NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US). Improving Food Safety Through a One Health Approach: Workshop Summary. Washington (DC): National Academies Press (US); 2012.

Cover of Improving Food Safety Through a One Health Approach

Improving Food Safety Through a One Health Approach: Workshop Summary.

Show details

A2ONE HEALTH AND HOTSPOTS OF FOOD-BORNE EIDS

, , and 3.

Summary

In this section, we focus on a One Health approach to food-borne emerging infectious diseases (EIDs), their causes, global patterns, and the drivers of their emergence. First, we review two case studies that show the complexity of food-borne pathogen emergence across the One Health domain. Second, we examine the composition of food-borne diseases with respect to their causal agents (pathogen type), their association with pathogens of zoonotic origin, and their apparent disassociation with pathogens that show drug resistance. Third, we analyze the socioeconomic, environmental, and ecological drivers of food-borne EID events. Finally, we use published, spatially explicit information on the drivers of disease emergence to produce a preliminary “hotspot” map that reveals the epicentres, or hotspots, of food-borne EID events globally.

Introduction

One Health's focus on the intersection of human, domestic animal, and environmental health is ideally suited to managing emerging zoonoses. However, the patterns of emergence are complex and poorly understood and for food-borne infections may involve multiple pathways. Food-borne infections can include directly transmitted or vector-borne diseases, for example, Rift Valley fever (Arzt et al., 2010). Single strains of drug-resistant microbes can infect livestock, wildlife, and humans (e.g., E. coli O157:H7) (Hughes et al., 2009; Nielsen et al., 2004; Rahn et al., 1997). Finally, viral pathogens that originate in wildlife may be driven to emerge by the intensification of livestock production (Pulliam et al., 2011) or by contamination of bush meat (Wolfe et al., 2005) or other food sources (Khan et al., 2011). Our ability to predict the emergence of food-borne infections is hampered by this complexity. However, recent efforts to analyse disease emergence (Jones et al., 2008; Taylor et al., 2001) have provided a strategy that can be adapted to analyzing the origins of food-borne infections.

Following our first efforts to predict global patterns of disease emergence (Jones et al., 2008), we have continued to compile data on human EID events and their drivers under the aegis of the U.S. Agency for International Development–funded Emerging Pandemic Threats PREDICT project (Daszak, 2011). In the updated database, when the EID events are classified according to their disease transmission modes (Figure A2-1), we find that food-borne pathogens are responsible for 14.9 percent of known EIDs.

A bar graph showing the proportion of EID events categorized by transmission mode

FIGURE A2-1

Proportion of EID events categorized by transmission mode.

In this section, we focus on food-borne EIDs, their causes, global patterns, and the drivers of their emergence. First, we review two case studies that show the complexity of food-borne pathogen emergence across the One Health domain. Second, we examine the composition of food-borne diseases with respect to their causal agents (pathogen type), their association with pathogens of zoonotic origin, and their apparent disassociation with pathogens that show drug resistance. Third, we analyze the socioeconomic, environmental, and ecological drivers of food-borne EID events. Finally, we use published, spatially explicit information on the drivers of disease emergence to produce a preliminary “hotspot” map that reveals the epicentres, or hotspots, of food-borne EID events globally.

Food-Borne, Wildlife-Origin Pathogens: Two Case Studies

Nipah virus (NiV) is a paramyxovirus that first emerged in Malaysia in 1999, causing encephalitis with a 40 percent case fatality rate in humans (Chua et al., 2000). The virus originated in fruit bats of the genus Pteropus but was first transmitted to domestic pigs, which amplified the virus via a rapidly spreading respiratory infection. Subsequent transmission to people occurred via droplets or fomites contaminated with pig saliva. The initial spillover of NiV seems to have occurred when fruit bats fed on mango and other fruit trees planted next to pigsties at the index farm as a source of additional income and to increase shade. The question remained: Why did it suddenly emerge in this pig farm and not in pig farms 20 years earlier or 20 years later?

To answer this question we analyzed pig production and the age structure of NiV dynamics within the index farm population (Pulliam et al., 2011). We produced a mathematical model, parameterized with detailed data from the index farms and other similar farms still in existence in Malaysia today. This model allowed us to re-create the conditions of the farm when NiV first emerged and to test hypotheses on the drivers of its emergence. Our analyses suggest that repeated introduction of NiV from bats changed infection dynamics in pigs. Initial viral introduction produced an epizootic that drove itself to extinction within 1 to 2 months. Subsequent introduction into a now partially immune population, coupled with the gradual loss of maternal antibodies in pigs born to sows infected in the initial outbreak, led to ideal conditions for pathogen persistence and a prolonged window of spillover to people and regional spread as infected pigs were sold. The structured, compartmentalized nature of the index farm was critical to the emergence of NiV and was a product of agricultural intensification.

A similar scenario surrounds the emergence of highly pathogenic influenza A/H5N1. This virus is able to infect wild waterbirds, domestic poultry, and humans, and its emergence is linked to both intensive production of poultry and the patterns of rice farming within Southeast Asia. When rice is double-cropped, it attracts ducks throughout the year and allows greater potential for new strains of influenza to cross over into pigs and for subsequent crossover of those strains (Gilbert et al., 2008). Analysis of the patterns of double-cropping in Southeast Asia shows that it is possible to predict the risk of its presence throughout the region based on the type of agricultural system (Gilbert et al., 2008). Poultry production in this region includes large intensive and small “backyard” farms, all connected via trade routes into markets and through the supply of breeding stock and their contact with wild birds. We have used a similar modeling approach for A/H5N1 to examine how farm size and connectivity matter as risk factors for the emergence of avian influenza. Our modeling shows that both factors interact to produce specific conditions conducive to outbreaks. When the vast majority of farms are of small size, outbreaks occur more frequently and last longer, but they involve few individual birds and therefore have a lower risk of infecting people. When farms are poorly connected these outbreaks die out because of stochastic factors. When large intensive farms predominate, outbreaks are very few in number, but their duration is relatively short because so many birds die in such a short space of time that the cause is rapidly recognized and the farm culled. The peak in duration and intensity of outbreaks occurs when there is a mixture of intensive and backyard farming. These are the conditions that occur most commonly in Southeast Asia because of the rapid growth of some economies and efforts to intensify poultry production.

Causes, Patterns, and Drivers of Food-Borne EIDs

How important are food-borne infections in the context of global disease emergence events? Going back to Figure A2-1, approximately 15 percent of human EID events are associated with food-borne transmission pathways. With 475 EID events in the updated database, this translates to 71 separate food-borne EIDs, at an average emergence rate of just under one completely new, previously unknown EID event per year reported globally.

When broken down by causal pathogen type (Figure A2-2), food-borne EIDs are usually bacterial in origin, with smaller proportions of protozoan and helminth-driven diseases. While bacteria are also the major causes of EIDs associated with the contaminated environment and fomites, food-borne EIDs are generally more common and therefore account for the highest number of EIDs of bacterial origin (50) among all of the transmission modes. Hence, when bacteria are the causal agent implicated in EID events, they are more likely to be food-borne than of any other transmission mode. In contrast to the other transmission mode groups, food-borne EIDs are very rarely viral, accounting for only one (1.4 percent) food-borne EID (hepatitis A) compared to ∼20 to 45 percent (average 30.9 percent) in the other groups. However, many viral pathogens (e.g., NiV and H5N1) are considered simply zoonotic because the role of food-borne transmission is either less well known or less well understood.

A bar graph showing the number of EID events per transmission by pathogen type

FIGURE A2-2

Number of EID events per transmission mode classified by pathogen type.

Our analyses suggest that the vast majority of food-borne EIDs are indeed zoonotic; in fact, an even higher proportion of food-borne EIDs are zoonotic (84.5 percent) than the background rate of all EIDs in the updated database (62.3 percent) and of any other transmission mode (Figure A2-3). Clearly, pathogens from animals entering the food-production chain are of significant concern for their potential to become EIDs.

A bar graph showing the number of EID events per transmission by zoonotic origin

FIGURE A2-3

Number of EID events per transmission mode categorized by zoonotic origin.

One of our earlier findings (Jones et al., 2008) was that a majority (54.3 percent) of human EIDs were bacterial/rickettsial in origin, reflecting a large number (20.8 percent of all EIDs) of new drug-resistant pathogen strains. We show above that if an EID was identified as being caused by bacteria, it was most likely to be food-borne, and similarly if an EID was linked with food it was most likely to be bacterial than of another transmission mode. Given the propensity of bacteria to develop drug resistance, and the abundance of food-borne infections of bacterial origin, is there any evidence that food-borne pathogens are contributing to new drug-resistant diseases?

Perhaps surprisingly, the answer is no: when EID events are split into categories reflecting the presence or absence of drug resistance (ignoring for a moment the secondary split on whether the pathogen was zoonotic or not), food-borne pathogens are very unlikely to be drug resistant (Figure A2-4). Although it is true that drug resistance is relatively infrequently observed across most transmission modes (the exception being fomite-associated EIDs), resistance is particularly infrequent in food-borne (as well as vector-borne) EIDs. Hence, even though bacteria are quite likely to cause food-borne EIDs and bacteria also cause the majority of new drug-resistant diseases, this is quite unlikely to occur together, resulting in very few drug-resistant food-borne EIDs. Why is drug resistance not more common in this group?

Two bar graphs showing the proportion of drug resistant and nonresistant EID events of zoonotic or non-zoonotic origin

FIGURE A2-4

Proportion of drug-resistant and nonresistant EID events of zoonotic (1), or non-zoonotic origin (0).

The answer may be related to whether the causal agent is zoonotic or not. Generally speaking, there is a low frequency of zoonotic EIDs that exhibit drug resistance (6.0 percent), regardless of the transmission mode (Figure A2-4). Non-zoonotic EIDs are far more likely to be associated with drug resistance (40.9 percent), again across all groups. This is consistent with the idea that new drug-resistant pathogens result from selection on our own circulating pathogens by the routine use of antimicrobial drugs, and not on the pathogens circulating in the food industry that originate in animals. In other words, even though most food-borne EIDs are caused by bacteria, which generally show high potential for becoming drug resistant, the fact that most food-borne EIDs are zoonotic means that the group is quite unlikely to experience strong selection pressure from routine drug administration in human patients. Obviously there are limitations to this type of analysis, particularly in how extensive the data are, but it is clear that this issue is an important target for future research.

Finally, what are the drivers of food-borne pathogens, and are they an ongoing concern for their EID potential? As we have seen, food-borne EIDs are common, usually zoonotic, usually bacterial, and not likely to exhibit drug resistance. So what factors are driving them to emerge? What factors are allowing them to enter and circulate within the food-production system to subsequently cause disease in humans? In Figure A2-5, we analyse the underlying drivers listed in a previous Institute of Medicine report (IOM, 2003) for food-borne EIDs and find that the vast majority of food-borne EIDs are associated with “technology and industry,” and to a lesser extent with “international trade and commerce” and “human susceptibility to infection.” This is consistent with previous studies that have suggested that outbreaks of food-borne infections are likely to be associated with changes in livestock production and centralization of slaughtering and processing (IOM, 2003; Tauxe, 1997). As a result of these analyses, we can hypothesize that the global distribution of food-borne EIDs is driven by a process of intensive production of livestock and food, not simply the number of livestock produced in a region.

A bar graph showing the association of food-borne EIDs with other drivers

FIGURE A2-5

Association of food-borne EIDs with other drivers. SOURCE: Following IOM (2003)

Food-Borne EID Hotspots

Our previous approach to predicting the future geographic origins of new EIDs (Jones et al., 2008) can be adapted for food-borne EIDs. This approach involves identifying the geographic and temporal origins of previous disease emergence events and correcting them for surveillance biases. We then identify correlations between these and purported socioeconomic (demography, travel, trade), environmental (climate, land cover), and ecological drivers (biodiversity, species interactions). Considering all EIDs together, these models suggest that surveillance should be directed toward regions of high biodiversity and dense human populations, which mainly occur in tropical and subtropical latitudes (Jones et al., 2008). When we adapt this approach to food-borne EID events and use the same drivers as in our earlier analysis, human population density and human population growth emerge as the most important in the emergence of novel food-borne outbreaks (Figure A2-6). This suggests that rapidly developing regions are the sites where most novel food-borne pathogens will emerge in future. This may appear to be in conflict with Figure A2-5; however, this is because the spatial analyses have so far been limited primarily by the availability of relevant spatial information. Human population density and growth are likely to be meaningful proxies for a range of other mechanistically more relevant drivers. One of our main goals more recently has thus been to improve our database of detailed drivers. We have, for example, begun to include spatial information on land-use change (cropping and pasture) and livestock density (including cattle, pigs, buffalo, goats, and sheep) into the predictive models. We are currently validating these new data sources for use in future models.

A world map showing the relative risk of food-borne EID events

FIGURE A2-6

Relative risk of food-borne EID events, based on Jones et al. (2008). Human population density and human population growth are the most important variables. SOURCE: Reprinted by permission from Macmillan Publishers Ltd: Nature, (Jones et al., 2008).

We conclude that food-borne EIDs are a common and important group within emerging diseases that emerge through complex pathways involving wildlife, livestock, and humans. They are therefore ideal candidates for a One Health approach but have rarely been considered in this way previously. Our analyses show that the majority of food-borne EIDs (1) are bacterial; (2) are, if bacterial, more likely to be food-borne than of any other transmission mode; (3) are zoonotic; (4) do not tend to be associated with drug resistance, perhaps because zoonotic pathogens in general show little tendency to become resistant; and (5) are driven by changes in human food-production systems, including intensification and centralization as human populations grow larger and more dense.

References

  • Arzt J, White WR, Thomsen BV, Brown CC. Agricultural diseases on the move early in the third millennium. Veterinary Pathology. 2010;47:15–27. [PubMed: 20080480]
  • Chua KB, Bellini WJ, Rota PA, Harcourt BH, Tamin A, Lam SK, Ksiazek TG, Rollin PE, Zaki SR, Shieh W, Goldsmith CS, Gubler DJ, Roehrig JT, Eaton B, Gould AR, Olson J, Field H, Daniels P, Ling AE, Peters CJ, Anderson LJ, Mahy BW. Nipah virus: A recently emergent deadly paramyxovirus. Science. 2000;288:1432–1435. [PubMed: 10827955]
  • Daszak P. Smart surveillance: Analyzing environmental drivers of emergence to predict and prevent pandemics. Ecohealth. 2011;7:S12–S13.
  • Gilbert M, Xiao XM, Pfeiffer DU, Epprecht M, Boles S, Czarnecki C, Chaitaweesub P, Kalpravidh W, Minh PQ, Otte MJ, Martin V, Slingenbergh J. Mapping H5N1 highly pathogenic avian influenza risk in Southeast Asia. Proceedings of the National Academy of Sciences of the United States of America. 2008;105:4769–4774. [PMC free article: PMC2290786] [PubMed: 18362346]
  • Hughes LA, Bennett M, Coffey P, Elliott J, Jones TR, Jones RC, Lahuerta-Marin A, McNiffe K, Norman D, Williams NJ, Chantrey J. Risk factors for the occurrence of Escherichia coli virulence genes eae, stx1 and stx2 in wild bird populations. Epidemiology and Infection. 2009;137:1574–1582. [PubMed: 19327201]
  • IOM (Institute of Medicine) Microbial threats to health: Emergence, detection, and response. Washington, DC: The National Academies Press; 2003.
  • Jones KE, Patel N, Levy M, Storeygard A, Balk D, Gittleman JL, Daszak P. Global trends in emerging infectious diseases. Nature. 2008;451:990–994. [PubMed: 18288193]
  • Khan MSU, Hossain J, Gurley ES, Nahar N, Sultana R, Luby SP. Use of infrared camera to understand bats' access to date palm sap: Implications for preventing Nipah virus transmission. Ecohealth. 2011;7:517–525. [PubMed: 21207105]
  • Nielsen EM, Skov MN, Madsen JJ, Lodal J, Jespersen JB, Baggesen DL. Verocytotoxin-producing Escherichia coli in wild birds and rodents in close proximity to farms. Applied and Environmental Microbiology. 2004;70:6944–6947. [PMC free article: PMC525191] [PubMed: 15528570]
  • Pulliam JR, Epstein JH, Dushoff J, Rahman SA, Bunning M, Jamaluddin AA, Hyatt AD, Field HE, Dobson AP, Daszak P. Agricultural intensification, priming for persistence and the emergence of Nipah virus: A lethal bat-borne zoonosis. Journal of the Royal Society Interface. 2011;9(66):89–101. [PMC free article: PMC3223631] [PubMed: 21632614] [Cross Ref]
  • Rahn K, Renwick SA, Johnson RP, Wilson JB, Clarke RC, Alves D, McEwen S, Lior H, Spika J. Persistence of Escherichia coli O157:H7 in dairy cattle and the dairy farm environment. Epidemiology and Infection. 1997;119:251–259. [PMC free article: PMC2808848] [PubMed: 9363025]
  • Tauxe RV. Emerging foodborne diseases: An evolving public health challenge. Emerging Infectious Diseases. 1997;3:425–434. [PMC free article: PMC2640074] [PubMed: 9366593]
  • Taylor LH, Latham SM, Woolhouse MEJ. Risk factors for human disease emergence. Philosophical Transactions of the Royal Society of London, B. 2001;356:983–989. [PMC free article: PMC1088493] [PubMed: 11516376]
  • Wolfe ND, Daszak P, Kilpatrick AM, Burke DS. Bushmeat hunting, deforestation and prediction of zoonotic emergence. Emerging Infectious Diseases. 2005;11:1822–1827. [PMC free article: PMC3367616] [PubMed: 16485465]
3

EcoHealth Alliance, 460 West 34th Street, New York, NY 10001, USA.

Copyright © 2012, National Academy of Sciences.
Bookshelf ID: NBK114504

Views

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...