2010 due

to currency fluctuations. This indicates that traders operate

on slim profit margins and were inclined to work around

official pathways or stop trading (Kerr et al., 2013). Traders

are recognized as an important source of information on

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 3

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia

animal movements, particularly as previous investigations

indicate that official movement permits did not capture

information on the significant unofficial livestock trade as

well as the movement drivers and trading practices that

influence disease risk (Kerr et al., 2012; O’Connell et al.,

2013). Although these investigations of trader activities

have provided valuable information on regional animal

movement, there remains a paucity of quantifiable infor-

mation on the large ruminant value chain within Cambo-

dia.

Methodology

To identify the required inputs for the BCA of a biannual

FMD vaccination control programme in Cambodia, a

three-phase analysis framework was developed involving

six steps (Fig. 1). Following framework development and

then by working backwards, known information sources

enabled identification of knowledge gaps, then identifica-

tion of specific steps where estimations were required.

Estimation of the financial impact per animal

The mean financial impact of FMD on a per animal basis

was calculated using previously reported estimated average

costs (Young et al., 2013b). This included four possible

outcomes and mean costs: (i) animal survival with treat-

ment USD 216.32; (ii) animal survival with treatment and

draught replacement USD 247.54; (iii) animal dies with

treatment USD 339.32 and (iv) animal dies with treatment

and draught replacement USD 370.54. Using Microsoft

Excel (2013), the four outcomes were proportionally

weighted, based on reported village morbidity (77.3%) and

mortality (7.3%) and use of draught (27%) (Young et al.,

2013b). Monte Carlo (MC) simulation was then applied to

the four mean cost outcomes using normal distribution

and one standard deviation to calculate the predicted mean

cost per animal of clinical FMD using 10 000 iterations in

@risk 6 (Palisade 2014).

Estimating the true FMD incidence in 2010

A reporting probability model was developed to describe

the reporting cascade of FMD from the smallholder farmer

(animal owner) to the central DAHP in Phnom Penh. The

model included the five stages involved in passing on FMD

outbreak information. These included the smallholder

farmer, the VAHW, the district office of animal health and

production (DOAHP), the provincial office of animal

health and production (POAHP), the national veterinary

research institute (NaVRI) and the central DAHP. The

model used a dichotomous scenario, of ‘yes’ – the FMD

report was passed further up the cascade resulting in a

successful report (SR), or ‘no’ – the FMD report was not

escalated resulting in a non-report (NR). The model was

used to calculate the estimated true incidence (ETI) by

Fig. 1. The three-phase analysis framework comprising six steps developed to evaluate the large ruminant value chain in Cambodia and to determine

the financial impact and BCA of potential FMD control in 2010.

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.4

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

multiplying each SR up the cascade. To apply sensitivity to

the model and account for assumptions, a MC simulation

using @risk 6 was applied to the cascade reporting model

for each reporting step to calculate the ETI of FMD using

the formula:

ETI ¼ ðReported FMD casesÞ=ðReporting rateÞ@risk incorporates MC simulation that is based on

inputs including minimum (Mi), most likely (Ml) and

maximum (Ma) values and defining the distribution, then

running the model through 10 000 simulations to provide

an output. A triangular distribution was used where pub-

lished literature provided input mean and range and one

standard deviation (SD) for Mi and Ma where only a singu-

lar value was available or assumptions were used. For esti-

mating the number of farmers who report FMD cases to

the VAHW, values of 25% (Mi), 53% (Ml) and 65% (Ma)

were used based on previous published surveys (Nampanya

et al., 2013). For the number of VAHW’s reporting a case

of FMD to the DOAHP, 72% (Ml) was used (Stratton,

2013) with Mi and Ma as one SD from the mean. In the

remaining three steps of the cascade (Fig. 2) a value of

two-thirds SR (66.6%) was determined to be the Ml with a

triangular distribution and one SD for Mi and Ma values.

The MC model was run using 10 000 iterations to calculate

the ETI.

Establishing the large ruminant value chain including a

farm-gate valuation

To gain further understanding of the large ruminant value

chain, a survey of 100 traders operating in Cambodia was

performed in December 2011 and January 2012. Traders

were purposively selected by DAHP project staff through

known networks, as no sampling frame was available. They

were advised that the information was for research pur-

poses only. Traders were asked to provide details on large

ruminant trading during the previous 12 months. The

structured survey included closed questions on trader’s

address, years trading, primary operating location and total

number of large ruminants purchased during the previous

12 months, broken down by

• Species (cattle or Asiatic buffalo)

• Breed

• Sex (male or female)

• Age groups (0–2, >2–8, >8 years)

• Body condition score (skinny, medium or fat)

The purchase price (per animal) and market sale price

(per kg) for meat from both species was recorded. The tra-

der survey data were summarized to provide mean sale val-

ues for each large ruminant species, sex, age group and

body condition score. Where large ruminant categories

were not represented in the survey, the mean purchase

value was predicted using the proportional variation in cat-

egories of the same breed where the full range of prices was

available. Traders were asked to identify the transport

methods used and what proportion of large ruminants was

sourced directly from farmers and from other traders.

Traders were questioned on costs relating to transport,

slaughter, animal movement levies, meat inspection and

market stall rental. Data from traders who traded more

than 100 large ruminants per year (considered ‘large trad-

ers’) were further summarized to provide value chain infor-

mation, including sources of large ruminants and transport

Fig. 2. Probability model to calculate the estimated true incidence (ETI) of FMD reporting.

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 5

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia

routes and methods. If traders identified more than one

source province, the first-named province was used due to

the assumption that the first was the primary source.

Finally, traders were asked a series of open-ended questions

to describe their views on current issues relating to the large

ruminant marketing and what possible solutions they

might suggest. The results of the trader survey were entered

into and analysed (Microsoft ExcelTM 2010) to provide a

descriptive analysis of the value chain. Purchase and sale

data were provided in Cambodian Riel and converted to

USD at a rate of 4018 Riel = 1 USD. Longitudinal cattle

and buffalo production surveys (Young et al., 2013a) were

used to provide an estimate of smallholder large ruminant

herd age and livestock class structure, to be matched to the

trader survey results. This analysis enabled calculation of

the national farm-gate valuation of large ruminants in

Cambodia.

Estimating the national financial impact of FMD on the

large ruminant industry

The national impact of FMD in Cambodia in 2010 was esti-

mated as:

Total impacts at farm-gate USD¼Mean financial impact per animal (calculated in Step 1)� ETI of FMD in2010ðcalculated in Step 2Þ

Estimating the cost of FMD control

Cost data from regular cattle and buffalo vaccinations in

BPHH project villages was used to prepare a budget for

biannual FMD vaccination at the village level (Table 1). To

vaccinate a village of 400 large ruminants generally requires

3 staff due to the limited availability of large ruminant han-

dling facilities making this a time and labour intensive

activity. The vaccination procedure involves DAHP staff

setting up a vaccination station in each village location with

farmers invited to bring their large ruminants for vaccina-

tion. Smallholder farmers will generally be available from

8:30 a.m. to 11:00 a.m. due to other household work and

employment commitments. On average, 100 large rumi-

nants are vaccinated per day for 4 days. The budget for the

village vaccination was developed and then scaled to calcu-

late the cost of a national herd control campaign by calcu-

lating the cost on a per animal basis and then multiplying

that cost by the national population.

Undertaking a benefit-cost analysis for FMD control using

biannual vaccination

A BCA was performed to determine the economic feasibil-

ity for conducting a national large ruminant biannual FMD

vaccination campaign over a 5-year duration. A benefit-

cost ratio (BCR) was determined as the decision criterion

and was calculated based on the formula (Dijkhuizen and

Morris, 1997):

BCR ¼ PV(benefits)=PV(costs)

PV = FV=ð1þ r=100Þn

r is the annual ‘interest rate’ (in %) and n is the number of

years in the future

The present value (PV) of a future cost or benefit (FV)

incorporates the discount rate to ensure costs and benefits

are comparable. The BCR was calculated assuming a major

outbreak (equivalent to the 2010 outbreak) occurs in either

year one, two, three, four or five (i.e. an incidence of 1.00,

0.50, 0.33, 0.25 and 0.20 over the 5-year control pro-

gramme). A discount rate of 8% was selected based on the

10-year average interest rate (1994–2014) in Cambodia of

4% (Trading Economics, 2014) and an additional 4% to

account for an estimated agricultural investment risk, and

applied to both the FMD vaccination programme cost and

predicted financial impact. MC simulation using defined

inputs (from steps 1 to 5) was applied to the BCA model

with 10 000 iterations for the five different incidence sce-

narios. A BCR ≥1 indicates the investment is worthwhile.

Results

Estimation of the financial impact per animal

The mean financial impact per animal of FMD in Cambo-

dia was USD 247.34 based on a proportional weighted out-

come using the four possible outcomes (Fig. 3).

Estimating the true incidence of disease in 2010

Using the probability model in Fig. 2 and applying MC

simulation, the estimated reported number of FMD cases

accounted for 11.3% of the ETI. Therefore the ETI of FMD

in Cambodia in 2010 was calculated as 545 109 individual

large ruminant cases.

Table 1. FMD vaccination budget in USD for a village with 400 large

ruminants

Item Unit cost (USD) Total (USD)

FMD vaccine $1.30 per animal 9 400 $520.00

Vehicle cost $100 per day 9 4 days $400.00

Equipment Cooler box, vaccination

guns, needles, recording sheets

$100.00

Accommodation $10 per person per

night 9 4 nights

$120.00

Per diem $10 per person per

day 9 3 people

$120.00

Total $1260.00

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.6

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

Evaluating the farm-gate value of large ruminants in

Cambodia

Surveyed traders had a mean of 10.0 years (range 1–32 years) trading experience. A descriptive summary of

traders who traded 100 or more large ruminants in the

12 months prior to the survey is provided (Table 2). The

mean sale prices recorded by the traders for each large

ruminant stock class is also provided (Table 3). Large

ruminant herd structure including age, sex and body condi-

tion was matched to longitudinal data to calculate the

national farm-gate valuation of USD 1 271 789 989.

Traders operated in multiple geographic areas with 49%

of traders sourcing animals from outside of their home

operating location and purchasing 64.7% of livestock

traded. All traders (100%) determine the purchase price of

large ruminants based on body condition score and general

appearance. Their primary operating locations included the

provinces of Kandal (18 traders), Kampong Thom (19),

Takeo (10), Kampot (14), Kampong Cham (33), Beantey

Men Chay (2), Phnom Penh (1), Peras Vihar (1) and Thai-

land (2). All but 5 traders stated that they were farmers as

well as traders. Districts where ‘large traders’ sourcing more

than 100 large ruminants in the 12-month period are

shown (Fig. 4) and value chain information is summarized

(Table 2).

Traders sourced large ruminants by multiple methods,

including: (i) 94% of traders purchase livestock directly

from farmers; (ii) 36% purchase from a middleman trader;

(iii) 6% purchase from a broker and (iv) 2% of trader’s

sourced livestock by ‘other’ means.

Twenty-three traders were based in Phnom Penh primar-

ily supplying large ruminants to the Choy Chhangva

Slaughterhouse in Phnom Penh. Large ruminants were

sourced from seven provinces including Kampong Cham,

Oddar Meanchey, Kampong Thom, Preah Vihear, Oddar

Meanchey, Stung Treng, Banteay Meanchey and one trader

sourced large ruminants from Thailand. Average beef prices

at six markets are provided (Table 4).

Traders were asked open questions about issues they

faced in their occupation. These were grouped where possi-

ble to highlight the most common issues (Table 5).

National financial impact of FMD to the large ruminant

industry

The estimated financial national impact on the farm-gate

valuation of large ruminants was calculated as the

ETI 9 (mean costs). This equated to USD 134 829 478

equivalent to 10.6% of the farm-gate valuation.

Estimated cost of FMD control in Cambodia

The cost of FMD control on an animal per year basis was

calculated as USD 3.15 per animal per vaccination, or USD

6.30 an animal per year. The total annual cost for the

national herd was USD 25 827 373.

Cost-benefit analysis for improved FMD control in

Cambodia using biannual vaccination

Results from a baseline model to determine the benefits

and costs for improved FMD control in Cambodia using

biannual vaccination are presented (Table 6). Both costs

and impacts were adjusted annually to account for the

discount rate. Only in the fifth year of the programme

did the BCR 95% CI drop below the decision criterion

of 1.

Fig. 3. Proportional weighted animal survival outcome of FMD infection.

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 7

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia

Table

2.Descriptive

summaryoftrad

erswhotrad

ed100ormore

largeruminan

tsin

the12monthspriorto

thesurvey

Trad

ers(n)

Source

province/Country

Sourcedistrict

Total

cattle

Total

buffalo

Prim

aryroute

(nTrad

ers)

Tran

sport

method

Averagetran

sport

cost(USD

)Slau

ghterhouse

Salepoint

10

Kam

pongCham

Cham

kaler,Tu

ongKmom,

Kam

pongSe

em,Se

iySo

ntuor,

KosSo

etin,KongMea

s,

Prey

Chor

2649

0Mek

ongRiver

(8),

Road

6,7an

d

8(2)

Truck,Boat

18.48(16.85boat,

24.97truck)

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

2Ban

teay

Mea

nchey

Svay

Chea

k,Tm

orPo

uk

340

36

Road

6A(2)

Truck

39.95

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

5Kam

pongTh

om

Sontuk,

Stuen

gsen,Stung,

Kam

pongSvay

768

95

Road

6A(5)

Truck

21.97

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

2Oddar

Mea

nchey

Ban

teyChhma,

Kokm

on

1160

0Road

5(1),

6A(2)

Truck

46.19

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

2Prea

hVihea

rRovean

g,Kuland

640

60

Road

64(2),

6A(2)

Truck

37.45

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

1StungTren

gPe

rasViher

600

0Road

7,8,6A(1)

Truck

67.42

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

1Th

ailand

Thailandan

dCam

bodainCountry

Border

(Ben

gtrak

uon)

100

0Road

5,6A(1)

Truck

29.96

ChoyChhan

gva

Slau

thterhouse

(PP)

Phnom

PenhMarke

t

8Kan

dal

Saan

g,Kosthom

878

0Road

21

Walk,

Truck,

Moto

4.99(m

oto),2.50

(walk),5.08(truck)

Kosthom

Slau

thterhouse

Kosthom

Marke

t

10

Kan

dal

Saan

g,Kosthom,Kea

nSvay

2045

0Road

21

Walk,

Truck,

Moto

7.07(m

oto),2.50

(walk),4.56(truck)

Saan

gSlau

thterhouse

Saan

gMarke

t

3Kam

pongTh

om

Stuen

gSa

en,Prasat

Balan

gk,

Prasat

Sambour,Kam

pongSvay

910

110

Road

6A

Truck,Moto

6.24(Truck),7.49

(Moto)

Kam

pongSvay

Slau

ghterhouse

Kam

pongSvay

Marke

t

2Prea

hVihea

rRovien

g950

0Road

6A

Truck

12.48

SuntukSlau

thterhouse

KompongTh

om

Marke

t

8Kam

pongTh

om

Stuen

gSa

en,Prasat

Balan

gk,

Prasat

Sambour,Kam

pongSvay,

Stoung,Sa

ndaa

n

1615

140

Road

6A

Truck,Walk,

Moto

4.65(M

oto),

2.50(w

alk)

Stuen

gSe

an

Slau

thterhouse

Stuen

gSe

anMarke

t

5Ta

keo

Trea

ng,Prey

Kab

bas,

Samrarong,Bati,

DounKae

v,Angko

rBorei

1234

0Road

2Truck,Walk,

Moto

14.98

Unkn

own

Trea

ngDistrict

4Ta

keo

KiriV

ong,Sa

mrarong

1670

0Road

2Truck,Walk

30.59

Unkn

own

KiriV

ongDistrictto

VeitNam

border

(Phnom

Den

t)

1Ta

keo

DounKae

v220

0Road

2Truck

37.45

Unkn

own

DounKae

vDistrict

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.8

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

Discussion

This study provides one of the first attempts to quantify

both the national impact of a large-scale FMD epizootic

and conduct a BCA of a national FMD control pro-

gramme. This BCA analysis supports the implementation

of a national biannual FMD vaccination control pro-

gramme in Cambodia, with sensitivity analysis using con-

fidence intervals only marginally dropping below 1 in the

fifth year of the programme. As the movement of live

animals is considered the highest risk factor for FMD

Table 3. Mean price of large ruminants in Cambodia between December 2011 and January 2012

Breed Sex Age Condition n traders

Total head

sold Lower price Upper price Mean price (USD)

L F 0–2 1 0 0 – – 255.09a

L F 0–2 2 5 131 249.69 374.53 335.46

L F 0–2 3 2 60 424.47 449.44 441.16

L F >2–4 1 2 21 174.78 174.78 174.78

L F >2–4 2 28 1200 149.81 624.22 413.34

L F >2–4 3 9 439 299.63 923.85 597.15

L F >4–10 1 2 63 249.69 249.69 249.69

L F >4–10 2 47 2966 324.59 749.06 505.44

L F >4–10 3 16 742 299.63 823.97 608.64

L M 0–2 1 0 0 – – 265.35a

L M 0–2 2 5 152 274.66 424.47 364.02

L M 0–2 3 2 110 499.38 499.38 499.38

L CM >2–4 1 2 33 199.75 199.75 199.75

L CM >2–4 2 27 2231 199.75 624.22 484.89

L CM >2–4 3 9 849 324.59 998.75 660.81

L CM >4–10 1 2 78 299.63 299.63 299.63

L CM >4–10 2 49 4443 299.63 873.91 633.40

L CM >4–10 3 16 1888 374.53 998.75 541.75

CB F >2–4 3 2 1400 523.04 523.04 523.04

CB F >4–10 3 2 100 560.40 560.40 560.40

CB CM >2–4 3 2 2440 523.04 523.04 523.04

CB CM >4–10 3 2 100 560.40 560.40 560.40

CH F 0–2 1 0 0 – – 184.01a

CH F 0–2 2 2 82 224.16 249.07 239.35

CH F 0–2 3 1 50 311.33 311.33 311.33

CH F >2–4 1 0 0 – – 325.22a

CH F >2–4 2 6 188 186.80 435.87 340.94

CH F >2–4 3 6 248 249.07 435.87 357.43

CH F >4–10 1 0 0 – – 291.38a

CH F >4–10 2 8 365 249.07 373.60 361.15

CH F >4–10 3 15 1367 311.33 498.13 447.62

CH M 0–2 1 0 0 – – 322.47a

CH M 0–2 2 2 57 249.07 373.60 358.31

CH M 0–2 3 1 50 398.51 398.51 398.51

CH M >4–10 1 0 0 – – 412.32a

CH M >4–10 2 1 15 435.87 435.87 435.87

CH M >4–10 3 1 25 460.77 460.77 460.77

CH CM >4–10 1 0 0 – – 304.52a

CH CM >4–10 2 8 596 273.97 473.23 379.53

CH CM >4–10 3 15 1247 373.60 560.40 473.02

CH CM >2–4 1 0 0 – – 240.60a

CH CM >2–4 2 6 207 211.71 498.13 323.00

CH CM >2–4 3 6 360 311.33 535.49 433.62

B F >2–4 2 1 10 186.80 186.80 186.80

B F >4–10 2 7 175 249.07 311.33 276.11

B CM >4–10 2 8 256 249.07 373.60 285.79

NB: L = Local breed, CB, Crossbred/Brahman, CH, Crossbred/Haryana, B, Asiatic Buffalo, aEstimated mean purchase value predicted using the pro-

portional variation in categories of the same breed where the full range of prices was available, F = female, M = entire male, CM = castrated male.

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 9

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia

transmission, consideration of control measures without

attempts to understand the smallholder large ruminant

value chain may be limited in value. In this analysis, the

farm-gate value of the national large ruminant herd was

USD 1.271 billion. This represents a significant store of

wealth predominantly held by smallholder farmers. There

is a paucity of information on the large ruminant value

chain, and baseline descriptive information on the large

ruminant market is needed to assist in allocation of

resources to both protect and develop this market.

Not surprisingly, a large variation in both trader prac-

tices and prices for livestock was reported, demonstrating

the heterogeneous nature of the large ruminant industry in

Cambodia, a not uncommon finding in smallholder live-

stock systems. Cattle beef prices were generally higher than

buffalo at market, and the highest beef price was reported

in Phnom Penh and Kandal Province, most likely reflecting

the highest demand for red meat in the capital. Traders

used multiple methods both to source and transport large

ruminants to markets, with almost all traders sourcing

directly from farmers, and approximately a third sourcing

from middlemen. This study identified five slaughterhouses

where surveyed traders reported over 100 trades in a 12-

month trading period. Further engagement with both trad-

ers and slaughterhouse operators could offer opportunities

for the development of meat processing, food safety, live-

stock recording systems, market trends, disease prevention

activities and biosecurity. While further research is needed,

particularly in value chain components beyond the farm

gate, this study provides information on large ruminant

trading, transport methods and routes, processing and sale

points in Cambodia. This information is important in

understanding where to apply disease control interventions,

including both specific locations and key stakeholders. The

farm-gate valuation provides a method to help quantify the

impact of the 2010 FMD epizootic on the collective small-

holder farmer.

The calculated total impact of the 2010 FMD epizootic

of USD 135 million represents 10.6% of the farm-gate

value. Flow-on effects further along the value chain are

likely to add to negative financial impacts of FMD and fur-

ther support the decision for FMD control using economic

evidence. Chronic impacts of FMD may also exacerbate the

losses. Food security and poverty are major issues within

the GMS, with the sustainable production of red meat

highlighted as a priority by the Cambodian government.

However, the ongoing secure supply of red meat from large

ruminants is subject to the introduction of improved con-

trol and prevention measures for diseases such as FMD and

HS. The allocation of limited resources for disease risk

management and control must be supported with accurate

evidence of disease cost to promote government policy for-

mation and facilitate support from donor organizations.

This paper enhances knowledge of the Cambodian large

ruminant value chain as well as proposes a model for

improved estimates of the costs of FMD and benefits of

FMD control in Cambodia. As impact per animal data was

sourced from four villages and the resultant potential of

sample size to lead to sampling bias and impact validity,

Monte Carlo simulation modelling was developed using

Microsoft Excel with@risk to apply sensitivity analysis.

Undertaking large-scale FMD impact studies presents a

range of both logistical (and potentially ethical) difficulties.

This should be acknowledged rather than restrictive in

efforts to develop models to quantify impacts, as ultimately

decisions for improved biosecurity including the use (or

non-use) of vaccination must be made. Indirect impacts

are difficult to assess, and the growing awareness of animal

welfare driving consumer preferences in on-farm manage-

ment must not be ignored, including within developing

countries. In smallholder systems, there is a huge range of

variability between farm-to-farm, village-to-village and so

on. The Cambodian agricultural calendar is determined by

a seasonal climate, and the timing of an outbreak would be

expected to cause variation in the direct and indirect

impact costs on affected smallholders. FMDV infection can

produce variable clinical signs and severity depending on

the strain, infective dose and individual susceptibility (Kit-

ching et al., 2005). Secondary infections may also occur,

leading to greater weight loss or increased mortality. Com-

bined outbreaks of FMD and HS have been reported in

India (Subash et al., 2004), and this scenario may have

occurred in Cambodia as both diseases are endemic.

The chronic impacts of FMD have been reported (Mi-

nett, 1948; Maqsood et al., 1958; Ghanem and Abdel-

Hamid, 2010) and reviewed by Arzt et al. (2011). During

the BPHH project, we were advised of cases of cattle post-

FMD that fit the Arzt et al. (2011) clinical description of

the ‘heat intolerance syndrome’, with farmers commenting

cattle ‘like to swim a lot’ and their ‘coats changed colour’.

Little is known about FMDV recovery in the GMS, with

assumptions that animals make a full recovery, although

there is very little data to support this or information on

how long it takes for return to the pre-disease healthy

weight in ‘the field’. Impacts on future fertility and repro-

duction, growth and lifetime productivity in the GMS are

also unknown and further research is required.

The notifiable disease reporting cascade was presented to

demonstrate the hierarchical nature of veterinary services

in GMS countries. While FMD is often described as being

endemic, this does not always reflect the nature of epizootic

outbreaks that have been observed such as in 2010–2011(Madin, 2011). A number of factors may impact this, such

as an outbreak of a new FMDV strain or subtype, waning

natural herd immunity and market forces leading to

changes in animal movement trends. Since the 2010–2011

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.10

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

epizootic, the Cambodian DAHP has implemented a dis-

ease reporting phone ‘hotline’ (FAO, 2013) to provide a

more direct reporting tool. The effectiveness of this strategy

requires evaluation, particularly as it may be limited in the

absence of a widespread biosecurity public awareness

campaign. Further, the reporting of cases of FMD may be

limited due to a number of reasons. Clinical diagnosis can

be a challenge due to difficulty to recognize the mild form

of the disease (Kitching, 2002) and differentiate it from

other infectious diseases such as HS. Farmers and VAHWs

may not have the knowledge to identify and differentiate

disease based on clinical signs and/or are unaware of the

requirement to report this notifiable disease. Recognition

of clinical signs and knowledge of infectious disease has

been shown to be low (Nampanya et al., 2012; Young et al.,

2014). In the event of widespread outbreaks, as was the case

in 2010, farmers decisions to report may be impacted by a

so-called diffusion of responsibility, not dissimilar to the

‘bystander effect’ reported in psychology literature where

large groups who witness an emergency are less likely to

intervene believing that others will (Darley and Latan�e,

1968). The estimation of underreporting of FMD cases

used both literature reported inputs and estimations, par-

ticularly in the three upper steps of the reporting cascade.

While accuracy in these assumptions would be a valid

query, a detail capture-recapture analysis in southern Cam-

bodia in 2009 reported only 5% of FMD cases were

reported (Vergne et al., 2012). Therefore, our ETI of

545 000 or 11.3% could be considered conservative. Vergne

et al. (2012) identified that only the most severely affected

villages were reported, as the case in Svay Rieng in 2009.

Provincial authorities were aware that many more villages

experienced clinical cases, and their failure to record all the

infected villages can hamper national and regional efforts

to combat and eradicate FMD (Vergne et al., 2012) with

more accurate notification protocols required.

Fig. 4. Districts identified by traders as the primary source of large ruminants along with primary transit routes.

Table 4. Average meat market price per kg

Mean meat market price (USD) per kg

Province Cattle Buffalo

Kandal 6.99 6.49

Kampong Thom 5.49 5.49

Takeo 5.74 5.49

Kampot 5.49 5.49

Phnom Penh 6.99 6.49

Kampong Cham 5.74 5.74

All Provinces 6.08 5.87

Table 5. Cambodian trader comments and issues

Cambodian trader comments & issues %

Low margin on trades 42

Lack formal domestic market 14

DAHP to improve farmer production and knowledge 14

Limited large ruminant supply 11

Suggest improve large ruminant breed 4

Reduce slaughter taxes 3

Vietnam sale payment issues 2

High transport licence costs 2

High transport costs 1

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 11

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia

The cost of biannual vaccination was based on staff from

the DAHP travelling to each village site and undertaking a

vaccination programme. This activity is costly, and greater

engagement with VAHWs and collaboration with the com-

mercial sector would be expected to lead to improved effi-

ciency and a reduction of cost. A government-led national

vaccination programme with the current human resources

available is not considered viable, with research of methods

of implementation and sustainability urgently needed.

FMD vaccine is costly and significant effort and technical

expertise is required to ensure vaccines provide efficacious

protection to circulating field strains by the collection of

evidence that successful immunomodulation has been

achieved. Education is needed to improve intervention suc-

cess and sustainability, particularly in cold chain manage-

ment of vaccine and farmer understanding of the benefits

of vaccination and improved biosecurity (Young et al.,

2013b). The model does not account for vaccine protection

levels, and instead assumes that herd immunity is achieved

with costs are allowed for the entire national herd. The fun-

damental concept of herd immunity is that it is not neces-

sary to immunize every individual large ruminant within a

population to be able to prevent its entry into that popula-

tion (Roeder and Taylor, 2007). Despite the model, it is

unlikely that any vaccination programme could be expected

to reach all large ruminants, particularly in year 1. Initial

attention should focus on FMD ‘hot spots’ or areas with

high reports and large ruminant density. However, defining

hot spots may be influenced by reporting bias in the

absence of a highly functional and effective veterinary ser-

vice. This indicates that surveillance should focus on areas

of dense large ruminant populations and where high levels

of trade in large ruminants occur. Careful consideration

should be given to areas for zoning under the OIEs guide-

lines, and in particular support (including financial) in the

implementation of such programmes. Applying funding for

vaccination without funding for effective implementation

capacity carries a high risk, with a suggested funding ratio

of 1:1 for vaccine cost and technical support. It should be

noted that this model did not account for FMD control

costs such as on-going disease surveillance.

The OIE defines an ‘outbreak’ as the occurrence of one

or more cases in an epidemiological unit. However, defin-

ing the epidemiological unit can present challenges that

vary depending on management and community structure.

We propose that in Cambodia the epidemiological unit be

defined as the village, as large ruminants are often taken

out daily in groups from multiple smallholdings (Young

et al., 2013a) and that biosecurity interventions should be

targeted at the village level due to the high level of comin-

gling and intravillage trades. The key indicator should be

the recognition of an infected village, rather than number

of cases per se, particularly as underreporting subsequently

results in reduced awareness of the full socio-economic

impacts of the disease outbreak. Use of social network

analysis research techniques could provide valuable infor-

mation on risk quantification of intravillage interactions

and contact rates.

The greatest impact on reducing FMD spread among vil-

lages is considered to involve strategies that reduce the like-

lihood of introductions through livestock purchases and

encourage villagers to take greater care when grazing live-

stock with animals from neighbouring villages, particularly

when sharing common water supplies (Cleland et al.,

1996). Previous research has also identified that biosecurity

education for traders and other market chain participants

would help to reduce TAD transmission in the GMS (Kerr

et al., 2013). However, further evidence is needed to sup-

port this, particularly as there is considerable potential for

traders to purchase and move diseased livestock at ‘salvage’

prices, capitalizing on animals with reduced body weight

and condition and potentially exacerbating disease trans-

mission through direct movement or fomites. Traders cur-

rently establish large ruminant purchase and sale prices on

visual assessment of body weight and condition (Henry

and Bush, 2013). VAHWs may also stand to benefit from

the result of a disease outbreak through the sale of drugs

for treatments, particularly of inappropriate antibiotics of

doubtful therapeutic value for a viral aetiology unless a sec-

ondary bacterial infection has occurred, incurring signifi-

cant losses to household incomes (unpublished

observations). A very different scenario exists for an out-

break of HS, where immediate antibiotic therapy is strongly

indicated to prevent mortalities. This indicates the need for

improved skills in differentiating FMD from HS in the field

and applying the appropriate interventions. Understanding

Table 6. Benefit-cost analysis over the 5-year biannual vaccination programme

Cost-benefit analysis Year 1 Year 2 Year 3 Year 4 Year 5

Annual Incidence 1.00 0.50 0.33 0.25 0.20

Annual cost USD 25 827 373 27 893 563 30 125 048 32 535 052 35 137 856

Accrual cost USD – 53 720 936 83 845 983 116 381 035 151 518 891

Financial impact of outbreak USD 134 829 478 145 615 836 157 265 103 169 846 311 183 434 016

Benefit Cost Ratio (BCR) 5.22 2.71 1.88 1.46 1.21

MC simulation BCR (95%CI) 6.05 (4.11–9.55) 3.14 (2.15–4.97) 2.17 (1.48–3.40) 1.69 (1.15–2.66) 1.40 (0.96–2.20)

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.12

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

motivations of stakeholders is an important step in manag-

ing disease control efforts. In order to ensure stakeholders

are motivated to report disease, there must be an actionable

result that addresses risk management. If reporting results

in very few emergency response disease control measures

due to limited resources such as vaccination programmes

and movement restrictions (Tum et al., 2011), VAHWs

would be expected to have limited motivation to report

infections to veterinary authorities (Vergne et al., 2012).

Conclusion

Large ruminants are important in smallholder livestock sys-

tems in the GMS and sustainable growth of this market has

significant potential to reduce rural poverty and food inse-

curity. However, significant transboundary animal disease

risks such as FMD need to be addressed to ensure small-

holder farmers have an opportunity to participate in grow-

ing regional economies of the GMS. This study provides a

framework for assessing financial impacts at the national

level and a BCA for FMD control by vaccination. This

framework can be updated with improved data and utilized

in other countries within the GMS. It is envisaged this

information will help direct policy and further TAD control

in Cambodia, and potentially, the GMS. The framework

can be applied to all countries within the GMS and input

data improved upon as future data and research is con-

ducted.

Acknowledgements

The ‘Best practice health and husbandry of cattle, Cambo-

dia’ (AH/2005/086) project was funded by the Australian

Centre for International Agricultural Research, and this

support is acknowledged as many of the results from this

and other recent research were used to develop the frame-

work used in this manuscript. The authors would particu-

larly like to thank the staff from the Department of Animal

Health and Production, Phnom Penh, for their diligent

field work and dedication to animal health and production

extension.

References

ADB 2013. Greater Mekong Subregion Overview. Available at:

http://www.adb.org/countries/gms/overview (accessed August

7, 2014)

Arzt, J., B. Baxt, M. J. Grubman, T. Jackson, N. Juleff, J. Rhyan,

E. Rieder, R. Waters, and L. L. Rodriguez, 2011: The patho-

genesis of foot-and-mouth disease II: viral pathways in swine,

small ruminants, and wildlife; myotropism, chronic syn-

dromes, and molecular virus-host interactions. Transbound.

Emerg. Dis. 58, 305–26.

Barasa, M., A. Catley, D. Machuchu, H. Laqua, E. Puot, D. Tap

Kot, and D. Ikiror, 2008: Foot-and-mouth disease vaccination

in south Sudan: benefit–cost analysis and livelihoods impact.

Transbound. Emerg. Dis. 55, 339–351.

Cleland, P. C., P. Chamnanpood, F. C. Baldock, and L. J. Glee-

son, 1996: Village level risk factors for foot-and-mouth disease

in northern Thailand. Prev. Vet. Med. 26, 253–261.

Cocks, P., R. Abila, A. Bouchot, C. Benigno, S. Morzaria, P.

Inthavong, V. L. Nguyen, N. Bourgeois-Luthi, A. Scoizet, and

S. Sieng, 2009: FAO ADB and OIE SEAFMD Study on Cross

Border movement and market chains of large ruminants and

pigs in the Greater Mekong SubRegion. Bangkok, Thailand,

October 2009.

Coker, R. C., B. M. Hunter, J. W. Rudge, M. Liverani, and P.

Hanvoravongchai, 2011: Emerging infectious diseases in

southeast Asia: regional challenges to control. Lancet 377,

599–609.

Darley, J. M., and B. Latan�e, 1968: Bystander intervention in

emergencies: diffusion of responsibility. J. Pers. Soc. Psychol. 8,

377–383.

Dijkhuizen, A. A., and R. S. Morris, 1997: Animal Health Eco-

nomics Principles and Applications. Post Graduate Founda-

tion in Veterinary Science, University of Sydney, PO Box

A985, Sydney South 1235, Australia.

FAO, 2013: Lessons from HPAI – A technical stocktaking of out-

puts, outcomes, best practices and lessons learned from the

fight against highly pathogenic avian influenza in Asia 2005–

2011. FAO Animal Production and Health Paper No. 176.

FAO, Rome, Italy.

Ferris, N. P., A. I. Donaldson, R. M. Shrestha, and R. P. Kit-

ching, 1992: A review of foot and mouth disease in Nepal.

Rev. Sci. Tech. 11, 685–698.

Fonnan, A. J., 1991: The field control of foot and mouth disease.

OIE-FAVA Symposium on the Control of Major Livestock

Diseases in Asia, Pattaya, Thailand, 8–9 November 1990, OIE,

Paris, pp. I74–I182.

Ghanem, M. M., and O. M. Abdel-Hamid, 2010: Clinical, hae-

matological and biochemical alterations in heat intolerance

(panting) syndrome in Egyptian cattle following natural foot-

and-mouth disease (FMD). Trop. Anim. Health Prod. 42,

1167–73.

Henry, L. A., and R. D. Bush, 2013: Understanding the market

chain: trader survey and determining animal values. In:

Young, J., L. Rast, S. Suon, and P. Windsor (eds), Cattle

Health, Production and Trade in Cambodia, p. 151. Proceed-

ings from three ACIAR-funded projects presented at an inter-

national workshop held in Phnom Penh, Cambodia, 7–8 June

2011. ACIAR Proceedings No. 138. Australian Centre for

International Agricultural Research, Canberra.

Kawasaki, M., J. R. Young, S. Suon, R. D. Bush, and P. A. Wind-

sor, 2013: The socioeconomic impacts of Haemorrhagic Septi-

caemia on smallholder large ruminant farmers in Cambodia.

Transbound. Emerg. Dis. doi: 10.1111/tbed.12174.

Kerr, J., S. Sieng, and A. Scoize, 2012: Working with traders to

understand livestock movements and spread of animal disease

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 13

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia

in Cambodia and Lao PDR. In: Adam, L. B., G. D. Gray, and

G. Murray (eds), Animal Biosecurity in the Mekong: Future

Directions for Research and Development, Siem Reap Cam-

bodia, ACIAR Proceeding 137, 59–64.

Kerr, J., N. Bourheois Luthi, S. Sieng, P. Inthavong, B. Madin,

and C. Hawkins 2013: Investigating trans-boundary livestock

trade and associated disease risk in the Greater Mekong

Subregion. In: Young, J., L. Rast, S. Suon, and P. Windsor

(eds), Cattle Health, Production and Trade in Cambodia, p.

151. Proceedings from three ACIAR-funded projects pre-

sented at an international workshop held in Phnom Penh,

Cambodia, 7–8 June 2011. ACIAR Proceedings No. 138.

Australian Centre for International Agricultural Research,

Canberra.

Kitching, R. P., 2002: Clinical variation in foot and mouth dis-

ease: cattle. Rev. Sci. Tech. 21, 499–504.

Kitching, R. P., A. M. Hutber, and M. V. Thrusfield, 2005: A

review of foot-and-mouth disease with special consideration

for the clinical and epidemiological factors relevant to predic-

tive modelling of the disease. Vet. J. 169, 197–209.

Knight-Jones, T. J. D., and J. Rushton, 2013: The economic

impacts of foot and mouth disease –What are they, how big

are they and where do they occur? Prev. Vet. Med. 112, 161–

173.

Knowles, N. J., J. He, Y. Shang, J. Wadsworth, B. Valdazo Gonz-

alez, H. Onosato, K. Fukai, K. Morioka, K. Yoshida, I. Cho, S.

Kim, J. Park, K. Lee, G. Luk, V. Borisov, A. Scherbakov, A. Ti-

mina, D. Bold, T. Nguyen, D. J. Paton, J. M. Hammond, X.

Liu, and D. P. King, 2012: Southeast Asian foot-and-mouth

disease viruses in eastern Asia. Emerg Infect Dis [serial on the

Internet]. Available at http://dx.doi.org/10.3201/eid1803.

110908 (accessed March 5, 2012)

Madin, B., 2011: An evaluation of Foot-and-Mouth Disease out-

break reporting in mainland South-East Asia from 2000 to

2010. Prev. Vet. Med. 102, 230–241.

MAFF, 2011: Ministry of Agriculture Forestry and Fisheries

Annual Report 2010–2011 and Work Plan 2011–2012. Minis-

try of Agriculture, Forestry and Fisheries, Phnom Penh.

MAFF (Ministry of Agriculture Forestry and Fisheries), 2012:

Ministry of Agriculture Forestry and Fisheries Annual Report

2011–2012 and Work Plan 2012–2013. Ministry of Agricul-

ture, Forestry and Fisheries, Phnom Penh, Cambodia, p. 110.

Maqsood, M. S., M. Ishaq, and M. Anwar, 1958: A heat intoler-

ance syndrome in cattle following an attack of foot-and-

mouth disease. Vet. Rec. 70, 299–300.

Minett, F. C., 1948: Panting in cattle; a sequel to foot and mouth

disease. J. Am. Vet. Med. Assoc. 113, 545–550.

MoP (Ministry of Planning), 2009: General Population Census

of Cambodia 2008 – National Report on Final Census Results.

National Institute of Statistics, Ministry of Planning Phnom

Penh, Cambodia.

Nampanya, S., S. Suon, L. Rast, and P. A. Windsor, 2012:

Improvement in smallholder farmer knowledge of cattle pro-

duction, health and biosecurity in southern Cambodia

between 2008 and 2010. Transbound. Emerg. Dis. 59, 117–127.

Nampanya, S., S. Khounsy, A. Phonvisay, J. R. Young, R. D.

Bush, and P. A. Windsor, 2013: Financial impact of Foot and

Mouth Disease on large ruminant smallholder farmers in the

Upper Mekong region. Transbound. Emerg. Dis. doi: 10.1111/

tbed.12183.

National Bank of Cambodia, 2012. Annual Report 2011, General

Directorate of Supervision. Phnom Penh, April 4, 2012.

O’Connell, J., J. R. Young, L. A. Henry, L. Rast, S. Suon, R. D.

Bush, and P. A. Windsor 2013. Assessment of current trends

in smallholder cattle trade in Cambodia. In: Young, J. R., L.

Rast, S. Suon, and P. A. Windsor (eds), Cattle Health, Pro-

duction and Trade in Cambodia. Proceedings from Three AC-

IAR- Funded Projects Presented at an International

Workshop Held in Phnom Penh, Cambodia, 7–8 June 2011.

ACIAR Proceedings No. 138. Australian Centre for Interna-

tional Agricultural Research, Canberra.

OIE Sub-Regional Representation for South-East Asia, 2011:

South East Asia and China Foot and Mouth Disease campaign

(SEACFMD) 2020~a roadmap to prevent, control and eradi-

cate foot and mouth disease freedom (by2020) in South-East

Asia and China, 2nd edn. Available at: http://www.srr-se-

asia-oie.com/uploads/tx_oiedownload/SEAFMD_2020.pdf

(accessed May 25, 2013).

Palisade, 2014: Software for risk and decision analysis. Available

at: www.palisade.com (accessed November 1, 2014).

Perry, B. D., L. J. Gleeson, S. Khounsy, P. Bounma, and S. D.

Blacksell, 2002: The dynamics and impact of foot and mouth

disease in smallholder farming systems in South-East Asia: a

case study in Laos. Rev. Sci. Tech. 21, 663–673.Roeder, P. L., and W. P. Taylor, 2007: Mass vaccination and

herd immunity: cattle and buffalo. Rev. Sci. Tech. 26, 253–263.

Rosenberg, F. J., V. M. Astudillo, and R. M. Goic, 1980: Regional

strategies for the control of foot and mouth disease: an eco-

logical outlook. In: Proc. of the 2nd Int. Symp. on Veterinary

Epidemiology and Economics, Australian Government Print-

ing Service, Canberra, pp. 587–596.

Rweyemamu, M. M., 1984: Foot and mouth disease control

strategies in Africa. Prev. Vet. Med. 2, 329–340.

Shankar, B., S. Morzarina, A. Fiorucci, and M. Hak, 2012: Ani-

mal disease and livestock-keeper livelihoods in Southern

Cambodia. Int. Dev. Plann. Rev. 34, 39–64.

Stratton, J. 2013. Village animal health workers as a link between

the government and farmers in?foot-and-mouth disease con-

trol in Cambodia. In: Young, J. R., L. Rast, S. Suon, and P. A.

Windsor (eds), Cattle Health, Production and Trade in Cam-

bodia. Proceedings from three ACIAR- funded projects pre-

sented at an international workshop held in Phnom Penh,

Cambodia, 7–8 June 2011. ACIAR Proceedings No. 138. Aus-

tralian Centre for International Agricultural Research, Canb-

erra.

Subash, C.V., M.K. Mahajan, G. Malik, and J.P. Dahiya, 2004:

An epidemiological study on Bovine haemorrhagic septicae-

mia in Haryana. Indian J. Anim. Res. 38, 14–19.

Suon, S., J. R. Young, and P. A. Windsor. 2013. Livestock infec-

tious disease status in Cambodia. In: Young, J. R., L. Rast, S.

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases.14

Benefit-Cost Analysis of FMD Control in Cambodia J. R. Young et al.

Suon, and P. A. Windsor (eds), Cattle Health, Production and

Trade in Cambodia. Proceedings from three ACIAR- funded

projects presented at an international workshop held in

Phnom Penh, Cambodia, 7–8 June 2011. ACIAR Proceedings

No. 138. Australian Centre for International Agricultural

Research, Canberra.

The World Bank, 2013: Data: Cambodia, The World Bank

Group. Available at http://data.worldbank.org/country/cam-

bodia (accessed June 5, 2013).

Trading Economics, 2014. Cambodia interest rate. Available at

http://www.tradingeconomics.com/cambodia/interest-rate

(accessed August 7, 2014).

Tum, S., C. Sar, and D. Bourn, 2011: Policy implications of Foot

and Mouth Disease in Cambodia (poster presentation). Pro-

ceedings of the International Conference on Animal Health

Surveillance. Epid�emiol. Sant�e Anim. 59–60, 441–443.

UNPD 2014. Human Development Report 2014 Sustaining

Human Progress: Reducing vulnerabilities and building resil-

ience. Available at: http://hdr.undp.org/en/content/

human-development-report-2014 (accessed August 7, 2014).

Vergne, T., V. Grosbois, B. Durand, F. Goutard, C. Bellet, D.

Holl, F. Roger, and B. Dufour, 2012: A capture-recapture

analysis in a challenging environment: assessing the epidemio-

logical situation of foot-and-mouth disease in Cambodia.

Prev. Vet. Med. 105, 235–243.

Windsor, P. A., 2011: Perspectives on Australian animal health

aid projects. Editor invited paper. Transbound. Emerg. Dis. 58,

375–386.

Windsor, P. A., P. G. Freeman, R. Abila, C. Benigno, V. Nim,

and A. Cameron, 2011: Control and eradication of Foot and

Mouth Disease in the Bicol surveillance buffer zone in the

Philippines. Transbound Emerg. Dis. 58, 421–433.

Young, J. R., L. Rast, S. Suon, R. D. Bush, L. A. Henry, and P. A.

Windsor, 2013a: The impact of best practice health and hus-

bandry interventions on smallholder cattle productivity in

southern Cambodia. Anim. Prod. Sci.. Available at http://dx.

doi.org/10.1071/AN13033 (accessed November 24, 2013).

Young, J. R., S. Suon, C. J. Andrews, L. A. Henry, and P. A.

Windsor, 2013b: Assessment of financial impact of foot and

mouth disease on smallholder cattle farmers in Southern

Cambodia. Transbound. Emerg. Dis. 60, 166–174.

Young, J. R., S. Evans-Kocinski, R. D. Bush, and P. A. Windsor,

2013c: Improving smallholder farmer biosecurity in the Me-

kong region through change management. Transbound.

Emerg. Dis. doi: 10.1111/tbed.12181.

Young, J. R., R. A. O’Reilly, K. Ashley, S. Suon, V. I. Leoung, P.

A. Windsor, and R. D. Bush, 2014: Impacts on rural liveli-

hoods in Cambodia following adoption of best practice health

and husbandry interventions by smallholder cattle farmers.

Transbound. Emerg. Dis. 61(Suppl. 1), 1–14.

© 2014 Blackwell Verlag GmbH • Transboundary and Emerging Diseases. 15

J. R. Young et al. Benefit-Cost Analysis of FMD Control in Cambodia