Social Network Analysis for Assessment of Avian Influenza Spread and Trading Patterns of Backyard Chickens in Nakhon Pathom, Suphan Buri and Ratchaburi, Thailand

From isolation to revival: trade recovery amid global health crises

BACKGROUND

The COVID-19 pandemic has highlighted the importance of designing effective trade recovery measures in response to global health events (GHEs). This study combines international trade risk management theory and multi-case comparative analysis of past GHEs to present a theoretical framework for designing national trade recovery measures for future events.

RESULTS

The research finds that during GHEs, trade risks shift to fundamental uncertainty, requiring spatial-temporal-subject dimension recovery measures. The study suggests changing the focus of trade recovery policy design from emergency-oriented and single-dimension measures to reserve-oriented and enduring-effect measures of comprehensive dimensions at micro- and macroeconomic levels.

CONCLUSION

The study contributes to the debate on managing trade risks in times of crisis, where there is a need to develop effective trade recovery measures that account for the complexities of global trade and the unique challenges of GHEs. The findings provide practical guidance for trade officials and policymakers to design measures in response to GHEs to improve a country's overall trade recovery.

Post-larvae movement network of marine shrimp during the 2013 outbreak of acute hepatopancreatic necrosis disease in Thailand

This study aimed to assess the movement patterns of post-larvae (PL) of whiteleg shrimp and black tiger shrimp, which were linked to the epidemic of acute hepatopancreatic necrosis disease (AHPND) in Thailand in 2013, using social network analysis. Hatcheries or nurseries were assigned as origin nodes, whereas cultivated ponds were designated as destination nodes. From 376 farms, data of 1300 ponds were collected from network data using a questionnaire. We found that 1477 nodes with 1315 ties appeared in this study. These were divided into a network of cases and controls that presented 538 nodes with 436 ties and 1013 nodes with 879 ties, respectively. The top four out-degree centrality values in the entire partial network, cases, and controls were used to analyze ego-alter similarity. The results showed that each PL provider played a different role in the association with the disease sender. A simulated network using exponential random graph models presented edges as a significant microstructure. In addition, our analysis indicated that the hatchery or nursery (PL provider) is an important node for AHPND occurrence. In conclusion, this study led us to understand the source of PL, which plays a key role in the surveillance, prevention, and control of AHPND.

Avian Influenza Risk Environment: Live Bird Commodity Chains in Chattogram, Bangladesh

In this paper, we identify behaviours in live bird commodity chains in Chattogram, Bangladesh, which may influence the risk of pathogen emergence and transmission: the nature of poultry trade, value appropriation and selling sick or infected birds. Examining the reasons why actors engage in these behaviours, we emphasise the politics of constraints within a context of real-world decisions, governed by existential and pragmatic agency. Focusing on contact zones and entanglement, analysing patron-client relationships and precarious circumstances, we argue that agency and structure specific to the Bangladeshi context produce a risk environment. Structural constraints may reinforce risky occupational practises and limit individual agency. Structural constraints need to be addressed in order to tackle animal and zoonotic disease risk along live animal commodity chains.

Social network analysis for better understanding of influenza

INTRODUCTION

The objective of this study is to improve the understanding of spatial spreading of complicated cases of influenza that required hospitalizations, by creating heatmaps and social networks. They will allow to identify critical hubs and routes of spreading of Influenza, in specific geographic locations, in order to contain infections and prevent complications, that require hospitalizations.

MATERIAL AND METHODS

Data were downloaded from the Healthcare Cost and Utilization Project (HCUP) - SID, New York State database. Patients hospitalized with flu complications, between 2003 and 2012 were included in the research (30,380 cases). A novel approach was designed, by constructing heatmaps for specific geographic regions in New York state and power law networks, in order to analyze distribution of hospitalized flu cases.

RESULTS

Heatmaps revealed that distributions of patients follow urban areas and big roads, indicating that flu spreads along routes, that people use to travel. A scale-free network, created from correlations among zip codes, discovered that, the highest populated zip codes didn't have the largest number of patients with flu complications. Among the top five most affected zip codes, four were in Bronx. Demographics of top affected zip codes were presented in results. Normalized numbers of cases per population revealed that, none of zip codes from Bronx were in the top 20. All zip codes with the highest node degrees were in New York City area.

DISCUSSION

Heatmaps identified geographic distribution of hospitalized flu patients and network analysis identified hubs of the infection. Our results will enable better estimation of resources for prevention and treatment of hospitalized patients with complications of Influenza.

CONCLUSION

Analyses of geographic distribution of hospitalized patients with Influenza and demographic characteristics of populations, help us to make better planning and management of resources for Influenza patients, that require hospitalization. Obtained results could potentially help to save many lives and improve the health of the population.

Use of network analysis multidrug-resistant tuberculosis contact investigation in Kanchanaburi, Thailand

OBJECTIVE

To characterise MDR-TB outbreak and incorporate social network analysis with contact investigation to detect case-contact linkages and clusters.

METHODS

MDR-TB cases registered in the district hospital between October 2012 and September 2015 were interviewed and their contacts were investigated. A relationship-based weighted network was constructed.

RESULTS

Among 43 interviewed MDR-TB cases, 20 (47%) were male, five (12%) were asymptomatic (and discovered incidentally) and 22 (51%) had underlying diseases. From the documented 115 contacts, 61 (53%) were household contacts and 49 (43%) were close (non-household) contacts; 70 (61%) were screened for TB using various tests. In this network, we prioritised 37 contacts connected with more than one MDR-TB patient. The largest cluster was identified in the pharmacy unit of the hospital.

CONCLUSION

This investigation yielded a significant number of MDR-TB contacts, and social network analysis facilitated the prioritisation for screening. Social network analysis is useful and feasible in this program setting and complements MDR-TB contact investigation.

The implementation of cattle market closure strategies to mitigate the foot-and-mouth disease epidemics: A contact modeling approach

Foot-and-mouth disease (FMD) is one of the most endemic diseases in livestock worldwide. The disease occurrence generally results in a huge economic impact. The virus may distribute across countries or even continents along the contact network of animal movements. The present study, therefore, aimed to explore a cattle movement network originated in Tak, a Thailand-Myanmar bordered province and to demonstrate how FMDV spread among the nodes of market, source and destination. Subsequently, we examined the effectiveness of market closure intervention. The market-market (M-M) network was constructed to highlight the inter-market connections and the FMDV was modeled to spread along the trade chain. Four market closure scenarios based on rapidness and duration of implementation were examined. Our results indicate that two of the three major markets located in the province were highly connected and a strongly connected component was identified. The intra-provincial animal movements, which were currently overlooked, should be moved into sights as most of the high-risk sources for FMD epidemics were recognized in a close proximity to the cattle markets. Simultaneously, remote destinations across the country were identified. The inter-provincial animal movement control must be strengthened once FMD outbreak is notified. Based on our simulations, closing markets with low inter-market connectivity may not prevent the spread of FMDV. A selective market closure strategy targeting highly connected markets together with cattle trader tracking system was an alternative approach. However, socio-economic consequences regarding this intervention must be considered.

Using social network analysis to characterize the collaboration network of backyard poultry trainers in ackCalifornia

In order to better understand collaboration among trainers in the backyard poultry community (i.e. feed store managers, youth development programs (i.e. 4-H), veterinarians, government agencies, extension resources and backyard poultry club leaders), Social Network Analysis (SNA) was used as a tool to better characterize and quantify the current collaboration network structure of backyard poultry trainers in California. Invited trainer attendees of two "Train-the-Trainers" poultry workshops (n = 67) held in Northern and Southern California were given a survey that asked them to list contacts that they collaborated with on backyard poultry (BYP) related work. The collaboration network in this study included a total of 109 trainers, 18 practitioners, and 32 individuals who are both trainers and practitioners for a total of 170 nodes (11 individuals did not have affiliation information available). In order to help identify central actors or collaboration leaders, the surveys were analyzed using Social Network Analysis (SNA), which allows for a quantitative analysis of relationships among various stakeholders. While the SNA showed that the existing collaboration network is disconnected with a clustering coefficient of 0.043 and median total degree centrality of 1 (range 9) and therefore not conducive for collaboration, key insights that could help restructure and improve the network were identified. As an example, among different poultry groups, 4-H was identified as the organization with the second highest median coverage score and fifth highest median centrality score. In addition, 4-H group leaders act as both trainers and practitioners. Consequently, outreach to 4-H group leaders throughout the state would potentially have the greatest impact with respect to overall coverage both inside and outside the 4-H network due to their high centrality and boundary spanning roles. Using SNA to strengthen the collaboration network infrastructure of backyard poultry trainers ultimately offers a more targeted approach toward extension for backyard poultry owners, which could ultimately facilitate communication and knowledge-sharing with BYP owners during a disease outbreak.

Exponential random graph models to evaluate the movement of backyard chickens after the avian influenza crisis in 2004-2005, Thailand

The aim of this study was to use exponential random graph models (ERGMs) to explain networks of movement of backyard chickens in provinces which had been hotspots for avian influenza outbreaks in Thailand during 2004-2005. We used structured questionnaires to collect data for the period January to December 2009 from participants who were involved in the backyard chicken farming network in three avian influenza hotspots (Ratchaburi, Suphan Buri, and Nakhon Pathom provinces) in Thailand. From 557 questionnaires, we identified nodes, points of entry and exit from nodes, and activities relating to backyard chicken farming and movement of chickens, and generated ERGMs based on non-festive periods (Model 1) and the Chinese New Year period (Model 2). In Model 1, k-star (the central node is connected to k other nodes) connections were predominant (P <  0.001). In Model 2, the frequency of movement increased by 10.62 times, k-star connections were still predominant (P <  0.001), and the model was scale-free. Hubs were formed from owners/observers in the arenas/training fields, farmers who raised chickens for consumption only, and traders. In conclusion, our models indicated that, if avian influenza was introduced during non-festive periods, the authorities would need to regularly restrict the movement of chickens. However, during high-frequency periods of movement of backyard chickens, authorities would also need to focus on the network hubs. Our research can be used by the relevant authorities to improve control measures and reduce the risk or lessen the magnitude of disease spread during an avian influenza epidemic.

Combining value chain and social network analysis as a viable tool for informing targeted disease surveillance in the rural poultry sector of Zambia

Diseases are among the greatest challenges to the rural poultry sector in sub-Saharan Africa. The lack of a sustainable poultry disease surveillance system and the possible existence of communities and occasions where the interaction between birds is high present an opportunity for targeted surveillance of poultry diseases in these regions. However, the establishment of such a system requires adequate knowledge of the sector in the targeted area. Zambia is an example of a developing country located in the tropics that faces the challenge of frequent poultry disease outbreaks. Consequently, an interview-based survey to study the poultry sector's market chain and social networks was conducted in Eastern Zambia to derive information required for configuring targeted surveillance. This survey involved a poultry value chain analysis that also included an assessment of trading practices to identify biosecurity hot spots within the chain that could be targeted for disease surveillance. A social network analysis of poultry movement within Eastern Zambia was also conducted using whole-network analysis and ego network analysis to identify poultry trade hubs that could be targeted for poultry disease surveillance based on their centrality within the network and their size and influence within their ego networks. Rural farmers, middlemen and market traders were identified as biosecurity risk hot spots whose poultry and utensils could be targeted for disease surveillance within the value chain. Furthermore, social network analysis identified four districts as poultry trade hubs that could be targeted for disease surveillance. This study is the first to formally describe poultry movement networks within Zambia and the surrounding region. Its findings provide data required to implement targeted surveillance in regions where resources are either inadequate or non-existent, and the results provide a deeper understanding of the cultural and practical constraints that influence trade in developing countries.

The use of social network analysis to examine the transmission of Salmonella spp. within a vertically integrated broiler enterprise

To better understand factors influencing infectious agent dispersal within a livestock population information is needed on the nature and frequency of contacts between farm enterprises. This study uses social network analysis to describe the contact network within a vertically integrated broiler poultry enterprise to identify the potential horizontal and vertical transmission pathways for Salmonella spp. Nodes (farms, sheds, production facilities) were identified and the daily movement of commodities (eggs, birds, feed, litter) and people between nodes were extracted from routinely kept farm records. Three time periods were examined in detail, 1- and 8- and 17-weeks of the production cycle and contact networks were described for all movements, and by commodity and production type. All nodes were linked by at least one movement during the study period but network density was low indicating that all potential pathways between nodes did not exist. Salmonella spp. transmission via vertical or horizontal pathways can only occur along directed pathways when those pathways are present. Only two locations (breeder or feed nodes) were identified where the transmission of a single Salmonella spp. clone could theoretically percolate through the network to the broiler or processing nodes. Only the feed transmission pathway directly connected all parts of the network.

Social network analysis of the movement of poultry to and from live bird markets in Bali and Lombok, Indonesia

Highly pathogenic avian influenza H5N1 has resulted in large losses to the Indonesian poultry sector. Evidence suggests that live bird markets (LBMs) play an important role in the epidemiology of the disease. Knowledge of the frequency and type of contact between the various poultry market players should allow animal health authorities to develop a better understanding of factors influencing virus transmission between Indonesian villages. A questionnaire-based cross-sectional survey was conducted in 17 LBMs on the neighbouring Indonesian islands of Bali and Lombok to investigate the movement patterns of poultry to and from markets. Using social network analyses, a network of contacts was created for each island from a total of 413 live poultry traders and 134 customers. Individual nodes with high degree and/or betweenness were identified in each network. The Lombok network was more dense and connected than the Bali network indicating that disease transmission would be more efficient in the Lombok network. Our findings indicate that whilst live poultry are typically transported over relatively short distances of approximately 10 km, it is not uncommon for traders and customers to travel in excess of 100 km to buy or sell poultry, which may facilitate the spread of disease over a large geographical area. This study highlights the different roles markets play in poultry movement networks and their potential for disease dissemination. The identification of highly influential market nodes allows authorities to target HPAI surveillance activities to locations where disease is more likely to be present, which is crucial in low-resource settings.

A traditional cattle trade network in Tak province, Thailand and its potential in the spread of infectious diseases

Cattle in Thailand are traded by cattle dealers in the cattle markets through informal negotiation processes. The present study aimed to explore a traditional cattle trade network and its potential in the spread of infectious diseases. A cross-sectional questionnaire-based survey was conducted in cattle markets located in Tak province, Thailand. A static weighted directed one-mode network was constructed and geographic locations of sources and destinations of animals were plotted. A total of 210 cattle dealers responded to the questionnaires. The cattle trade network contained 112 sources and destinations of animals and 168 trading activities. The highest generalised weighed in- and out-degree centralities and k-core were observed in the same cattle market at 53.4, 106.7 and 4, respectively. Spatially, cattle dealers could transport animals as far as 774.4 km from the original markets and natural barriers such as mountainous areas tended to dictate the trading routes, premises of animals and magnitude of trading activities. Traditional cattle trading patterns described in this study pose some important challenges to veterinary authorities in tailoring effective prevention and control measures for confronting infectious diseases. This study quantitatively analysed the traffic within a network of cattle trade and found a wide range of spatial movements and a high connectivity of markets to sources and destinations of animals, which may accommodate the spread of infectious diseases within the network. This peace time network analysis could help related authorities to better focus on possible infected premises and to immediately respond once an epidemic occurs.

Social Network Analysis of Cattle Movement in Sukhothai Province, Thailand: A Study to Improve Control Measurements

The aim of this study is to analyse the pattern of cattle movement in Sukhothai province, Thailand. A validated questionnaire was applied to 308 respondents related to cattle farming using one-step snowball sampling. The results showed that most of the nodes are farmers who move their animals in the province. The average normalized degree centrality and normalized closeness centrality were low (<0.01 and 0.04, resp.). We found that traders are the nodes with a high value of centrality. This corresponds with the cutpoint analysis results that traders are outstanding. In conclusion, the relevant authorities should focus on the nodes such as traders for controlling disease. However, a measure to detect disease in the early stages needs to be implemented.

Dynamics of avian coronavirus circulation in commercial and non-commercial birds in Asia--a review

It is essential to understand the latest situation regarding avian coronaviruses (ACoVs), commonly referred to as the well-known avian infectious bronchitis virus (IBV), given that new and diverse types of IBV are continually being identified worldwide, particularly ones that are isolated from commercial poultry and associated with a wide range of disease conditions. The existing IBVs continue to evolve in various geographic areas in Asia, which results in the recombination and co-circulation between IBV types. This makes it increasingly difficult to prevent and control IBV infections, despite routine vaccination. Some ACoVs have also been identified in other avian species and they may pose a threat of cross-transmission to commercial sectors. The present review provides an overview of IBV circulation and the dynamic emergence of new variants found throughout Asia via the recombination of IBV strains. In addition to commercial poultry, backyard poultry and free-ranging birds may serve as a ‘hub’ for ACoV transmission within a particular area. These birds may be capable of spreading viruses, either to areas of close proximity, or to remote places via migration and trade.

Collective resistance to HPAI H5N1 surveillance in the Thai cockfighting community: Insights from a social anthropology study

Farmers may organize themselves to collectively manage risks such as animal diseases. Our study shows some evidence of such organization among fighting cock owners in Thailand. Fighting cocks were specifically targeted by HPAI (Highly Pathogenic Avian Influenza) H5N1 surveillance and control measures in Thailand because they were thought to pose a high risk of spreading diseases. In this work, we used a social-anthropological approach to gain an inside view of the issues associated with HPAI H5N1 surveillance in the cockfighting community in Thailand. Based on a qualitative analysis of data collected through in-depth interviews and observation of cockfighters' practices, we found that fighting cock owners share a sense of belonging to the same community based on a common culture, values, interests, practices, and internal rules, including rules to manage poultry diseases. During the HPAI H5N1 outbreaks, these rules may have contributed to mitigating the potential risk associated with the intense movements of fighting cocks inside the country. Nevertheless, this community, despite the high awareness and know-how of its members regarding poultry diseases, has shown a strong reluctance to comply with HPAI surveillance programs. We suggest that this reluctance is due to important gaps between the logic and rationales underlying surveillance and those associated with cockfighting activities. Our study highlights the need for multi and trans-disciplinary research involving the social sciences to analyze interactions between stakeholders and the collective actions implemented by communities to face risks.

Modeling the dynamics of backyard chicken flows in traditional trade networks in Thailand: implications for surveillance and control of avian influenza

In Southeast Asia, traditional poultry marketing chains have been threatened by epidemics caused by the highly pathogenic avian influenza H5N1 (HPAI H5N1) virus. In Thailand, the trade of live backyard chickens is based on the activities of traders buying chickens from villages and supplying urban markets with chicken meat. This study aims to quantify the flows of chickens traded during a 1-year period in a province of Thailand. A compartmental stochastic dynamic model was constructed to illustrate trade flows of live chickens from villages to slaughterhouses. Live poultry movements present important temporal variations with increased activities during the 15 days preceding the Chinese New Year and, to a lesser extent, other festivals (Qingming Festival, Thai New Year, Hungry Ghost Festival, and International New Year). The average distance of poultry movements ranges from 4 to 25 km, defining a spatial scale for the risk of avian influenza that spread through traditional poultry marketing chains. Some characteristics of traditional poultry networks in Thailand, such as overlapping chicken supply zones, may facilitate disease diffusion over longer distances through combined expansion and relocation processes. This information may be of use in tailoring avian influenza and other emerging infectious poultry disease surveillance and control programs provided that the cost-effectiveness of such scenarios is also evaluated in further studies.