Sustaining Global Surveillance and Response to Emerging Zoonotic Diseases (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 Making the Case for Zoonotic Disease Surveillance “The difficulty of uncertainty is that we are dealing with things that are likely to emerge at some time and that need attention. We have to per- suade decision-makers to invest in surveillance systems and other actions to deal with these uncertainties in a flexible and responsive way without being able to tell them, with an absolute precision, when they are going to emerge and what their economic or social cost might be.” —Dr. David Nabarro Senior United Nations System Coordinator for Avian and Human Influenza Special Interview with the Committee (September , 00) Recent emerging zoonotic diseases have had significant impacts in in- dustrialized countries, despite well-developed health systems and sanitary infrastructures (Vorou et al., 2007; Jones et al., 2008; Murphy, 2008), and their impacts have been even more devastating for middle-income and developing countries. When emerging diseases become endemic, they not only continue to cause morbidity and mortality in human and animal populations, but also represent a threat of future epidemics if conditions for explosive transmission are reestablished. Emerging infectious disease trends suggest that the frequency of such disease events that are zoonotic in nature will not lessen in the future (McMichael, 2004; Woolhouse and Gaunt, 2007; Jones et al., 2008). If anything, with increasing human and animal populations and changing environments, the trends are more con- sistent with continual increases in the pace of emergence; however, it is simply unknown where or when they will occur (King, 2004; Morens et al., 2004). Disease surveillance represents the eyes and the ears of the global public health effort, systematically generating information that informs actions to contain, control, and mitigate the consequences in at-risk humans and animals. Detecting diseases early through surveillance and implementing early response measures can reduce the scope, magnitude, and cost of emergency response measures downstream. To better predict and prevent zoonotic disease outbreaks, scientific approaches are needed to gather and understand information about the nature of disease appearance and spread,  

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES and to understand genetic-, population-, social-, and ecological-level char- acteristics that enable zoonotic pathogens to jump species and spread easily to humans. National and international support is also critical in addressing this global issue. SOCIOECONOMIC FACTORS AFFECTING ZOONOTIC DISEASE EMERGENCE Humans and animals can serve as pathogen reservoirs and vectors, and pathogens that may have resided in one part of the world can be carried or spread across long distances to become established in another part of the world. Technological advances now allow humans, animals, animal products, and their disease vectors to circumnavigate the globe in the span of 24 hours. Distance is no longer a barrier to disease. For example, in the first half of 2003, the United States saw concurrent importation of two zoonotic agents never before seen in the country—severe acute respiratory syndrome (SARS) and human monkeypox—as well as the establishment of new geographical niches for West Nile virus (WNV), an agent new to the United States and now endemic across the country. That same year, the United States also dealt with its first diagnosed case of bovine spongiform encephalopathy (BSE) despite more than 10 years of broad preventive ef- forts by the government and industry. In 2008, international tourist arrivals reached 924 million (UNWTO, 2009), a number that is estimated to grow annually by 5 percent over the next 20 years (FAO et al., 2008). Globalization and Trade Today, more goods, people, technology, and financial resources flow between countries than ever before, making countries less self-reliant and more dependent on each other. The level of economic interdependence among countries has increased dramatically on a global scale, especially in the past decade, as illustrated in Figure 2-1. In 2008, total global trade stood at $32.5 trillion, almost equally divided between imports and exports (WTO, 2009). In 2008, the total value of food imported into the United States was $75 billion or about 7.5 percent of total imports (Collins, 2007), and more than 25,000 shipments of food regulated by the U.S. Food and Drug Administration1 arrived daily in the United States from more than 100 countries (Koonse, 2008). In particular, the international trade of live animals and animal products 1 The U.S. Food and Drug Administration inspects and monitors the safety of all foods, domestic and imported, except for meat, poultry, and egg products, which are regulated by the U.S. Department of Agriculture. 

Total Expor ts Total Impor ts Agricultural Expor ts Agricultural Imports FIGURE 2-1 Total trade versus total agricultural trade. Agricultural trade data only from 1961–2006. In 2008, total trade imports were valued at $16.1 trillion USD and total trade exports were valued at $16.4 trillion USD.  SOURCES: FAO (2009), WTO (2009). Figure 2-1.eps bitmap image broadside 

0 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES has sharply increased over the past decade (Figure 2-2). Increased trade brings increased movement of animals and animal products, thereby in- creasing the potential for disease emergence from zoonotic pathogens. The global food production system is highly competitive and increas- ingly mobile. With attractive export markets, it often pays for exporting countries to establish the necessary veterinary infrastructure to meet the sanitary requirements of the importing country, as shown by countries such as Thailand for poultry and Brazil for beef. However, even competitive mar- ket economies do not necessarily reward additional investments in animal health infrastructure or encourage disease surveillance to track changing risk factors that might signal the potential emergence of a new disease. This failure to build veterinary capacity is even more relevant in countries where the food-animal production sectors primarily serve the local economy. Only with time and adverse experience are some countries and companies now grappling with disease threats across their production and distribution sup- ply chains, including the possibility of full-fledged disease outbreaks. Evolving Animal Agriculture and Trade To remain economically viable in highly competitive environments and to produce affordable animal protein for the growing global popula- tion, there is continued pressure to seek out economies of size and scale, including expanding or establishing operations in those parts of the world offering favorable cost structures. Thus, the geographic distance between where animals are produced and where ultimate consumption occurs con- tinues to expand. North America currently supplies one quarter of global meat exports (FAO, 2006). Asia has approached the Americas in volume of poultry production in a little more than a decade (see Figure 2-3). Bra- zil is now the largest single country for poultry and beef exports, and its diversified export market enables the movement of products to more than 150 countries (FAO, 2009). Starting with more developed agricultural economies, such as the United States, but then spreading to other countries, much of the agricultural prod- ucts that flow into international trade originate from increasingly capital intensive enterprises and well-coordinated supply chains. On the supply side, improvements in technology, infrastructure, and animal health have all contributed to this growth. Along with improvements in other areas such as genetics, nutrition, and management, the growing recognition of animal and herd health programs has enabled expansion and growth of large-scale animal agriculture. Large-scale production with animal crowding and un- sanitary conditions in some settings has contributed to the use of antibiotics to fight disease, with secondary effects on selection for antibiotic-resistant microorganisms and environmental contamination. 

FIGURE 2-2 International agricultural trade (world imports + world exports) by commodity type, 1961–2006. 31 SOURCE: FAO (2009). 

 FIGURE 2-3 Trends in poultry production. SOURCE: FAO (2009). Figure 2-3.eps 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE For countries such as the United States, the recognition that herds free of selected diseases could be translated into broader social and economic benefits has led to the support and implementation of national disease eradication campaigns. Freedom from brucellosis and tuberculosis not only contributes to the improvement of human and animal health, but has also lowered production costs, thereby establishing an international marketing advantage over countries that are not elevating their level of sanitary health. The public investment in animal health infrastructure includes the capacity to carry out disease surveillance, diagnosis, and treatment, and helps to facilitate export growth by enabling the movement of disease-free animals and related products into new markets and countries. To ensure that such improvements are not jeopardized or compromised, imports of susceptible animals or products are restricted from those countries that have not elimi- nated disease or achieved comparable levels of sanitary health. This allows certain exporting countries to further grow production capacity for domes- tic and international markets, largely through the adoption of standards formulated through the World Organization for Animal Health (OIE).2 The higher level of sanitary infrastructure has provided benefits to both producers and consumers. Producers benefit through factors such as de- creased costs of production (e.g., the extra cost of raising healthier animals is compensated by survival, weight gain, and increased market price), real or perceived increases in product quality, and the ability to meet consumer demand. Consumers benefit from the reduced risk of exposure to zoonotic pathogens. In many parts of the world, the public investment in national animal health infrastructure has not been commensurate with agricultural devel- opment. South and Central America provide more than one-fourth of the world’s agricultural exports (WTO, 2008), yet only 5 percent or so of national government outlays go into agriculture support. Moreover, only 5 to 10 percent of that finds its way into animal and plant health programs, and that is for a limited array of existing pathogens and pests (Pomareda, 2001). In sub-Saharan Africa, where food-animal production contributes about 30 percent of the agricultural gross domestic product (GDP) and is a part of the livelihood of about 150 million people, public expenditure on food-animal production research and development is less than 10 percent of the total public agricultural research expenditure (World Bank, 2008a).3 In addition, private-sector expenditure for agricultural research is low, 2 The Office International des Epizooties (OIE) is also known as the World Organization for Animal Health. OIE formulates standards related to animal health through committees consisting of representatives from member countries that are later adopted in its general as- sembly. OIE is recognized as a technical reference organization on animal health by the World Trade Organization. 3 Adapted from agricultural expenditure data in the 2008 World Development Report. 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES although philanthropic organizations have more recently emerged to sup- port crop and livestock research. Emerging Market Economies In 2000, emerging market economies accounted for 56 percent of the global middle class. By 2030, that figure is expected to reach 93 percent; China and India alone will account for two-thirds of this expansion. Rising incomes and growing demand can increase total trade while altering exist- ing and/or creating new trade flows, resulting in new or changing risk fac- tors. For instance, rapidly growing economies fuel an increase in individual wealth, which also increases the demand for meat. In 2007, the average Chinese consumer ate 50 kg of meat, which is more than twice the amount consumed in 1985 (The end of cheap food, 2007). In 2008, an estimated 21 billion food animals were produced for a global population of 6.5 billion people (FAO et al., 2008). Market dynamics also led to more live animal auctions where animals are brought together and then shipped across great distances and traditional “wet markets” where local farmers market their live animals to local con- sumers. These trends contribute to an increase in animal densities and closer contact between humans and animals, with a considerably greater risk of dispersing pathogens. International trade can transcend geographical bar- riers that in the past may have naturally slowed the spread of disease. The global market economy can also amplify disease effects through market in- stability as characterized by price volatility, shifts in consumption patterns, and variability in supplies. International Wildlife Trade Globalization has also impacted the movement of live, wild animals. From 2000 to 2004, more than 1 billion live animals were legally im- ported into the United States from 163 countries (Jenkins et al., 2007; Marano et al., 2007). In 2007 alone, the U.S. Fish and Wildlife Service processed 188,000 wildlife shipments worth more than $2.8 billion, and recorded more than 200 million legally imported live wildlife (CRS, 2008a; Einsweiler, 2008). These animals and animal products were imported for zoo exhibitions, scientific research,4 food and products, and increasingly for the growing commercial pet trade, including many exotic animals (Marano, 4 The U.S. Centers for Disease Control and Prevention (CDC) prohibited the importation of most monkeys as companion animals in 1975, but some imported for research are now being sold in the pet trade. CDC and other enforcement agencies do not track where animals go after quarantine (Ebrahim and Solomon, 2006). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE 2008). Most of these animals are not required under U.S. law to be screened for zoonotic diseases before or after entering the country (Marano et al., 2007). The effect of this, compounded by the lack of coordination among U.S. government agencies involved in regulating different aspects of wildlife imports, are important reasons for the failure to prevent the introduction of new pathogens into the country (Stephenson, 2003). Some exotic animals and wildlife that are banned from import are able to enter through the illegal wildlife trade.5 These are likely to include less healthy, more risky animals that pose a greater threat to human health and security (CRS, 2008a; U.S. House of Representatives, 2008). Even so, most of the zoonotic diseases reported to be caused by wildlife trade involved imports of legal wildlife (see Appendix B on monkeypox). The European Union (EU) is the top global importer of wildlife and wildlife products by value at €2.5 billion in 2005 (Engler and Parry-Jones, 2007), and it is concerned that increasing demands for wildlife importa- tion is a driver of illegal and unsustainable trade. EU member states have concluded that a major barrier to wildlife trade law enforcement and implementation is their lack of a coordinated strategic approach to monitor compliance (Theile et al., 2004; Engler and Parry-Jones, 2007). A review of the socioeconomic factors that drive the wildlife trade in Southeast Asia, which is both a consumer of wildlife products and a key supplier, revealed the inadequacy of policies and interventions aimed at decreasing the illegal and unsustainable trade of wildlife (World Bank, 2008b). Although poor populations in this region are often involved in wildlife trade, they do not necessarily drive this trade; therefore interventions for poverty reduction are not likely to reduce wildlife exports. Instead, many experts consider that the increased disposable income in consumer countries is the major con- tributor of demand for Southeast Asian wildlife, parallel to the increased access to these markets (World Bank, 2008b). These observations only serve to highlight the complexity of market forces. On the supply side, the illegal logging industry and the bushmeat trade has facilitated the extraction of certain wildlife species and threatened local wildlife populations (Chomel et al., 2007). Refugee camps set up in response to humanitarian crises, such as northwestern Tanzania, have led to serious forest degradation and have provided people with a greater proximity to wildlife habitats to hunt bush- meat, resulting in a decline of wildlife populations (Jambiya et al., 2007). The lack of a single international mechanism that captures data on wildlife trade represents a serious shortcoming of current national and international policies aimed at preventing illegal and unsustainable international wildlife trade (Gerson et al., 2008). 5 The illegal wildlife trade is difficult to quantify, although some estimates range from $5 billion to more than $20 billion annually (CRS, 2008a). 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES The Need for Disease Surveillance in Food Animals Improved prevention and disease control efforts in food-animal health has led to multiple benefits for human and animal populations, including reduced human morbidity and mortality, enhanced food security, improved market access for products, economic gains, and savings on potential out- break costs (Caspari et al., 2007). Many countries have strengthened their border controls and quarantine procedures, but the advances and benefits in improving animal health through actions such as disease eradication, prevention, and education have not been uniform across all countries. However, as education has advanced and become more available, surveil- lance and prevention efforts have also advanced and become specialized in areas such as vaccines and diagnostics. Although significant investments are needed to build infrastructure and institutional and regulatory capacity, necessary investments have not yet been made to implement food-animal disease surveillance, diagnosis, and treatment. Countries such as the United States and Australia have made available significant financial and technical resources for international disease eradi- cation or control campaigns, especially in the past 5 years for the control of highly pathogenic avian influenza (HPAI) H5N1 in Southeast Asia. In 2006, the U.S. Agency for International Development (USAID) provided $161.5 million for disease surveillance and pandemic preparedness for avian influenza (CRS, 2008b). In 2009, USAID will award $260 million over 5 years for the Predict and Respond initiatives aimed at four regions of the world prone to zoonotic disease emergence (Grants.gov, 2009a,b). From 2003–2006, Australia’s Agency for International Development committed $152 million to combat avian influenza and other emerging and reemerg- ing zoonotic diseases (AusAID, 2009). The EU has supported major animal disease eradication campaigns in Asia and Africa: Specifically in Africa, the EU partnered with the Organization of African Unity in 1999, providing an overall budget of €72 million for 7 years for the Pan African Programme for the Control of Epizootics (PACE) (OAU-IBAR, 2009). PACE targeted establishing and strengthening sustainable animal disease surveillance in sub-Saharan Africa. HEALTH AND ECONOMIC IMPACTS OF ZOONOTIC DISEASES Human Health Human mortality resulting from emerging zoonotic diseases has been relatively low compared to other leading causes of death from infectious diseases, with the exception of the 1918 influenza pandemic and HIV/AIDS, a zoonosis that now transmits readily among humans. Between 2003 and 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE 2009, there were 421 confirmed human cases of avian influenza A(H5N1), and as of April 23, 2009, 257 deaths were reported to the World Health Or- ganization (WHO) (Figure 2-4). In contrast, between November 2002 and July 2003, 8,096 individuals were diagnosed with SARS, which resulted in 774 deaths (WHO, 2004). As shown in Table 2-1, none of the recent major emerging diseases has led to large fatality numbers. The number of people infected or number of fatal cases, however, are not the only concerns. Impacts on trade and movement of people, economic stability, and panic and societal disintegration based on perception of danger can be seriously disruptive to the global order. 450 30 0 425 40 0 Cumulative D eaths , 2 57 375 250 350 325 30 0 20 0 275 250 Cases 225 150 20 0 175 150 10 0 79 125 59 10 0 43 75 50 33 32 50 7 25 4 0 0 20 03 20 04 20 05 20 06 20 07 20 08 20 09 Total Year Vietnam Myanmar Djibouti Turkey Lao PDR China Thailand Cambodia Iraq Pakistan Indonesia Bangladesh Azerbaijan Nigeria Egypt FIGURE 2-4 Number of confirmed human cases and deaths of avian influenza A (H5N1) reported to the World Health Organization by country and year. Confirmed cases (left axis) and cumulative deaths reported (rights Figure 2-4 color.ep axis) as of April 23, 2009. SOURCE: WHO (2009). 

TABLE 2-1 Selected Examples of Recent Zoonotic Outbreaks of International Significance  Human Health Impacts Country Disease Period Host Cases Fatalities Animal Losses Economic Impact Malaysia Nipah virus September Swine, dogs, 265a 105a Swine: 1.1 millionb $617 millionb encephalitis 1998–April 1999 fruit bats United Bovine spongiform 1986–2009 Cattle 168 164 Cattle: 214,305d 1986–1996: Direct costs Kingdom encephalopathy cases of cases of OTM: more than $936 million per year; (BSE) vCJDc vCJDc 8 millione 1997–2000: $858 million per yearf United States BSE 2003–2007 Cattle 3 cases 3 cases Cattle: 3 $11 billionh of vCJDg of vCJDg China, Taiwan, Severe acute November 1, Civets 7,667i 718i N/A $13 billion or 0.5–1.1 Hong Kong, respiratory 2002, to July 31, percent of GDPj; and Singapore syndrome (SARS) 2003 East Asian economies: 2 percent of regional GDP or US$200 billionk Canada SARS November 1, Civets 251i 43i N/A 0.15 percent of GDP or 2002, to July 31, C$1.5 billionl 2003 Asia Highly pathogenic January 24, 2004, Poultry, wild 337m 222m Birds: Estimated Asia $10 billion avian influenza to January 7, fowl, and 250 million (December 2003 to (HPAI) 2009 mammals February 2006)n Africa HPAI November 30, Poultry, wild 56m 26m N/A N/A 2005, to January fowl, and 7, 2009 mammals 

Europe HPAI October 21, 2005, Poultry, wild 0 0 N/A N/A to January 7, fowl, and 2009 mammals Worldwide Severe HPAI N/A Poultry, wild N/A 1–71 N/A Up to $3 trilliono pandemic fowl, and milliono (estimate) mammals United States West Nile virus January 1999– Birds and 28,975p 1,124p Birds: >317 species $400 million (1999–2007); fever December 2008 mosquitoes have been infected; Louisiana: $20.1 million (vector) Horses: 23,755 (June 2002–February cases 2003)q India Plague August–October Rodents 693r 52r N/A $600 million–$2 billions 1994 Kenya, Rift Valley fever November 30, Sheep, cattle, 1,062t 315t N/A N/A Somalia, and 2006, to May 3, goats, water Tanzania 2007 buffalo; mosquitoes (vector) NOTES: N/A = not available; total = total cases worldwide; vCJD = variant Creutzfeldt-Jakob disease. aFAO (2002); WHO (2007a). bEstimate includes $35 million in compensation for the pigs destroyed; $136 million for the control program from the Department of Veterinary Services; $105 million in lost tax revenue from swine industry; $97 million for the 1.1 million pigs destroyed; $120 million due to loss of pig export trade; and $124 million for loss by pig farmers during the outbreak period (FAO and APHCA, 2002). cAndrews (2009). dCattle slaughtered as a result of passive disease surveillance in Great Britain as of July 10, 2009 (Defra, 2009a). eThe Over Thirty Month (OTM) Rule bans meat from cattle aged over 30 months, which are more likely to have developed a significant amount of BSE agent in any tissue, from being sold for human consumption (Defra, 2009b). fOECD and WHO (2003). gThe three cases of vCJD were acquired abroad (CDC, 2009).  continued 

TABLE 2-1 Continued 0 hLosses in U.S. exports resulting from BSE-related restrictions (USITC, 2008). iWHO (2004). jBrahmbhatt and Dutta (2008). kWorld Bank (2005). lDarby (2003). mWHO (2009). nBased on survey of the United Nations Economic and Social Commission for Asia and the Pacific (Elci, 2006). oMcKibbin and Sidorenko (2006); Burns et al. (2008). pCDC (2008). qZohrabian et al. (2004). rCDC (1994). sWorld Resources Institute, United Nations Environment Programme, United Nations Development Programme, and World Bank (1996); Cash and Narasimhan (2000); Gubler (2001). tWHO (2007b). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE BOX 2-1 Examples of the Underestimated Burden of Zoonotic Diseases Rhodesiense sleeping sickness: According to this study, the actual mortality from sleeping sickness during an epidemic in southeast Uganda was approxi- mately 12 times higher than reported. The authors considered that many sleep- ing sickness cases were likely to have been misdiagnosed as malaria in poorly resourced rural clinics and so were not properly treated; all such patients would have died (Odiit et al., 2005). Rabies: These studies estimated that the actual incidence of human rabies in Tanzania was 10 times higher than reported through passive disease surveillance. Worldwide, the number of rabies deaths annually was estimated to be 32 times higher than the number reported to the World Health Organization (Fèvre et al., 2005; Knobel et al., 2005). Leishmaniasis: The study reported that the actual incidence of visceral leish- maniasis in Bihar, India, was estimated to be 8 times higher than reported by passive disease surveillance (Singh et al., 2006). Experience from past events and future projections based on contempo- rary events warn that low mortality is not a given for all disease events. The 1918 pandemic influenza virus killed tens of millions of people in a short time period, with estimates from 20 million to more than 50 million. Projections on the potential human losses from HPAI H5N1, should it attain a similar virulence as the 1918 virus, indicate that a severe pandemic of H5N1 virus could kill as many as 1 in 40 infected individuals or some 71 million (Barry, 2005; McKibbin and Sidorenko, 2006). Approximately 1 million individuals could die under a mild scenario (modeled after the Hong Kong influenza of 1968–1969), and 14 million under a moderate scenario (based on the char- acteristics of the 1957 Asian influenza) (McKibbin and Sidorenko, 2006). Looking at the same data, others suggest that as many as 180–260 million could die in a worst-case scenario (Osterholm, 2005). Furthermore, zoonoses can impose a significant human and animal health burden locally and, in many cases, that burden is underestimated (see Box 2-1). Economic Impact The economic impact of disease outbreaks depends on several critical factors, including public understanding and response, type of disease, and market scope. Measuring the economic impact of emerging zoonotic infec- tions is complex because there are so many sources of losses and dispropor- tionate impacts on different sectors and geographic regions (Kimball and 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES Davis, 2006). Table 2-1 provides estimated economic impacts associated with outbreaks for selected zoonotic diseases. Emerging zoonotic diseases can cause economic losses as a result of morbidity and mortality among food animals, losses related to public inter- ventions, and market losses at household, national, and global levels. Food- animal morbidity and mortality losses can be the result of the disease itself, or result from preventive actions such as culling of diseased, suspected, or at-risk animals. As of January 2009, 61 countries reported outbreaks of HPAI H5N1 in poultry, of which slightly more than half were developing countries. More than 250 million birds have died or been culled since the onset of the disease; however, this accounts for less than 1 percent of the 52 billion birds slaughtered annually. However, in Vietnam, which has imple- mented probably the most severe culling policy against HPAI H5N1, 50 million or 12 percent of the total annual poultry stock died or was culled, heavily impacting household and national economies. Economic losses related to public interventions can be the result of efforts to prevent and eventually contain and eradicate the disease. Those efforts include quarantine and disease surveillance systems, hospital and medical services, and the cost and compensation for culling or eventual other losses experienced by the private sector. This can also include losses from unproductive “downtime” forced on affected poultry farms and mea- sures to reduce human morbidity and mortality. During the SARS outbreak, 866 employees of the U.S. Centers for Disease Control and Prevention participated in the human and animal health response, totaling 46,214 person-days at a cost of well over $20 million in salary alone. This in- cluded deployments to 10 foreign countries and 19 domestic ports of entry (Marano, 2008). In the course of the 1994 outbreak of plague in India, trade and travel restrictions were imposed internally and externally, which led to economic impacts that shocked the region’s stock markets with losses of nearly $2 billion (Price-Smith, 1998; Cash and Narasimhan, 2000; Gubler, 2001). That 1994 plague outbreak in India is described in more detail in Chapter 5. Similar travel and economic disruptions were seen with SARS: Figure 2-5 shows tourist arrivals in China and Thailand and compares the immediate impact of SARS with the 2004 Pacific Ocean Tsunami. Losses through the market can result from changes in consumption patterns and trade, which directly affect prices and can last long beyond the period of risk. The spread of HPAI H5N1 caused international chicken prices to fluctuate in major poultry markets in Europe, Africa, and the Middle East (FAO, 2006). The EU’s total ban of beef and cattle exports from the United Kingdom (UK) in March 1996 due to BSE (see Box 2-2) resulted in the loss of trade estimated at £700 million per year (DTZ Pieda Consulting, 1998; van Zwanenberg and Millstone, 2002; Kimball and Taneda, 2004). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE FIGURE 2-5 Tourist arrivals in China (left axis) and Thailand (right axis) between 2001–2006. SOURCE: Brahmbhatt (2006). Reproduced with permission from the World Bank. Figure 2-5 color.eps bitmap image The combined economic impact of these losses indicates that outbreaks and epidemics of zoonotic diseases can cause short- and long-term economic consequences due to significant disruption of economic activities (Hanna and Huang, 2004). Detailed breakdowns of economic losses as described above are generally not available, but as shown in Table 2-1, total losses from emerging zoonotic diseases over the past two decades exceed $200 billion. Economic losses would be even higher if one had reached a severe pandemic scenario, which would amount to as much as 4.8 percent of global GDP (Burns et al., 2008). The serious economic effects of pandemic A(H1N1) 2009 have yet to be realized presuming there is a major global winter outbreak in the northern hemisphere. As shown in Figure 2-6, about 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES BOX 2-2 The Economic Impact of Bovine Spongiform Encephalopathy Outbreaks in the United Kingdom, the United States, and Canada In 1986, the United Kingdom (UK) had a major outbreak of a novel disease in cattle, bovine spongiform encephalopathy (BSE) (Wells et al., 1987). By 1990, British scientists suggested a possible link between BSE and Creutzfeldt-Jakob disease (CJD); and thus, the UK government set up a new disease surveillance unit with the mandate to identify any change in the pattern of this disease that might be attributable to the emergence of BSE in humans. The existence of a novel variant of CJD (vCJD) was first reported in 1996. A series of experimental studies subsequently confirmed BSE transmissibility from animals to humans (The BSE Inquiry, 2000). The years it took for scientists to gather the necessary evidence to establish this linkage, however, delayed the introduction of measures to protect human and animal health. The costs associated with the BSE outbreaks in UK cattle from 1986 to 1996 were reviewed by the BSE Inquiry, a committee created to investigate the re- sponse of the government to this animal disease. Based on this review, the public sector and ultimately the taxpayers bore the brunt of the economic consequences of BSE. Public expenditures due to BSE increased in the areas of biomedical re- search, compensation payments, and operational overheads incurred by different government agencies. From 1986 to 1996, the total expenditure on BSE-related research was £61 million, while other government expenditures, including com- pensation schemes and running costs, amounted to approximately £227 million (The BSE Inquiry, 2000). The private sector also suffered, particularly the produc- tion side of the beef industry and businesses (The BSE Inquiry, 2000). Before the European Commission introduced a ban of UK beef and cattle exports on March 27, 1996, the economic impact suffered by the beef- and cattle-related industries were relatively minor. The Inquiry concluded that the BSE-related costs suf- fered by farmers and businesses accelerated the decline of the industry’s overall growth. The introduction of the 1996 ban resulted in the collapse of the industry that same year due to the loss of major export markets and related markets. The United States and Canada suffered immense economic losses after BSE-infected animals were detected in 2003. In the United States, the value of U.S. beef exports dropped from $3.1 billion in 2004 to $2.5 billion in 2007 after the detection of a BSE-infected cow in December 2003. Net revenues declined by $1.5–2.7 billion per annum over the same period, resulting in a total loss to the sector of $11 billion USD (USITC, 2008). In Canada, the subsequent ban of Canadian beef and cattle imports by the United States and many other countries following the detection of a BSE-infected cow in May 2003 resulted in a drop in the value of beef and cattle exports of more than $1 billion in 2003, while domestic cattle prices fell 50 percent (FAO, 2006). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE FIGURE 2-6 Economic impact of a potential human influenza pandemic by per- centage of GDP (x-axis). SOURCE: Brahmbhatt (2006). Reproduced with permission from the World Bank. 60 percent of the economic losses would be from efforts to avoid infection (e.g., minimizing face-to-face interactions). Although the economic impact estimates in the case of an influenza pandemic show a high mortality in humans, the largest impact might arise from the uncoordinated efforts of people to avoid infection and the economic losses resulting from the reduc- tion in the size and productivity of the world labor force due to illness and death (Brahmbhatt, 2006). Equity Impacts In many of the least developed countries, both culling and the high mortality of birds have had a major impact on the livelihoods of poultry- dependent households. The poorest strata of rural households in developing countries derive a higher portion of their income from food-animal produc- tion than higher income households (de Haan et al., 2001). The importance of food-animal production for the poor is even more pronounced in poultry. In South Asian countries, more than 90 percent of flocks and 50–65 percent of birds are kept under an extensive “backyard” system. Village household surveys in Vietnam showed that income from the poultry sector was im- portant for 99 percent of the poor households; losses because of death or 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES FIGURE 2-7 Household income and expenditure effects of a backyard poultry ban (percentage change in annual income). 2-7.eps Figure SOURCE: Otte et al. (2006). Reproduced with permission from FAO. bitmap image culling of their flocks amounted to an average of $69 per household. A ban on poultry would cause losses of up to 30 percent of the income for the poorest households (see Figure 2-7). In Egypt, the poorest quintile of the population, with a monthly income of $35, earned 52 percent of their in- come from poultry, but suffered on average a loss of $22 from HPAI H5N1. Losses from emerging zoonotic diseases therefore disproportionately affect the poor (Otte et al., 2006). DISEASE SURVEILLANCE TO MITIGATE EMERGENCY RESPONSE MEASURES AND COSTS In a functionally integrated disease surveillance system for human and animal health, there are various opportunities for preventing, detecting, and responding to zoonotic disease emergence and transmission. Through early detection, a timely and effective response to zoonotic diseases in animal populations can prevent or minimize the likelihood of transmission to hu- man populations (see Figure 2-8). After detecting a zoonotic disease event in either human or animal populations, surveillance data would inform human and animal health decisionmakers so they can plan, implement, and 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE evaluate responses to reduce morbidity and mortality from zoonotic infec- tions. Without capacity or willingness to activate an emergency response, surveillance merely occurs in a vacuum. Effective prevention and control of emerging zoonotic diseases require both disease surveillance and emergency response capabilities that include disseminating and communicating action- able disease surveillance information to officials who have the authority, motivation, and capability to implement a response. The relationship be- tween disease surveillance and emergency response is typically in the size and efficacy of the two efforts: The more effective and timely the disease surveillance, the more likely it is to avert a relatively large emergency re- sponse. Large and effective surveillance programs will detect the first sign of a problem, then, if the actionable information is supplied to the proper authorities, a relatively small and targeted emergency response may ef- fectively curtail spread and mitigate the threat. On the other hand, small and inadequate surveillance programs are likely to miss many new disease events, so by the time the disease is recognized, a much larger emergency response is necessary. Surveillance information on zoonotic diseases in humans and animals, however, is highly variable under different scenarios, making the response to these zoonotic threats also variable. Box 2-3 and Appendix B provide some examples of the imbalance in the surveillance-response dynamic. It is also important to recognize that the threshold of detection will vary with the capacity of the laboratory. For instance, a newly emerged agent may be readily identifiable through basic technology widely available, such as bacterial culture of Escherichia coli O157:H7. A slightly more sophisticated laboratory, with the capability of embryonated egg inoculation, may be able to identify a new strain of highly pathogenic avian influenza. Identification of a disease entity such as BSE, which requires advanced technology such as Western blotting or immunohistochemistry, will be beyond the capacity of most laboratories, even if surveillance for other more easily detectable agents is extensive. Using current approaches, the cost of emergency response is usually several times greater than the cost of disease surveillance. The more wide- spread the disease is before detection and implementation of response, the larger the cost of the control measures. Moreover, the case of HPAI H5N1 in Vietnam underscores the importance of continuous surveillance of this virus to prevent subsequent waves of outbreaks (see Appendix B). As dis- cussed in more detail in Chapter 6, the investment in a well-functioning global disease surveillance system and in early response capability is roughly estimated to amount to about $800 million per year, whereas the economic losses from emerging, highly contagious zoonotic diseases have reached more than $200 billion over the last decade. 

 Exposure in Exposure in Animals Humans Symptoms in Clinical Signs Humans in Animals Seek Veterinary Care Humans Seek Medical Care Number Affected AP AD1 AR AD2 AR AR3 AR HP HD1 HR HD2 HR HD3 HR Time FIGURE 2-8 Opportunities to prevent, detect, and respond to the emergence and transmission of zoonotic diseases. NOTES: The graph is a stylized representation of a zoonotic disease outbreak and is not meant to represent any specific infectious agent. The timeline along the X-axis indicates when during the outbreak various opportunities to intervene in animal and human populations occur, and is expanded Revised 2-8 to facilitate illustration. AP = Preventing the emergence of zoonotic diseases in animal populations. Examples would include vaccination programs (rabies, avian influenza, 11-9-09 other); effective live animal market sanitation and other management policies to prevent mixing of species; sanitation including manure and rodent/ pest management on farms; biosecurity measures; testing and treatment programs of companion and stray animals for zoonotic diseases (e.g., Echinococcus granulosus); Hazards Analysis of Critical Control Points procedures in food systems; and regular preventive visits to veterinarians for companion animals. AD = Detection of zoonotic diseases in animals: • AD1, detection before clinical signs occur would result from testing animal specimens at diagnostic laboratories in ongoing serosurveillance pro- grams, or in testing food animals at slaughter. 

• AD2, detection in animal populations after clinical signs occur at the local level, by farmers, food-animal production workers, household flock and companion animal owners, and the public (e.g., wild animal die-offs). • AD3, detection in animal populations after veterinary care is sought would occur with veterinarians and or animal health workers diagnosing and reporting disease. AR = Response to control disease transmission in animal populations. This would include rapid outbreak investigation to determine etiologic agent, risk factors for spread, and points for control. Other response measures would include vaccination of animal populations in the face of an outbreak, testing and slaughter of food-animal populations, mass depopulation programs, strengthening biosecurity programs, and treating and curing zoonotic disease infection in household animals. HD = Detection of zoonotic diseases in human populations: • HD1, detection in humans before symptoms occur. This would include detection from screening programs, serosurveillance, etc. • HD2/HD3, detection in humans after symptoms occur and before or at the time medical care is sought. Detection would occur by patient report, findings by village healthcare workers, at local health clinics, at district hospitals, at tertiary hospitals, by healthcare providers, and through laboratory testing. HP = Preventing the transmission of zoonotic disease agents to human populations. This would include effective use of hand washing, use of personal protective equipment in abattoir workers, animal disease control personnel, and other people in regular close contact with animal populations; use of universal precautions and personal protection equipment by healthcare workers; vaccination programs; safe food preparation at the household level; restrictions on the importation, movement, and ownership of exotic pets; and regular prevention check-ups of humans in close contact with animals and animal populations. HR = Response to control zoonotic disease in human populations. This would include outbreak investigation to identify etiologic agent, risk factors, points for control, treatment at diagnosis, instituting isolation and quarantine measures, targeted disease surveillance and vaccination programs, mass vaccination programs, targeted antimicrobial administration, and mass antimicrobial administration.  

0 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES BOX 2-3 Selected Examples of the Balance and Imbalance Between Disease Surveillance and Emergency Response for Past Outbreaks • imited surveillance not detecting a new disease, and once detection oc- L curred, linkage with control is slow, so that emergency response is futile because it is so widespread: HIV emerged in central Africa in the 1970s. Because of inadequate disease surveillance, authorities did not realize this was an emerging problem. The lack of recognition, combined with the long incubation period, allowed this disease to spread globally, so that it soon became the foremost infectious disease in many parts of the world. Then once recognized and associated with marginalized populations (homosexuals and drug abusers), effective control mea- sures were slow to develop. Had early recognition occurred and been combined with effective controls, there could have been an effective global emergency response that might have prevented the majority of human morbidity and mortality. • xample of surveillance detecting a new disease locally, but without ac- E tionable information shared regionally and globally so that when the global spread of disease occurs, a global emergency response is necessary and very costly: Severe acute respiratory syndrome (SARS) emerged in China, was not diagnosed until it moved to Hong Kong, and affected visitors from multiple con- tinents. The issue for SARS was the lack of actionable information at early stages of the outbreak. Disease surveillance at the local level may have been effective, but the information did not reach the level required to implement a timely global emergency response. By the time it was recognized globally as a serious emerging health threat, emergency responses on several continents had to be activated. • xample of surveillance detecting a disease, but then no follow-through with E appropriate emergency response, so the disease continues to spread: In 2004, disease surveillance for highly pathogenic avian influenza (HPAI) H5N1 in Southeast Asia highlighted the presence of the HPAI H5N1 strain in chickens and its association with human mortalities. There were two problems here. First, disease surveillance detected the disease in humans and poultry, but only after the Therefore, an effective global disease surveillance system can be ex- pected to reduce the emergence of zoonotic diseases in humans and provide early detection of zoonotic diseases in livestock, thereby reducing billions in economic losses. In most emerging zoonoses, if the disease had been recognized much earlier (as would happen with well-functioning disease surveillance systems), effective emergency responses, if any, would have been smaller and cost effective. However, global disease surveillance sys- tems have not been adequate to detect disease in timely fashion and limit impact, so more often than not massive and expensive emergency responses have been required. 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE disease had been observed in another region of the world. An emergency response was instituted that was weighted more toward surveillance and control in human populations rather than in poultry populations, thus allowing for continued spread and circulation in poultry. Second, because of the lack of integrated human, poultry, and wildlife expertise, considerable time was needed to identify disease transmis- sion mechanisms. In the meantime, the virus continued to circulate, and eventually spread across Asia into Europe and Africa. Several countries improved their dis- ease surveillance system after the first outbreak. For example, after two waves of HPAI H5N1, Thailand mounted an impressive disease surveillance system based on human and animal health village volunteers and about 1,000 joint (Ministries of Health and Agriculture) District Surveillance and Rapid Response teams, which has probably kept the third wave of outbreaks much more localized. Vietnam, after an initial delay in the reporting of the disease, also developed a community-based animal healthcare worker system for early alert, which has proven to be effective. • xample of good initial surveillance finding a disease, but delayed under- E standing of the disease epidemiology, then emergency response mounted is effective: A new disease caused by Nipah virus surfaced in Malaysia in 1999. In this case, disease surveillance highlighted the presence of a neurological disease in pig farmers. The disease was initially misdiagnosed as Japanese encephalitis. After some delay, the true causative agent, Nipah virus, was identified and linked to infected swine, leading to the culling of 1.2 million pigs. It took longer to identify the fruit bat reservoir and the presence of fruit trees on the pig farms as a predisposing factor, and major economic losses could have been prevented. Another example is human monkeypox in prairie dogs in the United States in 2003. Detection of the zoonotic hazard was quickly followed by emergency responses to contain the threat. Both of these examples are from countries with advanced economic and healthcare systems, so both disease surveillance and emergency response were effective. NOTE: For further details on surveillance and response of select zoonotic disease outbreaks, see Appendix B. The reality is that procuring funding for large, expensive emergency response measures is easier than funding continual disease surveillance for detecting future and unknown diseases. This is unfortunate because a well-designed emerging zoonotic disease surveillance system is what will ultimately result in less human morbidity and mortality and fewer adverse economic impacts globally. It is widely recognized that emergency response is essential. Yet it is penny-wise and pound-foolish to continually invest in large emergency responses without investing in effective disease surveillance systems that would lead to smaller, less costly control efforts. 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES UNDERSTANDING ZOONOTIC DISEASE AGENTS AND TRENDS TO PREDICT ZOONOTIC DISEASE EMERGENCE To accurately predict and detect when and where zoonotic pathogens might emerge, it is important to understand the biological pathways affect- ing their emergence. Data gathered from disease surveillance systems are crucial, enabling scientists to predict how and when pathogens may emerge and the extent of their spread and impact. This information allows decision- makers to more confidently allocate resources to prevent outbreaks from occurring. If a zoonotic disease outbreak should arise, such data become even more critical for informing effective control and response measures. The Biology of Pathogen Emergence Of approximately 1,400 species of human pathogens that are now recognized, more than 800 (nearly 60 percent) are known to be zoonotic (Woolhouse and Gaunt, 2007). Moreover, many nonzoonotic pathogens are known or believed to have origins in nonhuman animals (Table 2-2). Some of these have only recently emerged (e.g., HIV/AIDS, pandemic strains of TABLE 2-2 Examples of Human Pathogens with Evolutionary Origins in Nonhuman Hosts Disease Pathogen Original Host AIDS Human immunodeficiency virus-1 Chimpanzees AIDS Human immunodeficiency virus-2 Sooty mangabeys SARS SARS coronavirus Bats/palm civets Malaria Probably birds Plasmodium falciparum Malaria Asian macaques Plasmodium vivax Sleeping sickness Trypanosoma brucei subspp. Wild ruminants Diphtheria Probably domestic herbivores Corynebacterium diphtheriae Hepatitis Hepatitis B virus Apes Viral lymphoma Human T-lymphotropic virus-1 Primates (possibly Asian macaque) (Unknown) Human T-lymphotropic virus-2 Bonobos Respiratory infection Human coronavirus OC43 Bovine Influenza Influenza A virus Wildfowl Measles Measles virus Sheep/goats Mumps Mumps virus Mammals (possibly pigs) Smallpox Variola virus Ruminants (possibly camels) Typhus Rodents Rickettsia prowazeckii Plague Rodents Yersinia pestis Dengue fever Dengue fever virus Old World primates Yellow fever Yellow fever virus African primates SOURCE: Adapted from Wolfe et al. (2007). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE influenza A), and others have origins going back thousands or millions of years (e.g., plague, malaria). Zoonotic disease emergence from nonhuman animals may be viewed as a series of steps from primarily animal diseases, such as rabies that oc- casionally are transmitted to humans, all the way to diseases originating in animals, such as HIV-1 that jumped species to humans and successfully transmitted from human to human without further involvement of the orig- inal animal host. There are five stages in this “pathogen pyramid” wherein the barriers to pathogens progressing from one stage to the next are both biological (functional constraints, often at the molecular level, to infection and transmission) and ecological (restricted opportunities to infect humans or transmit between humans) (Wolfe et al., 2007; see Figure 2-5 in IOM and NRC, 2008). Overcoming these barriers may involve evolution of the pathogen, although increased opportunities to infect or transmit between humans can arise purely from changes in human behavior or demography (e.g., intensification of food-animal production and increased trade of exotic species—see Chapter 3) or from changes in pathogen ecology (e.g., altered distribution of the reservoir host or vector). The example of HIV-1 suggests that a pathogen can rapidly progress through the stages of the pyramid over time scales of decades. High variability in virus genomes might generate high functional diversity, producing human-infective vari- ants on a regular basis, some of which successfully “take off” in human populations (Woolhouse and Antia, 2008). Pathogen Discovery Analysis of emerging diseases from 1940 to the present demonstrates that the rate of emergence “events” rose significantly over this period (Jones et al., 2008) after correcting for trends in disease surveillance effort. The discovery of new human pathogen species continues at a rate of 3–4 spe- cies per year (see Appendix C). The discovery of new human pathogens has three components: (1) recognition of pathogens that have existed in humans for a long time, but have just been detected (e.g., hepatitis C); (2) pathogens that have existed for a long time, but have only recently had the opportunity to infect humans (e.g., Baboon cytomegalovirus); (3) newly evolved human pathogens that did not previously exist (e.g., pandemic A(H1N1) 2009 virus as a relatively recent example in humans; canine parvovirus as an animal example). Pathogens of all three kinds continue to be discovered. The majority of recent discoveries of new human pathogens are viruses (see Figure 2-9 and Appendix C) (Woolhouse et al., 2008). The discovery rate of human non-virus pathogens is much slower and mainly involves rickettsia and microsporidia. There is every reason to expect current trends 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES Novel species: 87 Bacteria, 13% Fungi, 15% Viruses, 67% Protozoa, 3% Helminths, 1% Prions, 1% FIGURE 2-9 Patterns of pathogen discovery. Percentage of novel pathogen species by type. 2-9 new SOURCE: Adapted from data by Woolhouse and Gaunt (2007). of virus discovery to continue in the immediate future (Woolhouse et al., 2008). Although the rate of virus discovery has historically been re- markably consistent since the advent of tissue culture, the introduction of new technologies such as polymerase chain reaction and the advent of high-throughput sequencing has led to a substantial increase in the global capacity to identify novel pathogens. That, coupled with a great deal of interest in pathogen discovery, makes it possible that the rate of discovery, particularly of viruses, will accelerate as new efforts are made through surveillance programs. The majority of newly discovered human pathogens are either zoonotic or have recent origins in nonhuman reservoirs. Most are associated with other mammalian hosts, a few with birds, and only rarely with other classes of vertebrates. The mammalian taxa most commonly associated with new zoonoses are ungulates, carnivores, and rodents. These patterns are similar to the known zoonoses; in other words, we share our new pathogens with the same kinds of reservoir with which we have always shared our patho- gens (Woolhouse and Gowtage-Sequeria, 2005). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE Many recent high-profile emerging zoonoses have spilled over from wildlife hosts to humans. The rate of emergence of these wildlife-origin zoo- notic diseases also appears to have increased significantly over the past six decades, and pathogens of wildlife origin represent the majority of emerging pathogens in the 1990s (Jones et al., 2008). Animal susceptibility studies performed in laboratories worldwide in collaboration with WHO, the Food and Agriculture Organization of the United Nations (FAO), and OIE quickly identified a novel coronavirus as the etiological agent that caused the 2003 SARS outbreak. Moreover, these studies revealed that a variety of wild and domestic animals were harboring this agent (WHO, 2003). Data Limitations and Information Gaps The committee identified several issues in terms of the data limitations. First, monitoring is subject to massive ascertainment biases. There are vast differences in the efforts invested in different places and at different times, leading to important gaps in information whether at the level of species discovery, emerging disease “events,” or disease outbreaks in humans. Adjusting for this bias is difficult. One-third of emerging disease events are reported from the United States, 10 times as many compared to China, India, Brazil, and other hotspot countries (see Figure 2-10), and that seems unlikely to represent the frequency of emerging events in these countries. Second, monitoring is ad hoc, not systematic, and is partly driven by re- sponses to the most recent events (e.g., clusters of discoveries in eastern Australia; spate of discovery of coronaviruses following the SARS out- break) and partly by availability of detection and identification technolo- gies. Third, determining the number of pathogens that have not yet been identified or detected in mammalian and other reservoir hosts is difficult. The inventory of species pathogenic to humans is incomplete but still grow- ing (Woolhouse and Gaunt, 2007). The inventory of species pathogenic to major domestic food-animal species, plus cats and dogs, is also incomplete. In addition, there is very limited knowledge of the pathogens for the vast majority of other mammal species, let alone birds or other vertebrates (Dobson and Foufopoulos, 2001). Fourth, questions on the frequency with which humans are exposed to animal pathogens (so-called “chatter”) and what fraction of those exposures are capable of crossing the species barrier to cause human infection remain unanswered. Fifth, the determination of what constitutes a species barrier and what characteristics allow patho- gens to overcome it (e.g., pathogen evolution, immunosuppression, new transmission routes) are important issues that need to be addressed. And sixth, whether a human infection that resulted from exposure to an animal pathogen can be transmitted (directly or via an indirect route) to another 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES 120 100 Number of events 80 60 40 20 0 USA Germany India Australia China Brazil UK DRC Japan Malaysia Egypt Mexico FIGURE 2-10 Patterns of reporting of emerging disease “events”: five countries Country reporting the highest number of “events” (left) and selected others (right). SOURCE: Woolhouse (2008a). Reproduced with permission from Macmillan Pub- lishers LTD: Nature. 2-10 human is simply not known for hundreds of pathogen species (Taylor et al., 2001). Domestic Animals and Wildlife Surveillance to Predict Zoonotic Disease Emergence Given the desire to more effectively predict where the next zoonotic disease will emerge, there are many gaps in knowledge of potential emerg- ing pathogens amidst the evidence of continuing events, underscoring the need for active disease surveillance in animal reservoirs for known zoonoses including domestic animals and also wildlife wherever possible. Improved disease surveillance is particularly important where the protection of human health depends wholly or partly on measures taken to prevent disease emer- gence or control disease in the reservoir (e.g., BSE, rabies, African sleeping sickness) and where the risk of outbreaks in humans is largely determined by the epidemiology of infection in the reservoir (e.g., Nipah virus, WNV, hantaviruses, plague). In addition, human resources and field capacity need to be developed to be able to conduct surveillance for zoonotic pathogens in animal reservoirs that often can be difficult to reach. Improved human resources and field capacity will greatly improve capacity to detect novel and emerging zoonoses. 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE Statistical Analysis, Modeling, and Predicting Future Trends Once a zoonotic pathogen has emerged and been identified, surveil- lance data are critical for enabling researchers to predict the extent and magnitude of the outbreak. Statistical methods are needed to make reli- able inferences and hypothesis testing from epidemiological findings and approaches (Jewell, 2003). The use of such analyses and disease models can better inform decisionmakers on how to effectively respond to disease outbreaks early on. Statistical analysis and dynamical modeling have a long history of providing insights into the importance of infectious diseases and their transmission dynamics, beginning when Daniel Bernoulli modeled smallpox transmission in 1760 (Bernoulli, 1766). With dramatic increases in both computational power and detailed data on human and animal diseases in recent years, statistical analyses and quantitative studies have been un- dertaken in the wide range of issues related to zoonoses. These analyses and modeling utilize data from a variety of sources, including those from surveys (e.g., Easterbrook et al., 2007), from routine sentinel disease sur- veillance, and from detailed experiments with randomized treatments to identify and characterize key features of the epidemiological system. An example is a study of the use of antibiotics in food-animals to reduce bac- terial illnesses in animals, thereby reducing subsequent human illness, with an associated risk of selecting for antibiotic-resistant bacteria, which could make food-associated human infections harder to treat (Singer et al., 2007). Synthesis of statistical and mathematical methods has allowed transmission models to be based on robustly estimated parameter values. However, most modeling studies have been limited in scope to one host and one pathogen, even though most pathogens have multiple hosts (Woolhouse et al., 2001). A good example of multihost modeling is the rabies study in the Serengeti ecosystem of Tanzania (Lembo et al., 2008). Uses of Statistical Analysis Key statistical principles include those of quantitative hypothesis test- ing, parameter estimation (with corresponding measures of parameter un- certainty), and model fitting/criticism. Specific statistical methods have been developed to allow the integrated analysis of data sources that vary in source, type (e.g., combining retrospective studies of known outbreaks and disease surveillance of key disease events) (Burkom, 2003), and qual- ity (rigor and relevance) (Turner et al., 2009). The analysis of all the rel- evant evidence relating to a particular disease can, however, lead to highly complex probability models. In such cases, particular care must be paid to model criticism and the detection of inconsistent or conflicting evidence 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES (Presanis et al., 2008), taking into consideration the assumptions of the model(s) underpinning the analyses. A limitation of traditional statistical modeling and analysis (e.g., regres- sion, survival analysis, and analysis of contingency tables) is that the insights are typically limited to comparisons and quantification of association rather than giving insights into the often complex mechanisms underlying the observed epidemiological patterns. Good models are difficult because epi- demic diseases and especially emerging epidemic diseases are multisystem, dynamic, nonlinear, stochastic processes. Models for causal inference were developed to overcome some of these limitations (Holland, 1986). There are several examples from recent emerging infectious disease investigations (including Hendra virus, Nipah virus, coral diseases, and avian influenza) where techniques designed to infer causation—including epidemiological causal criteria, strong inference, causal diagrams, model selection, and tri- angulation—were successfully applied (Plowright et al., 2008). Uses of Dynamical Modeling Dynamical models of disease transmission are those developed to repre- sent underlying epidemiological (and sometimes demographical) processes. Four main aims of such modeling have been identified (Anderson, 1988; Massad et al., 2005): (1) Enhancements to the logic and specification of current theories and concepts relating to disease transmission; (2) Genera- tion of new testable hypotheses through computer program-based (so-called in silico) experiments or simulation processes; (3) Prediction of the future course of an epidemic and/or the impact of preventive measures; and (4) Identification of types of epidemiological data needed to refine understand- ing of disease epidemiology and/or make better predictions. On the basis of the particular aims of the exercise, models are some- times applied retrospectively to interpret historical epidemiological data and are sometimes used prospectively to generate predictions. In practice, retrospective analysis often provides the basis for predictive modeling (see Box 2-4). Examples of retrospective or historical modeling of emerging zoonoses include analysis of both the recent past (e.g., modeling analy- sis of recent Ebola outbreaks [Chowell et al., 2004; Ferrari et al., 2005; Legrand et al., 2007]) and the more distant past (e.g., modeling of 1918 influenza pandemic [Mills et al., 2004; Sertsou et al., 2006; Vynnycky et al., 2007]). Predictive modeling is used to evaluate future scenarios as more or less likely, and to explore the possible benefits and/or risks of alternative realities. These alternatives could include alternative disease surveillance efforts (e.g., increased testing of live cattle for M. bovis or increased ef- forts to detect bovine tuberculosis in slaughtered cattle); various possible 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE culling policies designed to reduce future disease incidence; and alternative policies aimed at controlling or eradicating disease. Predictive modeling for disease emergence is a difficult and complex challenge. Although some data on which to model do exist, the biological and ecological characteristics needed for an outbreak to occur are unknown. Therefore, to improve the science behind any effort in modeling emergence, particularly of pathogens, it seems axiomatic that hypotheses need to be generated and data gathered to either strengthen and support or refute and abandon the premise being studied. The prospects of successfully predicting emergence events would be greatly enhanced by systematic data collection on the patterns of pres- ence and prevalence of infectious agents in animal populations, which means developing and implementing a systematic, ongoing integrated dis- ease surveillance program that is global in scope. A longitudinal study of the underlying factors driving disease emergence, including those associated with animal production systems and climate change, could provide valu- able information to such a program. To inform such a study, the pairing of complex mathematical models with remote sensing data could be useful to correlate environment with disease outbreaks and more accurately predict future disease events (Ford et al., 2009). Mathematical models have also been developed and deployed during ongoing epidemics to help advise control policies. Such “real-time” model- ing presents a number of challenges, including rapid collection and com- munication of input data, validating the process of model development, and generating formal estimates of model parameters from initially sparse data, noting that rigorous methods to fit such models to data are more complex and computationally burdensome than those required for traditional sta- tistical models. Even so, real-time modeling can inform the management of an epidemic: Examples include the 2001 epidemic of foot-and-mouth disease in the UK, the 2003 global SARS epidemic, and the 2009 influenza A(H1N1) pandemic. Projecting into the Future Projections are defined as “the numerical consequences of the assump- tions chosen. The numbers are conditional on the assumptions being ful- filled” (Keyfitz, 1972, p. 347). In the context of infectious disease, these assumptions could take the form of “if” circumstances: a closed popula- tion of a particular size, number of people encountering (and potentially infecting) each other randomly at a particular rate, and the introduction of a person infectious with a disease of a certain transmissibility into a fully susceptible population. Possible epidemic scenarios could be described, al- though the resulting incidence of disease on any subsequent day could not be derived with certainty due to random chance. Projections provide useful 

0 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES BOX 2-4 Simulation of Human Influenza Transmission in Thailand Using detailed demographic data (including population distribution and household size) and newly derived parameter estimates from reanalysis of historic data (including U.S. and UK 1918 pandemic mortality data), Ferguson and colleagues (2005) simu- lated human influenza transmission in Thailand to evaluate the potential effectiveness of targeted mass prophylactic use of antiviral drugs and social distancing to contain influenza. Figure a shows the time sequence (in days) of an epidemic, with spreading in a single simulation of an epidemic with R0 = 1.5. Red indicates presence of infected individuals, and green indicates the density of people who recovered from infection or died. Figure b shows the daily incidence of infection over time for R0 = 1.5 in the absence of control measures. Thick blue lines show the average for realizations result- ing in a large epidemic; grey shading represents 95 percent confidence limits of the incidence time-series. Multicolored thin lines show a sample of realizations, illustrating a large degree of stochastic variability. Box 2-3.eps bitmap image broadside information to predict (or forecast) the future, insofar as the assumptions (from model structure to parameter values) are realistic. Traditional statistical methods (most often time-series and regression) are sometimes used to provide short-term predictions of infectious dis- ease incidence, quantifying past trends, and projecting them forward (see Box 2-5). Temporal, seasonal, and spatial trends were quantified along with temporal correlation to predict the incidence of meningococcal disease in France, and the model was based solely on trends observed in the detailed incidence data available (Knorr-Held and Richardson, 2003). An alterna- tive approach is to predict incidence based on risk factors previously ob- served to be associated with incidence rates. For example, having previously 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE A similar study was published simultaneously in Science (Longini et al., 2005). The World Health Organization issued the following statement: The models provide additional information which will help WHO and public health officials in our Member States to improve pandemic influenza preparedness planning. . . . Several countries have already purchased stockpiles of antiviral drugs and WHO has taken steps to establish an international stockpile. . . . If we have a chance to reduce the scale of a pandemic with antivirals and other public health measures, the success of these interventions will depend on effective disease surveillance and early reporting in risk-prone countries. Before any stockpile can be used effectively, both must be strengthened. (WHO, 2005a) These influenza studies offered the authors’ most plausible set of transmission scenarios in order to inform policymakers, along with other available evidence. The next decisions are how much effort and what type to invest in planning for a serious future human and animal health crisis. Surveillance data are critical to underpin estimation of key epidemiological param- eters, which in turn determine which transmission scenarios are most plausible. SOURCE: Ferguson et al. (2005); WHO (2005a). shown an association between weather conditions and the presence of St. Louis encephalitis hemagglutination inhibition antibodies in wild birds, a hydrology model and a logistic regression model were combined to predict the incidence of human cases of St. Louis encephalitis, and these predictions were found to perform well looking 2 to 4 months ahead (Shaman et al., 2003, 2006). The predictions from transmission models under different scenarios can be compared to inform debate about the potential consequences (both risks and benefits) of alternative courses of action. In this context, math- ematical models have the advantage of transparency, since the basis for making predictions (for example, about the impact of control measures) is 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES BOX 2-5 Predicting an Outbreak Anyamba and colleagues (2006, 2007) observed that sea surface temperatures in the equatorial east Pacific ocean increased anomalously during July to October 2006, indicating El Niño conditions. Such conditions previously had been associated with excess rainfall in East Africa. Such rainfall was predicted to give rise to the normalized difference vegetation index (NDVI), and a Rift Valley fever (RFV) model, based on the NDVI data, indicated that in October to December 2006 there would be an elevated risk of RVF in northern Kenya, central Somalia, and subsequently Tanzania. Based on these results, early warning advisories were issued by the Food and Agriculture Organization of the United Nations and the World Health Organization to alert countries’ authorities in early November 2006 of the elevated risk of RVF out- breaks (WHO, 2007c; Anyamba et al., 2009). On this basis “the [U.S.] Department of Defense–Global Emerging Infections Surveillance and Response System and the Department of Entomology and Vector-borne Disease, United States Army Medical Research Unit–Kenya initiated entomological surveillance in Garissa, Kenya, in late November 2006, weeks before subsequent reports of unexplained hemorrhagic fever in humans in this area” (Anyamba et al., 2009, p. 957). DENG MAL RI CHOL MAL RVF Hotspots of potential elevated risk for disease outbreaks under El Niño conditions, Box 2-5.eps 2006–2007. bitmap image available for inspection, criticism, and change (Woolhouse, 2008b). Often, models will be the best evidence we have to inform decisionmaking. Models can also be used to gain insight into situations where an intervention was implemented and an unexpected result was obtained. As with any model- ing exercise (other factors being equal), a model that has been shown to 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE SOURCE: Anyamba et al. (2006). The index case was a patient in Kenya who experienced the onset of symptoms on November 30, 2006 (CDC, 2007; WHO, 2007c), and the Kenyan cases peaked in late December. From November 30, 2006, to March 12, 2007, 684 cases were reported in Kenya, including 155 deaths; 114 cases were reported in Somalia, including 51 deaths; and 290 cases were reported in Tanzania, including 117 deaths (WHO, 2007c). The model’s successful prediction of the epidemic enabled the affected countries to be forewarned of the increased risk (Kaplan, 2007). “The early warning enabled the government of Kenya, in collaboration with the World Health Organization, the United States Centers for Disease Control and Prevention, and the Food and Agriculture Or- ganization of the United Nations to mobilize resources to implement disease mitigation and control activities in the affected areas, and prevent its spread to unaffected areas” (Anyamba et al., 2009, p. 957). Non-fatal (n=491) Fatal (n=126) No. of Class Week of onset of symptoms Cases of Rift Valley fever meeting inclusion criteria by date of onset of symptoms, Box 2-4.eps Kenya, December 2006–February 2007 (n = 617). bitmap image produce accurate predictions has increased credibility compared with one that has only been shown to fit data well retrospectively. Dynamical mathematical models of disease transmission, in contrast to statistical models of trend or association, are better suited to longer term predictions and predictions of new and emerging threats. They also 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES have the potential to explore usefully “what if” scenarios, such as sudden changes in control policies or human behavior (e.g., due to travel restric- tions). There are considerable challenges posed by such studies. Infectious disease incidence depends directly on various factors of the particular dis- ease under study (e.g., population size, weather, or risk behaviors). Thus, making accurate predictions requires both accurately incorporating the roles of important drivers into the transmission model and making accurate predictions of how these drivers will behave in the future. Mathematical models are valuable tools for policymakers, but are best used as one com- ponent of the decisionmaking process, which should draw on all kinds of evidence available. INTERNATIONAL AND NATIONAL SUPPORT IS CRITICAL Zoonotic diseases can transcend boundaries and affect multiple coun- tries, thus the support of both the national and international community is critical for effectively responding to them. The control of HPAI H5N1 at the international and national levels has provided insight into how different actors cooperate and collaborate on zoonotic disease concerns. The expe- riences reported here are based mainly on independent evaluation reports from FAO, United Nations System Influenza Coordinator (UNSIC), and the World Bank. International Level At the international level, WHO, FAO, and OIE are the main play- ers in the international HPAI H5N1 arena.6 According to their respective mandates, WHO focuses on the human health aspects and FAO on the implementation of the standards and strategies that OIE sets for animal health. The scope and mode of operation of these three agencies is quite different. WHO has a significant country presence, which enables it to more directly affect national decisionmaking. FAO has a much-limited presence at field level, normally without any animal health expertise in its country offices. Finally, OIE has a 40-person staff, a limited number of regional representatives, and no specific country representation. These organizations (without the United Nations Children’s Fund, or UNICEF) cooperated well in the Codex Alimentarius Committee, which sets food safety standards. This committee was established by WHO and FAO, and now also has close relations with OIE. It was described by the recent Independent External 6 To support communication about HPAI H5N1 and its control, UNICEF was added as an additional technical agency, although its role and mandate were never clearly articulated. 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE Evaluation panel of FAO as an example of an effective partnership among international organizations (FAO, 2007). The start-up of the collaboration among the international agencies in addressing the HPAI H5N1 threat was difficult and slow, however. The first outbreaks of the current H5N1 strain of HPAI occurred in December 2003, with major outbreaks in 10 East Asian countries in January 2004. The first WHO strategy (2005b), without any discernable FAO or OIE input, was prepared in early September 2005. A joint FAO/OIE strategy prepared in collaboration with WHO was prepared by November 2005, or nearly 2 years after the outbreak (FAO et al., 2005). The reasons for the delays were caused by a lack of understanding of the mission of the involved agen- cies, lack of understanding on the epidemiology of the disease, difference of opinions among the agencies on how to respond, and the slow pace of resource mobilization. This delay led to a rather fragmented approach that was arguably one of the main factors in the slow donor response in providing financial sup- port, which caused donors to get involved in a bilateral fashion, based on the advice of their own technicians. The overwhelming number of missions of the technical agencies with large numbers of expatriate specialists and the complexity of procedures were also frequently mentioned at the country level as important issues (FAO, 2007). Starting in mid-2005, and in particular leading up to and following the Beijing Conference, the International Finance Institutions (IFIs) and the World Bank also became directly involved in the HPAI H5N1 cam- paign, although the World Bank had supported Vietnam with an earlier emergency loan. This opened a new set of constraints, which affected the implementation of the campaign, especially administrative and procedural aspects. These constraints became especially apparent in the cooperation between WHO, FAO, and the IFIs, where the respective roles of these United Nation’s agencies as cooperators for technical expertise and as contractors for services led to conflicts with the procurement rules of the IFIs. These administrative differences were exacerbated by differences in fiduciary requirements between the technical agencies and the IFIs (Willitts- King et al., 2008). The cooperation among WHO, FAO, OIE, and to some extent UNICEF significantly improved over time because of the major increase in funding, the strong pressure from donors, and the excellent coordination role of UNSIC. There are now weekly conference calls, and there is a stronger co- operation in the preparation of the strategy updates. The institutions work together in the preparation of Integrated National Action Plans. A mutual trust between the main day-to-day decisionmakers in these organizations has emerged. However, even now, the cooperation is mainly concerned with strategy development and planning, yet there are few joint activities 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES on implementing disease surveillance and control. The relationship between UNSIC and the technical agencies—especially WHO, which sees itself as the lead technical agency in human health and pandemic preparedness—is still a challenge (Willitts-King et al., 2008). At the individual level, the three agencies provided a rapid reaction. For example, FAO, with input from OIE, organized an international work- shop in East Asia on HPAI H5N1 only 3 weeks after the first outbreak. FAO became involved quite early with its Special Fund for Emergency and Rehabilitation Activities in the implementation of control measures. This flexible tool, with much lighter administrative requirements than normally demanded in FAO, provided FAO with the flexibility to respond early to the disease outbreaks. The lack of funds, however, caused the initial support that FAO provided in the affected and at-risk countries to be limited and restricted to strengthening disease surveillance systems, providing protective gear, and supporting epidemiological studies. Funding included almost no support in containing the disease, such as support of public administrations to be able to enforce movement control, compensation for culling, and vac- cination. Similarly, WHO focused on the stocking of antivirals, although it could have used its much greater country presence to raise greater aware- ness and train local staff in the epidemiology and control of HPAI H5N1. National Level At the national level, cooperation among the respective ministries of health, agriculture, and the environment in many countries is cumbersome at best. They often have separate human and animal disease reporting procedures and communication channels during a disease outbreak. En- vironmental agencies are the weakest in the public sector, and efforts to bring them together are often confronted with major transaction costs, bureaucratic delays, and competency issues. The main lessons learned from the HPAI H5N1 campaign point to the importance of political support for disease control and the existence of an institutional framework. Political support is crucial for disease control. The picture, which emerges from the reviews, shows ownership and political will at the highest levels to effectively plan and implement HPAI H5N1 campaigns. In several countries, this lack of ownership has led to inadequate interministerial collaboration; grossly insufficient national funding for human, veterinary, and wildlife services; and reluctance to share animal disease incidence information. These trends will severely affect the sustainability of future HPAI H5N1 activities. The institutional framework is another critical element. Key observa- tions that emerge from the reviews concern these factors: (1) the hierarchi- cal place of HPAI H5N1 campaigns in government, and experience in the 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE current campaigns seems to indicate that placement at a higher level (deputy prime minister, ministry of finance) than the line ministries of health or agriculture gives better results7; (2) decentralization, which, with some ex- ceptions,8 severely obstructs lines of command9; (3) the limited simulation testing and the general neglect in the preparation of most national prepared- ness and Integrated National Action Plans; and (4) the limited involvement of the private sector and, in particular, the nearly complete lack of use of private service providers (private veterinarians and paraveterinarians) under a sanitary mandate. Lessons Learned In an early phase of an emerging outbreak, countries need to de- fine a mutually agreed-upon strategy with the international organizations concerned and with other relevant institutions. As was the case with the HPAI H5NI control campaign, it is important to collaborate early on with institutions specialized in environmental health and wildlife. This could be the function of the current UNSIC, whose current mandate expires in De- cember 2010 and would have to be extended. Many developing countries lacked funding for investment in the surveillance of and response to HPAI H5N1. To avoid lack of funds to control an emerging disease at an early stage, sustainable funding is needed for highly infectious zoonotic diseases. To foster cooperation at the national level, governments need to establish special permanent, functional cross-sector coordination mechanisms, either through the exchange of memorandums of agreement between the different ministries and agencies involved, or a coordinating authority (e.g., special task force) above the sectoral human health, veterinary, and environmental agencies (e.g., the prime minister or deputy prime minister). In the case of an emerging disease outbreak, such institutions would define the control strategy, prepare contingency plans, and oversee their implementation; an option would be to let such a task force evolve into an independent agency. Finally, they need to cultivate a new style of leadership that promotes co- operation, teambuilding, and mentoring. This would need to be achieved through education and underpinned by incentive systems, which recognizes achievements in these areas rather than the current performance systems that often promote single department goals and individual achievements. 7 Other disease control campaigns (HIV/AIDS) find that strengthening line ministries might be more efficient. 8 For example, in India, where the identification of HPAI H5N1 was a national priority, with upfront government financial support and technical assistance from the central level, the full cooperation of the states was secured. 9 At the local level, early communication between the human and animal health authorities may reduce the likelihood of the spread of disease from animals to humans. 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES CONCLUSION Recent human outbreaks of zoonotic diseases have unavoidably re- sulted in increased attention to their impacts on national economies, inter- national trade, household livelihoods, and human morbidity and mortality. Recent socioeconomic changes and the increase in international trade have also been critical drivers of zoonotic disease emergence and spread. Disease surveillance is critical for detecting the emergence of zoonotic pathogens in human populations, preventing their spread between animal populations, and preventing transmission to human populations. The ear- lier an emerging pathogen can be detected and eliminated or controlled, the smaller the emergency response and cost will be. In addition, models of disease transmission have been successful in predicting future zoonotic disease outbreaks and trends. They have been used to make informed de- cisions on the relative risks and benefits of preventive measures aimed at managing the risk at low levels prior to infection. Data from surveillance systems are necessary for more accurately predicting future disease out- breaks. Accurately predicting or anticipating a disease outbreak enables local human and animal health authorities to implement prevention and control efforts, averting the need for costly emergency responses. Accurate prediction is important for preventing an outbreak altogether, decreasing an outbreak’s duration, and lessening its impact on national and household economies and on human health. The case for systematic and sustainable zoonotic disease surveillance, as presented in this chapter, is based on the committee’s conclusion that conditions promoting the driving forces for zoonotic disease emergence are intensifying (further discussed in Chapter 3), that technologies and approaches that could be employed to develop a global system are avail- able, and that the socioeconomic and health consequences for humans and animals are too enormous for inaction. REFERENCES Anderson, R. M. 1988. Epidemiological models and predictions. Trop Geogr Med 40(3): S30–S39. Andrews, N. J. 2009. Incidence of variant Creutzfeldt-Jakob disease diagnoses and deaths in the UK January –December 00. Statistics Unit, Centre for Infections, U.K. Health Protection Agency (HPA). London, UK: HPA. http://www.cjd.ed.ac.uk/cjdq60. pdf (accessed March 25, 2009). Anyamba, A., J. P. Chretien, J. Small, C. J. Tucker, and K. J. Linthicum. 2006. Developing global climate anomalies suggest potential disease risks for 2006–2007. Int J Health Geogr 5(60):60. 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE Anyamba, A., J. P. Chretien, J. Small, C. J. Tucker, P. B. Formenty, J. H. Richardson, S. C. Britch, and K. J. Linthicum. 2007. Forecasting the temporal and spatial distribution of a Rift Valley fever outbreak in East Africa: 00–00. Paper presented at the American Society of Tropical Medicine and Hygiene 2007 Conference, Philadelphia, PA, November 4–8. Anyamba, A., J. P. Chretien, J. Small, C. J. Tucker, P. B. Formenty, J. H. Richardson, S. C. Britch, D. C. Schnabel, R. L. Erickson, and K. J. Linthicum. 2009. Prediction of a Rift Valley fever outbreak. Proc Natl Acad Sci U S A 106(3):955–959. AusAID (Australian Agency for International Development). 2009. AusAID assistance to combat avian influenza and other emerging and resurging zoonotic diseases—Total commitments since 00. http://www.ausaid.gov.au/keyaid/avian/assistance.pdf (accessed April 11, 2009). Barry, J. M. 2005. The great influenza: The epic story of the deadliest plague in history. Lon- don, UK: Penguin Books. Bernoulli, D. 1766. Essai d’une nouvelle analyse de la mortalité causée par la petite vérole et des avantages de l’inoculation pour la prévenir. In Histoire de l’académie royale des sci- ences, avec mémoires de mathématique et de physique. Paris, France. Pp. 1–40. Brahmbhatt, M. 2006. Economic impacts of avian influenza propagation. Presented during the First International Conference on Avian Influenza in Humans, Institut Pasteur, Paris, France, June 29. Brahmbhatt, M., and A. Dutta. 2008. On SARS type economic effects during infectious disease control. Policy research working paper 4466. Washington, DC: The World Bank. Burkom, H. S. 2003. Biosurveillance applying scan statistics with multiple, disparate data sources. J Urban Health 80(2 Suppl 1):i57–i65. Burns, A., D. van der Mensbrugghe, and H. Timmer. 2008. Evaluating the economic conse- quences of avian influenza. Washington, DC: The World Bank. Cash, R. A., and V. Narasimhan. 2000. Impediments to global surveillance of infectious dis- eases: Consequences of open reporting in a global economy. Bull World Health Organ 78(11):1358–1367. Caspari, C., M. Christodoulou, and E. Monti. 2007. Prevention and control of animal diseases worldwide: Economic analysis—Prevention versus outbreak costs. Final report (Part I). Berlin, Germany: Agra CEAS Consulting. CDC (U.S. Centers for Disease Control and Prevention). 1994. International notes update: Human plague—India, 1994. MMWR 43(41): 761–762. CDC. 2007. Rift Valley fever outbreak—Kenya, November 2006–January 2007. MMWR 56(4):73–76. CDC. 2008. West Nile virus activity in the United States from 1999–2008. Statistics, Surveil- lance, and Control. Atlanta, Georgia: CDC. http://www.cdc.gov/ncidod/dvbid/westnile/ surv&control.htm (accessed January 26, 2009). CDC. 2009. Fact Sheet: Variant Creutzfeldt-Jakob disease. http://www.cdc.gov/ncidod/dvrd/ vcjd/factsheet_nvcjd.htm#surveillance (accessed February 13, 2009). Chomel, B. B., A. Belotto, and F. X. Meslin. 2007. Wildlife, exotic pets, and emerging zoono- ses. Emerg Infect Dis 13(1):6–11. Chowell, G., N. W. Hengartner, C. Castillo-Chavez, P. W. Fenimore, and J. M. Hyman. 2004. The basic reproductive number of Ebola and the effects of public health measures: The cases of Congo and Uganda. J Theor Biol 229(1):119–126. Collins, K. 2007. Statement of Keith Collins, chief economist, U.S. Department of Agriculture, before the U.S. House of Representatives Committee on Agriculture. October 18. http:// www.usda.gov/oce/newsroom/archives/testimony/2007a/Housetstoutlook10_17_07r2. doc (accessed April 6, 2009). 

0 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES CRS (U.S. Congressional Research Service). 2008a. International illegal trade in wildlife: Threats and U.S. policy, edited by L. S. Wyler and P. A. Sheikh. Washington, DC: Library of Congress. CRS. 2008b. Global health: Appropriations to USAID programs from FY00 through FY00, edited by T. Salaam-Blyther. Washington, DC: Library of Congress. Darby, P. M. 2003. The economic impact of SARS. Special Briefing, Canadian Tourism Research Institute. Ottawa, Ontario: The Conference Board of Canada. http://sso. conferenceboard.ca/documents.aspx?did=539 (accessed January 26, 2009). de Haan, C., T. J. van Veen, B. Brandenburg, J. Gauthier, F. Le Gall, R. Mearns, and M. Siméon. 2001. Livestock development: Implications for rural poverty, the environment, and global food security. Washington, DC: The World Bank. Defra (UK Department for Environment, Food and Rural Affairs). 2009a. Summary of passive surveillance reports in Great Britain. http://www.defra.gov.uk/vla/science/docs/ sci_tse_stats_gboverview.pdf (accessed April 11, 2009). Defra. 2009b. BSE statistics—Schemes. http://www.defra.gov.uk/animalh/bse/statistics/schemes. html (accessed April 11, 2009). Dobson, A., and J. Foufopoulos. 2001. Emerging infectious pathogens of wildlife. Philos Trans R Soc Lond B Biol Sci 356(1411):1001–1012. DTZ Pieda Consulting. 1998. Economic impact of BSE on the UK economy. Report to United Kingdom’s agricultural departments and Her Majesty’s Treasury. Edinburgh, UK: DTZ Pieda Consulting. Easterbrook, J. D., J. B. Kaplan, N. B. Vanasco, W. K. Reeves, R. H. Purcell, M. Y. Kosoy, G. E. Glass, J. Watson, and S. L. Klein. 2007. A survey of zoonotic pathogens carried by Norway rats in Baltimore, Maryland. Epidemiol Infect 135(7):1192–1199. Ebrahim, M., and J. Solomon. 2006. Exotic pets in the U.S. may pose health risk. The Associ- ated Press, November 27. Einsweiler, S. 2008. Monitoring the wildlife trade. Presentation, Fifth Committee Meeting on Achieving Sustainable Global Capacity for Surveillance and Response to Emerging Diseases of Zoonotic Origin, Washington, DC, December 1–2. Elci, C. 2006. The impact of HPAI of the H5N1 strain on economies of affected countries. In Proceedings of the Conference on Human and Economic Resources. First International Conference on Human and Economic Resources, Izmir, Turkey, May 24–25. Izmir, Tur- key: Izmir University of Economics. Pp. 104–117. Engler, M., and R. Parry-Jones. 2007. Opportunity of threat—The role of the European Union in global wildlife trade. Brussels, Belgium: Traffic Europe. FAO (Food and Agriculture Organization of the United Nations). 2002. Japanese encephali- tis/Nipah outbreak in Malaysia. FAO/WHO Global Forum on Food Safety Regulators, Marrakech, Morocco, January 28–30. Rome, Italy: FAO. http://www.fao.org/DOCREP/ MEETING/004/AB455E.HTM, (accessed January 26 2009). FAO. 2006. Impacts of animal disease outbreaks on livestock markets. Committee on Com- modity Problems. 21st Session of the Intergovernmental Group on Meat and Dairy Products, November 14. FAO. 2007. The challenge of renewal. Report of the Independent External Evaluation of the Food and Agriculture Organization of the United Nations (FAO) submitted to the Council Committee for the Independent External Evaluation of FAO (CC-IEE), Septem- ber. Rome, Italy: FAO. ftp://ftp.fao.org/docrep/fao/meeting/012/k0827e02.pdf (accessed July 19, 2009). FAO. 2009. FAOSTAT, FAO Statistical Division. Rome, Italy: FAO. FAO and APHCA (Food and Agriculture Organization of the United Nations Regional Office for Asia and Animal Production and Health Commission for Asia and the Pacific). 2002. Manual on the diagnosis of Nipah virus infection in animals. Bangkok, Thailand: FAO. 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE FAO, OIE, and WHO (Food and Agriculture Organization of the United Nations, World Or- ganization for Animal Health, and World Health Organization). 2005. A global strategy for the progressive control of highly pathogenic avian influenza (HPAI). http://www.fao. org/avianflu/documents/HPAIGlobalStrategy31Oct05.pdf (accessed July 19, 2009). FAO, OIE, WHO, UNSIC, UNICEF (Food and Agriculture Organization of the United Na- tions, World Organization for Animal Health, World Health Organization, United Na- tions System Influenza Coordinator, United Nations Children’s Fund), and World Bank. 2008. Contributing to one world, one health: A strategic framework for reducing risks of infectious diseases at the animal–human–ecosystems interface. Consultation document prepared for the Inter-ministerial Meeting on Avian and Pandemic Influenza, Sharm-el- Sheikh, Egypt, October 14. Ferguson, N. M., D. A. Cummings, S. Cauchemez, C. Fraser, S. Riley, A. Meeyai, S. Iamsirithaworn, and D. S. Burke. 2005. Strategies for containing an emerging influenza pandemic in Southeast Asia. Nature 437(7056):209–214. Ferrari, M. J., O. N. Bjornstad, and A. P. Dobson. 2005. Estimation and inference of R0 of an infectious pathogen by a removal method. Math Biosci 198(1):14–26. Fèvre, E. M., R. W. Kaboyo, V. Persson, M. Edelsten, P. G. Coleman, and S. Cleaveland. 2005. The epidemiology of animal bite injuries in Uganda and projections of the burden of rabies. Trop Med Int Health 10(8):790–798. Ford, T. E., R. R. Colwell, J. B. Rose, S. S. Morse, D. J. Rogers, and T. L. Yates. 2009. Using satellite images of environmental changes to predict infectious disease outbreaks. Emerg Infect Dis 15(9). http://www.cdc.gov/EID/content/15/9/1341.htm (accessed August 28, 2009). Gerson, H., B. Cudmore, N. E. Mandrak, L. D. Coote, K. Farr, G. Baillargeon. 2008. Monitor- ing international wildlife trade with coded species data. Conserv Biol 22(1):4–7. Grants.gov. 2009a. RFI Avian and Pandemic Influenza and zoonotic Disease Program PRE- DICT. file:///N:/Zoonotic%20Diseases/Public%20Access%20File/USAID%20Summary. RFI.Predict.4.16.09.htm (accessed September 24, 2009). Grants.gov. 2009b. RFI Avian and Pandemic Influenza and zoonotic Disease Program RESPOND. file:///N:/Zoonotic%20Diseases/Public%20Access%20File/USAID%20Summary.RFI. Predict.4.16.09.htm (accessed September 24, 2009). Gubler, D. J. 2001. Silent threat: Infectious diseases and U.S. biosecurity. Georgetown J of International Affairs II(2):15–23. Hanna, D., and Y. Huang. 2004. The impact of SARS on Asian economies. Asian Econ Pap 3(1):102–112. Holland, P. W. 1986. Statistics and causal inference. J Am Stat Assoc 81(396):945–960. IOM and NRC (Institute of Medicine and National Research Council). 2008. Achieving sus- tainable global capacity for surveillance and response to emerging disease of zoonotic origin: Workshop report. Washington, DC: The National Academies Press. Jambiya, G., S. H. A. Milledge, and N. Mtango. 2007. “Night time spinach”: Conservation and livelihood implications of wild meat use in refugee situations in Northwestern Tan- zania. Dar es Salaam, Tanzania: TRAFFIC East/Southern Africa. Jenkins, P. T., K. Genovese, and H. Ruffler. 2007. Broken screens: The regulation of live animal importation in the United States. Washington DC: Defenders of Wildlife. http://www. defenders.org/resources/publications/programs_and_policy/international_conservation/ broken_screens/broken_screens_report.pdf (accessed April 11, 2009). Jewell, N. P. 2003. Statistics for epidemiology. Boca Raton, Florida: Chapman & Hall/CRC Press. Jones, K. E., N. G. Patel, M. A. Levy, A. Storeygard, D. Balk, J. L. Gittleman, and P. Daszak. 2008. Global trends in emerging infectious diseases. Nature 451(7181):990–993. 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES Kaplan, K. 2007. Model successfully predicts Rift Valley fever outbreak. Agricultural Research Service, U.S. Department of Agriculture. http://www.ars.usda.gov/is/pr/2007/070216.htm (accessed January 29, 2009). Keyfitz, N. 1972. On future population. J Am Stat Assoc 67 (338):347–363. Kimball, A. M., and R. Davis. 2006. Costs of epidemics in APEC economies. In Plagues, power, and politics: Infectious disease and international policy, edited by A. Price-Smith. Toronto, Canada: Palgrave Publishers. Kimball, A. M., and K. Taneda. 2004. Emerging infections and global trade: A new method of gauging impact. Rev Sci Tech 23(3):753–760. King, L. J. 2004. Introduction—Emerging zoonoses and pathogens of public health concern. Rev Sci Tech 23(2):429–430. Knobel, D. L., S. Cleaveland, P. G. Coleman, E. M. Fevre, M. I. Meltzer, M. E. Miranda, A. Shaw, J. Zinsstag, and F. X. Meslin. 2005. Re-evaluating the burden of rabies in Africa and Asia. Bull World Health Organ 83(5):360–368. Knorr-Held, L., and S. Richardson. 2003. A hierarchical model for space-time surveillance data on meningococcal disease incidence. Appl Stat 52(2):169–183. Koonse, B. 2008. Regulators and aquaculture certification: Can we use it? Presentation, FAO Aquaculture Certification Workshop, Silver Spring, Maryland, May 29–30. http:// library.enaca.org/certification/washington08/presentation-koonse.pdf (accessed January 22, 2009). Legrand, J., R. F. Grais, P. Y. Boelle, A. J. Valleron, and A. Flahault. 2007. Understanding the dynamics of Ebola epidemics. Epidemiol Infect 135(4):610–621. Lembo, T., K. Hampson, D. T. Haydon, M. Craft, A. Dobson, J. Dushoff, E. Ernest, R. Hoare, M. Kaare, T. Mlengeya, C. Mentzel, and S. Cleaveland. 2008. Exploring reservoir dynam- ics: A case study of rabies in the Serengeti ecosystem. J Appl Ecol 45(4):1246–1257. Longini, I. M., Jr., A. Nizam, S. Xu, K. Ungchusak, W. Hanshaoworakul, D. A. Cummings, and M. E. Halloran. 2005. Containing pandemic influenza at the source. Science 309(5737):1083–1087. Marano, N. 2008. CDC’s role in preventing introduction of zoonotic diseases via animal importation. Presentation, Fifth Committee Meeting on Achieving Sustainable Global Capacity for Surveillance and Response to Emerging Diseases of Zoonotic Origin, Wash- ington, DC, December 1–2. Marano, N., P. M. Arguin, and M. Pappaioanou. 2007. Impact of globalization and animal trade on infectious disease ecology. Emerg Infect Dis 13(12):1807–1809. Massad, E., M. N. Burattini, L. F. Lopez, and F. A. Coutinho. 2005. Forecasting versus projection models in epidemiology: The case of the SARS epidemics. Med Hypotheses 65(1):17–22. McKibbin, W., and A. Sidorenko. 2006. Global macroeconomic consequences of pandemic influenza. Sydney, Australia: Lowy Institute. McMichael, A. J. 2004. Environmental and social influences on emerging infectious diseases: Past, present and future. Philos Trans R Soc Lond B Biol Sci 359(1447):1049–1058. Mills, C. E., J. M. Robins, and M. Lipsitch. 2004. Transmissibility of 1918 pandemic influ- enza. Nature 432(7019):904–906. Morens, D. M., G. K. Folkers, and A. S. Fauci. 2004. The challenge of emerging and re- emerging infectious diseases. Nature 430(6996):242–249. Murphy, F. A. 2008. Emerging zoonoses: The challenge for public health and biodefense. Prev Vet Med 86(3–4):216–223. OAU-IBAR (Organization of African Unity-Interafrican Bureau for Animal Resources). 2009. Pan-African programme for the control of epizootics. http://www.au-ibar.org/ ach_animhealth/pace.html (accessed April 11, 2009). 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE Odiit, M., P. G. Coleman, W. C. Liu, J. J. McDermott, E. M. Fèvre, S. C. Welburn, and M. E. Woolhouse. 2005. Quantifying the level of under-detection of trypanosoma brucei rhodesiense sleeping sickness cases. Trop Med Int Health 10(9):840–849. OECD and WHO (Organisation for Economic Co-operation and Development and World Health Organization). 2003. Food borne disease in the OECD countries: Present state and economic costs. Paris, France: OECD. Osterholm, M. T. 2005. Preparing for the next pandemic. Foreign Af 84(4):24–37. Otte, J., D. Roland-Holst, and D. Pfeiffer. 2006. HPAI control measures and household in- comes in Vietnam. Rome, Italy: FAO Pro-Poor Livestock Policy Initiative. Plowright, R. K., S. H. Sokolow, M. E. Gorman, P. Daszak, and J. E. Foley. 2008. Causal inference in disease ecology: Investigating ecological drivers of disease emergence. Front Ecol Environ 6(8):420–429. Pomareda, C. 2001. Propuesta de programa hemisférico de sanidad agropecuaria e inocuidad de alimentos. Presentada por el IICA a consideración de Organismos Internacionales de Financiamiento de Desarrollo, Agencias de Cooperación Bilateral. San José, Costa Rica. Presanis, A. M., D. De Angelis, D. J. Spiegelhalter, S. Seaman, A. Goubar, and A. E. Ades. 2008. Conflicting evidence in a Bayesian synthesis of surveillance data to estimate human immunodeficiency virus prevalence. J R Stat Soc Ser A 171(4):915–937. Price-Smith, A. T. 1998. Contagion and chaos: Infectious disease and its effects on global se- curity and development. University of Toronto Centre for International Studies Working Paper –00. Toronto, Ontario: Centre for International Studies. Sertsou, G., N. Wilson, M. Baker, P. Nelson, and M. G. Roberts. 2006. Key transmission parameters of an institutional outbreak during the 1918 influenza pandemic estimated by mathematical modelling. Theor Biol Med Model 3(38):38. Shaman, J., J. F. Day, and M. Stieglitz. 2003. St. Louis encephalitis virus in wild birds during the 1990 south Florida epidemic: The importance of drought, wetting conditions, and the emergence of Culex nigripalpus (diptera: Culicidae) to arboviral amplification and transmission. J Med Entomol 40(4):547–554. Shaman, J., J. Day, M. Stieglitz, S. Zebiak, and M. Cane. 2006. An ensemble seasonal forecast of human cases of St. Louis encephalitis in Florida based on seasonal hydrologic fore- casts. Clim Change 75(4):495–511. Singer, R. S., L. A. Cox, Jr., J. S. Dickson, H. S. Hurd, I. Phillips, and G. Y. Miller. 2007. Modeling the relationship between food animal health and human foodborne illness. Prev Vet Med 79(2–4):186–203. Singh, S. P., D. C. Reddy, M. Rai, and S. Sundar. 2006. Serious underreporting of vis- ceral leishmaniasis through passive case reporting in Bihar, India. Trop Med Int Health 11(6):899–905. Stephenson, J. 2003. Monkeypox outbreak a reminder of emerging infections vulnerabilities. JAMA 290(1):23–24. Taylor, L. H., S. M. Latham, and M. E. Woolhouse. 2001. Risk factors for human disease emergence. Philos Trans R Soc Lond B Biol Sci 356(1411):983–989. The BSE Inquiry. 2000. Economic impact and international trade. In The BSE Inquiry: The inquiry into BSE and variant CJD in the United Kingdom. Vol. 10. London, UK: Stationery Office. http://www.bseinquiry.gov.uk/report/volume10/chapted3.htm#273979 (accessed January 14, 2009). The end of cheap food. 2007. The Economist 385(8558):11–12. Theile, S., A. Steiner, and K. Kecse-Nagy. 2004. Expanding borders: New challenges for wild- life trade controls in the European Union. Brussels, Belgium: TRAFFIC Europe. Turner, R. M., D. J. Spiegelhalter, G. C. Smith, and S. Thompson. 2009. Bias modelling in evidence synthesis. J R Stat Soc Ser A 172(1):21–47. 

 GLOBAL SURVEILLANCE AND RESPONSE TO zOONOTIC DISEASES UNWTO (United Nations World Tourism Organization). 2009. Quick overview of key trends. UNWTO World Tourism Barometer 7(1):1–9. http://unwto.org/facts/eng/pdf/barometer/ UNWTO_Barom09_1_en_excerpt.pdf (accessed March 26, 2009). U.S. House of Representatives, Committee on Natural Resources. 2008. Poaching American security: Impacts of illegal wildlife trade. 110th Cong., March 5. USITC (U.S. International Trade Commission). 2008. Global beef trade: Effects of animal health, sanitary, food safety, and other measures on U.S. beef exports. Investigation no. 332-488. Washington, DC: USITC. http://hotdocs.usitc.gov/docs/pubs/332/pub4033.pdf (accessed February 18, 2009). van Zwanenberg, P., and E. Millstone. 2002. Mad cow disease 1980s–2000: How reassurances undermined precaution. In Late lessons from early warnings: The precautionary prin- ciple –000. Environmental issue report No 22. Copenhagen, Denmark: European Environment Agency. Vorou, R. M., V. G. Papavassiliou, and S. Tsiodras. 2007. Emerging zoonoses and vector-borne infections affecting humans in Europe. Epidemiol Infect 135(8):1231–1247. Vynnycky, E., A. Trindall, and P. Mangtani. 2007. Estimates of the reproduction numbers of Spanish influenza using morbidity data. Int J Epidemiol 36(4):881–889. Wells, G. A., A. C. Scott, C. T. Johnson, R. F. Gunning, R. D. Hancock, M. Jeffrey, M. Dawson, and R. Bradley. 1987. A novel progressive spongiform encephalopathy in cattle. Vet Rec 121(18):419–420. WHO (World Health Organization). 2003. Consensus document on the epidemiology of severe acute respiratory syndrome (SARS). WHO/CDS/CSR/GAR/2003.11. Geneva, Switzerland: WHO. WHO. 2004. Summary of probable SARS cases with onset of illness from  November 00 to  July 00. Geneva, Switzerland: WHO. WHO. 2005a. WHO welcomes pandemic influenza response modelling papers, August . http://www.who.int/mediacentre/news/statements/2005/s08/en/index.html (accessed March 26, 2009). WHO. 2005b. Responding to the avian influenza pandemic threat: Recommended strategic actions. Epidemic and Pandemic Alert and Response. WHO/CDS/CSR/GIP/2005.8 Ge- neva, Switzerland: WHO. WHO. 2007a. World health report 00. Geneva, Switzerland: WHO. WHO. 2007b. Rift Valley fever in Kenya, Somalia and the United Republic of Tanzania. Disease Outbreak News Alerts, May 9. Geneva, Switzerland: WHO. http://www.who. int/csr/don/2007_05_09/en/index.html (accessed February 10, 2009). WHO. 2007c. Outbreaks of Rift Valley fever in Kenya, Somalia and United Republic of Tan- zania, December 2006–April 2007. Wkly Epidemiol Rec 82(20):169–178. http://www. who.int/wer/2007/wer8220.pdf (accessed March 26, 2009). WHO. 2009. Cumulative number of confirmed human cases of avian influenza A/(HN) reported to WHO,  April 00 update. Geneva, Switzerland: WHO. http://www.pandemicflunews. us/cumulative-number-of-confirmed-human-cases-of-avian-influenza-ah5n1-reported-to- who-5/ (accessed April 23, 2009). Willitts-King, B., A. Smith, and L. Sims. 2008. Evaluation of United Nations System In- fluenza Coordination (UNSIC)—Final report. http://www.undg.org/docs/9411/UNSIC- evaluation-final-report-submitted-July-22-2008.pdf (accessed March 25, 2009). Wolfe, N. D., C. P. Dunavan, and J. Diamond. 2007. Origins of major human infectious diseases. Nature 447(7142):279–283. Woolhouse, M. E. 2008a. Epidemiology: Emerging diseases go global. Nature 451(7181): 898–899. Woolhouse, M. 2008b. Exemplary epidemiology. Nature 453(7191):34. 

 MAKING THE CASE FOR zOONOTIC DISEASE SURVEILLANCE Woolhouse, M., and R. Antia. 2008. Emergence of new infectious diseases. In Evolution in Health and Disease, 2nd ed., edited by S. C. Stearns and J. K. Koella. Oxford, UK: Ox- ford University Press. Pp. 215–228. Woolhouse, M. E., and E. Gaunt. 2007. Ecological origins of novel human pathogens. Crit Rev Microbiol 33(4):231–242. Woolhouse, M. E., and S. Gowtage-Sequeria. 2005. Host range and emerging and reemerging pathogens. Emerg Infect Dis 11(12):1842–1847. Woolhouse, M. E., L. H. Taylor, and D. T. Haydon. 2001. Population biology of multihost pathogens. Science 292(5519):1109–1112. Woolhouse, M. E. J., R. Howey, E. Gaunt, L. Reilly, M. Chase-Topping, and N. Savill. 2008. Temporal trends in the discovery of human viruses. Proc R Soc B 275(1647):2111–2115. World Bank. 2005. East Asia update: Countering global shocks. November. Washington, DC: The World Bank. http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/ EASTASIAPACIFICEXT/EXTEAPHALFYEARLYUPDATE/0,,contentMDK:20708543 ~ menuPK:550232~pagePK:64168445~piPK:64168309~theSitePK:550226,00. html (accessed October 17, 2008). World Bank. 2008a. World development report 00: Agriculture for development. Washing- ton, DC: The World Bank. World Bank. 2008b. What’s driving the wildlife trade? A review of expert opinion on economic and social drivers of the wildlife trade and trade control efforts in Cambodia, Indonesia, Lao PDR, and Vietnam. Washington, DC: The World Bank. World Resources Institute, United Nations Environment Programme, United Nations Develop- ment Programme, and the World Bank. 1996. Box 2.3 The Black Death revisited: India’s 1994 plague epidemic. In World Resources –: The urban environment. New York: Oxford University Press. WTO (World Trade Organization). 2008. International trade statistics. Geneva, Switzerland: WTO. WTO. 2009. International Trade Statistics retrieved May 27, 2009, from WTO Statistics Database. Geneva, Switzerland: WTO. Zohrabian, A., M. I. Meltzer, R. Ratard, K. Billah, N. A. Molinari, K. Roy, R. D. Scott II, and L. R. Petersen. 2004. West Nile virus economic impact, Louisiana, 2002. Emerg Infect Dis 10(10):1736–1744.