Preventive Veterinary Medicine 98 (2011) 204–208
Contents lists available at ScienceDirect
Preventive Veterinary Medicine
journa l homepage: www.e lsev ier .com/ locate /prevetmed
Identifi th h
influen farm
epidem
Foloruns issch
Jorge A. H
a National Vete
b Department o alenaan,
c Department o Carolin
d Poultry Refer Science
e Department of Infectious Diseases and Immunology, Faculty of Veterinary Medicine, Utrecht University, Yalenaan, The Netherlands
f Department of Large Animal Clinical Sciences, College of Veterinary Medicine, University of Florida, Gainesville, FL, USA
a r t i c l e i n f o
Article history:
Received 21 Fe
Received in re
Accepted 4 No
Keywords:
Avian influenz
H5N1
Risk factors
Nigeria
Poultry
a b s t r a c t
1. Introdu
In 1996,
with the hig
(HPAI H5N
onset of a m
wide as the
logical prop
other HPAI
genotype (X
Thevirusha
∗ Correspo
Vom, Nigeria.
E-mail add
0167-5877/$ –
doi:10.1016/j.bruary 2010
vised form 3 November 2010
vember 2010
a
We conducted a matched case–control study to evaluate risk factors for infection with
highly pathogenic avian influenza (HPAI) H5N1 virus in poultry farms during the epidemic
of 2006–2007 in Nigeria. Epidemiologic data were collected through the use of a question-
naire from 32 case farms and 83 control farms. The frequency of investigated exposure
factors was compared between case and control farms by using conditional logistic regres-
sion analysis. In the multivariable analysis, the variables for (i) receiving visitors on farm
premises (odds ratio [OR] =8.32; 95% confidence interval [CI] = 1.87, 36.97; P<0.01), (ii)
purchased live poultry/products (OR=11.91; 95% CI =3.11–45.59; P<0.01), and (iii) farm
workers live outside the premises (OR=8.98; 95% CI =1.97, 40.77; P<0.01) were identified
as risk factors for HPAI in poultry farms. Improving farm hygiene and biosecurity should
help reduce the risk for influenza (H5N1) infection in poultry farms in Nigeria.
© 2010 Elsevier B.V. All rights reserved.
ction
a geese farm was infected in Guangdong, China
hly pathogenic H5N1 strain of avian influenza
1). That particular incident was to mark the
ajor catastrophe in the poultry industry world
virus from the farm acquired several new bio-
erties and became the progenitor of several
H5N1 viruses-the key among which is the Z-
u et al., 1999; Guan et al., 2002; Li et al., 2004).
s caused thedeathor cullingof over 300million
nding author at: National Veterinary Research Institute,
Tel.: +234 703 312 8949.
ress: daydupe2003@yahoo.co.uk (F.O. Fasina).
multiple avian species and spread to at least 50 countries,
including 467 human infections and 282 human fatalities
(OIE, 2010; WHO, 2010).
Several factors have been determined to be responsible
for the dissemination of influenza virus in poultry. Bavinck
et al. (2009) showed that raising of multiple species in
backyard poultry and living in close proximity to infected
commercial farm premises increases the risk of infection.
Backyard poultry have also been found to be an impor-
tant source of spread and persistence of HPAI H5N1 in
South East Asia (Tiensin et al., 2005). Thompson et al.
(2008) also found out that large population of birds within
farm premises, poor biosecurity and presence of outside
birds, especially the Egyptian geese, were associated with
increasing the risk of infection of ostriches with the H5
influenza virus. Layer type operation has been associated
see front matter © 2010 Elsevier B.V. All rights reserved.
prevetmed.2010.11.007cation of risk factors associated wi
za H5N1 virus infection in poultry
ic of 2006–2007
o O. Fasinaa,b,e,∗, Ariel L. Rivasc, Shahn P.R. B
ernandezf
rinary Research Institute, Vom, Nigeria
f Farm Animal Health, Faculty of Veterinary Medicine, Utrecht University, Y
f Population Health and Pathobiology, College of Veterinary Medicine, North
ence Centre, Department of Production Animal Studies, Faculty of Veterinaryighly pathogenic avian
s, in Nigeria during the
opd, Arjan J. Stegemane,
The Netherlands
a State University, NC, USA
, University of Pretoria, South Africa

F.O. Fasina et al. / Preventive Veterinary Medicine 98 (2011) 204–208 205
with infection in the Netherlands (Thomas et al., 2005).
Finally, a study by Nishiguchi et al. (2007) proved that
introduction of end-of-lay chickens, sharing of farm equip-
ment, poor biosecurity and proximity to infected farms
are importa
Japan.
Nigeria
infection in
was first de
Northern N
other comm
36 states of
(National V
imately 711
of 1,264,19
reported an
early 2007.
In Afric
constituent
origin (Duc
Couacy-Hym
et al., 2009
virus infec
Although p
logic frame
H5N1 viru
et al., 200
study findi
not apply
ations in c
socioecono
populations
study were
H5N1 virus
2006–2007
2. Materia
2.1. Study l
Field an
Nigeria from
time-period
disease out
lates of HP
in Nigeria (
tigation wa
where one
virus infect
selected so
regionswou
2 states fro
the Central
tory); and 2
This design
tative of al
(those locat
states, case
below.
Spatial
on poul
ected ar
Selectio
se far
ted far
s wou
swere
2). All the 32 farms met the double criteria of a
farm clinical signs – sudden heightened mortality,
noea, coughing, wheezing, decreased food and water
e, decreased or cessation in egg production; and pos-
virological tests conducted at the National Veterinary
rch Institute, Vom (Joannis et al., 2008).
Selection of controls
ntrol farms were randomly selected from a group of
s located in the same states that reported infection
06–2007 (sample frame: 2111). Serum samples had
collected and tested at the same time the epidemic
red (to monitor other diseases), which produced
egative test results forHPAIH5N1virus. Control farms
selected to generate a ratio of 2.59:1 (83 farms).
e control farms were matched individually to case
s usingflock size (<500or >500; P-value formeasure of
iation between cases and controls = 0.015, two-tailed
% confidence interval, McNemar test value =5.898).
Epidemiologic questionnaire
structured questionnaire was designed, validated,
pplied during farm visits for data collection of epi-nt risk factors for infection with HPAI H5N2 in
was the first African country to report HPAI
February 2006, when the HPAI H5N1 virus
tected in a commercial poultry farm in Kaduna,
igeria. The virus thereafter spread rapidly to
ercial and backyard poultry and infected 25 of
Nigeria including the Federal Capital Territory
eterinary Research Institute records). Approx-
,000 birds of various species died and a total
1 were culled. One human fatality was also
d confirmed in Lagos, South-West Nigeria in
a, despite reports on the genetic/biological
s/properties of HPAI H5N1 viruses of African
atez et al., 2006, 2007; Njouom et al., 2008;
ann et al., 2008; Cattoli et al., 2009; Fasina
), the epidemiologic aspects of HPAI H5N1
tion in poultry have not been investigated.
revious studies have provided an epidemio-
work for investigation of risk factors for HPAI
s in poultry (Henzler et al., 2003; Thomas
5; Bouma et al., 2009; Busani et al., 2009),
ngs and proposed prevention strategies may
to the conditions of Nigeria in view of vari-
limatic conditions, poultry practices, and the
mic and cultural orientations of agricultural
in this geographic region. The aims of this
to investigate and identify risk factors for HPAI
infection in poultry farms in Nigeria during
.
ls and methods
ocation and time period
d laboratory investigations were conducted in
February 2006 to October 2007. During this
, a total of 1205 poultry farms with suspected
breaks were investigated, and 299 virus iso-
AI H5N1 were recovered from 26 of 36 states
Fasina et al., 2009). This epidemiologic inves-
s conducted within a subset of the 26 states
or more poultry farms reported HPAI H5N1
ions in 2006–2007. Seven stateswere randomly
that at least two states from each of the affected
ld be investigated. Thefinal selection included:
m the North (Kano, Kaduna); 3 states from
region (Bauchi, Plateau, Federal Capital Terri-
states from the South (Ogun, Lagos; see Fig. 1).
was meant not to assess the farms represen-
l Nigerian poultry farms, but high-risk farms
ed within infected states). Within the selected
and control farms were selected as indicated
Fig. 1.
study
the inf
2.2.
Ca
infec
state
naire
(n=3
case
dysp
intak
itive
Resea
2.3.
Co
farm
in 20
been
occur
seron
were
Th
flock
assoc
at 95
2.4.
A
and arepresentation of case and control farms in a case–control
try farms in Nigeria. Sampled populations were taken from
eas using stratified random sampling approach.
n of cases
ms were randomly selected from the list of
ms so that five farms from each of the selected
ld be sampled (n=35). Of those, question-
completedand retrieved fromthirty-two farms