Preventive Veterinary Medicine 149 (2018) 92–97

Contents lists available at ScienceDirect

Preventive Veterinary Medicine

journal homepage: www.elsevier.com/locate/prevetmed

Risk factors for Newcastle disease in broiler farms in Israel T

a,b,⁎ b a
A. Wiseman , E.M. Berman , E. Klement
a
Koret School of Veterinary Medicine, The Hebrew University of Jerusalem, P.O. Box 12, Rehovot, 76100, Israel
b
The Poultry Health Laboratories, The Egg and Poultry Board, P.O. Box 443, Kiryat Malachi, 83103, Israel

A R T I C L E I N F O A B S T R A C T

Keywords: Following a large outbreak of Newcastle disease (ND) in Israel, a cross-sectional study was conducted in the
Newcastle disease broilers sector. The aim of the study was to find geographical and farm related risk factors for ND. Information
Broiler was available on 96% of the broiler farms in Israel. Of these, farms diagnosed with ND in the years 2010–2012
Risk factors were compared with the other farms. Risk factors for ND were analyzed, using Generalized Estimating Equation
Israel
models. Six variables were found to be associated with the risk for ND outbreak: a distance of less than 300 m
GIS
from another farm (OR = 1.77, 95% CI 1.07–2.93), a distance of less than 6000 m from a national border
(OR = 2.00, 95% CI 1.22–3.30), farm location in the Ha’amakim district (OR = 2.46, 95% CI 1.32–4.61), village
type: a Moshav (village) vs. Kibbutz (Cooperative village) (OR = 1.96, 95% CI 1.04–3.70), and carcass disposal
in an uncovered bin (OR = 1.96, 95% CI 1.18–3.26). A distance of less than 800 m from an inter-city road was
found to be a protective factor (OR = 0.60, 95% CI 0.37–0.98). The results of this study provide information that
may be used to improve surveillance and control of ND.

1. Introduction
Newcastle Disease (ND) is an important viral disease of domestic
poultry and wild birds, characterized by gastrointestinal, respiratory
and neurological signs. ND affects poultry worldwide. The disease
causes economic losses due to mortality, morbidity, growth retardation
and drop in egg production, as well as performance losses associated
with post-vaccination reactions. Significant losses are caused by inter-
national trade restrictions on poultry products applied to countries
which are infected by ND (Cornax et al., 2012). Due to its significance
many efforts and resources are invested in prevention and control of ND
(Miller et al., 2015).
The cause of ND is the Newcastle disease virus (NDV), an RNA virus
from the family Paramyxoviridae. Clinical signs caused by NDV infec-
tions of poultry range from in-apparent to rapidly fatal disease, de-
pending upon the pathotype of the virus involved and the host condi-
tion and its immune status. NDV strains are classified into highly
virulent (velogenic), intermediate (mesogenic) or low virulent (lento-
genic) based on pathogenicity tests or the amino acid composition in
the cleavage site of the viral fusion protein. Lentogenic strains are
widely used as live vaccines. Hence the OIE definition of ND outbreak is
infection of birds with mesogenic or velogenic strains of NDV (OIE,
2012).
The first outbreak of Newcastle in Israel was recorded in 1937
(before the State of Israel was established) and since then Israel has
been intermittently infected by the virus (Neriah, 2001). Constant im-
provement in breeding methods and biosecurity, as well as im-
plementation of control measures such as mandatory vaccination, de-
claration of infected areas, and culling of infected flocks, failed to
prevent the occurrence of outbreaks of the disease, which still causes
significant economic losses in Israel.
During the last three decades, five epidemics of ND occurred in
Israel. These epidemics were usually characterized by a sharp accu-
mulation of outbreaks, occurring in many farms, followed by a gradual
decrease in the number of outbreaks during the following years (Fig. 1).
Similar to other areas of the world, the epidemics are usually seasonal
(Antipas et al., 2012; Awan et al., 1994), and in Israel occur mostly
during the winter months. In this study we analyzed a major epidemic
which peaked in 2011–2012. During the years 2010–2012, ND was
diagnosed in Israel in 393 flocks. Of these, 53% were diagnosed in
broiler flocks, 26% in commercial table egg layer flocks and the re-
maining in turkey flocks (6%), breeding flocks (4%) and backyard
flocks and wild birds (11%). During the study period (2010–2012), the
monthly flock incidence in the broiler sector ranged from 0 to 12.2%
and the cumulative yearly flock incidence was 0.28%, 5.92% and
2.48% in 2010, 2011 and 2012 respectively. The average age at diag-
nosis was 32 days (Wiseman, 2015).
All commercial poultry flocks in Israel are vaccinated against NDV
according to a defined program using a combination of live and in-
activated vaccines. The vaccination protocol for broilers during the

⁎
Corresponding author at: The Poultry Health Laboratories, The Egg and Poultry Board, P.O. Box 443, Kiryat Malachi, 83103, Israel.
E-mail address: anat@epb.org.il (A. Wiseman).

https://doi.org/10.1016/j.prevetmed.2017.11.009
Received 18 June 2017; Received in revised form 7 November 2017; Accepted 8 November 2017
0167-5877/ © 2017 Elsevier B.V. All rights reserved.
A. Wiseman et al.
years of the study included a live vaccine by spray on the day of hatch,
inactivated vaccine by subcutaneous injection at 10–12 days of age, and
another live vaccine by aerosol at 17–21 days of age (Gallili and Ben-
Nathan, 1998; Israel Veterinary Services and Animal Health, 2014). An
additional live vaccine was administered at the age of 10–12 days in
several farms. In Israel, even though it is mandatory to vaccinate
backyard and petting zoo poultry against NDV, vaccine coverage in this
sector is low.
The surveillance measures for ND in Israel includes active and
passive surveillance. All movement of poultry between farms is condi-
tional on a flock's minimal anti-NDV antibody titer (tested by HI-he-
magglutination inhibition). Breeding flocks are routinely tested by HI-
NDV every two months. Broiler farms are tested by HI-NDV once a year
in order to monitor the implementation and the performance quality of
the vaccinations program against ND. The passive surveillance includes
reports from producer and field veterinarians when ND is suspected,
and reports from the poultry health laboratories on positive tests
(mainly real time reverse transcriptase PCR and virology).
There are only a few studies aimed at assessing risk factors for ND in
developed countries. A survey that was conducted in Australia fol-
lowing several ND outbreaks (East et al., 2006), found geographic lo-
cation and a distance of less than 0.5 km to the nearest-neighbor
poultry farm, as statistically significant risk factors for positive serology
to NDV. In other production sectors (layers and breeders) increased risk
was associated with location, ownership, distance to the nearest-
neighbor poultry farm, increasing age, and the presence of wild birds on
the farm. A far greater number of studies were performed in developing
countries where poultry production is based on small backyard flocks.
Some of the risk factors are different in developing countries and de-
veloped countries and some are common. The following risk factors
were found in Ethiopia: reduced frequency of removing poultry litter,
receiving or buying replacement birds as opposed to hatching at the
farm itself, flock size, and using an open water source for the poultry
drinking water. The use of a grain supplement was found as a protective
factor (Chaka et al., 2013). Additional risk factors were found in rural
African poultry production systems and included seasonality, poor
physical condition of the birds, concurrent disease, breed (local breeds
are more resistant), high percentage of young birds in a flock, poultry
sheltering, environmental pollution, poultry transport and commerce
and the use of poultry litter for field fertilization. Poultry vaccination
was found as a protective factor (Antipas et al., 2012; Miguel et al.,
2013).
This is the first study aiming to identify and analyze risk factors for
ND in broiler flocks in Israel. This study focused on geographical and
farm variables. Recently, we showed that herd immunity is a significant
protective factor for the risk of ND outbreak (Wiseman and Berman,
93
Preventive Veterinary Medicine 149 (2018) 92–97
Fig. 1. Number of reported flocks diagnosed with ND in Israel by year
1985–2015. Columns with solid fill represent the study period 2010–2012.
2017). We chose to conduct the study on the broiler sector because half
of the ND cases during the years 2010–2012 were diagnosed in this
sector and because of the higher reliability of data generated from this
poultry sector. The Israeli broiler sector is advanced, introducing
modern husbandry practices as they become available. In 2013, 43% of
the farms were part of integrated operations. 45% of houses were
tunnel ventilated. All farms were all-in-all out with an average of 3.0
houses per farm. The average farm housed 65,000 chicks per flock
(Wiseman, 2015).
2. Materials and methods
2.1. Study design
A cross-sectional study was designed to compare all the broiler
farms diagnosed with ND during 2010–2012 with all the remaining
broiler farms.
2.2. Sources of information
In order to avoid information and memory biases, we analyzed only
variables for which data existed in available databases. Four sources of
information were used in this study.
1 Data on in-farm variables were collected from the results of a broiler
farms survey of the Israel Veterinary Services and Animal Health. In
2012, the veterinary services conducted a survey in all broiler farms
prior to the implementation of poultry farms and houses business
licensing regulations on August 2013 (Israel Ministry of Agriculture
and Rural Development, 1981). In the survey, inspectors visited the
broiler farms and filled out a questionnaire according to the business
licensing regulations demands. This study used only the ques-
tionnaires filled out in the first visit to a farm as part of the survey
during 2012.
2 Data on diagnosed ND cases were collected from the database of the
Poultry Health Laboratories of Israel Egg and Poultry Board (Berman
and Samberg, 1991). This database records veterinary data on all
commercial flocks.
3 Geographical data were collected from the Israel Ministry of
Agriculture GIS database. This database includes geographical data
(longitude and latitude) of each farm, district maps, village type and
location of the farm relative to the village.
4 GIS layers of roads and borders in Israel available in the Hebrew
University GIS unit were used to assign distance of each poultry
farm from main roads and borders.
A. Wiseman et al.
Table 1
Average, median, minimum, maximum, and interquartile range values for the distance and density variables for broiler farms in Israel.
Variable Average
Minimal distance from a border(meters)a 8297
Minimal distance from other poultry farm (meters)b 718
Poultry houses density per km2.b 4.4
Minimal distance from an inter-city road. (meters) 1011
Number of houses in the farm 3.0
a
Distance from a border of the state of Israel or a distance from the west bank border ('green line'). For study purposes, Israeli farms located in the west bank (n = 8) are considered as
having a zero distance from the border.
b
Since no information about the poultry farms in the Palestinians authority was available, Israeli farms in the west bank (n = 8) were not included in the analysis of the variables
minimal distance from other poultry farms and poultry houses density. The poultry houses density is the average density per broiler farm of poultry houses (from all sectors) per km2.
2.3. Study population
In the Israeli regulation all broilers in one farm must be of a single
age (never more than 10 days in range) raised by a single operator
(Israel Ministry of Agriculture and Rural Development, 1981). In this
study a flock is defined as all the broilers in all the houses on one farm.
In 2013, 727 broiler farms were registered in the Egg and Poultry
Board database. The study included 697 of the registered farms (96%)
for which GIS data was available. Data from the Broiler farms survey
existed for 584 of these farms.
2.4. Case definition
A case was defined as all the broiler flocks diagnosed with ND
during 2010–2012 (Fig. 1). An ND outbreak was defined according to
the OIE definition for reporting an outbreak of ND (OIE, 2012). The
diagnosis of cases was based on virus isolation and a real time reverse
transcriptase polymerase chain reaction (rRT-PCR) or rRT-PCR test
alone, regardless of clinical signs. The rRT-PCR method differentiates
virulent viruses from low-virulence (vaccine) viruses (Fuller et al.,
2009; Wise et al., 2004).
2.5. Virology
Virus isolation was performed by injection of a suspension of tra-
cheal and cloacal swabs into the allantoic cavity of 9-11-day-old
Specific pathogen free (SPF) embryonated chicken eggs (OIE, 2012).
The swabs were tested in pools of 5 swabs (1–2 pools for every source)
and each pool was injected into 5 eggs. The virus isolation was followed
by a hemagglutination assay (HA) and rRT-PCR on allantoic fluid in
order to confirm the diagnosis and differentiate virulent (velogenic)
viruses from low-virulent (lentogenic) viruses.
2.6. Real time reverse transcriptase PCR (rRT-PCR)
Total RNA was extracted from suspension of tracheal and cloacal
swabs using the QIAmp Viral RNA Mini Kit (Qiagen, USA) according to
the manufacturer’s instructions. The rRT-PCR assay that targeted the
matrix (M) gene was performed according to Wise et al. (Wise et al.,
2004), and sub-typing by rRT-PCR that targeted the fusion protein was
performed according to Fuller et al. (Fuller et al., 2009).
2.7. Geographical data processing
The geographical data were processed using Arc Map™ 10 software.
All poultry houses in the Agriculture Ministry GIS database appear as
separate polygons. Prior to data extraction, houses defined as inactive
were deleted from the file. In order to check the minimal distance from
a farm, all houses belonging to a same farm were aggregated to create
one polygon per farm (using dissolve function). A minimal distance
from a road/border/other farm was measured using the proximity tool
“near”. The calculation of density of poultry houses per km2 was
94
Preventive Veterinary Medicine 149 (2018) 92–97
Median Minimum Maximum Interquartile range
6637 0 30,635 9422
343 0 8972 685
3.0 1.0 28.8 3.3
787 5 4546 1264
2.0 1 35 3.0
performed by replacing each house polygon with a point (feature to
point function) and calculating point density to each house. Then an
average density per farm was calculated. The geographical variables
were calculated as continuous variables. The data for the broiler farms
were extracted from Arc Map for further analysis in SPSS.
2.8. Variables analyzed in the study
The following continuous variables were analyzed in the study: 1.
Minimal distance from an inter-city road. 2. Minimal distance from a
border. 3. Minimal distance from other poultry farms (from any sector).
4.Poultry house density (from all sectors) per km2. 5. Number of houses
in the farm. These continuous variables have an asymmetrical dis-
tribution. The medians, averages, minimum, maximum, and inter-
quartile range values of these variables are noted in Table 1.
The categorical variables that were analyzed in the study are: 1. The
sector type of the nearest poultry farm (for most broiler farms (70%) the
nearest farm belonged also to the broiler sector). 2. District: Israel is
divided into 5 districts by the Israeli Ministry of Agriculture. 3. Village
type: Kibbutz vs Moshav. A Kibbutz is a type of cooperative village
unique to Israel. 29% of broiler farms in Israel are located in a Kibbutz.
There is a significant difference between Kibbutz and other villages (or
Moshav) in the variables: number of houses in the farm and minimal
distance from other poultry farms. Kibbutz farms are larger and sig-
nificantly distant from other farms, as shown in Table 2. 4. The location
of the farm relative to the village (farm in the village or farm far from
the village). 5. Presence of a fence. 6. Presence of a gate. 7. The con-
dition of the concrete floor. All farms had concrete floors in the houses
but in 51% of the farms the floors were broken and needed repair. 8.
Presence of a concrete skirt extending at least one meter from external
walls of the poultry house. 9. Carcass disposal method. Most farms used
closed methods of carcass disposal such as underground waste con-
tainers or closed garbage bins. 22% of the farms used open carts or open
waste bins. 15% of the farms didn’t have a means of carcass disposal
(Wiseman, 2015). 10. Cleanliness of the farm yard (absence of weeds,
poultry litter, unnecessary equipment and feathers). 11. Method of feed
filling. Feed silos can be filled from outside the farm fence or the feed
truck must enter the farm. Variables 5–11 were retrieved from the 2012
broiler farms survey of the Israeli Veterinary Services and Animal
Health. According to the regulation of poultry farms and houses busi-
ness licensing, broiler farms should have a fence, gate, concrete or
Table 2
broiler farms in Israel: number of houses in a farm and minimal distance from other
poultry farm by village type.
village type Number of 1–2 houses in a Minimal distance from other
farms farm farm is less than 300 m
Moshav 498 313 (63%)a 281 (56%)a
Kibbutz 199 40 (20%)b 32 (16%)b
a,b
for each column, values in the different rows bearing different superscripts are sig-
nificantly different by Chi-squared test, P < 0.001.
A. Wiseman et al.
asphalt floor and concrete skirt, and a facility for carcasses disposal.
During survey time, most farms did not comply with the regulation
requirements.
2.9. Statistical methods for data analysis
The statistical analysis was performed by SPSS™ 25 software.
Continuous variables were converted to dichotomous variables by se-
parating them to two categories; below or above the median. Most
categorical variables (except the district and sector type of nearest
poultry farm variables) were also reduced to dichotomous variables.
In order to control for possible clustering and over dispersion, a
Generalized Estimating Equation (GEE) model with a logit link function
was used, in which the village name was defined as the subject variable.
The independent variables were initially tested by univariable GEE.
Variables with P < 0.1 in the univariable analysis were included in the
multivariable model. Multicollinearity between the independent vari-
ables was tested using SPSS™ linear regression model multicollinearity
test. The multicollinearity assessment was based on variance inflation
factor (VIF).
3. Results
One hundred and sixty six (23.8%) of the 697 registered farms for
which GIS data were available were diagnosed with ND at least once
during the years 2010–2012, one hundred and forty one (24.1%) farms
out of the 584 farms for which data from the broiler farms survey exist
were diagnosed with ND at least once. Fig. 2A shows all broiler farms in
the study by division to cases (all the broiler farms diagnosed with ND
during 2010–2012) and controls (all the remaining broiler farms).
Fig. 2B illustrates the poultry house density in Israel.
3.1. Univariable analyses
Sixteen variables were tested by univariable GEE (Table 3). Six
variables had P < 0.1 and were inserted into the multivariable model.
3.2. Multivariable analysis
Due to the correlation between the variables village type and
number of houses in the farms and the correlation between the vari-
ables minimal distance from a border/other poultry farm and district;
the variables number of houses in the farms and district were added to
the multivariable analysis in order to control possible confounding
95
Preventive Veterinary Medicine 149 (2018) 92–97
effect. The variable poultry house density was found to be non-sig-
nificant in the initial multivariable GEE model and was removed from
the final model (Table 4). The GEE model used data from 553 farms out
of the 697 farms in the study. The VIF values of the variables in the
multicollinearity test ranged from 1.01 to 1.35.
The following variables were associated with an increased risk for
ND: Minimal distance of less than 300 m from other farm, Minimal
distance of less than 6000 m from a border, location in the Ha'amakim
district, village type Moshav, and an open carcass bin. The variable
minimal distance of less than 800 m from an inter-city road was found
to be a protective factor.
4. Discussion
Identification of distance from other farms as a risk factor in our
study is consistent with what is known in the literature (East et al.,
2006; Alexander and Jones, 2008). East et al. found in their study that a
distance greater than 500 m from other farm was associated with a
decreased risk of a farm to be seropositive for NDV. In our study the
distance was set to 300 m; this is the minimal distance allowed between
farms according to Israeli regulation from 1981 (Israel Ministry of
Agriculture and Rural Development, 1981). The regulations refer to
new farms. Since many of the farms are older, this value is close to the
median value. The virus can travel through the air especially when the
air is humid and cool (Alexander, 1988). Animals such as cats, dogs and
rodents can wander between neighboring farms and spread the virus
among farms with low biosecurity. Another explanation is that close
farms will have more horizontal contacts increasing the risk of me-
chanical transfer of the virus through people, cars and equipment (East
et al., 2006). This explanation may be less significant in Israel; where
due to its small size, contacts such as different suppliers and feed trucks
move all over the country.
We found that distance of less than 6000 m from a border was as-
sociated with an increased risk for ND. NDV can spread between
countries, especially through traffic of people, poultry, and trade con-
nections (Miguel et al., 2013). There is no such traffic in Israel’s in-
ternational borders (with the countries Egypt, Jordan, Syria, and Le-
banon). In this study we considered the border of the west bank ('green
line' – pre 1967 border) as an international border as Israel does not
have control or knowledge regarding Palestinian poultry farms. It is
known however that there is legal and illegal traffic of poultry and
poultry products between the two sides of the green line. All traffic,
especially the illegal traffic, raise the risk of transmitting infectious
diseases. Israel’s neighboring countries do not always report ND
Fig. 2. Israel map, A: location of broiler farms (96% of the broiler farms in
Israel) by type: case (farms diagnosed with ND during 2010–2012) and
control (all other broiler farms). Some points are overlapping others. B:
poultry houses density in Israel per km2. Map B was based on all active
houses from all poultry sectors.
A. Wiseman et al. Preventive Veterinary Medicine 149 (2018) 92–97

Table 3
GEE univariable analysis of risk factors for Newcastle disease in broiler farms in Israel, 2010–2012.

Variable (n) Category (n) Percentage among cases (n) Percentage among controls (n) OR 95% CI (OR) p

Minimal distance from other poultry farm (697)a <= 300 m (313) 51.2 (85) 42.9 (228) 1.47 0.96–2.23 0.073
>300 m (384) Ref

Minimal distance from an inter-city road (697)a <= 800 m (351) 43.4 (72) 52.5 (279) 0.67 0.43–1.06 0.086
>800 m (346) Ref

Minimal distance from a border (697)a <= 6000 m (309) 53.0 (88) 41.6 (221) 1.72 1.14–2.60 0.010
>6000 m (388) Ref

Poultry houses density (689)a >3 houses per km2 (345) 41.4 (67) 52.6 (277) 1.60 1.02–2.52 0.042
<=3 houses per km2 (344) Ref

Number of houses in the farm (697) 1–2 (353) 47.0 (78) 51.8 (275) 0.84 0.60–1.20 0.338
>2 (344) Ref

Sector type of the nearest farm (689) Broilers (482) 68.1 (112) 70.4 (370) Ref
Commercial layers (97) 12.7 (21) 14.3 (76) 0.91 0.52–1.59 0.739
Turkey meat (45) 7.2 (11) 6.6 (34) 1.07 0.52–2.19 0.855
Breeders (42) 7.8 (13) 5.6 (29) 1.40 0.63–3.12 0.405
Other (23) 4.2 (7) 3.0 (16) 1.39 0.51–3.77 0.524

District (697) Tzafon (256) 33.7 (56) 37.7 (200) Ref

Ha'amakim (121) 22.9 (38) 15.6 (83) 1.61 0.89–2.92 0.119
Mercaz (83) 7.8 (13) 13.2 (70) 0.66 0.31–1.39 0.268
Shfela va'Har (85) 9.0 (15) 13.2 (70) 0.81 0.39–1.71 0.585
Negev (152) 26.5 (44) 20.3 (108) 1.41 0.82–2.45 0.218

village type (697)a Moshav (498) 78.9 (131) 69.1 (367) 1.69 1.05–2.70 0.030
Kibbutz (199) Ref

Location of the farm relative to the village (697) Farm in the village (243) 58.4 (97) 67.2 (357) 1.43 0.93–2.18 0.100
Farm far from the village (454) Ref

Fence (566) No (415) 71.4 (100) 73.9 (315) 0.90 0.57–1.43 0.668
Yes (151) Ref

Gate (447) No (304) 67.9 (76) 68.1 (228) 1.05 0.65–1.69 0.853
Yes (143) Ref

Concrete floor (534) Impaired (271) 44.0 (59) 53.0 (212) 0.72 0.46–1.12 0.146
Proper (263) Ref

Concrete skirt around the house (568) No (357) 64.5 (89) 62.3 (268) 1.14 0.75–1.73 0.553
Yes (211) Ref

Type of carcass bin (553)a Uncovered bin (124) 32.6 (44) 19.1 (80) 1.90 1.20–3.00 0.006
Covered bin (429) Ref

Clean farm yard (562) No (345) 64.0 (87) 60.6 (258) 1.16 0.76–1.78 0.485
Yes (217) Ref

Method of feed filling (481) Internal (91) 20.8 (25) 18.3 (66) 0.87 0.49–1.54 0.625
External (390) Ref

a
P < 0.10. Variable was included in initial multivariable GEE regression model.

outbreaks but it is reasonable to assume that ND is present in these
countries. Wild birds crossing borders can carry and spread NDV and
other infectious diseases; but the role of wild birds in transferring NDV
is not clear.
Village type was found to be significantly associated with risk for
ND. Due to the difference in the village structure, Kibbutz farms are
larger and more isolated than Moshav farms. Usually, in a Kibbutz there
are one or two farms with one producer and a small number of em-
ployees (1–2). On the other hand, in a Moshav there is a large number
of small and densely situated farms, each farm owned by a different
producer. This structure results in a much higher traffic of poultry,
poultry workers, feed trucks, and other suppliers in the Moshav. The
high traffic rate may increase the risk for disease transfer. In the mul-
tivariable models, the variable village type remains significant after
adjustment by the GEE subject variable, name of village and the vari-
ables number of houses in the farm and distance from other poultry
farms. This might indicate that there are other inherent differences in
farming habits between the two village types. These differences are
probably not related just to vaccination adequacy as in a recent study it
was shown that Moshav villages were at higher risk for ND outbreaks
despite control of level of herd immunity (Wiseman and Berman, 2017).
In this study, location of a farm in the Ha'amakim district was as-
sociated with increased risk for ND. East et al. also showed association
between geographical region and NDV seroprevalence (East et al.,
2006). The difference could be climate related or more probably related
to differences of the human population. Customs, commerce connec-
tions and people and poultry traffic through a certain area, for example
through the two sides of the ‘green line’, could be risk factors for ND.
Another explanation for the increased risk is that the 2010–2012 out-
break began in the Ha'amakim district.
Distance of less than 800 m from an inter-city road was associated
with decreased risk for ND. This result is contrary to what was ex-
pected. The hypothesis was that proximity to a road which carries
trucks of poultry, carcasses and litter would increase the risk of the
virus traveling through the air and infecting farms near the road. There
are a few studies from developing countries testing proximity to a road
as a risk factor for other diseases, such as avian influenza (Gilbert and
Pfeiffer, 2012; Paul et al., 2014). In these studies, proximity to a road
was found as a risk factor because areas containing main roads are
highly densely populated areas, with more farms and more

96
A. Wiseman et al. Preventive Veterinary Medicine 149 (2018) 92–97

Table 4 that carcass disposal method by open bins should not be permitted.
Generalized Estimating Equation model (logit link function) depicting the significant risk When planning new farms, the risk of Newcastle (and other diseases)
factors for clinical infection by Newcastle disease virus during 2010–12 in a sample of 553
should be taken into consideration. The fact that some variables were
broiler poultry farms in Israel. The number of subject variable levels (village name) was
292 with 1–26 measurements per subject. not found to be significant in this study does not mean that they are not
important. In Israel and other countries, the accepted biosecurity
Variable Category (n) OR 95% CI (OR) p measures and management practices should be followed.
Minimal distance from other <=300 m (235) 1.77 1.07–2.93 0.026
poultry farma Acknowledgments
>300 m (318) Ref

Minimal distance from an <=800 m (278) 0.60 0.37–0.98 0.043

We are thankful to our colleagues in the poultry health department
inter-city roada of the Israel Veterinary Services and the ministry of agriculture and
>800 m (275) Ref rural development for their cooperation. We would also like to thank
Minimal distance from a <=6000 m (231) 2.00 1.22–3.30 0.006 Mr. Adi Ben-Nun, from the Hebrew University GIS unit, for his help in
borderb data processing.
>6000 m (322) Ref

Districtb,c Tzafon (180) Ref References

Ha'amakim (98) 2.46 1.32–4.61 0.005
Mercaz (65) 0.80 0.34–1.89 0.610 Alexander, D.J., 1988. Newcastle disease: methods of spread. Newcastle Disease.
Shfela va'Har (74) 0.85 0.34–2.09 0.717 Springer, pp. 256–272.
Negev (136) 1.78 0.91–3.47 0.093 Antipas, B.-B.B., Bidjeh, K., Youssouf, M.L., 2012. Epidemiology of Newcastle disease and
Number of houses in the 1–2 (258) 0.67 0.42–1.08 0.098 its economic impact in chad. Eur. J. Exp. Biol. 2, 2286–2292.
Awan, M.A., Otte, M.J., James, A.D., 1994. The epidemiology of Newcastle disease in
farm
rural poultry: a review. Avian Pathol. 23, 405–423.
>2 (295) Ref
Berman, E.M., Samberg, Y., 1991. A national medical database for poultry in Israel.
Village typea Moshav (388) 1.96 1.04–3.70 0.039 Abstracts, S24.3. Presented at the XXIV World Veterinary Congress. Rio de janeiro,
Kibbutz (165) Ref Brazil.
Chaka, H., Goutard, F., Roger, F., Bisschop, S.P., Thompson, P.N., 2013. Household-level
Type of carcass bina Uncovered bin 1.96 1.18–3.26 0.010 risk factors for Newcastle disease seropositivity and incidence of Newcastle disease
(124) virus exposure in backyard chicken flocks in Eastern Shewa zone, Ethiopia. Prev. Vet.
Covered bin (429) Ref Med. 109, 312–320.
Cornax, I., Miller, P.J., Afonso, C.L., 2012. Characterization of live LaSota vaccine strain-
a
P < 0.05. induced protection in chickens upon early challenge with a virulent Newcastle dis-
b
P < 0.01. ease virus of heterologous genotype. Avian Dis. 56, 464–470. http://dx.doi.org/10.
c
Overall P-value for the significance of the District variable = 0.009 (Wald chi- 1637/10043-122011-Reg.1.
East, I., Kite, V., Daniels, P., Garner, G., 2006. A cross-sectional survey of Australian
square = 13.46).
chicken farms to identify risk factors associated with seropositivity to Newcastle-
disease virus. Prev. Vet. Med. 77, 199–214.
Fuller, C.M., Collins, M.S., Alexander, D.J., 2009. Development of a real-time reverse-
opportunities for the virus to travel between the farms. This explanation transcription PCR for the detection and simultaneous pathotyping of Newcastle dis-
does not match the results of this study. There is probably another ease virus isolates using a novel probe. Arch. Virol. 154, 929–937.
Gallili, G.E., Ben-Nathan, D., 1998. Newcastle disease vaccines. Biotechnol. Adv. 16,
factor which is absent from the multivariable model that may explain 343–366.
the correlation between the distance from a road and the risk for ND. Gilbert, M., Pfeiffer, D.U., 2012. Risk factor modelling of the spatio-temporal patterns of
The type of carcass disposal was obtained from the broiler farms highly pathogenic avian influenza (HPAIV) H5N1: a review. Spat. Spatio-Temporal
Epidemiol. 3, 173–183.
survey performed by the Israel Veterinary Services. 22% of the farms in Israel Ministry of Agricultureand Rural Development, 1981. Business Licensing
the survey used open garbage bins for carcass disposal. The use of Regulations (terms for Poultry Farms and Houses Licensing). 1981 [WWW
uncovered bins as opposed to other methods was associated with an Document]. http://www.moag.gov.il/NR/rdonlyres/11CCA783-36A8-443B-A9BB-
C162B891BD7E/0/takanot.pdf (Accessed 29 May 2016).
increase of the risk for ND. When an open garbage bin is used, the Israel Veterinary Services and Animal Health, 2014. Broilers Vaccination Program.
carcasses are exposed and attract animals such as cats, rats and ravens [WWW Document]. URL http://www.vetserv.moag.gov.il/Vet/Yechidot/Ofot/
that could transfer the Newcastle virus between nearby farms tochniot_hisunim/default.htm (Accessed 24 May 2016).
Miguel, E., Grosbois, V., Berthouly-Salazar, C., Caron, A., Cappelle, J., Roger, F., 2013. A
(Alexander, 1988). It seems that this method increases the risk of in-
meta-analysis of observational epidemiological studies of Newcastle disease in
fection of the farm itself and the surrounding farms as well. African agro-systems, 1980–2009. Epidemiol. Infect. 141, 1117–1133.
Since August 2013 all broiler farms were required to meet the Miller, P.J., Dimitrov, K.M., Williams-Coplin, D., Peterson, M.P., Pantin-Jackwood, M.J.,
biosecurity measures listed in the regulation of poultry farms and Swayne, D.E., Suarez, D.L., Afonso, C.L., 2015. International biological engagement
programs facilitate newcastle disease epidemiological studies. Front. Public Health 3,
houses business licensing before they could populate the farm with a 235.
new flock. In the years 2013, 2014 and 2015, the cumulative yearly Neriah, A., 2001. Veterinary Medicine in Eretz-Israel −Fifty Years of Veterinary Medicine
flock incidence in the broiler sector has decreased to 0.31%, 0.30% and (1917–1967). The Cattle Breeders Association.
OIE, 2012. Newcastle Disease. OIE Terrestrial Manual, pp. 1–19 OIE Terr. Man. 2012,
0.27% respectively, despite the fact that it increased in the table egg 2012.
layer sector (The poultry health laboratories, 2016). Paul, M.C., Gilbert, M., Desvaux, S., Andriamanivo, H.R., Peyre, M., Khong, N.V.,
The limitation of our study is the relatively small number of vari- Thanapongtharm, W., Chevalier, V., 2014. Agro-environmental determinants of avian
influenza circulation: a multisite study in Thailand, Vietnam and Madagascar. PLoS
ables tested in the study. Since our research was retrospective, we could One 9, e101958.
not collect data on many important biosecurity and management The poultry health laboratories, 2016. The Poultry Health Laboratories Annual Report
practices variables. On the other hand, the use of existing databases 2014–2015.
Wise, M.G., Suarez, D.L., Seal, B.S., Pedersen, J.C., Senne, D.A., King, D.J., Kapczynski,
prevented information and memory biases. D.R., Spackman, E., 2004. Development of a real-time reverse-transcription PCR for
detection of newcastle disease virus RNA in clinical samples. J. Clin. Microbiol. 42,
5. Conclusions 329–338.
Wiseman, A., Berman, E.M., 2017. Herd immunity to Newcastle disease virus in broiler
flocks in Israel. Avian Pathol. 46, 396–402.
In the light of study results, it is recommended that new farms will Wiseman, A., 2015. Identification and Analysis of Risk Factors for Newcastle Disease in
not be constructed at a distance of less than 300 m from other farms and Broiler Farms in Israel (Thesis). The Hebrew university of Jerusalem, Israel.

97