Zoonoses and Public Health
ORIGINAL ARTICLE
Predictors for Risk Factors for Spread of Avian Influenza Viruses by Poultry Handlers in Live bird markets in Uganda H. Kirunda1, K. K. Mugimba2, B. Erima3, D. Mimbe3, D. K. Byarugaba3 and F. Wabwire-Mangen4 1 2 3 4
National Livestock Resources Research Institute, Tororo, Uganda College of Veterinary Medicine, Animal Resources and Biosecurity, Makerere University, Kampala, Uganda Makerere University Walter Reed Project, Kampala, Uganda School of Public Health, College of Health Sciences, Makerere University, Kampala, Uganda
Impacts
• Live bird markets in Uganda are popular for poultry trade, but could pose
risk for introduction and spread of avian influenza in animals and humans.
• The study investigated socio-demographic characteristics of bird handlers •
that were predictors for risk factors in introduction and spread of avian influenza viruses in live bird markets in Uganda. The study revealed that educational background, residence and region of bird handlers were significant predictors for risky hygiene and management practices that increased risk of spread of influenza viruses, providing useful information for development of strategies for prevention of avian influenza outbreaks.
Keywords: Avian influenza viruses; bird handlers; live bird markets; risky practices; risk factor Correspondence: H. Kirunda. National Livestock Resources Research Institute (NaLIRRI), P.O. Box 96, Tororo, Uganda. Tel.: +256772927430; E-mail: [email protected] Received for publication May 9, 2014 doi: 10.1111/zph.12151
334
Summary Live bird markets (LBMs) are essential for marketing poultry, but have been linked to many outbreaks of avian influenza (AI) and its spread. In Uganda, it has been observed that demographic characteristics of poultry traders/handlers influence activities and decision-making in LBMs. The study investigated the influence of socio-demographic characteristics of poultry handlers: age, sex, religion, educational background, level of income, location of residence and region of operation on 20 potential risk factors for introduction and spread of AI in LBMs. Study sites included 39 LBMs in the four regions of Uganda. Data was collected using a semi-structured questionnaire administered to 424 poultry handlers. We observed that background of education was a predictor for slaughter and processing of poultry in open sites. Location of residence was associated with slaughter of poultry from open sites and selling of other livestock species. Region influenced stacking of cages, inadequate cleaning of cages, feeders and drinkers, and provision of dirty feed and water. Specifically, bird handlers with secondary level of education (OR = 12.9, 95% CI: 2.88– 57.4, P < 0.01) were more likely to be involved in open site slaughter of poultry than their counterparts without formal education. Comparatively, urbanite bird handlers were less likely to share poultry equipment (OR = 0.4, 95% CI: 0.22–0.63, P < 0.01) than rural resident handlers. Poultry handlers in Northern were 3.5 times more likely to practise insufficient cleaning of cages (OR = 3.5, 95% CI: 1.52–8.09) compared to those in Central region. We demonstrated that some socio-demographic characteristics of poultry handlers were predictors to risky practices for introduction and spread of AI viruses in LBMs in Uganda.
© 2014 Blackwell Verlag GmbH Zoonoses and Public Health, 2015, 62, 334–343
H. Kirunda et al.
Introduction
Factors for Spread of Influenza in Bird Markets in Uganda
da et al. (2011) particularly observed that the level of education, income levels and sex of the poultry traders influenced activities and decision-making in LBMs in Uganda. While several public health strategic interventions are required for effective prevention of occurrence of AI in any country (WHO, 2006), epidemiological information on spread of these viruses in Uganda is largely lacking. This study was designed to establish existence of factors that could increase risk for introduction and spread of AI viruses in the LBMs in Uganda.
Avian influenza (AI) viruses are endemic in many parts of Asia and in Egypt. The wide genetic diversity and the potential for recombination with human influenza strains continue to pose major animal biosecurity and public health concerns (Li et al., 2010; Fournie et al., 2011). Although measures for effective control of AI epidemics in developed countries exist, the high financial outlay makes such strategies ineffective in resource poor settings, where most poultry is raised by smallholder owners. As poultry has been associated with introduction and spread of avian influenza viruses in Asia, Europe and Africa (Kung et al., 2007), ineffective measures may create conditions that favour silent spread of the virus within the poultry sector (Yupiana et al., 2010). Live bird markets (LBMs) are essential for marketing poultry in many developing countries, but have been shown to play an important role in the introduction and spread of AI viruses in birds and humans (Kung et al., 2007). The role of poultry movements and trading activities in the occurrence of AI has particularly been confirmed in recent studies (Desvaux et al., 2011). Live birds are presumed to constitute the highest risk because of virus replication, virus shed into the environment and movement over long distances (Cardona et al., 2009). Contact with infected poultry or surfaces and objects contaminated by poultry droppings increase chances of exposure of AI viruses to humans (Aditama et al., 2011). In Uganda, live birds are sold in the numerous LBMs that are spread across the country. In these markets, bird handlers (poultry handlers/traders) are of different sex, age groups, levels of education and income levels (Emuron et al., 2010). Some of these markets operate on a daily basis, while others operate on designated days, once a week or fortnightly. Poultry handlers in urban areas may confine birds in cages, but those in rural areas mainly operate in the open with birds mostly restrained on the ground. Whereas no outbreak of HPAI has been reported in Uganda, a recent study by Kirunda et al. (2014) has reported occurrence of AI viruses in poultry sold in LBMs in the country. In previous studies, LBMs have been reported to be sources of AI virus infection in humans (Wan et al., 2011). Existing literature indicates that age, sex/gender, education/knowledge and religion can influence the health behaviours, hygiene practices and utilization of advice in humans (Amanda et al., 2009; McNamara et al., 2010; Huang et al., 2013), leading to the hypothesis that hygiene and management practices by poultry handlers in LBMs in Uganda could equally be associated with these demographic characteristics. Emuron et al. (2010) and Natukun-
All quantitative data was entered in EpiData and analysed using SPSS statistical program. Analysis was performed at univariate, bivariate and multivariate levels for
© 2014 Blackwell Verlag GmbH Zoonoses and Public Health, 2015, 62, 334–343
335
Materials and Methods Study area The study was undertaken in 39 LBMs. These included markets observed to have the lowest level of biosecurity in an earlier study by Kirunda et al. (2014). The markets in the current study were located in 29 of 112 districts in the four regions of Uganda including Central, Eastern, Northern and Western. The sampled markets were in the rural and urban parts of the country where birds were sold either in live or slaughtered form. Study design and sample size determination This cross-sectional study was undertaken in 2 months from March to April, 2011. Data were collected using a semi-structured questionnaire. All consenting live poultry handlers among those met in a market on the day of visit were interviewed. In the study, only four poultry handlers among those asked to participate declined to be interviewed. A total of 424 bird handlers consented to the interview. Study variables The seven independent study variables included the sociodemographic factors of the poultry handlers including age, sex, religion, level of education, level of income, location of residence and region of operation. The 20 outcome variables included hygiene and management practices previously associated with increased risk of introduction and spread of AI viruses (Table 1 and Fig. 1). These were predictors of poor biosecurity in a study by Kirunda et al. (2014), which are only a few of those reported in other African countries (FAO, 2013). Data management and analysis
Factors for Spread of Influenza in Bird Markets in Uganda
H. Kirunda et al.
Table 1. Socio-demographic characteristics of bird handlers in live bird markets in Uganda Regional proportions of bird handlers with the characteristic Characteristic
Risk factors for spspread of avian influenza
Sex Male Female Age group Adolescent Adult Religion Catholic Islam Anglican (Protestant) Pentecostal Educational background No education Primary education Secondary education+ Residence Rural area Town area Possession of extra business Had other business Had no other business
Central (n = 50) (%)
Eastern (n = 102) (%)
Northern (n = 175) (%)
Western (n = 97) (%)
90.0 10.0
99.0 1.0
100 –
100 –
10.0 90.0
3.9 96.1
– 100
– 100
48.0 18.0 10.0 24.0
20.6 45.1 10.8 23.5
46.9 25.1 22.9 5.1
40.2 13.4 29.9 16.5
0.0 76.0 24.0
6.9 84.3 8.8
6.3 85.7 8.0
6.2 84.5 9.3
18.0 82.0
62.7 37.3
100.0 –
68.0 32.0
16.0 84.0
22.5 77.5
15.4 84.8
25.8 74.2
Never frequently washed hands Never wore protective clothing Never disinfected returned troughs Shared equipment Sold other livestock species Stored feed in open containers Feed/water provided were dirty Never cleaned troughs Never cleaned cages Never lined stacked cages Cages were stacked Slaughtered/processed birds Unsold birds returned to farm/home Never separated sick birds Never separated birds by species Never quarantined new birds Birds from un-reliable source Never kept records >20 birds in a cage Allowed buyer
ORIGINAL ARTICLE
Predictors for Risk Factors for Spread of Avian Influenza Viruses by Poultry Handlers in Live bird markets in Uganda H. Kirunda1, K. K. Mugimba2, B. Erima3, D. Mimbe3, D. K. Byarugaba3 and F. Wabwire-Mangen4 1 2 3 4
National Livestock Resources Research Institute, Tororo, Uganda College of Veterinary Medicine, Animal Resources and Biosecurity, Makerere University, Kampala, Uganda Makerere University Walter Reed Project, Kampala, Uganda School of Public Health, College of Health Sciences, Makerere University, Kampala, Uganda
Impacts
• Live bird markets in Uganda are popular for poultry trade, but could pose
risk for introduction and spread of avian influenza in animals and humans.
• The study investigated socio-demographic characteristics of bird handlers •
that were predictors for risk factors in introduction and spread of avian influenza viruses in live bird markets in Uganda. The study revealed that educational background, residence and region of bird handlers were significant predictors for risky hygiene and management practices that increased risk of spread of influenza viruses, providing useful information for development of strategies for prevention of avian influenza outbreaks.
Keywords: Avian influenza viruses; bird handlers; live bird markets; risky practices; risk factor Correspondence: H. Kirunda. National Livestock Resources Research Institute (NaLIRRI), P.O. Box 96, Tororo, Uganda. Tel.: +256772927430; E-mail: [email protected] Received for publication May 9, 2014 doi: 10.1111/zph.12151
334
Summary Live bird markets (LBMs) are essential for marketing poultry, but have been linked to many outbreaks of avian influenza (AI) and its spread. In Uganda, it has been observed that demographic characteristics of poultry traders/handlers influence activities and decision-making in LBMs. The study investigated the influence of socio-demographic characteristics of poultry handlers: age, sex, religion, educational background, level of income, location of residence and region of operation on 20 potential risk factors for introduction and spread of AI in LBMs. Study sites included 39 LBMs in the four regions of Uganda. Data was collected using a semi-structured questionnaire administered to 424 poultry handlers. We observed that background of education was a predictor for slaughter and processing of poultry in open sites. Location of residence was associated with slaughter of poultry from open sites and selling of other livestock species. Region influenced stacking of cages, inadequate cleaning of cages, feeders and drinkers, and provision of dirty feed and water. Specifically, bird handlers with secondary level of education (OR = 12.9, 95% CI: 2.88– 57.4, P < 0.01) were more likely to be involved in open site slaughter of poultry than their counterparts without formal education. Comparatively, urbanite bird handlers were less likely to share poultry equipment (OR = 0.4, 95% CI: 0.22–0.63, P < 0.01) than rural resident handlers. Poultry handlers in Northern were 3.5 times more likely to practise insufficient cleaning of cages (OR = 3.5, 95% CI: 1.52–8.09) compared to those in Central region. We demonstrated that some socio-demographic characteristics of poultry handlers were predictors to risky practices for introduction and spread of AI viruses in LBMs in Uganda.
© 2014 Blackwell Verlag GmbH Zoonoses and Public Health, 2015, 62, 334–343
H. Kirunda et al.
Introduction
Factors for Spread of Influenza in Bird Markets in Uganda
da et al. (2011) particularly observed that the level of education, income levels and sex of the poultry traders influenced activities and decision-making in LBMs in Uganda. While several public health strategic interventions are required for effective prevention of occurrence of AI in any country (WHO, 2006), epidemiological information on spread of these viruses in Uganda is largely lacking. This study was designed to establish existence of factors that could increase risk for introduction and spread of AI viruses in the LBMs in Uganda.
Avian influenza (AI) viruses are endemic in many parts of Asia and in Egypt. The wide genetic diversity and the potential for recombination with human influenza strains continue to pose major animal biosecurity and public health concerns (Li et al., 2010; Fournie et al., 2011). Although measures for effective control of AI epidemics in developed countries exist, the high financial outlay makes such strategies ineffective in resource poor settings, where most poultry is raised by smallholder owners. As poultry has been associated with introduction and spread of avian influenza viruses in Asia, Europe and Africa (Kung et al., 2007), ineffective measures may create conditions that favour silent spread of the virus within the poultry sector (Yupiana et al., 2010). Live bird markets (LBMs) are essential for marketing poultry in many developing countries, but have been shown to play an important role in the introduction and spread of AI viruses in birds and humans (Kung et al., 2007). The role of poultry movements and trading activities in the occurrence of AI has particularly been confirmed in recent studies (Desvaux et al., 2011). Live birds are presumed to constitute the highest risk because of virus replication, virus shed into the environment and movement over long distances (Cardona et al., 2009). Contact with infected poultry or surfaces and objects contaminated by poultry droppings increase chances of exposure of AI viruses to humans (Aditama et al., 2011). In Uganda, live birds are sold in the numerous LBMs that are spread across the country. In these markets, bird handlers (poultry handlers/traders) are of different sex, age groups, levels of education and income levels (Emuron et al., 2010). Some of these markets operate on a daily basis, while others operate on designated days, once a week or fortnightly. Poultry handlers in urban areas may confine birds in cages, but those in rural areas mainly operate in the open with birds mostly restrained on the ground. Whereas no outbreak of HPAI has been reported in Uganda, a recent study by Kirunda et al. (2014) has reported occurrence of AI viruses in poultry sold in LBMs in the country. In previous studies, LBMs have been reported to be sources of AI virus infection in humans (Wan et al., 2011). Existing literature indicates that age, sex/gender, education/knowledge and religion can influence the health behaviours, hygiene practices and utilization of advice in humans (Amanda et al., 2009; McNamara et al., 2010; Huang et al., 2013), leading to the hypothesis that hygiene and management practices by poultry handlers in LBMs in Uganda could equally be associated with these demographic characteristics. Emuron et al. (2010) and Natukun-
All quantitative data was entered in EpiData and analysed using SPSS statistical program. Analysis was performed at univariate, bivariate and multivariate levels for
© 2014 Blackwell Verlag GmbH Zoonoses and Public Health, 2015, 62, 334–343
335
Materials and Methods Study area The study was undertaken in 39 LBMs. These included markets observed to have the lowest level of biosecurity in an earlier study by Kirunda et al. (2014). The markets in the current study were located in 29 of 112 districts in the four regions of Uganda including Central, Eastern, Northern and Western. The sampled markets were in the rural and urban parts of the country where birds were sold either in live or slaughtered form. Study design and sample size determination This cross-sectional study was undertaken in 2 months from March to April, 2011. Data were collected using a semi-structured questionnaire. All consenting live poultry handlers among those met in a market on the day of visit were interviewed. In the study, only four poultry handlers among those asked to participate declined to be interviewed. A total of 424 bird handlers consented to the interview. Study variables The seven independent study variables included the sociodemographic factors of the poultry handlers including age, sex, religion, level of education, level of income, location of residence and region of operation. The 20 outcome variables included hygiene and management practices previously associated with increased risk of introduction and spread of AI viruses (Table 1 and Fig. 1). These were predictors of poor biosecurity in a study by Kirunda et al. (2014), which are only a few of those reported in other African countries (FAO, 2013). Data management and analysis
Factors for Spread of Influenza in Bird Markets in Uganda
H. Kirunda et al.
Table 1. Socio-demographic characteristics of bird handlers in live bird markets in Uganda Regional proportions of bird handlers with the characteristic Characteristic
Risk factors for spspread of avian influenza
Sex Male Female Age group Adolescent Adult Religion Catholic Islam Anglican (Protestant) Pentecostal Educational background No education Primary education Secondary education+ Residence Rural area Town area Possession of extra business Had other business Had no other business
Central (n = 50) (%)
Eastern (n = 102) (%)
Northern (n = 175) (%)
Western (n = 97) (%)
90.0 10.0
99.0 1.0
100 –
100 –
10.0 90.0
3.9 96.1
– 100
– 100
48.0 18.0 10.0 24.0
20.6 45.1 10.8 23.5
46.9 25.1 22.9 5.1
40.2 13.4 29.9 16.5
0.0 76.0 24.0
6.9 84.3 8.8
6.3 85.7 8.0
6.2 84.5 9.3
18.0 82.0
62.7 37.3
100.0 –
68.0 32.0
16.0 84.0
22.5 77.5
15.4 84.8
25.8 74.2
Never frequently washed hands Never wore protective clothing Never disinfected returned troughs Shared equipment Sold other livestock species Stored feed in open containers Feed/water provided were dirty Never cleaned troughs Never cleaned cages Never lined stacked cages Cages were stacked Slaughtered/processed birds Unsold birds returned to farm/home Never separated sick birds Never separated birds by species Never quarantined new birds Birds from un-reliable source Never kept records >20 birds in a cage Allowed buyer
