PA

EJa

b

c

a

ARR1A

KNAPIVT

0d

Preventive Veterinary Medicine 104 (2012) 301– 308

Contents lists available at SciVerse ScienceDirect

Preventive Veterinary Medicine

j ourna l ho me pag e: ww w.elsev i er .com/ locate /prev etmed

revalence and incidence of Newcastle disease and prevalence ofvian Influenza infection of scavenging village chickens in Timor-Lesté

duardo Serrãoa, Joanne Meersa, Robert Pyma, Richard Coplandb, Debbie Eaglesc,oerg Henninga,∗

School of Veterinary Science, University of Queensland, Gatton 4343, Queensland, AustraliaSchool of Agriculture and Food Sciences, University of Queensland, Gatton 4343, Queensland, AustraliaCSIRO Australian Animal Health Laboratory, Private Bag 24, Geelong, VIC 3220, Australia

r t i c l e i n f o

rticle history:eceived 22 April 2011eceived in revised form5 December 2011ccepted 29 December 2011

eywords:ewcastle diseasevian Influenzarevalencencidenceillage chickenimor-Lesté

a b s t r a c t

A longitudinal study to monitor prevalence and incidence of antibodies against Newcas-tle disease (ND) virus and prevalence of antibodies against Avian Influenza (AI) virus inscavenging village chickens was conducted in 20 villages within 4 districts of Timor-Lesté.A total of 3600 blood samples was collected from 1674 individual birds in 300 householdchicken flocks during three sampling periods (December 2008–February 2009, March–May2009, and June–August 2009). The mean interval between household visits was 101.6 ± 1.9days. None of the birds enrolled in the study was vaccinated against ND or AI. A haemag-glutination inhibition (HI) test was used to determine antibody titres against ND virus anda competitive ELISA and HI tests were used to detect antibody against AI virus.

The bird-level ND seroprevalence pooled across all samplings (adjusted for clusteringby households) was 4.4% (95% CI 3.5–5.2). The bird-level ND seroprevalence in each of thethree sampling periods (adjusted for clustering by household) was 3.0% (95% CI 2.0–4.0),6.6% (95% CI 5.1–8.0) and 3.6 (95% CI 2.5–4.6), respectively. A total of 12.6% individual birdstested ND seropositive at least once over the total study period (95% CI 10.5–14.7).

The flock-level ND seroprevalence (at least one bird tested had antibodies against NDvirus) pooled across all samplings was 15.9% (95% CI 13.5–18.3). A total of 35.3% flocks hada minimum of one bird being ND seropositive at least once over the study period.

The bird-level incidence rate for the period between the first and the second samplingand between the second and the third sampling was 5.6 (95% CI 4.1–7.5) and 0.5 (95% CI0.5–3.8) per 10,000 bird-years-at-risk, respectively.

A total of 1134 serum samples from the last sampling period between June and August2009 was tested for antibodies against AI virus. Only 4 samples tested Influenza A posi-tive, indicating a bird-level seroprevalence level for Influenza A of 0.4% (CI 0.0–0.7%). TheseInfluenza A positive samples were further tested for HI antibodies against AI virus subtypesof H5N1, H5N3, H7N3 and H9N2, but all tested negative, suggesting that the influenza anti-bodies in those four birds resulted from exposure to low pathogenic AI viruses of different

H subtypes.

Our results indicate thawas a higher risk of infecprior or subsequent to thiAI viruses was detected in

∗ Corresponding author. Tel.: +61 7 5460 1839.E-mail address: j.henning@uq.edu.au (J. Henning).

167-5877/$ – see front matter © 2012 Elsevier B.V. All rights reserved.oi:10.1016/j.prevetmed.2011.12.018

t village chickens in Timor-Lesté are exposed to ND virus; theretion during the early months of 2009 than either immediatelys. No evidence of infection of village chickens with H5, H7 or H9

this study.© 2012 Elsevier B.V. All rights reserved.

erinary

302 E. Serrão et al. / Preventive Vet

1. Introduction

Timor-Lesté (East Timor) is young developing countryin South-east Asia that restored its independence only in2002. Scavenging village chickens (Manu Timor) are oneof the economically and socially most important types oflivestock in Timor-Lesté. According to the Ministry of Agri-culture of Timor-Lesté (MAP, 2005) approximately 670,000village chickens were kept in the country in 2005. Villagechickens mostly scavenge for their feed around householdyards and are usually not provided with any veterinary care(Menge et al., 2005). Disease transmission between flocksand attacks by predators are common in the village envi-ronment (Mwalusanya, 1998). Anecdotally, flock ownersin Timor-Lesté indicated losses of individual birds or thewhole flock during certain periods of the year, however,detailed information on village chicken health in Timor-Lesté was not available before the commencement of thisstudy.

Newcastle disease (ND) is considered to be one of themajor causes of mortality of village chickens and is animportant factor inhibiting improvement of village chickenproduction in developing countries (Norby, 1989; Aini,1990; Yongolo, 1996; Spradbrow, 2005; Henning et al.,2009), but the ND situation in Timor-Lesté is unknown.Highly Pathogenic Avian Influenza (HPAI) caused by H5and H7 subtypes of Avian Influenza (AI) virus is anotherdevastating poultry disease and is of economic significanceto poultry industries worldwide (Easterday et al., 1997).HPAI caused by H5N1 virus is endemic in the neighbouringcountry of Indonesia, where almost all provinces have beendeclared as infected, including West Timor (Nusa TenggaraTimur), which shares a border with Timor-Lesté (Samaan,2007). Although no HPAI outbreaks have been reported inTimor-Lesté to date, the HPAI infection status of birds in thecountry had not been reported. Therefore we conducted alongitudinal serological survey to estimate prevalence andincidence of ND and prevalence of AI infection in scaveng-ing village chickens in Timor-Lesté.

2. Materials and methods

2.1. Study area and selection of households

Because village chicken production and ND exposure areknown to vary by altitude and agro-ecological zone, respec-tively (Dessie et al., 2010; Dessie and Ogle, 2001; Zelekeet al., 2005), we based our study in areas of varying cli-mate, vegetation and altitude. Of the total of 13 districtsin Timor-Lesté, we selected four districts (Liquica, Aileu,Manatuto and Lautem) (Fig. 1) covering a wide range ofdifferent environments and altitudes that might influencethe management of village chickens and possibly exposureof birds to ND and AI viruses. Some characteristics of eachdistrict are presented in Table 1.

For each district, a sampling frame containing all villagenames within the district was obtained from the National

Statistics Office in Timor-Lesté. Within each district, fivevillages were selected using random numbers generated inMicrosoft Excel 2003 (Microsoft Corporation); this resultedin 20 villages being enrolled in the study. The village head

Medicine 104 (2012) 301– 308

man of each village was then visited and a list of all sub-villages was obtained (a Timor-Lesté village has between2 and 32 sub-villages). Two sub-villages were randomlyselected from each village by drawing numbers from ahat. If either of these sub-villages proved to be difficult orimpossible to access by vehicle (i.e. it is well recognisedthat some sub-villages are inaccessible because of lack ofa road connection or following flooding in the rainy sea-son), they were excluded from the study and a replacementsub-village was randomly selected as above. Of the 40 sub-villages initially selected, only 6 had to be excluded on thebasis of inaccessibility and were replaced. Village head menkeep a demographic record books with all household own-ers listed on a sub-village level and these lists were used asa sampling frame.

Sample size calculations were conducted to estimatebird-level prevalence of ND antibodies in unvaccinatedvillage chicken flocks. Sample size was calculated basedon two-stage sampling, with household flocks selected atthe first stage and chickens selected at the second stageusing the two-stage prevalence survey option in the SurveyToolbox software program (Cameron, 1999). The follow-ing parameters were used in the sample size calculations:(a) as no previous serological study has been conductedin Timor Lesté, the estimated prevalence of ND antibod-ies among village chicken flocks was assumed to be 50%;(b) the within-flock variance was set to be 0.10, consid-ering that ND spreads quickly within unvaccinated flocks(Spradbrow, 2000); (c) the between-flock variance was setto 0.20, based on observations of large variation of ND anti-body titres between regions (Alders et al., 1994); (d) theprecision was set to 5% and the confidence level was set to95%; (e) the average number of households per sub-villageacross the 40 selected sub-villages was 100 (estimatedfrom the sampling frame); (f) the total number of house-holds in the 40 selected sub-villages was 4000 (estimatedfrom the sampling frame); (g) the cost for sampling onehousehold and one chicken was estimated to be 2 and 0.05,respectively. In total, 307 study households were requiredwith four chickens per household to be sampled.

For each of the 40 selected sub-villages a pool of 30household owners was also randomly selected from thedemographic record book using random numbers gener-ated in Microsoft Excel 2003. These household owners werevisited sequentially to confirm criteria for their inclusion inthe study (a household had to have at least four chickensand all birds had not to be vaccinated against ND). If lessthan four chickens were present in a household or if birdshad been vaccinated against ND the next household ownerwas chosen from the pool of 30 household owners. Follow-ing this procedure a total of 7–8 eligible households wereenrolled from each sub-village, totalling 15 households pervillage (or 300 households across all 20 villages) that wereparticipating in the study.

2.2. Data collection

Households enrolled in the study were visited on threesampling occasions by an animal scientist accompanied bytwo animal health workers. The first sampling was con-ducted between December 2008 and February 2009, the

E. Serrão et al. / Preventive Veterinary Medicine 104 (2012) 301– 308 303

Fig. 1. Map of Indonesia and Timor Lesté (indicated in red) (top figure) and map of districts of Timor Lesté and locations of 300 village chicken households(red dots) (bottom figure) involved in a survey for antibodies against Newcastle disease (ND) virus in unvaccinated village chickens during three samplingperiods between December 2008 and August 2009. (For interpretation of the references to color in this figure legend, the reader is referred to the webv

ssihtsev

ersion of this article.)

econd sampling from March until May 2009 and the thirdampling from June 2009 until August 2009, thereby ensur-ng that the interval in days was the same for each house-old visit. At each visit an interview was conducted with

he flock owner and blood samples were collected fromelected village chickens. The duration of data collection inach household ranged between 30 and 45 min. For everyisit each household owner received a compensation of

2 US$ for making birds available for sample collections andfor their participation in the interviews.

2.3. Interviews

Two types of questionnaires were used, a baselinequestionnaire for the first sampling and a follow-up ques-tionnaire for subsequent samplings. These questionnaires

304 E. Serrão et al. / Preventive Veterinary Medicine 104 (2012) 301– 308

Table 1Characteristics of districts in Timor-Lesté, where the exposure status of village chickens to Newcastle disease and Avian Influenza virus was monitored in2008/2009.

Districts Range of altitudes (m)a Number ofvillagesb

Number ofhouseholdsb

Number of villagechickens in 2001c

Aileu 350–1500 37 7745 16,535Manatuto No data 29 8338 21,328Liquica 0–1266 23 11,063 97,514Lautem 0–1297 34 12,998 36,079

Timor (2

a Ministry of State Administration and Territorial Management of East

b The National Statistics Directorate, Timor-Lesté (2004).c Cruz and Da (2003).

recorded household details and information on the feeding,housing, productivity, health and trade of chickens (datanot shown).

2.4. Sample collection

At the first visit, 4 birds were randomly selected (2adults and 2 growers) as the birds the flock owner was firstable to catch. Replacement birds were selected at subse-quent visits following the same procedure. Birds were wingtagged with numbered aluminium tags (type 893 for adultbirds and type 898 for grower birds, National Band and TagCo, USA) using an applicator. Blood samples were collectedfrom the brachial (wing) vein with a 2.5 ml syringe and a25 gauge needle from adult birds and with a 1 ml syringeand a 23 gauge needle from growers.

2.5. Sample processing and testing

The blood samples were allowed to clot in the syringesand the serum was removed within 1–3 h of sample col-lection. Serum from the syringe was poured into a 1.5 mlmicrocentrifuge tube and centrifuged in a mini centrifugeat 6000 rpm for 15 s for further separation of the remainingred blood cells (Young et al., 2002). Serum was then storedat 4 ◦C until testing.

The ND antibody testing was conducted in Timor Lesté.The haemagglutination (HA) test was carried out to preparea suspension containing 4 HA units of ND virus antigen (V4strain)/25 �L, which were then used in the haemaggluti-nation inhibition (HI) test to determine ND antibody titre(Allan and Gough, 1974). A suspension of 1% chicken redblood cells (RBC) was used for both the HI and HA tests(Allan and Gough, 1974). Birds with log 2-transformed HItitres ≥3 were considered to be antibody positive (seropos-itive) (Allan and Gough, 1974). If at least one bird in thehousehold flock sample tested ND seropositive, the flockwas considered to be ND seropositive.

Serum samples from the last sampling were tested inthe Australian Animal Health Laboratory (AAHL), Geelong,Australia for antibodies against Avian Influenza A virusand further sub-types of the AI virus. The presence ofAvian Influenza A antibodies was tested using a compet-itive enzyme-linked immunosorbent assay (C-ELISA) and

the presence of antibodies against sub-types of AI viruswas tested using HI tests (OIE, 2004). The antigens used inthe HI tests were derived from H5N1 virus from Vietnam(A/chicken/Vietnam/8/2004), H5N1 virus from Indonesia

002).

(A/chicken/Konawe Selatan/BBVM204/2005), H9N2 virusfrom Malacca (A/chicken/Malacca/4905/2003), H5N3 virusfrom Australia (A/duck/Victoria/1462/2008) and H7N3virus from Australia (A/chicken/Victoria/224/1992).

For the Influenza A ELISA, the presence of antibodies inthe serum was detected by the binding of a monoclonalantibody against a conserved epitope on the nucleoproteinof a purified A/Pr/8/34 Influenza A virus antigen. The com-petitive binding between the two antibodies is determinedas percent inhibition (PI). PI values greater than 70% areconsidered positive, values less than 40% are negative andvalues of 40–60% are considered inconclusive.

2.6. Data management and statistical analysis

Serology results were compiled in Microsoft Excel 2007(© 2006 Microsoft Corporation, USA) and summarised in aMicrosoft Access database 2007 (© 2006 Microsoft Corpo-ration, USA). Prevalence estimates for antibodies againstND were calculated in STATA Intercooled 11 (StatisticalSoftware, © 2009 Stata Corporation LP, USA) using the–prop- command. Standard errors for bird-level estimateswere adjusted for the clustering by household flock usingthe –vce- command in STATA Intercooled 11. For flock-levelprevalence estimations, a flock was considered positivewhen at least one bird tested in the household had anti-bodies against ND.

Bird-level incidence rates, adjusted for birds leavingand entering the study population in-between samplings(Dohoo et al., 2009), were based on 10,000-bird-days atrisk. The 95% approximate confidence intervals for theincidence rates were calculated in WinPepi Version 10.5(Abramson, 2004); computer programs for epidemiologist.

3. Results

3.1. Bird-level prevalence for antibodies against ND

During the initial sampling, 1200 village chickens wereselected, but between samplings some birds lost their wingtags and replacement birds had to be selected at subse-quent samplings. A total of 369 and 105 replacement birdswere selected during the second sampling and the thirdsampling, respectively. Therefore a total of 1674 village

chickens from 300 households was monitored over thestudy period from December 2008 to August 2009.

Based on the questionnaire data collected, reasons forthe bird replacements were further investigated. A total

E. Serrão et al. / Preventive Veterinary Medicine 104 (2012) 301– 308 305

Fig. 2. Bird level prevalence (with 95% confidence intervals) for antibodiesagainst Newcastle disease (ND) virus in previously sampled unvacci-nated village chickens (“Marked birds”) and replacement unvaccinatedvillage chickens (“Replacement birds”) in Timor Lesté during three sam-pling periods between December 2008 and August 2009. Birds withlog 2-transformed haemagglutination inhibition antibody titres ≥3 werec

orudhto

oh

h(t(2pb1ib(pn7Ataw(wtt

Fig. 3. Flock level prevalence (with 95% confidence intervals) for antibod-ies against Newcastle disease (ND) virus in unvaccinated village chickensin Timor-Lesté during three sampling periods between December 2008

results. Only 4 samples out of the remaining 1134 sam-ples tested seropositive for Influenza A virus; therefore the

onsidered to be seropositive.

f 65% (308/474) of the missing wing-tagged birds wereeportedly lost to predators, a total of 7% (32/474) died fromnknown causes, 3% (12/474) died from suspect Newcastleisease, 3% (13/474) had been sold or consumed, 1% (5/474)ad been stolen and the remaining 22% (104/474) wing-agged birds were reported as missing for unknown reasonsr had lost their wing tags between visits.

A total of 1200 blood samples was obtained during eachf the three samplings. The mean time interval betweenousehold visits was 101.6 ± 1.9 days.

Bird-level ND seroprevalence (adjusted for clustering byousehold) was 3.0% (95% CI 2.0–4.0) at the first samplingDecember 2008–February 2009), 6.6% (95% CI 5.1–8.0) athe second sampling period (March–May 2009) and 3.6%95% CI 2.5–4.7) at the third sampling period (June–August009). The bird-level seroprevalence pooled across all sam-ling was 4.4% (95% CI 3.5–5.2). From the individual 1674irds monitored from December 2008 to August 2009,2.6% (95% CI 10.5–14.7) tested positive at least once dur-

ng this study period. The bird-level seroprevalence variedetween birds sampled previously and replacement birdsFig. 2). At the second sampling the ND bird-level sero-revalence of the 369 replacement birds (birds that wereot sampled at the first sampling) was 4.9% (CI 2.6–7.1) and.3% (CI 5.5–9.2) for the birds sampled previously (N = 831).t the third sampling the ND bird-level seroprevalence of

he 105 replacement birds (birds that were not sampledt the first or second sampling) was 30.5% (CI 20.3–40.6),hile only 1.0% (CI 0.42–1.59) of previously sampled birds

N = 1095) were ND antibody positive. Fifty of the birds thatere seropositive (log 2-transformed antibody titre ≥3) at

he second sampling were seronegative (log 2 titre <3) athe third sampling.

and August 2009. Birds with log 2-transformed haemagglutination inhi-bition antibody titres ≥3 were considered to be seropositive. A flock wasconsidered positive when at least one bird in the flock was seropositive.

3.2. Flock-level prevalence for antibodies against ND

The mean household flock size was 17.5 birds, rang-ing from 4 to 105 birds. Flock-level ND seroprevalenceat the first, second and third sampling was 11.3% (95%C.I. 7.7–14.9), 23.0% (95% C.I. 18.2–27.8), and 13.3% (95%C.I. 9.5–17.2), respectively (Fig. 3). The flock-level sero-prevalence pooled across all sampling was 15.9% (95% CI13.5–18.3).

From the individual 300 flocks monitored fromDecember 2008 to August 2009 a total of 106 flocks wastested at least once positive during this study period indi-cating a flock-level period seroprevalence of 35.3% (95% C.I.29.9–40.7).

3.3. Bird-level incidence rate for antibodies against ND

Out of 808 seronegative birds at the first sampling, atotal of 45 birds seroconverted by the second sampling,resulting in an incidence rate of 5.6 (95% C.I. 4.1–7.5) per10,000 bird-years-at-risk for this period. Of 1037 seronega-tive at the second sampling, 5 birds seroconverted betweenthen and the third sampling, resulting in an incidence rateof 0.5 (95% C.I. 0.5–3.6) per 10,000 bird-years-at-risk forthis period.

3.4. Bird-level prevalence for antibodies against AvianInfluenza virus

Of the 1200 serum samples from the last sampling(June–August 2009), 66 had insufficient amounts of serumfor testing in the Influenza A ELISA or provided inconclusive

bird-level seroprevalence for Influenza A was 0.4% (95%C.I. 0.0–0.7). The positive Influenza A samples came from 4

erinary

306 E. Serrão et al. / Preventive Vet

different farms that were located within 3 study districts(Aileu, Liquica and Lautem). These 4 serum samples werefurther tested in the HI test using antigen against H5N1(Vietnamese clade 1), H5N1 (Indonesian clade 2), H5N3 (Aus-tralian clade), H7N3 (Australian clade) and H9N2 (Malaccaclade) antigens, but all tested negative.

4. Discussion

Our results indicate that only a small proportion of vil-lage chickens in Timor-Lesté had protective antibody titresagainst Newcastle disease virus. Bird-level seroprevalenceranged between 3 and 7% at the three sampling occa-sions in our study – this is low in comparison to bird-levelseroprevalence estimates from other countries. Henninget al. (2008) reported from Myanmar that bird-level sero-prevalence (log 2-transformed titre ≥3) in village chickensranged from 26 to 52% between sampling months. Theseroprevalence was also low compared to a survey carriedout by Biswas et al. (2009) during the period of January2002–March 2006 in village chickens in Bangladesh wherean ND seroprevalence of 88% was reported. Our resultsare also dissimilar to estimates from Nigeria and Ethiopiawhere seroprevalences of 15% and 19.8%, respectively havebeen reported (Abubakar et al., 2008; Zeleke et al., 2005).Also the flock-level seroprevalence of 15.9% shown in thisstudy is low in comparison to 89% ND flock-level sero-prevalence in village chickens estimated for Bangladeshover a period from January 2002 to March 2006 (Biswaset al., 2009). The low level of exposure to the ND fieldviruses in our study compared to the results reported fromMyanmar, Bangladesh, Nigeria and Ethiopia might be dueto the relative isolation of villages in Timor Lesté. Only 4of the 20 study villages were located near main roads, andthere would likely be few opportunities for contact of studybirds with other scavenging poultry (which could poten-tially be infected) or for frequent visits by poultry traders(who could potentially spread ND virus) in the majorityof study villages. Alternatively the lack of widespread andwell-established poultry markets in Timor-Lesté and thelack of sufficient disposable farmer income to permit newbird purchases might have resulted in few additions ofnew poultry to the household flocks. Thus compared toother countries fewer opportunities for the introduction ofnew birds into household flocks might exist in Timor-Lesté,which is considered a major cause of ND virus infection ofnon-infected flocks (Alexander, 1988). About 28% (N = 474)of birds had to be replaced and a large proportion of these(65%) was lost due to predation. The possibility that someof them might have been infected with ND virus cannot beexcluded. Weakened from the disease, these birds mighthave not been able to save themselves from predation.

It would have been impossible to sample villages inthe whole country, in terms of time and human resourcesavailable and security concerns in certain areas. Thereforewe used the approach of selecting representative districts

within the country and then randomly selecting villagesand sub-villages. As only 6 inaccessible sub-villages had toreplaced, the possible bias related to these replacementscan be considered as small.

Medicine 104 (2012) 301– 308

Exposure to ND field virus in our study was highest dur-ing the period between the first and the second sampling,hence from the months of December throughout Februaryto the months of March throughout May. The rainy seasonstarts in Timor-Lesté in the month of December and lastsuntil the month of March, with the average rainfall rangingbetween 150 mm and 350 mm during this period (Barnettet al., 2003). During the rainy season, birds might be eithermore frequently exposed to ND field virus or more suscep-tible to ND infection due to stress associated with exposureto these extreme conditions. Considering that the concen-tration of ND virus is maintained in soil for a longer periodof time under 100% relative humidity compared to 0–15%humidity (Boyd and Hanson, 1958), it is likely that virussurvival in our study villages was higher during the rainyseason than at other times in the year and thus a possiblelikelihood of exposure of chickens was greater.

Seroprevalence of marked birds decreased at the thirdsampling. Although antibodies to ND following challengewith virulent virus can persist in some birds up to 11months (Fontanilla et al., 1994), as with other acute viralinfections ND antibody titres will gradually decline overtime in the absence of continued exposure to the virus. Thetime of the exposure to ND virus in our seropositive studybirds is unknown. Thus, the 50 birds that were seroposi-tive at the second sampling could have been exposed to NDvirus many months prior and were finally falling below thecut-off (log 2-transformed titre ≥3) at the third sampling. Inaddition, the possibility that some of the seropositive birdsat the second sampling were sold, consumed or lost theirwing tags cannot be excluded. In fact 25 birds seropositiveat the second sampling were replaced at the third sampling.It is possible that these seropositive birds were more likelyto be older birds or poor-performing birds that the ownerspreferred to sell or consume.

The HI test cannot distinguish between antibodiesdeveloped against velogenic, mesogenic or lentogenic NDviruses. Thus it is possible that the seropositive birds in thestudy had been exposed to mildly pathogenic ND virusesrather than velogenic viruses. In order to characterisethe viruses responsible for the antibody responses, virusisolation and molecular typing would be required. Unfor-tunately, facilities for these procedures were not availablein Timor Lesté.

No infection with HPAI subtypes H5, H7 or H9 was foundin this study. Of the 1134 serum samples tested, only foursamples were positive for Avian Influenza A, providing abird-level AI seroprevalence of only 0.4%. All further testsusing antigens of the H5N1 (Vietnamese and Indonesianclades), H5N3 (Australian clade), H7N3 (Australian clade)and H9N2 (Malacca clade) antigens were negative, suggest-ing that the influenza antibodies in those four birds resultedfrom exposure to low pathogenic AI viruses of different Hsubtypes. It is surprising that there is no HPAI in Timor-Lesté while H5N1 HPAI is endemic in the neighbouringcountry of Indonesia. Almost all provinces of Indonesia areinfected including the Timor Lesté neighbouring province

of West Timor (Samaan, 2007). Indonesia has the highestincidence of HPAI in the world in poultry and humans andreported the deaths of a total of 141 people from H5N1infection up until February 2011 (WHO, 2011). However,

erinary

tA

TbrK2wiwbi

itlctinaato(wi

mrbswhlHTal

A

AIAoWtDTo(cDIca

E. Serrão et al. / Preventive Vet

he possibility that the four samples tested positive forvian Influenza A are false positives cannot be excluded.

There is a constant danger that HPAI might spread intoimor-Lesté, because Timor-Lesté and West-Timor share aorder stretching 228 km in length. In fact, in 2006 it waseported that thousands of chickens died from HPAI nearupang, the capital city of West Timor-Indonesia (IRIN,008) which is only few hours by road from the borderith Timor Lesté. Since there is no physical border and lim-

ted security patrols between the two countries, and withidespread smuggling of chickens across the border, the

order is unquestionably a prospective portal for HPAI virusntroduction into Timor-Lesté.

Nevertheless, the HPAI control/prevention programmplemented in Timor-Lesté by the Ministry of Agricul-ure and Fisheries since 2007 (including a ban on imports ofive chickens, eggs and other related products from affectedountries and a public awareness campaign) is most likelyhe reason that no HPAI outbreaks have yet been reportedn Timor-Lesté. Another possible reason is that there areo commercial broiler and layer operations in Timor Lesténd all village chicken flocks are small. This is a reason-ble assumption because in intensive poultry operations,he stocking density is a factor that allows rapid spreadf AI viruses and provides great potential for an epidemicLeibler et al., 2009). This might also apply to ND viruses,hich could also explain the low observed ND prevalence

n our study compared to findings in other countries.Limitations of this study were the interval of almost 2.5

onths between samplings. The mountainous terrain, pooroad conditions and infrastructure in Timor-Lesté, com-ined with shortage of experienced staff to conduct theampling prevented more frequent field visits. However,e ensured that the sampling interval was the same for allouseholds. With the limited funding available, data col-

ection could only be conducted over a period of 9 months.owever we consider that this first ND and AI survey inimor-Lesté has provided valuable information which willssist in strengthening the veterinary services and surveil-ance work in this young country.

cknowledgements

We are grateful for financial support from the Food andgriculture Organisation (FAO), the Australian Centre for

nternational Agricultural Research (ACIAR), the Australiannimal Health Laboratory (AAHL) Geelong and the Schoolf Veterinary Science-University of the Queensland (UQ).e would also like to thank FAO officers in Timor-Lesté,

he field animal health workers of the National Livestockivision-the Ministry of Agriculture and Fisheries (MAF)imor-Lesté, the final-year students from the Departmentf Animal Science, the National University Timor Lorosa’eUNTL) and Dr. Zuhara Bensink, UQ for their invaluableontributions to the project. We are grateful to Dr. Peter

aniels and other AAHL staff for assisting with the Avian

nfluenza HI and ELISA testing. Finally, we extend our sin-ere gratitude to the village and sub-village coordinatorsnd household farmers from 20 villages in Timor-Lesté

Medicine 104 (2012) 301– 308 307

– without their cooperation and contributions, this studywould not have been possible.

References

Abramson, J.H., 2004. WINPEPI (PEPI-for-Windows): computer pro-grams for epidemiologists. Epidemiol. Perspect. Innov. I (6),Retrieved 29 March 2011 from: http://www.epi-perspectives.com/content/pdf/1742-5573-1-6.pdf.

Abubakar, M.B., Ei-Yuguda, A.D., Baba, S.S., 2008. Serological evidence ofinfluenza virus infections in domestic animals and birds in North-Eastern Nigeria. J. Food Agric. Environ. 6 (1), 67–70.

Alexander, D.J., 1988. Methods of spread. In: Alexander, D.J. (Ed.), New-castle Disease. Kluwer Academic Publishers, Boston, pp. 256–272.

Aini, I., 1990. Indigenous chicken production in South East Aisa. WorldPoult. Sci. J. 46, 51–57.

Alders, R.G., Inoue, S., Katongo, J.C., 1994. Prevalence and evaluation ofHitchner B1 and V4 vaccines for the control of Newcastle disease invillage chickens in Zambia. Prev. Vet. Med. 21, 125–132.

Allan, W.H., Gough, R.E., 1974. A standard haemagglutination inhibitiontest for Newcastle Disease, (2) Vaccination and challenge. Vet. Rec. 95,147–149.

Barnett, J., Dessai, S., Jones, R., 2003. Climate Change inTimor-Leste: Science, Policy and Planning. University of Mel-bourne, CSIRO, Australia, Retrieved 29 March 2011 from:http://geography.exeter.ac.uk/SurajeDessai/TimorReport.pdf.

Biswas, P.K., Barua, H., Udin, G.M.N., Biswas, D., Ahad, A., Debnath, N.C.,2009. Serosurvey of five viruses in chickens on smallholdings inBangladesh. Prev. Vet. Med. 88, 67–71.

Boyd, R.J., Hanson, R.P., 1958. Survival of Newcastle disease virus in nature.Avian Dis. 2, 82–93.

Cameron, A., 1999. Survey Toolbox for Livestock Diseases, A PracticalManual and Software Package for Active Surveillance in DevelopingCountries, ACIAR Monograph 54. Australian Centre for InternationalAgricultural Research.

Cruz, C.J., Da, 2003. Livestock development in East Timor. In: Agriculture:New Directions for a New Nation East Timor (Timor-Leste), Proceed-ings of a Workshop 1–3 October 2002. Australian Centre for Interna-tional Agricultural Research, Dili, East Timor, pp. 11–16, Retrieved 29March 2011 from: http://purl.pt/915/1/cd1/ta100/ta101.pdf.

Dessie, T., Ogle, B., 2001. Village poultry production systems in the centralhighlands of Ethiopia. Trop. Anim. Health Prod. 33 (6), 521–537.

Dessie, T., Moges, F., Mellesse, A., 2010. Assessment of village chickenproduction system and evaluation of the productive and reproduc-tive performance of local chicken ecotype in Bure district, North westEthiopia. Afr. J. Agric. Res. 5 (13), 1739–1748.

Dohoo, I., Martin, W., Stryhn, H., 2009. Veterinary Epidemiology Research.AVC Inc, University of Prince Edward Island, Canada.

Easterday, B.C., Hinshaw, V.S., Halvorson, D.A., 1997. Influenza. In: Calnek,B.W. (Ed.), Diseases of Poultry. , 10th ed. Iowa State University Press,Ames, Iowa, pp. 583–605.

Fontanilla, B.C., Silvano, F., Cumming, R., 1994. Oral vaccination againstNewcastle disease of village chickens in the Philippines. Prev. Vet.Med. 19, 39–44.

Henning, J., Morton, J., Hla, T., Meers, J., 2009. Evaluation of strategies toimprove village chicken production-controlled field trials to assesseffects of Newcastle disease vaccination and altered chick rearing inMyanmar. Prev. Vet. Med. 90, 17–30.

Henning, J., Morton, J., Hla, T., Meers, J., 2008. Mortality rates adjustedfor unobserved deaths and association with Newcastle disease virusserology among unvaccinated village chicken. Prev. Vet. Med. 85,241–252.

Integrated Regional Information Networks (IRIN), 2008. Porous borderincreases avian flu risks, Timor-Leste. Retrieved 29 March 2011 from:http://www.irinnews.org/Report.aspx?ReportId=78953.

Leibler, J.H., Otte, J., Roland-Holst, D., Pfeiffer, D.U., Magalhaes, R.S., Rush-ton, J., Graham, J.P., Silbergeld, E.K., 2009. Industrial food animalproduction and global health risks: exploring the ecosystems andeconomics of avian influenza. EcoHealth 6, 58–70.

Menge, E.O., Kosgey, I.S., Kahi, A.K., 2005. Bio-economic model to supportbreeding of indigenous chickens in different production system. Int. J.Poult. Sci. 4 (11), 827–839.

Ministry of State Administration and Territorial Management of East

Timor, 2002, Districts of Timor-Lesté. Retrieved 29 March 2011 from:http://www.estatal.gov.tl/English/Municipal/Municipal main.html.

Mwalusanya, N.A., 1998. Studies on productivity and nutritional statusof local chickens under village management conditions. MSc Thesis.Royal Veterinary and Agricultural University, Copenhagen, Denmark.

erinary

.

308 E. Serrão et al. / Preventive Vet

Norby, E., 1989. Modern approaches to live virus vaccine. Adv. Vet. Sci.Comp. Med. 33, 267.

World Organization for Animal Health (OIE), 2004. Avian influenza.In: Manual of Diagnostic Tests and Vaccines for Terrestrial Ani-mals, 5th ed. OIE, Paris, France, Retrieved 29 March 2011 from:http://web.oie.int/fr/normes/mmanual/A 00037.htm.

Samaan, G., 2007. Avian Influenza H5N1 in Indonesia, Regional HealthForum. 11 (1). Retrieved 29 March 2011 from: http://www.searo.who.int/LinkFiles/Regional Health Forum AVI Indonesia.pdf.

Spradbrow, P.B., 2000. Poultry as a tool in poverty eradication and pro-motion of gender equality. In: Dolberg, F., Henning Petersen, P. (Eds.),

Epidemiology of Newcastle Disease and the Economics of its Control.DSR Forlag, Frederiksberg, pp. 165–173.

Spradbrow, P.B., 2005. Appropriate vaccination and therapies for ruralflocks in developing countries and their relevance to developed coun-tries. World Poult. Sci. J. 61 (1), 47–54.

Medicine 104 (2012) 301– 308

The National Statistics Directorate, Timor-Lesté, 2004. Timor-Leste:private households by sucos. Retrieved 29 March 2011 from:http://dne.mof.gov.tl/census/tables/national/household/table3 4.htm

World Health Organization (WHO), 2011. Cumulative Number of Con-firmed Human Cases of Avian Influenza A/(H5N1) reported to WHO.Retrieved 29 March 2011 from: http://www.who.int/csr/disease/avian influenza/country/cases table 2011 02 25/en/index.html.

Yongolo, M.G.S., 1996. Epidemiology of Newcasatle disease in villagechicken in Tanzania. PhD dissertation. Sokoine University of Agricul-ture, Morogoro, United Republic of Tanzania.

Young, M., Alders, R., Grimes, S., Spradbrow, P., Dias, P., da Silva, A., Lobo,

Q., 2002. Preparing serum. In: Controlling Newcastle Disease in VillageChickens: A laboratory Manual. ACIAR Monograph No. 87, pp. 71–72.

Zeleke, A., Sori, T., Gelaye, E., Ayelet, G., 2005. Newcastle disease in villagechickens in the Southern and Rift Valley Districts in Ethiopia. Int. J.Poult. Sci. 4 (7), 507–510.