Livestock Research for Rural Development 26 (10) 2014 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Seroprevalence and risk factors for Salmonella gallinarum infection in smallholder layers in Mwanza City, Tanzania

Q T Bura, H B Magwisha1 and R H Mdegela2

Tanzania Veterinary Laboratory Agency, P. O. Box 129, Mwanza, Tanzania
1 Tanzania Veterinary Laboratory Agency, P. O. Box 9254, Dar Es Salaam, Tanzania.
2 Department of Veterinary Medicine and Public Health, Faculty of Veterinary Medicine, Sokoine University of Agriculture, P. O. Box 3021, Morogoro, Tanzania


A cross sectional study was carried out to establish the seroprevalence and risk factors for Salmonella gallinarum infection in smallholder layers flocks in Mwanza City, Tanzania from December 2013 to January 2014. A total of 315 layers from 63 flocks were randomly selected for collection of blood samples. A questionnaire survey and direct flock observation to determine risk factors were made in 63 sample flocks. Whole Blood Agglutination (WBA) and Tube Agglutination (TA) tests were carried out to establish the individual layers and flock-level seroprevalence of S. gallinarum.


The individual layers (n=315) seroprevalence established using WBA and TA tests were 28% and 15% respectively and flock-level (n=63) seroprevalence were 62% (95% C.I = 50 – 73) and 51% (95% C.I = 39 – 63), respectively.  Several variables were investigated in this survey in relation to sero-status of S. gallinarum infection. A logistic regression analysis indicated that, two variables were significantly associated with sero-status: presence of other birds (OR = 11.1; 95% CI = 2.7 – 45.3) and presence of multiple flocks of layers (OR = 7.8; 95% CI = 2.0 – 30.8). In conclusion, this study demonstrated relatively high birds exposure to S. gallinarum and responsible risk factors in smallholder layers flocks. The improvement of good management practices such as avoiding other birds and multiple flocks and biosecurity measures such as disinfection of equipment should be put in place to alleviate the risk of S. gallinarum infection.

Keywords: Flocks, Fowl typhoid, Ilemela, Nyamagana


 Salmonella enterica subspecies enterica serovar Gallinarum denoted, as S. gallinarum is a worldwide poultry pathogen of considerable economic importance, particularly in countries with a developing poultry industry. The serovar gallinarum is a causative agent of Fowl typhoid (FT), a disease of mature fowls that results in either acute enteritis with greenish diarrhoea or in  a chronic form that leads to reduced egg production due to genital tract infection (Proux et al 2002). It is a septicaemic disease primarily affecting chickens and turkeys, but other fowls are also susceptible (Zanella 2007). The disease is well controlled in many countries that comply with repeated serological testing and removal of positive reactor birds from the flocks (Zanella 2007).


Salmonella gallinarum is a Gram-negative bacterium, belonging to the family Enterobacteriaceae. This serovar is distinct from the rest of the known Salmonella serovars because it is non-motile, and adapted to fowls (Wu et al 2005). It is comprised of two related biovars namely Gallinarum and Pullorum. Both biovars possess somatic (O) antigens 1, 9, 12 and they belong to the serogroup D in Kauffman white scheme (Grimont and Weill 2007). The two biovars can be differentiated using biochemical and molecular tests (Paiva 2009; Ribeiro et al 2009; Batista et al 2013). As far as biochemical test is concerned, the main characteristics assessed are the capacity to use dulcitol by gallinarum, but not by pullorum, and ornithine decarboxylation by pullorum, but not by gallinarum (Paiva 2009). The conventional serological tests cannot differentiate these two serovars (Paiva 2009). Serology is useful in detection of both gallinarum and pullorum because they are not excreted extensively in the feces, unlike many other Salmonella serotypes that are more frequently associated with human food poisoning (Oliveira et al 2004; Soria et al 2012). The same antigen prepared using pullorum standard strain is globally used to detect both gallinarum and pullorum using agglutination test, however, sensitivity of the test is poor particularly for the latter (Proux et al 2002). Despite the poor sensitivity of this test, it has been used as a flock test in the control and eradication of S. gallinarum-pullorum infection in many countries (Oliveira et al 2004).


Salmonella gallinarum can be transmitted vertically and horizontally (Waiswa et al 2006). The vertical transmission is through eggs to chicks. The horizontal transmission is through movement of contaminated equipment and materials such as egg trays, manure, litter, clothes, feed, carcasses and birds (Soria et al 2012). The infected birds, such as carriers or reactors, wild birds and vermin (rodents) as well as scavengers, are by far the most important means of perpetuation and spread of the bacteria.


In Tanzania, fowl typhoid is greatly important in commercial chickens, with experience of frequent outbreaks in hatcheries and commercial layers (Minga et al 2001). Economically, fowl typhoid is the most important disease affecting the commercial chicken industry and has a high incidence in Tanzania (Minga and Nkini 1986; Mdegela 1998). It causes great economic losses through mortalities and, reduced egg production and growth rate.


Despite all that, there are limited detailed studies that attempted to investigate risk factors that are responsible for the status of this disease in Tanzania. The present study therefore was designed to establish seroprevalence and responsible risk factors for S. gallinarum infection in smallholder layers flocks of Mwanza City. Findings from this study would contribute knowledge for preventive and control measures of FT that will reduce the bird’s exposure to S. gallinarum.

Materials and Methods

Study area


This study was conducted in Mwanza City from December 2013 to January 2014. The City is comprised of two districts namely Ilemela and Nyamagana with 9 and 12 administrative wards respectively. It is located on the southern shores of Lake Victoria in northwestern Tanzania between 2ᵒ15ˈ and 2ᵒ45ˈS and 32ᵒ45ˈ and 33ᵒ0ˈE (Fig. 1).

Figure 1. A map of Mwanza City (left) and Tanzania (right) showing location of study flocks
Study design and sampling strategy


A cross sectional study to establish seroprevalence and responsible risk factors for S. gallinarum infection in layers was employed in this survey. From 21 wards of the study area, 125 layers flocks of different age groups and flock sizes were identified as a sampling frame with the assistance from District Veterinary Officers (DVOs).  The flock age was categorised as starter (day 1 – 8weeks old), growers (9 weeks – 20 weeks old) and layers (>20 weeks old), and each flock had layers of the same age. The flock size was categorised as small (≤200 birds), medium (201- 500 birds) and large (>500 birds). Because the same bird keepers in different locations owned some flocks, the sample flocks were purposively raised in order to get good representation of the information on the risk factor from different bird keepers. From the sampling frame, 63 sample flocks were randomly selected whereby 51 of different bird keepers owned them and they belonged in to 16 wards.  Five percent of each sample flock was randomly sampled with a maximum of five individual layers for flocks with more than 100 birds. The sample size was estimated following the formula,  , where n = number of samples, t = reliability coefficient for an alpha level of 0.05 (t = 1.96),  SD =  √[p(1-p)]2, where p = estimated prevalence of S. gallinarum in layers (18.4%) as reported by Mdegela et al (2000) and L = acceptable margin of error or precision (L = 5%) as described by Bartlett et al (2001) and Stevenson (2012).


The interviews using structured questionnaires were conducted to gather information on the responsible risk factors for S. gallinarum infection in all sample flocks. The respondents were owners, attendant, or any appropriate family member older than 18 years. Authors conducted the interviews using local language (‘Swahili’) and responses were consistently associated to expected outcome in all variables. The variables were selected based on the literature review of previous study concerning the risk factors for S. gallinarum infection in layers (Nasinyama et al 1997; Ahmed et al 2008; Mbuko et al 2009; Snow et al 2010).


Sample collection


Each selected flock was physically visited once to collect blood samples and the information on responsible risk factors. In case of multiple flocks, which were single-housed, only single flock per house was sampled. A total of 315 blood samples were collected from layers. About two - 5millilitre (ml) of blood was collected aseptically using a 21-Gauge sterile needle and syringe from the wing vein. Syringes with blood were kept in slanted position for about 6 - 12 hours. Sera were decanted and placed into 1.5 ml serum tubes, stored at - 20°C. Using a cold box with ice packs, all decanted sera were shipped to Sokoine University of Agriculture (SUA), Veterinary Microbiology Laboratory for further analysis by Tube Agglutination test (TA) for confirmation (OIE 2012).


Whole Blood Agglutination (WBA) and Tube Agglutination (TA) tests


The WBA and TA tests were performed according to the procedures described by OIE (2012). Antigens used for both tests were locally produced from local S. gallinarum strain available at SUA, Veterinary Microbiology Laboratory as described by OIE (2012). For WBA test, about 20µl of blood was placed immediately on a clean white tile and equal volume of S. gallinarum coloured antigen was poured on the blood and mixed thoroughly using toothpick followed by gentle rocking. Identification of positive cases was based on formation of clumps within two minutes.


On the other hand, the TA test was carried out using macrotitre plates instead of tubes. Using micro titre pipette, 100µl of Phosphate Buffer Solution (PBS) was added to each of the 10 wells of the plate. Equal volume of serum was added to first well and thoroughly mixed. The two-fold dilution was performed in all wells and equal amount of the mixture was discarded from the last wells. This was followed by adding100µl of colourless S. gallinarum antigen suspension into all wells. The mixtures were then incubated for 24 hours in a water bath at room temperature (30°C). Results were read based on presence or absence of agglutination. A clear supernatant fluid indicated a positive reaction based on agglutination. The cut-off point for positive sample was 1:64.


Structured interview


The interview using structured questionnaires was conducted to gather information on the responsible risk factors for S. gallinarum infection in all sample flocks. The responses were based on the layers husbandry and management practices’ issues as given in table 1.

Table 1: Summary of questionnaire administered to establish risk factors for S. gallinarum infection in layers in Mwanza City, Tanzania


Expected response

Layers keeping time

Months of being in layers keeping

All in all out practice


Types of birds kept

Mentioning all types of birds kept

Number of layer flocks

Discrete variable like 1, 2, 3 and 4

Flock size

Number of layers kept

Age of layer birds

Time spent by the layers in the flocks (in months)

Origin of layer chicks

Name of source or hatchery of the layers in flocks

Origin of feeds

Own made at home, machine or purchased readymade

Feed and water equipment

Raised up or put on the floor

Equipment disinfection


Flock attendance

Presence of single attendant per flock or not

Flocks attended by single person

Discrete variable like 1, 2, 3 and 4

Personal restriction to flock entrance

Yes/No – aided by direct observation

Use of footbath with disinfectant

Yes/No – aided by direct observation

Strength of poultry house (‘Banda’)

Presence or absence of gaps - aided by direct observation

Constraints in poultry keeping

Open-ended – open description

Any occurrence of outbreaks


Signs presented by FT

Open-ended – open description

How FT was taken care

Open-ended – open description

Carcass disposal

Burning, burying or scavenged

Litter and manure disposal

Burning, burying or gardening

Accessibility of veterinary services


Data analysis


Data on sero-status (positive or negative) were entered in Microsoft - Excel 7, analysed for individual layers and flock-level seroprevalence based on both WBA and TA test results. Data for risk factors were entered, edited and analysed using Epi Info 6 (Coulombier et al 2001). A univariate and multivariate logistic regression analysis was performed, whereby odds ratio (OR) and probability (p-values) at 95% confidence level were established for biological and statistical association between dependent and predictive investigated variables. A logistic regression model was built to identify the variables that are at least moderately associated to dependent variable (covariates), and to assess the association of covariates with the dependent variable.  The model was fitted by backward stepwise selection of variables (McDonald 2009; James et al 2013). Retaining of variables in the model was based on the likelihood ratio test p values (p<0.25 for the first model and p<0.05 for the final model). The potential confounding effects of those variables not retained in the final model were assessed by refitting each variable in succession into the final model and observing the percentage change in the odds ratios of the retained variables. The variables that resulted into more than 25% change in OR were considered as confounders in this survey. The goodness of fit for logistic regression model was assessed as described by Hosmer et al (2013).


Individual layers and Flock-level Seroprevalence


Out of 315 samples tested, 28% and 15% tested positive with WBA and TA tests, respectively. Using WBA test, the individual layers seroprevalence was found high in Mbugani (80%); low in Pamba (5%) and no reactors were found in Igogo ward. On the other hand, using TA test individual layers seroprevalence was found high in Nyamanoro (60%) and low in Nyakato (4%) ward; reactors were not found in Igogo and Pamba wards (Fig. 2). At the flock-level, out of 63 flocks, 62% (95% C.I = 50 – 73) and 51% (95% C.I = 39 – 63) tested positive with WBA and TA tests, respectively as given in table 2. The flock positivity was based on the criterion of finding at least one positive individual layer in a flock.


Based on district level, the flock-level seroprevalence was 65% (95% C.I = 49 – 78) and 58% (95%C.I = 39 – 74) using WBA test in Ilemela (n=37) and Nyamagana (n=26) districts, respectively. Using TA test, the flock-level seroprevalence was 49% (95%C.I = 33 – 64) and 54% (95%C.I = 35 – 71) in Ilemela and Nyamagana districts, respectively as given in table 2.

Figure 2. Individual layers seroprevalence for S. gallinarum infection based on WBA and TA test results by ward

Table 2: Flock seroprevalence for S. gallinarum infection based on WBA and TA test results by district



WBA (%)

TA (%)



65 (95%C.I = 49 – 78)

49 (95%C.I = 33 – 64)



58 (95%C.I = 39 – 74)

54 (95%C.I = 35 – 71)

Overall Prevalence


62 (95%C.I = 50 – 73)

51 (95%C.I = 39 – 63)

95% C.I = 95% Confidence interval; N = Number of flocks.

Variables influencing seroprevalence


The variables that significantly influences seropravence of S. Gallinarum at p < 0.25 during univariate analysis (table 3) were subjected in to multivariate logistic regression analysis and final model results are given in tables 4 and 5. The presence of other birds such as local chickens, ducks and turkeys in the premises of layers flocks increases a chance of S. gallinarum infection in layers by 11.1 folds (OR = 11.1; 95% C.I = 2.7 – 45.3). In addition, multiple flocks of layers belonging to the same owner have 7.8 times a chance of contracting S. gallinarum infection (OR = 7.8; 95% C.I = 2.0 – 30.8) than single layers flock (Table 5).

Table 3: Univariate regression analysis for variables considered as risks associated with S. gallinarum infection in layers showing seroprevalence, p-values and odds ratios

Variable and level


TA %


OR (C.I)

All in all out












0.971 (0.357 – 2.63)

Layers breed

    Rhode Island Red

    Black Australop










1.96 (0.493 – 7.75)

Disposal of litter/Manure













Equipment disinfection













0.255 (0.0889 – 0.729)

Flock age
















0.667 (0.147 – 3.03)

0.25 (0.04 – 1.56)

Flock size
















0.222 (0.0571 – 0.865)

0.159 (0.0425 – 0.593)

Keeping time
















0.244 (0.0731 – 0.818)

0.209 (0.0489 – 0.888)

Multiple flocks












5.41 (1.64 – 17.9)

Origin/source of feeds

    Machine mixed


    Home made













0.923 (0.316 – 2.69)

0.577 (0.104 – 3.19)

Poultry house strength












1.5 (0.494 – 4.58)













0.446 (0.161 – 1.24)

Source  of chicks

    Outside TZ

    Within TZ

    Within Mwanza City













1.6 (0.288 - 8.9)

1.71 (0.581 - 5.06)


Different age groups












0.542 (0.0841 – 3.49)

Use of foot bath












0.839 (0.156 – 4.51)

Other birds












8.03 (2.25 – 28.7)

Personal restriction












0.727 (0.255 -       2.07)

Specific attendant












0.303 (0.0961 - 0.959)

N = Number of flocks, OR = odds ratio, C.I = confidence interval, % = percentage, TZ = Tanzania

Table 4. Multivariate regression analysis for the retained variables in the model with *likelihood ratio (d.f= 8) p value ? 0.25 in the univariable regression analysis

Variable (level)


95% C.I





Equipment disinfection (yes/no)


0.0514 – 1.79





Flock age (Layers/Growers)


0.0186 – 4.68





Flock age (Starter/Growers)


0.00551 – 2.17





Flock size (Medium/Large)


0.0542 – 2.54





Flock size (Small/Large)


0.0265 – 1.38





Keeping time (Medium/Long)


0.0242 – 1.35





Keeping time (Short/Long)


0.0113 – 2.16





Multiple flocks (Yes/No)


2.14 – 156





Records (Yes/No)


0.107 – 4.44





Other birds (Yes/No)


3.31 – 194












Likelihood ratio p value







*Likelihood ratio (d.f = 8) p value represents the degree of deviation of the predictor variable from the model;
d.f represents degree of freedom (McDonald 2009).

Table 5: The final regression model for the retained variables with *likelihood ratio (at d.f = 2) p value ? 0.05 in Table. 4 that have shown significant association with S. gallinarum infection in layers.

Variable (level)


95% C.I





Multiple flocks (Yes/No)


2 – 30.4





Other birds (Yes/No)


2.72 – 45.3












Likelihood ratio







*Likelihood ratio (d.f = 2) p value represents the degree of deviation of the predictor variable from the model;
d.f represents degree of freedom (McDonald 2009).


This study has established the seroprevalence and risk factors for S. gallinarum infection in smallholder layers flocks in Mwanza City. The TA test was used as confirmatory test and therefore the discussion in this survey mainly focused on the flock-level seroprevalence obtained using this test in relation to responsible risk factors. The individual level was not deeply considered because the information on the responsible risk factors was collected based on the flock-level as it was not possible to get the epidemiological information for each individual layers. The seroprevalence of S. gallinarum has been reported elsewhere in the world. In Bangladesh, Hossain et al (2010) reported a highest individual seroprevalence of S. gallinarum-pullorum (37.6%) in adult compared to (16.7%) in the young and highest in large flocks (34.3%) compared to small flocks (21.3%). Studies in Nigeria indicated 18.6% flock seroprevalence of S. gallinarum (Okwori et al 2007).  In Uganda, Waiswa et al (2006) reported the individual seroprevalence of S. gallinarum of 27.9% in non-vaccinated chickens. In Tanzania, Mdegela et al (2000) recorded individual seroprevalence of S. gallinarum 18.4% in commercial layers. The results of the present study showed 51% flock-level seroprevalence and 15% individual layers seroprevalence. There was no significant difference in flock-level seroprevalence between Ilemela and Nyamagana districts (P>0.05). This can be accounted to the reason that, the two districts have the same climatic condition, bird’s keepers and type of birds kept.


On the other hand, some studies have investigated risk factors for S. gallinarum infection elsewhere in the world. In Uganda, downtime, flock size and bird type were associated with S. gallinarum infection (Nasinyama et al 1997). Snow et al (2010) reported size of holding, production type, source of feed and the distance to a nearest farm as risk factors, and that the presence of cats and dogs reduced the risk of Salmonella infection in layers in Great Britain. A study by Mbuko et al (2009) associated age, type and poultry species with the prevalence of fowl typhoid in Nigeria. The present survey investigated several variables that were thought to have significant association with S. gallinarum infection in layers. Results of the survey showed that, there was no significant difference in seroprevalence between Rhode Island Red and Black Australop layers (p>0.05). This finding agrees with earlier results on the study conducted in Morogoro, Tanzania by Wambura et al (2006). On top of that, there was no significant difference in seroprevalence between the flocks kept where equipment are disinfected or not disinfected. However, the study indicated that, disinfection has some protective effects to S. gallinarum infection (OR = 0.303) as given in table 4. Moreover, the S. gallinarum seroprevalence was relatively high in large flocks although there was no significant difference. These findings were comparable with those reported in Uganda by Nasinyama et al (1997) and in Bangladesh by Hossain et al (2010). Because in large flocks, there is a requirement of more labour forces and equipment to manage flocks effectively, the possibility of pathogens contamination is high as compared to small flocks regardless of the housing type.


In addition, there was no significant difference in seroprevalence of S. gallinarum between flock age groups (p>0.05) although the seroprevalence was high in growers (67%) and low in starter (33%) as compared to layers (58%). Mbuko et al (2009) reported similar trend of results in Nigeria. However, these findings contradicted the findings reported by Ahmed et al (2008) who reported high seroprevalence in starter birds in Bangladesh. In this present study, the high seroprevalence in growers and layers may be accounted to horizontal transmission between flocks and other birds because of poor biosecurity in relation to the bird time stay in the flocks. On top of that, the flock seroprevalence was high (75%) in flocks that bird keepers had long time of involvement in poultry keeping activities although there was no significant difference. The reason might be increased accumulation of pathogens in the environment and hence more contamination of the fomites with increase in keeping time.


No significant difference was observed between flocks that used own made feeds at home, own made but mixed in the public mixer machine and purchased readymade feeds from the animal feed mills (p>0.05). However, S. gallinarum seroprevalence was relatively high in those flocks who used own made feeds at home (57%) and low in those who purchased readymade feeds (43%). Likewise, no significant difference of S. gallinarum infection was observed between flocks and other variables investigated in this survey.


According to the final model, regression analysis showed a significant association between S. gallinarum infection and presence of other birds (OR = 11.1 and p = 0.0008).  Other birds such as local chickens, ducks, pigeons, broilers and turkeys can as well be a source of S. gallinarum infection. For instance, in local chicken, ducks and pigeons, S. gallinarum infection can occur and they can be potential carriers as well. Although the most known important source of S. gallinarum infection are hatcheries as reported by Waiswa et al (2006), under poor biosecurity situation, horizontal transmission between bird-to-bird and flock-to-flock cannot be under estimated. Horizontal transmission can occur through respiratory and oral routes. Birds can ingest bacteria through contaminated feeds, water and litter because of environmental contaminations mainly by faeces from the infected or carrier birds.


Furthermore, presence of multiple flocks of layers in the same premise (OR = 7.81 and p = 0.00311) has shown the significant association with S. gallinarum infection in layers. This finding supported the facts that, horizontal transmission can occur between one flock and another because of poor biosecurity measures on the farms. For instance, frequent change of attendants and/or dirty attendants, vermin, other birds and wild birds, sharing of equipment between flocks, lack of footbath and personal restriction (Waiswa et al 2006). Of all 63 visited flocks, 71% were attended by any of the family member, so there was no specific attendant for each flock.


Some variables have showed confounding effects to the above two significantly associated variables with the S. gallinarum infection after successional refitting in the final model: flock age, flock size, equipment disinfection, keeping time and keeping records were the confounding variables for other birds variable while flock size and equipment disinfection confounded multiple flocks variable in this survey.


Findings from this study have clearly demonstrated that S. gallinarum exposure to birds is prevalent in smallholder layers in Mwanza City, and therefore, poultry keepers need to abide to the principles of good hygienic poultry husbandry practices. The institutionalisation of repeated serological testing and removal of positive reactor birds in hatcheries and smallholder flocks is highly recommended. Layers should be kept in isolated flocks in order to prevent horizontal transmission of S. gallinarum between flock and flock or other birds.


I wish to thank the Southern African Development Community – Trans-boundary Animal Diseases (SADC-TADS) secretariat for funding this survey. I am sincerely gratefully to Mwanza City poultry keepers and technical staff of Tanzania Veterinary Laboratory Agency (TVLA), Mwanza Centre and Departments of Veterinary Medicine and Public Health, Microbiology and Parasitology of Sokoine University of Agriculture for their co-operation. Finally yet importantly, I am thankful to the Director of Veterinary Services for Ministry of Livestock and Fisheries Development of Tanzania for allowing this study to be undertaken.


Ahmed A K M, Islam M T, Haider M G and Hossain M M 2008 Seroprevalence and pathology of naturally infected Salmonellosis in poultry with isolation and identification of causal agents.  Journal of the Bangladesh Agricultural University, 6(2): 327–334.

Bartellet J E, II Kotrlik J W and Higgins C C 2001 Organisational Research: Determining Appropriate Sample Size in Survey Research. Information Technology, Learning and Performance Journal, 19(1): 43–50.

Batista D FA, Neto O C F, Lopes P D, Almeida A M, Barrow P A and Berchieri A Jr 2013 Polymerase chain reaction assay based on ratA gene allows differentiation between Salmonella enterica subsp. enterica serovar Gallinarum biovars Gallinarum and Pullorum. Journal of Veterinary Diagnostic Investigation, 25(2): 259 –262.

Coulombier D, Fagan R, Hathcock L and Smith C 2001 Epi Info 6 version 6.04. A word Processing, Database and Statistical Programme for Public Health. Centres for Disease Control and Prevention, Atlanta.

Grimont P A D and Weill F X 2007 Antigenic formulae of the Salmonella serovars. WHO Collaborating Centre for reference and research on Salmonella. Institut Pasteur, Paris France. Pp. 6 and 166.

Hosmer D W, Lemeshow S and Sturdivant R X 2013 Applied Logistic Regression (3rd ed). John Wiley & Sons, Inc., Hoboken, New Jersey; pp. 1-225.

Hossain K M 2010 Seroprevalence of Salmonella and Mycoplasma gallisepticum Infection in Chickens in Rajshahi and surrounding Districts of Bangladesh. International Journal of Biology, 2(2): 74-80.

James G, Witten D, Hastie T and Tibshirani R 2013 An Introduction to Statistical Learning, with application in R. (Edited by Casella  G, Fienberg S and Olkin I), Springer Science+Business Media New York. Pp 145 – 153; 220 – 224.

Mbuko I J, Raji M A, Ameh J, Saidu L, Musa W I and Abdul P A 2009 Prevalence and seasonality of fowl typhoid disease in Zaria-Kaduna State, Nigeria. Journal of Bacteriology Research 1(1): 001-005.

McDonald J H 2009 Handbook of Biological Statistics (2nd ed.). Sparky House Publishing, Baltimore, Maryland. pp. 247-255.

Mdegela R H 1998 Epidemiology of Salmonella gallinarum infection in chickens in Tanzania. MSc Thesis, Royal Veterinary and Agricultural University, Copenhagen.

Mdegela R H, Yongolo M G S, Minga U M and Olsen J E 2000 Molecular epidemiology of S. gallinarum in chickens in Tanzania. Avian Pathology, 29(5): 457-463.

Minga U M and Nkini R P 1986 Disease as a major constraint to chicken production in Tanzania. In: Proc. of the 13th Scientific Conference of the Tanzania Society of Animal Production, 13: 284–292.

Minga U M, Mtambo M M A, Katule A M, Mutayoba S K, Mwalusanya N A, Lawrence P, Mdegela R H and Olsen J E 2001 Improving the Health and Productivity of the Rural Chicken in Africa: Research and Development Efforts in Tanzania. In: Proceedings of SADC Planning Workshop on Newcastle Disease Control in Village Chickens, (eds. Alders, R.G. and Spradbrow, P.B.).pp. 6 - 9 March 2000 Maputo Mozambiquie, 134–139.

Nasinyama G W, McEwen S A, Waltner-Toews D, Gyles C L, Wilson J, Opuda-Asibo J and Poppe C 1997 Factors associated with occurrence of Salmonella in broiler and layer flocks of chickens in Kampala district, Uganda. Épidémiologie et Santé Animale, 4(24): 31-32.

OIE 2012 Fowl typhoid and Pullorum disease. In: Manual of Diagnostic Tests and Vaccines for Terrestrial Animals. OIE, Paris.

Okwori A E J, Hashimu G A, Adetunji J A, Okeke I O and Junaid S A 2007 Serological Survey of Salmonella gallinarum Antibody In Chickens Around Jos, Plateau State, Nigeria. Online Journal of Health and Allied Sciences, 6(2).

Oliveira G H, Berchieri A Jr, Montassier H J and Fernandes A C 2004 Assessment of Serological Response of Chickens to Salmonella Gallinarum and Salmonella Pullorum by Elisa. Brazilian Journal of Poultry Science, 6(2): 111 – 115.

Paiva J B, Calvillini J S, Silva M D, Almeida M A, Angela H L and Berchieri Junior A 2009 Molecular differentiation of Salmonella Gallinarum and Salmonella Pullorum by RFLP of fliC gene from Brazilian isolates. Brazilian Journal of Poultry Science, 11(4): 271–276.

Proux K, Humbert F, Jouy E, Houdayer C, Lalande F, Oger A and Salvat G 2002 Improvements required for the detection of Salmonella Pullorum and Gallinarum. Canadian Journal of Veterinary Research, 66(3): 151–157.

Ribeiro S A M, Paiva J B, Zotesso F, Lemos M V F and Berchieri A Jr 2009 Molecular differentiation between Salmonella enterica subsp enterica serovar Pullorum and Salmonella enterica subsp enterica serovar Gallinarum. Brazilian Journal of Microbiology, 40:184-188

Snow L C, Davies R H, Christiansen K H, Carrique-Mas J J, Cook A J C and Evans S J 2010 Investigation of risk factors for Salmonella on commercial egg-laying farms in Great Britain, 2004-2005. Veterinary Record, 166:579-586.

Soria M C, Soria M A and Bueno D J 2012 Comparison of 2 culture methods and PCR assays for Salmonella detection in poultry faeces. Poultry Science, 91 :616–626.

Stevenson M 2012 An Introduction to Veterinary Epidemiology. EpiCentre, IVABS, Massey University, Palmerston North New Zealand. pp. 70 –77.

Waiswa P, Lubowa N M and Nakavuma J L 2006 Seroprevalence and excretion status of Salmonella pullorum-gallinarum in chicken around Kampala Uganda. Africa Journal of Animal and Biomedical Sciences, 1 (1): 46-55.

Wambura P N, Makuri H G, Hoza A S, Tuntufye H and Mellau L S B 2006 Application of Microagglutination test for detection of antibodies to Salmonella gallinarum in commercial layers chickens. Bulletin of Animal Health and Production in Africa, 54: 124-130.

Wu K Y, Liu G R, Liu W Q, Wang A Q, Zhan S, Sanderson K E, Johnston R N and Liu S L 2005 The Genome of Salmonella enterica Serovar Gallinarum: Distinct Insertions/Deletions and Rare Rearrangements. Journal of Bacteriology, 187(14): 4720–4727.

Zanella A 2007 Poultry Disease Manual: Characteristics and Control of Infections. Neuva Biological Products, Bologna. 93pp.

Received 5 July 2014; Accepted 12 September 2014; Published 3 October 2014

Go to top