Food

A notice of availability of the Food and Drug Administration (FDA) draft risk assessment on the relationship between Vibrio parahaemolyticus in raw molluscan shellfish and public heath was published in the Federal Register of January 19, 2001 (66 FR 5517). A comment period was established during which FDA actively sought comments, suggestions, and additional data sources. The results of the draft risk assessment were presented for clarification during a public meeting on March 20, 2001 (66 FR 13544). Comments were submitted to the FDA Docket (No. 99N-1075) from nine institutions or individuals. The data and information acquired during the comment period were reviewed and used, as appropriate, to further enhance the risk assessment.

We appreciate the time and effort expended to submit these comments, and have addressed these in this revised risk assessment to the best of our ability. A summary of the modifications made to the draft risk assessment in response to the comments, new data and modeling techniques is provided below. A more detailed discussion of our response to the public comments can be found in Appendix 2.

Modifications Made to the 2001 Draft Vibrio parahaemolyticus Risk Assessment

Topic	Modifications
Assumptions	Additional information was obtained that further the following assumptions: Growth rates of pathogenic and non-pathogenic V. parahaemolyticus are similar; Time required for refrigerated oysters to cool down to temperatures that do not support the growth of V. parahaemolyticus is variable and may range from 1 to 10 hours.
Additional Data/ Information	Prevalence of total and pathogenic V. parahaemolyticus at harvest for Pacific Northwest region (PNW) and Gulf Coast regions; Relationship between water temperature and V. parahaemolyticus levels in oysters; Time-to-refrigeration after harvest for the PNW region.
Modeling techniques	Included intertidal harvesting in the PNW as an additional harvest region; Evaluated mitigation effect of specific reduction levels of V. parahaemolyticus in addition to types of interventions; Included regression-based sensitivity analysis; Added two additional uncertainty parameters (total V. parahaemolyticus in oysters based on water temperature and dose-response relationship) to the examination of factors that influence risk predictions; Oyster meat weights at retail were used rather than those at harvest; Comparison of the model-predicted number of illnesses using both retail survey and epidemiological data

CONTRIBUTORS (2004 Version)

Team Leader: Marianne Miliotis

Project Manager: Marianne Miliotis

Risk Analysis Coordinator: Sherri Dennis

Scientific Advisor: Robert Buchanan

Team Members

Modeling:

John Bowers
Mark Walderhaug

Exposure Assessment:

David Cook
Angelo DePaola
Elisa Elliot
Charles Kaysner
Marleen Wekell
William Watkins

Hazard Characterization:

Donald Burr

Epidemiology:

Karl Klontz
Marianne Ross

Technical Editing:

Robert Buchanan
Sherri Dennis
Louise Dickerson
Lori Pisciotta

ACKNOWLEDGEMENTS (2004 Version)

The Vibrio parahaemolyticus Risk Assessment team greatly appreciates the efforts of the following individuals who provided us with comments, information and assistance for this risk assessment:

Linda Andrews (Mississippi State University)
Enrico Buenaventura and Klaus Schalle (Canadian food Inspection Agency)
Colleen Crowe, Patti Griffin, Arthur Liang, John Painter, Donald Sharp, Cynthia (Stover) Smith, and Robert Tauxe (CDC)
Jessica DeLoach, Kathryn Lofi, Ned Therien, Jennifer Tibaldi, and Patti Waller (Washington State Department of Health)
Robin Downey (Pacific Coast Shellfish Growers Association)
Jeffrey Farber (Health Canada)
Lee Hoines (Washington State Department of Fish and Wildlife)
Mahendra Kothary (FDA/CFSAN)
Donald Kraemer (FDA/CFSAN)
Jeanette Lyon (FDA/CFSAN)
Sherri McGarry (FDA/CFSAN)
Michael Morrissey (Oregon State University)
Lori Pisciotta (FDA/CFSAN)
John Schwarz (Texas A&M University at Galveston)
Jessica Tave (FDA)
Ben Tall (FDA/CFSAN)
FDA Regional Shellfish Experts (Marc Glatzer, Jeremy Mulnick, Tim Sample)
Kirk Wiles (Texas State Health Department)

We are also deeply grateful to Sharon Edelson Mammel for evaluating the quality of data used in the model and to Louis Michael Thomas, Linda Shasti, and Aesha Minter, JIFSAN student interns, for assembling the references cited in the document. We also thank CDC staff for their assistance in providing the epidemiological data used for the dose-response model and the data analysis used to compare the model predictions to the epidemiological data. Our appreciation also goes to David Acheson (FDA), Robert Buchanan (FDA), Don Kraemer (FDA), Angela Ruple (NOAA Fisheries), and Richard Whiting (FDA) for reviewing and providing suggestions to improve the risk assessment documents. The team is also appreciative of the in depth review and evaluation of the model conducted by Clark Carrington (FDA) and Darrel Donahue (University of Maine).

CONTRIBUTORS (2001 Version)

The Vibrio parahaemolyticus risk assessment team members:

Team Leaders: Marianne Miliotis and William Watkins

Exposure Assessment - Harvest Module:
Marleen Wekell (Section Lead), Atin Datta, Elisa Elliot, Walter Hill, Charles Kaysner, Brett Podoski

Exposure Assessment - Post-Harvest Module:
Angelo DePaola and David Cook (Section Co-leads), George Hoskin, Susan McCarthy, William Watkins

Exposure Assessment - Consumption Module:
Michael DiNovi

Epidemiology:
Marianne Ross (Section Lead), Karl Klontz, Debra Street, Babgaleh Timbo

Hazard Characterization/Dose-Response:
Donald Burr (Section Lead), John Bowers, Mahendra Kothary, Wesley Long, Marianne Miliotis, Ben Tall, Mark Walderhaug

Modeling:
John Bowers (Section Lead), Mark Walderhaug

ACKNOWLEDGEMENTS (2001 Version)

The following people provided the V. parahaemolyticus team with comments, information and assistance we needed to accomplish this risk assessment:

Haejung An (Oregon State University)
Fred Angulo, Mary Evans, Nicholas Daniels, Paul Mead and Malinda Kennedy (Centers for Disease Control)
Robert Buchanan (FDA/CFSAN)
Mercuria Cumbo (Department of Marine Resources, Maine)
Sherri Dennis (FDA/CFSAN)
Paul Distefano (FDA/CFSAN)
Robin Downey (Pacific Coast Shellfish Growers Association)
Jan Gooch (National Oceanographic Service)
Michael Kelly (University of British Columbia)
Bill Kramer (Environmental Protection Agency)
Ken Moore, Sandra Sharp (Interstate Shellfish Sanitation Conference)
Mitsuaki Nishibuchi (Kyoto University, Japan)
Gary Richards (USDA/ARS)
Tina Rouse (FDA/CFSAN)
Angela Ruple (National Marine Fisheries Service)
Patricia Schwartz (FDA/CFSAN)
FDA Regional Shellfish Experts
FDA Shellfish Sanitation Team
Molluscan Shellfish Institute
National Advisory Committee for Microbiological Criteria for Food (NACMCF)
Shellfish Industry
State Shellfish Experts
State Health Departments

The team would especially like to thank the FDA/CFSAN Offices and Risk Assessment Consortium members for intensive review of the document in December, as well as Federal employees from other agencies, and Special Government Experts, for review of the document in May. We are also deeply grateful to Lauren Posnick for her outstanding contribution in preparing the interpretive summary of this document, Carolyn Jeletic for excellent technical editing of this document, and Faye Feldstein for assisting with assembling all the references.

Interpretive Summary

PREFACE

This Interpretive Summary provides an overview of the 2004 Food and Drug Administration (FDA) Vibrio parahaemolyticus risk assessment. Its purpose is to briefly describe, in non-technical language, the material covered in the complete risk assessment. This includes background information on Vibrio parahaemolyticus, the techniques and data used to develop the risk assessment, the results of the risk assessment, and the interpretation, implications and limitations of those findings. A full understanding of the risk assessment requires the reader to consider the complete risk assessment. The complete risk assessment may be obtained on the Internet. A printed copy will be provided upon request. Requests may be faxed to the CFSAN Outreach and Information Center at 1-877-366-3322.

INTRODUCTION

Vibrio parahaemolyticus is a bacterium that occurs naturally in coastal marine waters and estuaries (where rivers flow into the sea). It is recognized world-wide as a significant cause of bacterial seafood-borne illness. The United States Centers for Disease Control and Prevention (CDC) estimates that of the approximately 7,880 Vibrio illnesses each year in the United States, approximately 2,800 are estimated to be associated with Vibrio parahaemolyticus and raw oyster consumption. Vibrio parahaemolyticus is normally present in many types of raw seafood, including fish, crustaceans, and molluscan shellfish. It multiplies and colonizes in the gut of filter-feeding shellfish such as oysters, clams, and mussels. Not all strains of Vibrio parahaemolyticus cause illness; on the contrary, pathogenic strains represent a small percentage of the total Vibrio parahaemolyticus present in the environment or seafood.

FDA conducted this "product pathway" risk assessment to characterize the factors influencing the public health impact associated with the consumption of raw oysters containing pathogenic Vibrio parahaemolyticus. This is referred to as a "product pathway" risk assessment because the factors that influence the risk associated with Vibrio parahaemolyticus in oysters are examined from harvest through post-harvest handling to consumption. The risk assessment was conducted in response to outbreaks in 1997 and 1998 in the United States involving more than 700 cases of Vibrio parahaemolyticus illness. These outbreaks renewed concern for this pathogen as a serious foodborne threat to public health and raised concerns about the effectiveness of the risk management guidance available at that time.

SCOPE AND GENERAL APPROACH

This risk assessment was initiated in January 1999 and a draft risk assessment was made available for public comment in 2001. The draft risk assessment has been modified to take into account public comments, to incorporate additional scientific data and knowledge that has become available since 2001, and to take advantage of improvements in modeling techniques. Modifications made to the draft risk assessment are provided in Summary Table 1.

Summary Table 1. Modifications Made to the 2001 Draft *Vibrio parahaemolyticus* Risk Assessment
Topic	Modifications
Assumptions	Additional information was obtained that further support the following assumptions: Growth rates for pathogenic and non-pathogenic Vibrio parahaemolyticus are similar; Time required for refrigerated oysters to cool down to temperatures that do not support the growth of Vibrio parahaemolyticus is variable and may range from 1 to 10 hours.
Additional Data/Information	Prevalence of total and pathogenic Vibrio parahaemolyticus at harvest for Pacific Northwest (PNW) and Gulf Coast regions; Relationship between water temperature and Vibrio parahaemolyticus levels in oysters; Time-to-refrigeration after harvest for the PNW region.
Model/Modeling Techniques	Included intertidal harvesting in the PNW as an additional harvest region; Evaluated mitigation effect of specific reduction levels of Vibrio parahaemolyticus in addition to types of interventions; Included regression-based sensitivity analysis; Added two additional uncertainty parameters (total Vibrio parahaemolyticus in oysters based on water temperature and dose-response relationship) to the examination of factors that influence risk predictions; Oyster meat weights at retail were used rather than those at harvest; Comparison of the model-predicted number of illnesses using both retail survey and epidemiological data.

This risk assessment is based on a quantitative simulation model. The focus is on raw oysters, because that is the food in the United States predominately linked to outbreaks of illness associated with this pathogen since 1997. The risk assessment examines events occurring from oyster harvest to consumption that influence the levels of Vibrio parahaemolyticus likely to be present in raw oysters at the time of consumption. The levels of Vibrio parahaemolyticus in oysters at the time of consumption are influenced by the harvest methods and environmental conditions, as well as the handling of oysters after harvest. These practices and conditions vary considerably among different geographic areas and at different times of the year. Therefore, the model was constructed to predict illnesses for each harvest region and season in the United States. The likelihood and severity of illness following exposure to pathogenic Vibrio parahaemolyticus from consumption of raw oysters was estimated. Once developed, the baseline model was used to develop "what-if" scenarios to evaluate the likely impact of potential intervention strategies on the exposure to pathogenic Vibrio parahaemolyticus from consumption of raw oysters.

The risk assessment had two main objectives:

determine the factors that contribute to the risk of becoming ill from the consumption of pathogenic Vibrio parahaemolyticus in raw oysters
evaluate the likely public health impact of different control measures, including the effectiveness of current and alternative microbiological standards

Summary Figure 1 depicts a schematic representation of the components of the Vibrio parahaemolyticus risk assessment model. The Exposure Assessment model was separated into three modules: harvest, post-harvest, and consumption. The model outputs from the Exposure Assessment were then combined with the Dose-Response model to relate these exposures to public health outcomes. The model inputs are expressed as distributions instead of single point estimates (such as a mean). Using a distribution allows a range of values, each with a specific frequency of occurrence, to be included in the model. Distributions are commonly used in simulation modeling to account for the inherent biological variability in nature and our uncertainty of the "true" values, resulting in a more accurate prediction of the risk.

Data for this risk assessment were obtained from many sources including published and unpublished scientific literature and reports produced by various organizations such as State shellfish control authorities, the CDC, the shellfish industry, the Interstate Shellfish Sanitation Conference (ISSC), and state health departments. In some instances, the conduct of the risk assessment required that assumptions be made when data were incomplete for the purposes of modeling. To the extent possible, research undertaken to address the data gaps identified in the 2001 draft risk assessment have been incorporated into the model. The criteria used to select data for the risk assessment modeling are described in detail in the complete risk assessment.

For the risk assessment, 6 harvest regions and 4 seasons (winter, spring, summer, and fall) were considered separately in the model for a total of 24 region/season combinations (i.e., there were predictions of illnesses for 24 regions/seasons). The oyster harvest regions included: Gulf Coast (Louisiana), Gulf Coast (non-Louisiana), Mid-Atlantic, Northeast Atlantic, Pacific Northwest (Dredged), and Pacific Northwest (Intertidal). In the Gulf Coast, the harvest duration (i.e., time between removal of the oysters from the water to unloading them at the dock) for Louisiana is longer than for other states in that region (Florida, Mississippi, Texas, and Alabama). Since harvest duration can affect the levels of Vibrio parahaemolyticus in raw oysters, the Gulf Coast was divided into these two distinct regions. The Pacific Northwest was also divided into two regions, but in this case it was based on harvest methods, intertidal versus dredged. Oysters harvested in intertidal areas are typically exposed to higher temperatures before refrigeration than those harvested using dredging, leading to the need to define two harvest practice-based regions within the Pacific Northwest.

Summary Figure 1. Schematic Representation of the Vibrio parahaemolyticus Risk Assessment Model [The light grey boxes with black lettering show the Harvest Module, the gray boxes with black lettering

Summary Figure 1. Schematic Representation of the Vibrio parahaemolyticus Risk Assessment Model [The light grey boxes with black lettering show the Harvest Module, the gray boxes with black lettering show the Post-Harvest Module, the dark grey boxes with white lettering show the Consumption Module, the white boxes with black lettering show the Dose-Response model, and the white boxes with dark black outline show the Risk Characterization. Vp= Vibrio parahaemolyticus]

HAZARD IDENTIFICATION

Vibrio parahaemolyticus is a salt tolerant bacterium and a normal inhabitant of the marine environment. This bacterium is found in many types of seafood, including fish, crustaceans, and molluscan shellfish. It was first isolated in 1950 and implicated in an outbreak of food poisoning in Japan. In the United States, the first confirmed outbreak of Vibrio parahaemolyticus illness occurred in Maryland in 1971. Since 1997, several large outbreaks, associated with the consumption of raw oysters, have been reported in the United States. These outbreaks are shown in Summary Table 2.

Summary Table 2. Outbreaks of Illnesses from *Vibrio parahaemolyticus* Associated with Consumption of Raw Oysters in the United States
Year	Location	Number of Cases
1997	Pacific Northwest^a	209
1998	Pacific Northwest^a	48
1998	Texas	416^b
1998	Northeast Atlantic	10
2002	New York	7
2002	New Jersey	11
2004	Alaska	46
^a The Pacific Northwest includes California, Oregon, Washington State, and British Columbia. ^b 296 cases in Texas and 120 cases in other states that were traced back to oysters harvested from Texas.

Human illnesses from ingestion of Vibrio parahaemolyticus have been well documented. Any exposed individual can become infected with Vibrio parahaemolyticus and develop illness. The most common clinical manifestation of Vibrio parahaemolyticus infection is gastroenteritis, an inflammation of the gastrointestinal tract. Gastroenteritis is usually an illness of short duration and moderate severity that is characterized by diarrhea, vomiting, and abdominal cramps. Vibrio parahaemolyticus infections can also lead to septicemia, a severe, life-threatening disease caused by the multiplication of pathogenic microorganisms and/or the presence and persistence of their toxins in circulating blood. Individuals with underlying chronic medical conditions (such as diabetes, alcoholic liver disease, hepatitis, and those receiving immunosuppressive treatments for cancer or AIDS) do not appear to be at a higher risk of acquiring the initial infection than otherwise healthy people. However, individuals with underlying chronic conditions do appear to have a higher risk of the initial infection developing into septicemia.

The CDC estimates that of the total Vibrio illnesses in the United States (average 7,880 per year), there are approximately 4,500 Vibrio parahaemolyticus illnesses and of those approximately 2,800 are estimated to be associated with raw oyster consumption. There have been reports of Vibrio parahaemolyticus illness associated with various types of cooked and raw seafood including crayfish, lobster, shrimp, crab, oysters, and clams. Vibrio illnesses associated with cooked seafood are likely due to inadequate heating or recontamination after cooking. Although thorough cooking destroys Vibrio, oysters are often eaten raw, which may explain why it is the most common seafood associated with Vibrio infection in the United States. Epidemiological data indicate that consumers of raw oysters are 2.8 times more likely to experience Vibrio parahaemolyticus illness compared to non-raw oyster eaters. Food intake surveys indicate that raw shellfish is not a commonly consumed food in the United States: only 10 to 20% of the population consumes raw shellfish at least once a year. Among oyster consumers, raw oysters are typically eaten approximately once every 6 weeks and the typical serving size ranges from 6 to 24 oysters, with 12 being the most frequent.

Not all strains of Vibrio parahaemolyticus cause illness; on the contrary, pathogenic strains generally represent a small percentage of the total Vibrio parahaemolyticus present in the environment or seafood. Pathogenic Vibrio parahaemolyticus strains are more likely to produce symptomatic infections and have one or more distinctive traits that are generally absent in non-pathogenic strains. Two important virulence indicators are the ability to produce thermostable direct hemolysin (TDH) and the ability to produce a related toxin, thermostable related hemolysin (TRH). Hemolysin is an enzyme that breaks down red blood cells on a blood agar plate, which is referred to as the Kanagawa phenomenon. The vast majority of Vibrio parahaemolyticus strains isolated from the stools of patients with Vibrio parahaemolyticus gastroenteritis are TDH-positive (TDH⁺). The role of traits other than TDH has not yet been determined. Therefore, for the purposes of this risk assessment, pathogenic Vibrio parahaemolyticus is defined as strains that are TDH⁺.

Vibrio parahaemolyticus infections occur throughout the year, peaking in spring and summer. Cases are most often associated with the regions of the country within close proximity to marine environments. The geographical distribution of cases attributed to oysters from specific harvest areas likely reflects the propensity for individuals in close proximity to coastal areas to consume raw shellfish. Likewise, the volume of oysters harvested in the U.S. each year varies by season. Approximately 66% of the annual oyster harvest occurs in the winter and fall with the remainder in spring and summer. There are also regional differences in the oyster harvest volume; the Gulf Coast accounts for approximately one-half of the oyster harvest, the Pacific Northwest about a fourth, and less than a tenth from the Mid-Atlantic region. In addition, regional climatic differences (e.g., water temperatures) and post-harvest handling practices influence the levels of Vibrio parahaemolyticus in shellfish and consequently the potential for illness.

HAZARD CHARACTERIZATION

In a quantitative risk assessment, the Hazard Characterization typically entails the determination of a dose-response relationship for a specified population, relating the incidence of an identified adverse effect with the level of exposure to a particular microorganism (or substance). This dose-response relationship is often expressed as a relation between different levels of exposure and the likelihood (or probabilities) that such exposures will result in illness. For this risk assessment, a quantitative relationship was developed to predict the number and severity of illnesses resulting from ingestion of pathogenic Vibrio parahaemolyticus.

A quantitative dose-response model for Vibrio parahaemolyticus was developed based on human clinical feeding studies. The model extrapolates the observed illness rates from the studies to doses and illness rates that are more likely to be encountered with contaminated oysters. Next, the dose-response curve was adjusted to account for the estimates of the annual illness burden (2,800 cases per year) as determined by CDC. This approach is typically referred to as "anchoring" to epidemiological data. There is uncertainty in the dose-response relationship because of the limited data from the clinical studies. This uncertainty was accounted for in the model by multiple curve-fitting of the data.

Summary Figure 2 shows the dose-response model. Using the most likely estimate of the dose-response curve (i.e., the dashed line), the probability of illness is approximately 0.5 (50%) for a dose of approximately 100 million (i.e., 1x10⁸) Vibrio parahaemolyticus cells/serving. This means for every 100 individuals eating a serving of oysters that contains 1x10⁸cells of pathogenic Vibrio parahaemolyticus, approximately 50 individuals will become ill. At lower exposure levels (1x10³ or 1x10⁴), the probability of illness is much lower (<0.001). Using the risk assessment results and available epidemiological data, the likelihood that a Vibrio parahaemolyticus illness (gastroenteritis) will lead to septicemia was determined for healthy and immunocompromised individuals. (See section entitled, Risk Characterization, for the results of the assessment.)

Summary Figure 2. The Dose-Response Model for Vibrio parahaemolyticus (Vp) [The solid line is the best estimate of the model fit to pooled human feeding studies. The dashed line shows the shift adjust

Mean Dose (Vp cells per serving)

Summary Figure 2. The Dose-Response Model for Vibrio parahaemolyticus (Vp) [The solid line is the best estimate of the model fit to pooled human feeding studies. The dashed line shows the shift adjustment so that the model predictions agree with epidemiological surveillance data. MLE denotes the maximum likelihood estimate. ID₅₀ is the dose corresponding to a 50% probability of gastroenteritis.]

EXPOSURE ASSESSMENT

The purpose of the Exposure Assessment is to determine the likelihood of ingesting pathogenic Vibrio parahaemolyticus from consumption of raw oysters, and the likely level of exposure. Insufficient data are available on the levels of pathogenic Vibrio parahaemolyticus in raw oysters at the moment of consumption. Therefore, the model predicts these levels using available data on the factors that influence the levels of the pathogen present in oysters at harvest. These factors include the environmental conditions that contribute to the likely presence of Vibrio parahaemolyticus in oysters at harvest and the impact of post-harvest handling and processing practices on the growth or decline of Vibrio parahaemolyticus in oysters prior to consumption. In addition, the frequency of oyster meals and the amount of oysters consumed per serving were considered.

The Exposure Assessment was divided into three modules that reflect the chain of events from oyster harvest to consumption: Harvest, Post-Harvest, and Consumption. The levels of total and pathogenic Vibrio parahaemolyticus in oysters were estimated for each handling or processing event. The predicted levels of Vibrio parahaemolyticus from each module were used as inputs for the subsequent module (e.g., results from the Harvest module served as the input to the Post-Harvest module). Because Vibrio parahaemolyticus levels may be affected by climate and region-specific oyster harvesting practices, modeling was conducted separately for each of the 24 harvest region/season combinations described in the "Risk Assessment Framework" section above.

Harvest Module. In the Harvest Module of the Exposure Assessment model, factors identified as potentially influencing the variation of levels of Vibrio parahaemolyticus at the time of harvest were evaluated and the effects of those factors that could be suitably quantified were incorporated into the quantitative simulation model.

The available data suggest that a number of factors can affect the presence and growth of Vibrio parahaemolyticus in oysters at the time of harvest. Once present in the environment, Vibrio parahaemolyticus levels are affected primarily by water temperatures and to a lesser extent by salinity levels. Such factors as the amount of zooplankton in the shellfish growing area, the rate of tidal flushing, levels of dissolved oxygen in the water, and the presence of pollutants have less certain effects on Vibrio parahaemolyticus levels. Oyster-specific factors, such as the physiology and health of the oyster also contribute to the ability of Vibrio parahaemolyticus to colonize and grow in the oysters. Bacteriophages, toxins, or other proteins produced by bacterial strains that infect or colonize oysters at the same time as Vibrio parahaemolyticus may affect the survival of the Vibrio parahaemolyticus. In addition, the percentage of the total Vibrio parahaemolyticus that is pathogenic may vary. Several studies suggest that the average percentage of pathogenic Vibrio parahaemolyticus is higher on the West Coast than in other areas of the country.

Although a number of potential factors affecting Vibrio parahaemolyticus levels at the time of harvest were identified, there were little data available to quantify the effects of most of these factors. Furthermore, accompanying analyses indicated that in most instances water temperature is overwhelmingly the primary determinant that controls Vibrio parahaemolyticus levels in oysters. Water salinity was not included as a variable in the model because preliminary modeling indicated that the small variability in water salinity in the major commercial harvest regions was not a strong determinant of Vibrio parahaemolyticus prevalence and growth in oysters. Additionally, the impact on the model of varying water salinity was overshadowed by the impact of varying water temperatures. Levels of pathogenic Vibrio parahaemolyticus in oysters at-harvest were predicted using data on: 1) the relationship between total Vibrio parahaemolyticus in oysters and water temperature, 2) water temperature distributions, and 3) the ratio of pathogenic to total Vibrio parahaemolyticus in oysters.

The relationship between total Vibrio parahaemolyticus levels in oysters and water temperature was modeled based on the assumption that Vibrio parahaemolyticus may be present at levels below the sensitivity of the analytical method (e.g., less than the limit of detection) but not actually zero, even at low temperatures. The distribution of pathogenic Vibrio parahaemolyticus in oysters at harvest was determined using the distribution of total Vibrio parahaemolyticus in oysters at harvest and the appropriate pathogenic percentage for each region (i.e., 2.3% for the Pacific Northwest and 0.18 % for the Gulf Coast, Mid-Atlantic, and Northeast Atlantic regions).

Summary Table 3 provides the predicted mean levels of Vibrio parahaemolyticus at harvest for each of the 24 region/season combinations. Across all regions, the predicted levels are much higher in the warmer months compared to the cooler months. The predicted levels for the Gulf Coast region are considerably higher than the other regions due to the warmer water temperatures. During the summer, the levels of Vibrio parahaemolyticus in the mid-Atlantic and Northeast Atlantic are higher than those of the Pacific Northwest (when harvest occurs by dredging). Even during the summer, air temperatures in the Pacific Northwest are cooler, on average, than in the Gulf and Atlantic regions. However, exposure to higher temperatures for longer time periods, such as occurs during intertidal harvest in some Pacific Northwest areas, allows for additional growth, resulting in an increase of total and pathogenic Vibrio parahaemolyticus to levels higher than that of the Northeast Atlantic region.

Summary Table 3. Predicted Mean Levels of *Vibrio parahaemolyticus* per gram in Oysters At-Harvest
Region	Level	Summer^a	Fall^a	Winter^a	Spring^a
Gulf Coast (Louisiana)^b	Total	2,100	220	52	940
Gulf Coast (Louisiana)^b	Pathogenic	4	<1	<1	2
Gulf Coast (Non-Louisiana)^b	Total	2,100	220	52	940
Gulf Coast (Non-Louisiana)^b	Pathogenic	4	<1	<1	2
Mid-Atlantic	Total	780	51	3	200
Mid-Atlantic	Pathogenic	1	<1	<1	<1
Northeast Atlantic	Total	230	33	4	42
Northeast Atlantic	Pathogenic	<1	<1	<1	<1
Pacific Northwest (Dredged)^c	Total	5	<1	<1	<1
Pacific Northwest (Dredged)^c	Pathogenic	<1	<1	<1	<1
Pacific Northwest (Intertidal)^d	Total	650	2	<1	61
Pacific Northwest (Intertidal)^d	Pathogenic	15	<1	<1	1
^a Predicted mean levels of total and pathogenic Vibrio parahaemolyticus per gram of raw oysters. Values rounded to 2 significant digits. ^b Note: the values for Louisiana and non-Louisiana areas are the same because the water temperature is similar for these regions. Differences in the Gulf Coast states occur in the post-harvest portion of the model (See Summary Table 4). ^c Predicted mean levels when oyster reefs are submerged. ^d Predicted mean levels after intertidal exposure.

Post-Harvest Module. The Post-Harvest Module of the Exposure Assessment model predicts the effects of typical industry practices on Vibrio parahaemolyticus densities in oysters during transportation, distribution, and storage from harvest through retail. After oysters are harvested, levels of Vibrio parahaemolyticus can increase or decline in oysters during handling and storage before consumption. After harvesting, oysters are typically stored unrefrigerated on the oyster boat for a period of time ranging from a few hours to more than half a day. The potential growth of Vibrio parahaemolyticus in the oysters during this period of unrefrigerated holding is a function of the air temperature at the time of harvest and the length of time oysters are unrefrigerated. Once the oysters are placed under refrigeration (e.g., during transport or after arrival at wholesalers), the rate of growth slows until oysters reach a "no-growth" temperature (i.e., below 10 °C) for Vibrio parahaemolyticus. The length of time during which Vibrio parahaemolyticus growth occurs after the start of refrigeration and the (reduced) rate of growth during this period of time were estimated. When held at a refrigeration temperature of 45 °F (7.2 °C), levels of Vibrio parahaemolyticus decrease slowly as cells die under this storage condition; this effect was included in the Post-Harvest model. The post-harvest levels are carried forward to the Consumption Module where the dose levels of Vibrio parahaemolyticus consumed are modeled.

Summary Table 4 provides the predicted mean levels for total and pathogenic Vibrio parahaemolyticus in oysters post-harvest. These results, in comparison to those shown in Summary Table 3, are indicative of the effects of current post-harvest handling and processing practices on Vibrio parahaemolyticus levels. The predicted total and pathogenic Vibrio parahaemolyticus levels in oysters post-harvest are highest in both the Louisiana and non-Louisiana Gulf Coast regions because the levels at-harvest were the highest and ambient temperature is much higher in this region than in the other regions, allowing for more growth. Predicted levels in the Gulf Coast (Louisiana) are higher than those in the Gulf Coast (non-Louisiana), reflecting a longer time from harvest to refrigeration. The type of harvesting also has an impact on the levels of Vibrio parahaemolyticus. In the Pacific Northwest, the typically longer exposure to warmer air temperatures during intertidal harvesting can elevate oyster temperatures, allowing for additional growth of Vibrio parahaemolyticus during intertidal harvesting.

Summary Table 4. Predicted Mean Levels of Total and Pathogenic *Vibrio parahaemolyticus* per gram in Oysters Post-Harvest
Region	Level	Summer^a	Fall^a	Winter^a	Spring^a
Gulf Coast (Louisiana)	Total	60,000	5,700	290	23,000
Gulf Coast (Louisiana)	Pathogenic	100	10	<1	39
Gulf Coast (Non-Louisiana)	Total	42,000	2,500	135	16,000
Gulf Coast (Non-Louisiana)	Pathogenic	73	4	<1	28
Mid-Atlantic	Total	12,000	310	1	4,200
Mid-Atlantic	Pathogenic	21	<1	<1	7
Northeast Atlantic	Total	2,500	52	1	510
Northeast Atlantic	Pathogenic	4	<1	<1	<1
Pacific Northwest (Dredged)^b	Total	100	<1	<1	9
Pacific Northwest (Dredged)^b	Pathogenic	2	<1	<1	<1
Pacific Northwest (Intertidal)^c	Total	1,700	4	<1	150
Pacific Northwest (Intertidal)^c	Pathogenic	38	<1	<1	4
^a Predicted mean levels of total and pathogenic Vibrio parahaemolyticus per gram of raw oysters. Values rounded to 2 significant digits. ^b Predicted mean levels when oyster reefs are submerged. ^c Predicted mean levels after intertidal exposure.

Consumption Module. The Consumption Module of the Exposure Assessment model estimates the levels of total and pathogenic Vibrio parahaemolyticus in a single serving of an oyster meal. The number of oyster meals or servings eaten, the quantity of oysters consumed per serving, and the pathogenic Vibrio parahaemolyticus/g oyster at consumption are included in this module. The number of servings eaten refers to the oysters harvested from a specific region. As such, the risk assessment model predicts illness associated with oysters harvested from specific regions but does not predict illness associated with the location (region) where the oysters were consumed or illness reported. Summary Table 5 provides the mean predicted levels of total and pathogenic Vibrio parahaemolyticus at consumption.

Summary Table 5. Predicted Mean Levels of Total and Pathogenic *Vibrio parahaemolyticus* per Serving of Oysters at Consumption
Region	Level	Summer^a	Fall^a	Winter^a	Spring^a
Gulf Coast (Louisiana)	Total	12,000,000	1,200,000	58,000	4,600,000
Gulf Coast (Louisiana)	Pathogenic	21,000	2,000	98	7,900
Gulf Coast (Non-Louisiana)	Total	8,500,000	500,000	27,000	3,200,000
Gulf Coast (Non-Louisiana)	Pathogenic	15,000	880	47	5,600
Mid-Atlantic	Total	2,500,000	62,000	280	850,000
Mid-Atlantic	Pathogenic	4,300	110	<1	1,500
Northeast Atlantic	Total	500,000	11,000	300	100,000
Northeast Atlantic	Pathogenic	860	17	<1	180
Pacific Northwest (Dredged)^b	Total	21,000	46	2	1,900
Pacific Northwest (Dredged)^b	Pathogenic	460	1	<1	43
Pacific Northwest (Intertidal)^c	Total	330,000	800	3	30,000
Pacific Northwest (Intertidal)^c	Pathogenic	7,500	17	<1	740
^a Predicted mean levels of total and pathogenic Vibrio parahaemolyticus per serving of raw oysters. Values rounded to 2 significant digits. ^b Predicted mean levels when oyster reefs are submerged. ^c Predicted mean levels after intertidal exposure.

RISK CHARACTERIZATION

The Risk Characterization combines the results of the Exposure Assessment model with the Dose-Response model to predict the number of illnesses and the severity of illness associated with different regions and seasons. Estimates of the uncertainty associated with these predictions of risk and illness burden (i.e., upper and lower bounds) are also determined. For simplicity, the results of these regional and seasonal predictions of illness are presented below as the mean of the distribution (i.e., the mean number of predicted illnesses). A detailed description of the uncertainty distributions can be found in the complete risk assessment. Sensitivity analyses were conducted to evaluate the importance of the various input factors on the model results. The model was validated by comparing the results to a retail study and epidemiological data.

Predicted Illness Burden

Risk per Serving. The "risk per serving" is the risk of an individual becoming ill (gastroenteritis alone or gastroenteritis followed by septicemia) when he or she consumes a single serving of oysters. As shown in Summary Table 6, the predicted risk per serving is highest for the Gulf Coast (Louisiana) region and lowest for Pacific Northwest (Dredged) region. Within a region, the risk per serving is highest for the warmer months and lowest for the cooler months. For example, for the Northeast Atlantic, the risk per serving in the winter is on the order of 1x10^-8, meaning only one illness in every 100 million servings. For this same region, the risk per serving in the summer is approximately 3 orders of magnitude higher (one illness in every 100,000 servings). For the Pacific Northwest region during the summer and spring, the risk per serving is higher for oysters harvested by intertidal compared with dredged methods.

Summary Table 6. Predicted Mean Risk per Serving Associated with the Consumption of Pathogenic *Vibrio parahaemolyticus* in Raw Oysters
Region	Mean Risk Per Serving^a
Region	Summer	Fall	Winter	Spring	Total
Gulf Coast (Louisiana)	4.4 x 10^-4	4.3 x 10^-5	2.1 x 10^-6	1.7 x 10^-4	6.6 x 10^-4
Gulf Coast (Non-Louisiana)^b	3.1 x 10^-4	1.9 x 10^-5	1.1 x 10^-6	1.2 x 10^-4	4.5 x 10^-4
Mid-Atlantic	9.2 x 10^-5	2.2 x 10^-6	1.1 x 10^-8	3.1 x 10^-5	1.3 x 10^-4
Northeast Atlantic	1.8 x 10^-5	4.0 x 10^-7	1.1 x 10^-8	3.6 x 10^-6	2.2 x 10^-5
Pacific Northwest (Dredged)	1.0 x 10^-5	2.6 x 10^-8	8.1 x 10^-10	8.7 x 10^-7	1.1 x 10^-5
Pacific Northwest (Intertidal)^c	1.4 x 10^-4	3.9 x 10^-7	1.7 x 10^-9	1.3 x 10^-5	1.5 x 10^-4
^a Risk per serving refers to the predicted risk of an individual becoming ill (gastroenteritis alone or gastroenteritis followed by septicemia) when he or she consumes a single serving of raw oysters. ^bIncludes oysters harvested from Florida, Mississippi, Texas, and Alabama. The time from harvest to refrigeration in these states is typically shorter than for Louisiana. ^cOysters harvested using intertidal methods are typically exposed to higher temperature for longer times before refrigeration compared with dredged methods.

Risk per Annum. The "risk per annum" is the predicted number of illnesses (gastroenteritis alone or gastroenteritis followed by septicemia) in the United States each year. As shown in Summary Table 7, for each region, the highest number of predicted cases of illnesses is associated with oysters harvested in the summer and spring and the lowest in the winter and fall. Of the total annual predicted Vibrio parahaemolyticus illnesses,approximately 92% are attributed to oysters harvested from the Gulf Coast (Louisiana and non-Louisiana states) region in the spring, summer and fall and from the Pacific Northwest (intertidal) region in the summer. The lower numbers of illnesses predicted for the Northeast Atlantic and Mid-Atlantic oyster harvests are attributable both to the colder water temperatures and the smaller harvest from these regions. The harvesting practice also has an impact on the illness rate. Intertidal harvesting in the Pacific Northwest poses a much greater risk than dredging in this region (192 vs. 4 illnesses per year). This is likely attributable to elevation of oyster temperatures during intertidal exposure leading to Vibrio parahaemolyticus growth.

Summary Table 7. Predicted Mean Annual Number of Illnesses Associated with the Consumption of *Vibrio parahaemolyticus* in Raw Oysters
Region	Mean Annual Illnesses^a
Region	Summer	Fall	Winter	Spring	Total
Gulf Coast (Louisiana)	1,406	132	7	505	2,050
Gulf Coast (Non-Louisiana)^b	299	51	3	193	546
Mid-Atlantic	7	4	<1	4	15
Northeast Atlantic	14	2	<1	3	19
Pacific Northwest (Dredged)	4	<1	<1	<1	4
Pacific Northwest (Intertidal)^c	173	1	<1	18	192
TOTAL	1,903	190	10	723	2,826
^a Mean annual illnesses refers to the predicted number of illnesses (gastroenteritis alone or gastroenteritis followed by septicemia) in the United States each year. ^b Includes oysters harvested from Florida, Mississippi, Texas, and Alabama. The time from harvest to refrigeration in these states is typically shorter than for Louisiana. ^c Oysters harvested using intertidal methods are typically exposed to higher temperature for longer times before refrigeration compared with dredged methods.

Severity of Illness. The predicted number of cases of septicemia was determined for the total United States population as shown in Summary Table 8. The number of predicted cases of septicemia was calculated by multiplying the mean number of predicted illnesses (Summary Table 7) by the probability of gastroenteritis progressing to septicemia (0.0023). The calculation of the probability of gastroenteritis progressing to septicemia is described in the complete risk assessment. Since most of the cases of illness are predicted to be associated with the Gulf Coast (Louisiana) harvest, this is also the harvest that would be expected to be associated with the highest number of cases of septicemia.

Anyone exposed to Vibrio parahaemolyticus can become infected and develop gastroenteritis. However, compared to the healthy population, there is about a 40-fold higher probability of an infected individual with a concurrent underlying chronic medical condition developing septicemia. The model predicts about 7 cases of septicemia each year for the total population, of which 2 would be expected to occur in healthy individuals and 5 would be expected to occur among the immunocompromised population.

Summary Table 8. Predicted Mean Number of Cases of *Vibrio parahaemolyticus* Septicemia Associated with the Consumption of Raw Oysters
Region	Mean Annual Cases of Septicemia^a
Region	Summer	Fall	Winter	Spring	Total
Gulf Coast (Louisiana)	3	<1	<1	1	4
Gulf Coast (Non-Louisiana)^b	<1	<1	<1	<1	1
Mid-Atlantic	<1	<1	<1	<1	<1
Northeast Atlantic	<1	<1	<1	<1	<1
Pacific Northwest (Dredged)	<1	<1	<1	<1	<1
Pacific Northwest (Intertidal)	<1	<1	<1	<1	<1
TOTAL	4	<1	<1	2	7
^a Calculated by multiplying the probability of septicemia (0.0023) by the mean predicted number of illnesses (see Summary Table 7). ^b Includes oysters harvested from Florida, Mississippi, Texas, and Alabama. The typical time from harvest to refrigeration of oysters for these states is shorter than for Louisiana.

Sensitivity Analysis

A sensitivity analysis was conducted to determine which model input factors have the strongest influence on the predicted probability of illness. A representative example of this type of evaluation is shown in Summary Figure 3. The graph (referred to as a Tornado Plot) shows the rank and magnitude of influence of factors (from highest to lowest) on the probability of illness. For example, in the Gulf Coast (Louisiana) Summer harvest, the model prediction of risk is influenced the most by the level of Vibrio parahaemolyticus in the environment and secondly by the percent of pathogenic Vibrio parahaemolyticus in oysters at the time of harvest. The length of time oysters are unrefrigerated after harvest and air temperature are also important factors. The ranking is similar for all regions, except for intertidal-harvested oysters in the Pacific Northwest. For the Pacific Northwest intertidal harvest, the second and third most influential factors are air and oyster temperatures. Thus, for this region, higher levels of risk are associated with oysters that have been collected on warm sunny days. Since the levels of Vibrio parahaemolyticus decrease during cold storage, the length of time the oysters are refrigerated is negatively correlated with the risk for all regions and seasons and the factor points to the left rather than to the right on the Tornado Plot.

Summary Figure 3. Tornado Plot of Influential Variability Factors of Vibrio parahaemolyticus (Vp) Illness per Serving of Raw Oysters in the Gulf Coast (Louisiana) Summer Harvest

Summary Figure 3. Tornado Plot of Influential Variability Factors of Vibrio parahaemolyticus (Vp) Illness per Serving of Raw Oysters in the Gulf Coast (Louisiana) Summer Harvest

Model Validation

Exposure predictions were validated by comparing predicted Vibrio parahaemolyticus levels in oysters at the time of consumption to data from a 1998-1999 survey of Vibrio parahaemolyticus levels in oysters at retail conducted collaboratively by the Interstate Shellfish Sanitation Conference (ISSC) and the FDA (Summary Figure 4). These data were not used in the development of the risk assessment model. In general, the mean Vibrio parahaemolyticus levels predicted by the model compared well with the mean levels from the ISSC/FDA survey, particularly for the Gulf and Mid-Atlantic summer when the risk of illness is highest. For the Pacific Northwest, the model predictions are higher than the ISSC/FDA estimates, but there is substantial uncertainty associated with the ISSC/FDA data for this region due to the relatively small number of samples. Based on the generally good agreement between model-predicted V. parahaemolyticus densities and observed densities at retail, the exposure assessment portion of the model is considered to be validated.

Summary Figure 4. Observed log10 Density of Total Vibrio parahaemolyticus at Retail Compared to Model Predictions for the Gulf Coast (Louisiana and non-Louisiana states)

Summary Figure 4. Observed log₁₀ Density of Total Vibrio parahaemolyticus at Retail Compared to Model Predictions for the Gulf Coast (Louisiana and non-Louisiana states) [The error bars indicate one standard deviation above and below either the model predictions (square box) or observed values (filled circle).]

The corresponding validation of the risk estimates based on a comparison of the risk assessment predictions and available epidemiological data showed a higher degree of uncertainty. The surveillance data reported to CDC are the only data available to validate the model predictions of illness for each region and season. Temporally, the model predictions and CDC data both indicate that the risk of illness is higher in the spring and summer than in the winter and fall. However, agreement between the surveillance data and the regional predictions of risk were less clear cut, though both showing similar trends (e.g., the highest number of illnesses are associated with Gulf Coast oysters followed by Pacific Northwest oysters). In part, this uncertainty reflects the fact that the surveillance data indicate where (location) the illness occurred and the model predicts illnesses attributed to where (region) oysters were harvested. It is difficult to trace the oysters that caused an illness back to the harvest region. Because of the intrinsic difference in what the two systems measure (location of illness occurrence vs. harvest region of oysters that cause illness), full validation of the regional model predictions of illness based on regional surveillance data will require additional research and targeted surveillance initiatives with more thorough traceback data.

WHAT-IF SCEENARIOS

The risk assessment model can be used to estimate the likely impact of intervention strategies on the predicted number of illnesses. The impact of different harvesting methods, seasons (i.e., water and air temperatures), time until refrigeration, and length of storage before consumption were parameters considered in the baseline model. By changing one or more of the input parameters and measuring the resulting change in the model outputs, the likely impact of new or different processing procedures or regulatory actions can be evaluated. These changes to the baseline model are commonly referred to as "what-if" scenarios. The what-if scenarios evaluated include the following: reducing levels of Vibrio parahaemolyticus in oysters (representing various post-harvest mitigation controls); reducing time-to-refrigeration; re-submersion of intertidally harvested oysters; and sample-based control plans.

Reducing Levels of Vibrio parahaemolyticus in Oysters.

Post-harvest mitigation control scenarios included an evaluation of treatments that reduce levels of Vibrio parahaemolyticus in oysters. The reduction levels represent a range of potential mitigation controls: immediate refrigeration (i.e., cooling immediately after harvest); 2-log reduction (e.g., freezing and cold storage); and 4.5-log reduction (e.g., mild heat treatment, irradiation, or ultra high hydrostatic pressure). The effectiveness of immediate refrigeration may be expected to vary both regionally and seasonally and is typically approximately 1-log reduction.

Measures that control or reduce the levels of Vibrio parahaemolyticus in oysters reduced the predicted risk of illness associated with this pathogen (Summary Table 9). Treatment such as immediate refrigeration decreased the number of predicted illnesses by approximately 10-fold. The effect of immediate refrigeration is less pronounced in the cooler regions than in the warmer Gulf Coast. Treatment causing a 2-log decrease in the levels of Vibrio parahaemolyticus in oysters reduces the probability of illness by approximately 100-fold. Treatment causing a 4.5-log decrease in the number of Vibrio parahaemolyticus bacteria reduces predicted illness to an extent that makes it unlikely that illnesses would be observed.

Summary Table 9. Predicted Mean Number of Illnesses per Annum from Reduction of Levels of Pathogenic *Vibrio parahaemolyticus* in Oysters
Region	Predicted Mean Number of Illnesses per Annum
Region	Baseline	Immediate Refrigeration^a	2-log Reduction^b	4.5-log Reduction^c
Gulf Coast (Louisiana)	2,050	202	22	<1
Gulf Coast (Non-Louisiana)	546	80	6	<1
Mid-Atlantic	15	2	<1	<1
Northeast Atlantic	19	3	<1	<1
Pacific Northwest (Dredged)	4	<1	<1	<1
Pacific Northwest (Intertidal)	192	106	2	<1
TOTAL	2,826	391	30	<1
^a Represents refrigeration immediately after harvest; the effectiveness of which varies both regionally and seasonally and is typically approximately 1-log reduction. ^bRepresents any process which reduces levels of Vibrio parahaemolyticus in oysters 2-log, e.g., freezing. ^c Represents any process which reduces levels of Vibrio parahaemolyticus in oysters 4.5-log, e.g., mild heat treatment, irradiation, or ultra high hydrostatic pressure.

Reducing the Time-to-Refrigeration

For this scenario, the impact of "rapid" cooling (i.e., using ice or an ice slurry after harvest) such that oysters would be chilled to a "no-growth" temperature (<10 °C) within 1 hour of harvest were compared to "conventional" cooling (i.e., refrigeration after harvest) such that up to 10 hours were presumed for oysters to reach the no-growth temperature. For the Gulf Coast Louisiana/ Summer harvest, the greatest reductions were predicted for shorter times to refrigeration and using cooling with ice compared to cooling under conventional refrigeration (Summary Figure 5). Predicted reduction in Vibrio parahaemolyticus illnesses from oysters cooled within 1 hour after harvest ranged from 86% (conventional refrigeration) to 97% (cooling with ice). The lower ambient temperatures associated with the other regions result in predicted reductions that are less dramatic.

Summary Figure 5. Predicted Effectiveness of Two Different Methods of Cooling on Vibrio parahaemolyticus Risk for the Gulf Coast Region (Louisiana and non-Louisiana) Summer Harvest [Errors bars denote

Summary Figure 5. Predicted Effectiveness of Two Different Methods of Cooling on Vibrio parahaemolyticus Risk for the Gulf Coast Region (Louisiana and non-Louisiana) Summer Harvest [Errors bars denote central 95% of uncertainty distribution about the mean % reduction.]

Re-submersion of Intertidally Harvested Oysters

As an example of a harvest practice scenario, the impact of overnight submersion of oysters was evaluated. The model predicts the levels of Vibrio parahaemolyticus in intertidally-harvested oysters, e.g., oysters are placed into baskets and removed after the tide rises, a typical practice in the Pacific Northwest. Vibrio parahaemolyticus levels can increase in oysters during intertidal exposure but overnight submersion of the oysters in water has been shown to reduce these levels. Delaying harvest until near the end of the tidal cycle, just before oysters are exposed again, was predicted to reduce the risk of illness by approximately 90%. Research is needed to determine whether the predicted level of reduction can be achieved when oysters are stacked in baskets.

Sample-Based Control Plans

The FDA/ISSC recommends that the levels of Vibrio parahaemolyticus in oysters not exceed 10,000 cells/gram and the ISSC interim control plan (ICP) recommends monitoring of oyster meats for the presence of Vibrio parahaemolyticus. Prior to 2001, ISSC recommended that shellfish harvest waters be re-sampled for pathogenic Vibrio parahaemolyticus if the levels of total Vibrio parahaemolyticus in oyster meats at harvest exceed 10,000 cells/gram. In 2001, the ICP was revised to recommend that shellfish harvest waters be re-sampled for pathogenic Vibrio parahaemolyticus if the levels of total Vibrio parahaemolyticus in oyster meats at harvest are above 5,000 cells/gram.

The incidence of illness was evaluated assuming that it was possible to identify and exclude oysters from the raw market which contained various specified levels of Vibrio parahaemolyticus either at harvest or at retail. The Gulf Coast region (Louisiana)/Summer harvest is presented here as an example. As shown in Summary Figures 6 and 7, restricting the levels of Vibrio parahaemolyticus in oysters either at-harvest or at-retail reduces the number of predicted illnesses, but requires diversion of oysters from the raw market (or modification of handling practices to reduce post-harvest Vibrio parahaemolyticus growth). For the Gulf Coast region (Louisiana) summer harvest, in the absence of subsequent post-harvest mitigation, excluding oysters containing >10,000 Vibrio parahaemolyticus/g at the time of harvest is predicted to prevent approximately 16% of illnesses with an impact of approximately 3% of the oyster harvest(Summary Figure 6). However, excluding oysters containing >10,000 Vibrio parahaemolyticus at-retail reduced predicted illness by 99% but would require approximately 43% of the oyster harvest to be diverted from the raw market consumption (or subjected to preventive controls). The impact of compliance with different "at-harvest" and "at-retail" (i.e., after refrigeration) control levels was also evaluated. As might be expected, the effectiveness of a specific (or hypothetical) control level to reduce illnesses depend was proportional to the extent of compliance with that level.

Summary Figure 6. Potential Effect of Control of Total Vibrio parahaemolyticus Bacterium per gram At-Harvest for the Gulf Coast Region (Louisiana) Summer Harvest

Summary Figure 6. Potential Effect of Control of Total Vibrio parahaemolyticus Bacterium per gram At-Harvest for the Gulf Coast Region (Louisiana) Summer Harvest

Summary Figure 7. Potential Effect of Control of Total Vibrio parahaemolyticus per gram At-Retail for the Gulf Coast Region (Louisiana) Summer Harvest

Summary Figure 7. Potential Effect of Control of Total Vibrio parahaemolyticus per gram At-Retail for the Gulf Coast Region (Louisiana) Summer Harvest

Summary Table 10. Effect of Compliance with *Vibrio parahaemolyticus* Control Levels in Oysters for the Gulf Coast Region (Louisiana)/ Summer Harvest
Control Level^a	Compliance Level (%)	Percentage Illnesses Averted^b
Control Level^a	Compliance Level (%)	At-Harvest^c	At-Retail^d
100	50	65	74
100	100	98	100
1,000	50	37	69
1,000	100	68	100
5,000	50	14	63
5,000	100	28	100
10,000	50	8	60
10,000	100	16	99
^a Control level is the maximum theoretical level of total Vibrio parahaemolyticus per gram of oyster. Currently FDA/ISSC recommends a guidance level of 10,000 cells/gram and the ISSC interim control plan recommends monitoring oysters if levels are above 5,000 cells/gram. ^b The percentage of illnesses that would be prevented in comparison to the baseline model predictions. ^c At-harvest refers to a guidance level applied at the time of harvest and before oyster refrigeration. ^d At-retail refers to a guidance level applied after oyster refrigeration and storage.

CONCLUSIONS

This risk assessment included an analysis of the available scientific information and data in the development of a model to predict the public health impact of pathogenic Vibrio parahaemolyticus in raw oysters. The assessment focuses on comparing the relative risk among different geographic regions, seasons, and harvest practices. The scientific data and the mathematical models developed during the risk assessment facilitate a systematic evaluation of strategies to reduce the public health impact of pathogenic Vibrio parahaemolyticus associated with the consumption of raw oysters.

Although the risk assessment modeled sporadic Vibrio parahaemolyticus illnesses, steps taken to reduce sporadic cases would be expected to reduce the size and frequency of outbreaks. The proportional reduction would depend on the virulence of the outbreak strain and on the survivability and growth of the strain following post-harvest treatments. Mitigation or control measures aimed at decreasing levels of Vibrio parahaemolyticus in oysters will also likely decrease levels of other species in the Vibrio genus (or family), such as Vibrio vulnificus.

Below are the responses to the questions that the risk assessment team was charged with answering.

What is known about the dose-response relationship between consumption of Vibrio parahaemolyticus and illnesses?

Although an individual may become ill from consumption of low levels of Vibrio parahaemolyticus, it is much more likely that he or she will become ill if the level is high. The probability of illness is relatively low (<0.001%) for consumption of 10,000 Vibrio parahaemolyticus cells/serving (equivalent to about 50 cells/gram oysters). Consumption of about 100 million Vibrio parahaemolyticus cells/serving (500 thousand cells/gram oysters) increases the probability of illness to about 50%.
Anyone exposed to Vibrio parahaemolyticus can become infected and develop gastroenteritis. However there is a greater probability of gastroenteritis developing into septicemia (and possibly death) among the subpopulation with concurrent underlying chronic medical conditions.
The model predicts about 2,800 Vibrio parahaemolyticus illnesses from oyster consumption each year. Of infected individuals, approximately 7 cases of gastroenteritis will progress to septicemia each year for the total population, of which 2 individuals would be from the healthy subpopulation and 5 would be from the immunocompromised subpopulation.
This risk assessment assumed that pathogenic strains of Vibrio parahaemolyticus are TDH⁺ and that all strains possessing this characteristic are equally virulent. Modifications can be made to the risk assessment if data become available for new virulence determinants. For example, data from outbreaks suggest that fewer microorganisms of Vibrio parahaemolyticus O3:K6 are required to cause illness compared to other strains.

What is the frequency and extent of pathogenic strains of Vibrio parahaemolyticus in shellfish waters and in oysters?

Levels of pathogenic Vibrio parahaemolyticus usually occur at low levels in shellfish waters.
Levels of pathogenic Vibrio parahaemolyticus in oysters at the time of harvest are only a small fraction of the total Vibrio parahaemolyticus levels.

What environmental parameters (e.g., water temperature, salinity) can be used to predict the presence of Vibrio parahaemolyticus in oysters?

The primary driving factor to predict the presence of Vibrio parahaemolyticus in oysters is water temperature. Salinity was a factor evaluated but not incorporated into the model. Salinity is not a strong determinant of Vibrio parahaemolyticus levels in the regions that account for essentially all the commercial harvest. Other factors such as oyster physiology and disease status may also be important but no quantifiable data were available to include these factors in the model.
There are large differences in the predicted levels of Vibrio parahaemolyticus in oysters at harvest among regions and seasons. For all regions, the highest levels of Vibrio parahaemolyticus were predicted in the warmer months of summer and spring and the lowest levels in the fall and winter.
Overall, the highest levels of total and pathogenic Vibrio parahaemolyticus were predicted for the Gulf Coast (Louisiana) and the lowest levels in the Pacific Northwest (Dredged) harvested oysters.
After harvest, air temperature is also an important determinant of the levels of Vibrio parahaemolyticus in oysters. Vibrio parahaemolyticus can continue to grow and multiply in oysters until they are adequately chilled.
Levels of Vibrio parahaemolyticus are lower in oysters after harvest in the cooler vs. warmer months. This means that reducing the time between harvest and cooling will be more important in the summer and spring than in the fall and winter.

How do levels of Vibrio parahaemolyticus in oysters at harvest compare to levels at consumption?

With no mitigation treatments, levels of Vibrio parahaemolyticus are higher in oysters at consumption than at harvest. The difference between Vibrio parahaemolyticus densities at-harvest versus at-consumption is largely attributable to the extent of growth that occurs before the oysters are cooled to no-growth temperatures.
Levels of Vibrio parahaemolyticus in oysters vary by region and season and are highest during the summer.
During intertidal harvest, oysters are exposed to higher temperatures for longer times, allowing additional growth of Vibrio parahaemolyticus in oysters and leading to higher predicted risk of illness.
Preventing growth of Vibrio parahaemolyticus in oysters after harvest (particularly in the summer) will lower the levels of Vibrio parahaemolyticus in oysters and, as a consequence, lower the number of illnesses associated with the consumption of raw oysters.

What is the role of post-harvest handling on the level of Vibrio parahaemolyticus in oysters?

Post-harvest measures aimed at reducing the Vibrio parahaemolyticus levels in oysters reduced the model-predicted risk of illness associated with this pathogen.
Reducing the time between harvest and chilling has a large impact on reducing levels of Vibrio parahaemolyticus in oysters and the number of illnesses. Predicted reductions were greater for shorter times to refrigeration using ice (oysters reach no-growth temperature in 1 hour) compared to cooling under conventional refrigeration (which may take up to 10 hours until oysters reach a no-growth temperature).

What reductions in risk can be anticipated with different potential intervention strategies?

Overall. The most influential factor affecting predicted risk of illness is the level of total Vibrio parahaemolyticus in oysters at the time of harvest. Intervention strategies should be aimed at reducing levels of Vibrio parahaemolyticus and/or preventing its growth in oysters after harvest. These strategies, either at-harvest or post-harvest, may need to consider regional/seasonal differences. For example, the use of ice on harvest boats to cool oysters to the no-growth temperature of Vibrio parahaemolyticus will have a larger impact on reducing illnesses in the summer than in the winter when air temperatures are cooler and Vibrio parahaemolyticus levels are lower.
Regional/Seasonal Differences. The risk of Vibrio parahaemolyticus illness is increased during the warmer months of the year, with the magnitude of this increase a function of the extent to which the growing waters (and air temperature) are at temperatures that support the growth of the pathogen (e.g., temperatures above 10 °C). For each region, the predicted numbers of illnesses are much higher for the summer compared to the winter months. Intervention measures that depend on cooling oysters to no-growth temperatures for Vibrio parahaemolyticus may be more important in warmer seasons and regions.
The risk of Vibrio parahaemolyticus illness is substantial in the Gulf Coast region where water temperatures are warm over a large part of the year as compared to the Northeast Atlantic region where water temperatures support the growth of Vibrio parahaemolyticus only during a relatively small portion of the year. A difference is seen among the regions due to different harvesting methods. Within the Gulf Coast, the predicted number of illnesses is much higher in Louisiana compared to other states in this region because the harvest boats in Louisiana are typically on the water longer, i.e., leading to a longer time from harvest to refrigeration. Harvest volume is also a determining factor; in the summer, Louisiana accounts for approximately 77% of the Gulf Coast harvest. This is also seen in the Pacific Northwest by comparing intertidal versus dredged harvesting. Intertidal harvesting accounts for 75% of the Pacific Northwest harvest and exposes oysters to higher temperatures longer, allowing greater growth of Vibrio parahaemolyticus. Overnight submersion for a single tidal cycle, reduces levels of Vibrio parahaemolyticus in oysters and the risk of illness
Post-Harvest Treatments. Post-harvest treatments that reduce levels of Vibrio parahaemolyticus by 2 to 4.5-logs were found to be effective for all seasons and regions, with the most pronounced effects seen for regions and seasons with higher baseline risk. The model shows that any treatment that causes at least a 4.5-log decrease in the number of Vibrio parahaemolyticus bacteria reduces the probability of illness to such an extent that few illnesses would be identified by epidemiological surveillance. However, some outbreak strains (e.g., O3:K6) are more resistant to mitigations than endemic pathogenic Vibrio parahaemolyticus strains, and the duration or extent of treatment may need to be more stringent to achieve an equivalent degree of reduction. Studies have shown that both Vibrio parahaemolyticus and Vibrio vulnificus respond similarly to control measures such as ultra high pressure, mild heat treatment, and freezing. Therefore, mitigations aimed at decreasing levels of Vibrio parahaemolyticus will also likely decrease levels of Vibrio vulnificus.
The model also demonstrated that if oysters are not refrigerated soon after harvest, Vibrio parahaemolyticus rapidly multiply resulting in higher levels. For example, the model indicates that for the Gulf Coast there is a significant reduction (~10-fold) in the probability of illness when the oysters are placed in a refrigerator immediately after harvest. Less pronounced reductions are predicted for the other regions. Predicted reduction in illness is less in colder seasons because oysters harvested in cooler weather are already at or below the temperature threshold for Vibrio parahaemolyticus growth and as such refrigeration has little additional impact on levels of Vibrio parahaemolyticus.
At-Harvest and At-Retail Controls. Controlling the levels of Vibrio parahaemolyticus in oysters at-harvest or at-retail (after refrigeration and storage) drastically reduces the number of predicted illnesses but would require diversion of oysters from the raw market or modification of handling practices to reduce post-harvest Vibrio parahaemolyticus growth. For the Gulf Coast (Louisiana) region in the summer, excluding all oysters with at least 10,000 Vibrio parahaemolyticus/g at-harvest would reduce illness by approximately 16% at a loss of approximately 3% of the total harvest from the raw consumption market; and this same control level at-retail would reduce illness by about 99% with a 43% loss from the raw oyster market (or subjected to preventive controls). The effectiveness of the control level either at-harvest or at-retail to reduce illnesses depends on the extent of compliance with that control level.
In a sample-based control strategy, a reasonable surrogate for pathogenic Vibrio parahaemolyticus may be total levels of this microorganism. Criteria for rejection of oysters based on the levels of this surrogate might have to vary by region. For example, an at-harvest control criterion based on total Vibrio parahaemolyticus levels in the Pacific Northwest might need to be more stringent than in the Gulf Coast because the incidence of pathogenic strains appears to be higher in the Pacific Northwest. However, in an outbreak, the ratio of pathogenic to total Vibrio parahaemolyticus may not be the same or consistent, and the model does not evaluate how well total Vibrio parahaemolyticus would serve as a surrogate for pathogenic Vibrio parahaemolyticus in an outbreak situation.

In conclusion, the risk assessment illustrates that the levels of Vibrio parahaemolyticus at the time of harvest play an important role in causing human illness. However, other factors that either reduce or allow growth of Vibrio parahaemolyticus in oysters are also important in determining the number of illnesses. For example, shortening the time-to-refrigeration of oysters in the summer controls growth of Vibrio parahaemolyticus in oysters and subsequently reduces illnesses associated with this microorganism.

The results of this risk assessment are influenced by the assumptions and data sets that were used to develop the Exposure Assessment and Dose-Response models. The predicted risk of illness among consumers of raw oysters could change as a result of future data obtained from continuing surveillance studies. It is anticipated that periodic updates to the model will continue to reduce the degree of uncertainty associated with the factors that influence the risk. This risk assessment provides an understanding of the relative importance of and interactions among the factors influencing risk. It can be used to facilitate the formulation of effective guidance and requirements for the industry and in the evaluations of risk mitigation strategies. This Interpretive Summary provides a brief, non-technical description of the materials covered, but a full understanding requires the reader to consider the complete risk assessment.

EXECUTIVE SUMMARY

Background

The Food and Drug Administration (FDA) conducted a quantitative risk assessment to characterize the factors influencing the public health impact associated with the consumption of raw oysters containing pathogenic Vibrio parahaemolyticus. This effort was initiated in January 1999 and a draft risk assessment was made available for public comment in 2001. The risk assessment was conducted in response to four outbreaks in 1997 and 1998 in the United States involving over 700 cases of illness. These outbreaks renewed concern for this pathogen as a serious foodborne threat to public health and raised new concerns about the effectiveness of risk management guidance available at that time. These outbreaks also raised questions about the criteria used to close and reopen shellfish waters to harvesting and the FDA guidance for the maximum number of V. parahaemolyticus per gram in shellfish. FDA decided to conduct a quantitative risk assessment to provide new insights into how to better manage the presence of this pathogenic microorganism in shellfish.

This risk assessment focused on raw oysters, because that is the food in the United States predominately linked to illness from this pathogen. The risk assessment gathers available knowledge of V. parahaemolyticus in a systematic manner, and includes sophisticated, mathematical models. The levels of the pathogen in oysters were estimated beginning with harvest of the oysters through post-harvest handling, processing, and storage to predict human exposure from consumption of raw oysters and subsequent illnesses. The number of illnesses (on a per serving and a per year basis) were predicted for six regions in the United States and each season for a total of 24 region/season combinations. Total cases of illness include both gastroenteritis and septicemia. In addition, the probability of gastroenteritis progressing to septicemia in individuals with underlying medical conditions (such as diabetes, alcoholic liver disease, hepatitis, and those receiving immunosuppressive treatments for cancer or AIDS) was compared to that of healthy individuals. Once developed, the baseline model was used to develop "what-if" scenarios to evaluate the likely impact of potential intervention strategies on the exposure to pathogenic V. parahaemolyticus from consumption of raw oysters.

Vibrio parahaemolyticus is a gram-negative, salt tolerant bacterium that occurs naturally in estuaries. It has been long recognized as an important bacterial seafood-borne pathogen throughout the world. It was first isolated and implicated in an outbreak of food poisoning in Japan in 1950. Vibrio parahaemolyticus has been associated with outbreaks and individual cases of illness in the United States since 1969. These bacteria are normally present in many types of raw seafood, including fish, crustaceans, and molluscan shellfish. The microorganism concentrates, colonizes, and multiplies in the gut of filter-feeding molluscan shellfish such as oysters, clams, and mussels. Not all strains of V. parahaemolyticus cause illness; on the contrary, pathogenic strains represent a small percentage of the total V. parahaemolyticus present in the environment or seafood.

Scope and General Approach

This risk assessment is a quantitative product pathway analysis in which the key steps from harvest through post-harvest handling and processing to consumption were modeled. The likelihood of illness following exposure to pathogenic V. parahaemolyticus from consumption of raw oysters was calculated. The levels of V. parahaemolyticus in oysters at the time of consumption are influenced by the harvest methods and conditions, as well as the handling of oysters after harvest. These practices and conditions vary considerably among different geographic areas and at different times of year. The baseline risk assessment model was also used to estimate the likely impact of intervention strategies (referred to as "what-if" scenarios) on the predicted number of illnesses.

The risk assessment considered six oyster harvest regions and four seasons for a total of 24 region/season combinations. The oyster harvest regions included: Gulf Coast (Louisiana), Gulf Coast (non-Louisiana), Mid-Atlantic, Northeast Atlantic, Pacific Northwest (Dredged) and Pacific Northwest (Intertidal). In the Gulf Coast, the harvest duration (i.e., the time between removal of the oyster from the water to unloading them at the dock) for Louisiana is typically much longer than for other states in that region (Florida, Mississippi, Texas, and Alabama). Since harvest duration can affect the levels of V. parahaemolyticus in raw oysters, the Gulf Coast was divided into two distinct regions. Likewise, the Pacific Northwest was divided into two distinct regions, but in this case it was based on harvest methods, dredging and intertidal. Oysters harvested in intertidal areas are typically exposed to higher temperatures before refrigeration than those harvested using dredging. For the intertidal harvest method, oysters are hand-picked when oyster reefs are exposed during the tide cycle and left in baskets until the tide rises to a sufficient depth to allow a boat to retrieve the basket.

The risk assessment had two main objectives:

determine the factors that contribute to the risk of becoming ill from the consumption of pathogenic V. parahaemolyticus in raw oysters; and
evaluate the likely public health impact of different control measures, including the effectiveness of current and alternative microbiological standards.

Data for this risk assessment were obtained from many sources, including both published and unpublished scientific literature and reports produced by various organizations such as State shellfish control authorities, the Centers for Disease Control and Prevention (CDC), the shellfish industry, the Interstate Shellfish Sanitation Conference (ISSC), and State Health Departments. In some instances the conduct of the risk assessment required that assumptions be made when data were incomplete. To the extent possible, research was specifically undertaken during the period between issuing the original draft and the current version to address data gaps previously identified. These new data have been incorporated into the risk assessment.

Results

The model predicts illnesses (gastroenteritis alone and gastroenteritis followed by septicemia) associated with the consumption of V. parahaemolyticus in raw oysters for the 24 region/season combinations. Summary Table 1 provides the risk on a "per serving basis" (i.e., the risk of becoming ill per serving of raw oysters) and Summary Table 2 provides the risk on a "per annum basis" (i.e., the predicted number of illnesses per year).

Region	Mean Risk Per Serving^a
Region	Summer	Fall	Winter	Spring	Total
Gulf Coast (Louisiana)	4.4 x 10^-4	4.3 x 10^-5	2.1 x 10^-6	1.7 x 10^-4	6.6 x 10^-4
Gulf Coast (Non-Louisiana)^b	3.1 x 10^-4	1.9 x 10^-5	1.1 x 10^-6	1.2 x 10^-4	4.5 x 10^-4
Mid-Atlantic	9.2 x 10^-5	2.2 x 10^-6	1.1 x 10^-8	3.1 x 10^-5	1.3 x 10^-4
Northeast Atlantic	1.8 x 10^-5	4.0 x 10^-7	1.1 x 10^-8	3.6 x 10^-6	2.2 x 10^-5
Pacific Northwest (Dredged)	1.0 x 10^-5	2.6 x 10^-8	8.1 x 10^-10	8.7 x 10^-7	1.1 x 10^-5
Pacific Northwest (Intertidal)^c	1.4 x 10^-4	3.9 x 10^-7	1.7 x 10^-9	1.3 x 10^-5	1.5 x 10^-4

^a Risk per serving refers to the predicted risk of an individual becoming ill (gastroenteritis alone or gastroenteritis followed by septicemia) when he or she consumes a single serving of raw oysters.
^bIncludes oysters harvested from Florida, Mississippi, Texas, and Alabama. The time from harvest to refrigeration in these states is typically shorter than for Louisiana.
^cOysters harvested using intertidal methods are typically exposed to higher temperature for longer times before refrigeration compared with dredged methods.

Summary Table 2. Predicted Mean Annual Number of Illnesses Associated with the Consumption of *Vibrio parahaemolyticus* in Raw Oysters
Region	Mean Annual Illnesses^a
Region	Summer	Fall	Winter	Spring	Total
Gulf Coast (Louisiana)	1,406	132	7	505	2,050
Gulf Coast (Non-Louisiana)^b	299	51	3	193	546
Mid-Atlantic	7	4	<1	4	15
Northeast Atlantic	14	2	<1	3	19
Pacific Northwest (Dredged)	4	<1	<1	<1	4
Pacific Northwest (Intertidal)^c	173	1	<1	18	192
TOTAL	1,903	190	10	723	2,826

^a Mean annual illnesses refers to the predicted number of illnesses (gastroenteritis alone or gastroenteritis followed by septicemia) in the United States each year.
^b Includes oysters harvested from Florida, Mississippi, Texas, and Alabama. The time from harvest to refrigeration in these states is typically shorter than for Louisiana.
^c Oysters harvested using intertidal methods are typically exposed to higher temperature for longer times before refrigeration compared with dredged methods.

Below are the responses to the questions that the risk assessment team was charged with answering.

What is known about the dose-response relationship between consumption of Vibrio parahaemolyticus and illnesses?

Although an individual may become ill from consumption of low levels of V. parahaemolyticus, it is much more likely that he or she will become ill if the level is high. The probability of illness is relatively low (<0.001%) for consumption of 10,000 V. parahaemolyticus cells/serving (equivalent to about 50 cells/gram oysters). Consumption of about 100 million V. parahaemolyticus cells/serving (500 thousand cells/gram oysters) increases the probability of illness to about 50%.
Anyone exposed to V. parahaemolyticus can become infected and develop gastroenteritis. However there is a greater probability of gastroenteritis developing into septicemia (and possibly death) among the subpopulation with concurrent underlying chronic medical conditions.
The model predicts about 2,800 V. parahaemolyticus illnesses from oyster consumption each year. Of infected individuals, approximately 7 cases of gastroenteritis will progress to septicemia each year for the total population, of which 2 individuals would be from the healthy subpopulation and 5 would be from the immunocompromised subpopulation.

What is the frequency and extent of pathogenic strains of Vibrio parahaemolyticus in shellfish waters and in oysters?

Levels of pathogenic V. parahaemolyticus usually occur at low levels in shellfish waters.
Levels of pathogenic V. parahaemolyticus in oysters at the time of harvest are only a small fraction of the total V. parahaemolyticus levels.

What environmental parameters (e.g., water temperature, salinity) can be used to predict the presence of Vibrio parahaemolyticus in oysters?

The primary driving factor to predict the presence of V. parahaemolyticus in oysters is water temperature. Salinity was a factor evaluated but not incorporated into the model. Salinity is not a strong determinant of V. parahaemolyticus levels in the regions that account for essentially all the commercial harvest. Other factors such as oyster physiology and disease status may also be important but no quantifiable data were available to include these factors in the model.
There are large differences in the predicted levels of V. parahaemolyticus in oysters at harvest among regions and seasons. For all regions, the highest levels of V. parahaemolyticus were predicted in the warmer months of summer and spring and the lowest levels in the fall and winter.
Overall, the highest levels of total and pathogenic V. parahaemolyticus were predicted for the Gulf Coast (Louisiana) and the lowest levels in the Pacific Northwest (Dredged) harvested oysters.
After harvest, air temperature is also an important determinant of the levels of V. parahaemolyticus in oysters. Vibrio parahaemolyticus can continue to grow and multiply in oysters until they are adequately chilled.
Levels of V. parahaemolyticus are lower in oysters after harvest in the cooler vs. warmer months. This means that reducing the time between harvest and cooling will be more important in the summer and spring than in the fall and winter.

How do levels of Vibrio parahaemolyticus in oysters at harvest compare to levels at consumption?

With no mitigation treatments, levels of V. parahaemolyticus are higher in oysters at consumption than at harvest. The difference between V. parahaemolyticus densities at-harvest versus at-consumption is largely attributable to the extent of growth that occurs before the oysters are cooled to no-growth temperatures.
Levels of V. parahaemolyticus in oysters vary by region and season and are highest during the summer.
During intertidal harvest, oysters are exposed to ambient air temperatures for longer times, allowing additional growth of V. parahaemolyticus in oysters and leading to higher predicted risk of illness.
Preventing growth of V. parahaemolyticus in oysters after harvest (particularly in the summer) will lower the levels of V. parahaemolyticus in oysters and, as a consequence, lower the number of illnesses associated with the consumption of raw oysters.

What is the role of post-harvest handling on the level of V. parahaemolyticus in oysters?

Post-harvest measures aimed at reducing the V. parahaemolyticus levels in oysters reduced the model-predicted risk of illness associated with this pathogen.
Reducing the time between harvest and chilling has a large impact on reducing levels of Vibrio parahaemolyticus in oysters and the number of illnesses. Predicted reductions were greater for shorter times to refrigeration and ice (oysters reach no-growth temperature in 1 hour) compared to cooling under conventional refrigeration (which may take up to 10 hours until oysters reach a no-growth temperature).

What reductions in risk can be anticipated with different potential intervention strategies?

Overall. The most influential factor affecting predicted risk of illness is the level of total V. parahaemolyticus in oysters at the time of harvest. Intervention strategies should be aimed at reducing levels of V. parahaemolyticus and/or preventing its growth in oysters after harvest. These strategies, either at-harvest or post-harvest, may need to consider regional/seasonal differences.
Regional/seasonal Differences. The risk of V. parahaemolyticus illness is increased during the warmer months of the year, with the magnitude of this increase a function of the extent to which the growing waters (and ambient air temperatures) are at temperatures that support the growth of the pathogen (e.g., temperatures above 10°C). For each region, the predicted numbers of illnesses are much higher for the summer compared to the winter months. Intervention measures that depend on cooling oysters to no-growth temperatures for V. parahaemolyticus may be more important in warmer seasons and regions.
The risk of V. parahaemolyticus illness is substantial in the Gulf Coast region where water temperatures are warm over a large part of the year as compared to the Northeast Atlantic region where water temperatures support the growth of Vibrio parahaemolyticus only during a relatively small portion of the year. A difference is seen among the regions due to different harvesting methods. Within the Gulf Coast, the predicted number of illnesses is much higher in Louisiana compared to other states in this region because the harvest boats in Louisiana are typically on the water longer, i.e., leading to a longer time from harvest to refrigeration. Harvest volume is also a determining factor; in the summer, Louisiana accounts for approximately 77% of the Gulf Coast harvest. This is also seen in the Pacific Northwest by comparing intertidal versus dredged harvesting. Intertidal harvesting accounts for 75% of the Pacific Northwest harvest and exposes oysters to higher temperatures longer, allowing greater growth of V. parahaemolyticus. Overnight submersion for a single tidal cycle, reduces levels of V. parahaemolyticus in oysters and the risk of illness.
Post-Harvest Treatments. Post-harvest treatments that reduce levels of V. parahaemolyticus by 2 to 4.5-logs were found to be effective for all seasons and regions, with the most pronounced effects seen for regions and seasons with higher baseline risk. The model shows that any treatment that causes at least a 4.5-log decrease in the number of V. parahaemolyticus bacteria reduces the probability of illness to such an extent that few illnesses would be identified by epidemiological surveillance. However, some outbreak strains (e.g., O3:K6) are more resistant to mitigations than endemic pathogenic V. parahaemolyticus strains, and the duration or extent of treatment may need to be more stringent to achieve an equivalent degree of reduction. Studies have shown that both V. parahaemolyticus and V. vulnificus respond similarly to control measures such as ultra high pressure, mild heat treatment, and freezing. Therefore, mitigations aimed at decreasing levels of V. parahaemolyticus will also likely decrease levels of V. vulnificus.
The model also demonstrated that if oysters are not refrigerated soon after harvest, Vibrio parahaemolyticus rapidly multiply resulting in higher levels. For example, the model indicates that for the Gulf Coast there is a significant reduction (~10-fold) in the probability of illness when the oysters are placed in a refrigerator immediately after harvest. Less pronounced reductions are predicted for the other regions. Predicted reduction in illness is less in colder seasons because oysters harvested in cooler weather are already at or below the temperature threshold for V. parahaemolyticus growth and as such refrigeration has little additional impact on levels of V. parahaemolyticus.
At-Harvest and At-Retail Controls. Controlling the levels of V. parahaemolyticus in oysters at-harvest or at-retail (after refrigeration and storage) drastically reduces the number of predicted illnesses but would require diversion of oysters from the raw market or modification of handling practices to reduce post-harvest V. parahaemolyticus growth. For the Gulf Coast (Louisiana) region in the summer, excluding all oysters with at least 10,000 V. parahaemolyticus/g at-harvest would reduce illness by approximately 16% with an impact of approximately 3% of the total harvest; and this same control level at-retail would reduce illness by about 99% with a 43% loss from the raw consumption market. The effectiveness of the control level either at-harvest or at-retail to reduce illnesses depends on the extent of compliance with that control level.
In a sample-based control strategy, a reasonable surrogate for pathogenic V. parahaemolyticus may be total levels of this microorganism. Criteria for rejection of oysters based on the levels of this surrogate might have to vary by region. For example, an at-harvest control criterion based on total V. parahaemolyticus levels in the Pacific Northwest might need to be more stringent than in the Gulf Coast because the incidence of pathogenic strains appears to be higher in the Pacific Northwest. However, in an outbreak, the ratio of pathogenic to total V. parahaemolyticus may not be the same or consistent, and the model does not evaluate how well total Vibrio parahaemolyticus would serve as a surrogate for pathogenic V. parahaemolyticus in an outbreak situation.

Conclusions

Although the risk assessment modeled sporadic V. parahaemolyticus illnesses, steps taken to reduce sporadic cases from TDH⁺ strains could also proportionally reduce the size of outbreaks. However, some outbreak strains (e.g., O3:K6) may be more resistant to mitigations than endemic V. parahaemolyticus strains and may also require fewer cells to cause illness. The risk assessment illustrates that the levels of V. parahaemolyticus at-harvest play an important role in causing human illness. However, other factors that either reduce or allow growth of V. parahaemolyticus in oysters are also important in determining the number of illnesses. For example, shortening the time-to-refrigeration of oysters in the summer controls growth of V. parahaemolyticus in oysters and subsequently reduces illnesses associated with this microorganism.

The results of this risk assessment are influenced by the assumptions and data sets that were used to develop the Exposure Assessment and Dose-Response models. The predicted risk for illness among consumers of raw oysters and the most significant factors which influence the incidence of illness could change as a result of future data obtained from continuing surveillance studies. It is anticipated that periodic updates to the model when new data and knowledge become available will continue to reduce the degree of uncertainty associated with the factors that influence the risk, and that this will assist in making the best possible decisions, policies, and measures for reducing the risk posed by V. parahaemolyticus in raw oysters. This risk assessment provides an understanding of the relative importance and interactions among the factors influencing risk. It will hopefully provide a useful tool to facilitate the formulation of effective guidance and requirements and the evaluation of risk mitigation strategies.

I. INTRODUCTION

The Food and Drug Administration (FDA) conducted this risk assessment on the public health impact of Vibrio parahaemolyticus transmitted by raw oysters. This is a "product pathway" risk assessment and provides a systematic evaluation of the factors affecting V. parahaemolyticus in oysters and the sequence of events leading to consumer illnesses.

Background

Vibrio parahaemolyticus is a marine bacterium that occurs naturally in the estuarine environment and can accumulate in filter-feeding molluscan shellfish. This microorganism was first identified as a foodborne pathogen in Japan in the 1950s. It has been associated with outbreaks and individual cases of illness in the United States since 1969. In 1997 and 1998, over 700 cases of illness from four outbreaks were associated with consumption of raw oysters in three regions of the country, the Gulf Coast, Pacific Northwest, and Northeast. These outbreaks renewed concern for this pathogen as a serious foodborne threat to public health and raised new concerns about the effectiveness of current risk management guidance.

The Centers for Disease Control and Prevention (CDC) estimates that each year there are approximately 2,800 cases of V. parahaemolyticus illness associated with the consumption of raw oysters. The most common clinical manifestation of V. parahaemolyticus infection is gastroenteritis. In at-risk populations (individuals with underlying chronic medical conditions), infection can lead to more serious outcomes (septicemia and death).

FDA announced the initiation of this risk assessment in 1999 in the Federal Register (FDA, 1999). The public was invited to comment on the planned assessment and submit scientific data and information for use in the assessment. The advice and recommendations of the National Advisory Committee on Microbiological Criteria for Foods (NACMCF) were sought on the assumptions and the model structure to be used. During the conduct of this risk assessment, FDA solicited the technical advice and opinions of scientific experts both within and outside of the Federal government. The availability of the draft risk assessment was announced in the Federal Register (Federal Register Docket No. 99N 1075) in January 2001 (FDA, 2001). A comment period was established during which FDA actively sought comments, suggestions, and additional data sources. The draft risk assessment was presented to stakeholders and other interested parties during a public meeting on March 20, 2001. The risk assessment report and model were modified based on the public comments received and availability of new data. The revised document and model were subjected to extensive review. A chronology of the technical and scientific review involved in the development of this risk assessment is provided in Appendix 1. A summary of the modifications made to the 2001 model is provided in Appendix 2.

Scope

This risk assessment is a quantitative product pathway analysis in which the key steps from harvest through post-harvest handling and processing to consumption were modeled. The likelihood of illness following exposure to pathogenic V. parahaemolyticus from consumption of raw oysters was calculated. The levels of V. parahaemolyticus in oysters at the time of consumption can be influenced by the harvest methods and handling of oysters after harvest and these practices may vary considerably in different geographic areas and at different times of year. The impact of regional and seasonal conditions on the predicted risk was evaluated.

The risk assessment had two main objectives: (1) to determine the factors that contribute to the risk of becoming ill from the consumption of pathogenic V. parahaemolyticus in raw oysters and (2) to evaluate the likely public health impact of different control measures, including the effectiveness of current and alternative microbiological standards.

The risk assessment addresses the following questions:

What is known about the dose-response relationship between consumption of V. parahaemolyticus and illnesses?
What is the frequency and extent of pathogenic strains of V. parahaemolyticus in shellfish waters and in oysters?
What environmental parameters (e.g., water temperature, salinity) can be used to predict the presence of V. parahaemolyticus in oysters?
How do levels of V. parahaemolyticus in oysters at-harvest compare to levels at consumption?
What is the role of post-harvest handling on the level of V. parahaemolyticus in oysters?
What reductions in risk can be anticipated with different potential intervention strategies?

Risk Assessment Overview

The Vibrio parahaemolyticus Risk Assessment follows the risk assessment structure of the Joint Food and Agriculture Organization/World Health Organization Expert Consultation on the Application of Risk Analysis to Food Standards Issues (FAO/WHO, 1998). The structure consists of four components: (1) hazard identification, (2) hazard characterization, (3) exposure assessment, and (4) risk characterization. Figure I-1 shows the organization and components of the risk assessment including the types of data and modeling techniques used.

Hazard Identification

The Hazard Identification component of a microbial risk assessment is the identification of the pathogenic organism that may be present in a particular food or group of foods that are capable of causing adverse health effects. The hazard on which this risk assessment is focused is pathogenic V. parahaemolyticus in raw oysters. The adverse health effect considered is the number of illnesses characterized by gastroenteritis and septicemia. See Chapter II: Hazard Identification for details.

Hazard Characterization/Dose Response/Severity Assessment

The Hazard Characterization component of a microbial risk assessment is often referred to as Dose-Response because it characterizes the relationship between the level of exposure to a pathogen (the dose) and the likelihood of an adverse health effect for individuals and populations (the response). For this risk assessment, a quantitative relationship was developed to predict the number and severity of illnesses resulting from ingesting different amounts of pathogenic V. parahaemolyticus. The Dose-Response model was developed using human clinical volunteer feeding studies and epidemiological surveillance data. See Chapter III: Hazard Characterization for details.

Exposure Assessment

The Exposure Assessment component of a microbial risk assessment defines the frequency and likely level of exposure to a pathogenic microorganism. In this risk assessment, the likelihood of exposure to pathogenic V. parahaemolyticus from consumption of raw oysters was evaluated. The Exposure Assessment was divided into three modules: Harvest, Post-Harvest, and Consumption. The levels of V. parahaemolyticus in oysters at the time of consumption can be influenced by the harvest methods and handling of oysters after harvest and these practices may vary considerably in different geographic areas and at different times of year.

Oysters are harvested in the United States from the Gulf Coast, Mid-Atlantic, Northeast Atlantic, and Pacific Northwest. In the Gulf Coast, the harvest duration for Louisiana is typically much longer than for other states in that region (Florida, Mississippi, Texas, and Alabama), therefore it was divided into two distinct regions: Gulf Coast (Louisiana) and Gulf Coast (Non-Louisiana). Likewise, the Pacific Northwest was divided into two distinct regions: Pacific Northwest (Intertidal) and Pacific Northwest (Dredged). In the Pacific Northwest, oysters are harvested by two methods: dredging and intertidal. For the intertidal harvest method, oysters are hand-picked when oyster reefs are exposed during the tide cycle and left in baskets until the tide rises to a sufficient depth to allow a boat to retrieve the basket. The risk assessment considered six oyster harvest regions and four seasons, for a total of 24 region/season combinations. See Chapter IV: Exposure Assessment for details.

Risk Characterization

Risk Characterization is the integration of the Dose-Response relationship with the Exposure Assessment to predict the probability of potential adverse outcomes for individuals or populations. For this risk assessment, the likelihood and severity of illness (gastroenteritis alone or gastroenteritis followed by septicemia) from the consumption of raw oysters containing pathogenic V. parahaemolyticus was predicted on both a per serving and a per annum basis. The uncertainties associated with the predicted risk estimates were also determined. See Chapter V: Risk Characterization for details.

Using the Model as a Tool: "What-If" Scenarios

The baseline risk assessment model can be used to estimate the likely impact of intervention strategies on the predicted number of illnesses. "What-if" scenarios were conducted by changing one or more model inputs and measuring the resulting change to the model outputs. Various control measures and mitigation strategies were evaluated. See Chapter VI: What-If Scenarios for details.

		Chapter II: Hazard Identification
		Characteristics of Vibrio parahaemolyticus Endpoints of concern: Gastroenteritis, Septicemia Susceptible populations Food considered: Raw Oysters Incidence : Outbreaks; Sporadic Cases; Seasonality
Chapter III: Hazard Characterization			Chapter IV: Exposure Assessment (Harvest, Post-Harvest, Consumption)
Data: Human clinical studies Surveillance Host susceptibility	Modeling: Dose-response curves Adjustment factor(s)		Data: Water temperature Total vs. pathogenic Vibrio parahaemolyticus in oysters Time-to-refrigeration Air temperature Growth rates Oysters consumed/serving	Modeling: Pathogenic Vibrio parahaemolyticus levels in oysters at harvest Growth between harvest and refrigeration Pathogenic Vibrio parahaemolyticus in raw oysters at consumption
		Chapter V: Risk Characterization
		Number of illnesses: per serving and per annum Severity of illness (gastroenteritis vs. septicemia) Uncertainty and variability analysis Model validation

		Chapter VI: 'What-If' Scenario
		4.5-log₁₀ reduction (heat; ultra high pressure) 2-log₁₀ reduction (freezing) approximately 1-log₁₀ reduction (immediate cooling) Impact of time-to-refrigeration after harvest Sample-based control plans

Figure I-1. Overview of Vibrio parahaemolyticus Risk Assessment Document

II. HAZARD IDENTIFICATION

The Hazard Identification component of a microbial risk assessment is the identification of the pathogenic microorganism that is capable of causing adverse health effects and is present in a particular food or group of foods. The hazard on which this risk assessment is focused is pathogenic V. parahaemolyticus in raw oysters and the adverse health effects include gastroenteritis and septicemia.

Vibrio parahaemolyticus

Vibrio parahaemolyticus is a Gram-negative, halophilic bacterium that occurs naturally in estuaries and is recognized as an important bacterial seafood-borne pathogen throughout the world (Fujino et al., 1953; Sakazaki, 1973). Vibrio spp. are found in the estuarine environment in the tropical and temperate zones (Joseph et al., 1983). These bacteria are normally present in many seafoods, including fish, crustaceans, and molluscan shellfish. They concentrate in the gut of filter-feeding molluscan shellfish such as oysters, clams, and mussels where they multiply and cohere.

The genome of V. parahaemolyticus was sequenced (Makino et al., 2003) and was found to consist of two circular chromosomes of 3,288,558 bp and 1,877,212 bp, and contains 4,832 genes. Although V. parahaemolyticus is phylogenetically close to V. cholerae, comparison of the V. parahaemolyticus genome with that of V. cholerae showed there are many rearrangements within and between the two chromosomes. Chromosome 1 does not differ much in size between the two genomes (33 vs. 30 Mb), but chromosome 2 is much larger in V. parahaemolyticus than in V. cholerae. Genes for the type III secretion system (TTSS) identified in the genome of V. parahaemolyticus are not found in V cholerae. The TTSS is a central virulence factor of diarrhea-causing bacteria such as Shigella spp., Salmonella spp., and enteropathogenic Escherichia coli, which cause gastroenteritis by invading or intimately interacting with intestinal epithelial cells. These results suggest that V. parahaemolyticus and V. cholerae use different mechanisms to establish infection.

Serotypes
Isolates of V. parahaemolyticus can be differentiated by serotyping. The system for identifying V. parahaemolyticus serotypes is based on the different antigenic structures of the lipopolysaccharides groups (referred to as O groups) and capsular types (referred to as K types) (Joseph et al., 1983). Thirteen O groups and 71 K types have been identified by commercial antisera (Iguchi et al., 1995). Of these, 11 O groups and 38 K types have been isolated from V. parahaemolyticus strains collected in the United States (Fishbein et al., 1974). In a recent study, 27 different O:K serotypes were found among 178 strains isolated from various sources including seafood, sediment and clinical samples (DePaola et al., 2003a).

Historically, V. parahaemolyticus infections have been characterized by sporadic cases caused by multiple, diverse serotypes. However, three serotypes (O4:K12, O1:K56, and O3:K6) predominated in outbreaks associated with the consumption of raw molluscan shellfish in 1997 and 1998. The serotypes isolated from patients in the 1997 outbreak in the Pacific Northwest included O4:K12 and O1:K56 (Daniels et al., 2000a). In outbreaks in 1998 in Texas and New York, the serotype O3:K6 was the predominant isolate and principal cause of illness. Prior to the 1998 outbreak, the O3:K6 serotype had only been reported in Asia; this was the first time it was reported in the United States. This serotype may have a lower infectious dose then other pathogenic V. parahaemolyticus strains (Daniels et al., 2000b).

Strains
Strains of V. parahaemolyticus are isolates of the same serotype that have been characterized or distinguished from each other. Not all strains of V. parahaemolyticus cause illness in humans; in fact, the majority of strains isolated from the environment or seafood are not pathogenic. For the purpose of this risk assessment, pathogenic strains of V. parahaemolyticus are those that produce thermostable direct hemolysin (TDH). TDH is an enzyme that lyses (breaks down) red blood cells on Wagatsuma blood agar plates, which is referred to as the Kanagawa phenomenon. The role of the toxin in illness is not known.

Illnesses Caused by Vibrio parahaemolyticus

The most common clinical manifestation of V. parahaemolyticus infection is gastroenteritis, an inflammation of the gastrointestinal tract. Gastroenteritis is usually a self-limited illness with moderate severity and short duration (Barker, 1974; Barker and Gangarosa, 1974; Hlady, 1997; Levine et al., 1993). A summary of clinical symptoms associated with V. parahaemolyticus gastroenteritis infection is presented in Table II-1. Symptoms of illness include explosive watery diarrhea, nausea, vomiting, abdominal cramps, and less frequently headache, fever and chills. Diarrhea may also be characterized by full-blown dysentery with blood and pus and superficial ulceration on proctoscopic examination (Carpenter, 1995).

Table II-1. Clinical Symptoms Associated with Gastroenteritis Caused by *Vibrio parahaemolyticus*
Symptoms	Incidence of Symptoms
Symptoms	Median	Range
Diarrhea	98%	80 to 100%
Abdominal cramps	82%	68 to 100%
Nausea	71%	40 to 100%
Vomiting	52%	17 to 79%
Headache	42%	13 to 56%
Fever	27%	21 to 33%
Chills	24%	4 to 56%

Source of data: Barker and Gangarosa, 1974; Levine et al., 1993

On rare occasion, infection can lead to septicemia. Septicemia is a severe, life-threatening, systemic disease caused by the multiplication of pathogenic microorganisms and/or the presence and persistence of their toxins in the circulating blood. It is characterized by fever or hypotension and the ability to isolate the microorganism from the blood. In cases of septicemia, subsequent symptoms can include swollen, painful extremities with hemorrhagic bullae (Hlady, 1997; Klontz, 1990). Death may also occur subsequent to the occurrence of septicemia.

Duration of illness can range from 2 hours to 10 days (Barker and Gangarosa, 1974; Barker et al., 1974). Information from several United States outbreaks revealed that the incubation period ranges from 12 to 96 hours with a median of approximately 15 to 24 hours (CDC, 1998; CDC, 1999a; Lowry et al., 1989; Nolan et al., 1984).

At-Risk Populations

Any exposed individual can become infected with V. parahaemolyticus and develop illnesses (such as gastroenteritis). However, infected individuals with underlying chronic medical conditions often develop septicemia. Therefore, although all raw shellfish consumers are "at risk" for infection, there is a subpopulation of individuals with increased risk of severe disease.

Individuals with Chronic Medical Conditions. Chronic medical conditions include liver disease, immunodeficiency, peptic ulcer disease, diabetes, alcoholism, hematological disease, gastric surgery, heart disease, renal disease, cancer or malignancy, treatment with corticosteroids, and transplant recipients (Klontz, 1990; Klontz, 1997; Angulo and Evans, 1999).

The percentage of the population that is at increased risk for development of septicemia from V. parahaemolyticus infection is not known precisely. The Center for Science in the Public Interest reported that approximately 20% of the United States population (60 million) have immunocompromised conditions and are at increased risk for V. vulnificus septicemia (CSPI, 1997). However, it is not known how many of these individuals consume raw oysters. Based on studies showing that certain persons are at greatest risk for illness from raw-oyster associated V. vulnificus infection (Desenclos et al., 1991 and Klontz, 1990), it was estimated that approximately 7% of the population have immunocompromising health conditions associated with increased risk of infection (Klontz, 1997). Analysis of epidemiological surveillance data (Angulo and Evans, 1999) indicates that approximately 30% of 107 cases of gastroenteritis were identified in individuals with underlying chronic illnesses. However, immunocompromised individuals may be over represented in case series data because of a "reporting phenomenon" driven by the severity of illness. An immunocompromised individual may be more likely to seek medical care for the symptoms of V. parahaemolyticus illness than an otherwise healthy individual with the same symptoms.

Raw Shellfish Consumers. Surveys conducted by FDA in 1993 and 1998 indicate that consumption of raw shellfish is not uniformly distributed in the United States population (Levy and Fein, 1999). For example, a higher percentage of men consume raw oysters than women (16% vs. 7%) , and raw shellfish consumption is higher for those living along the coastline of the United States than for those living inland (22% vs. 13%). The trend in raw shellfish consumption, as evidenced in the 1998 FDA survey, is toward lowered consumption of raw shellfish. This may be the result of education efforts by the Agency concerning the risks associated with the consumption of raw or undercooked protein foods, such as beef, chicken, eggs, and shellfish.

Annual Incidence

In 1999, CDC conducted a comprehensive evaluation of the national burden of infectious food-related illnesses in the United States. The total annual incidence of Vibrio illness was estimated as 7,880 illnesses and of that 65% were estimated to be food related (Mead et al., 1999). This estimate was based on the frequency of reported cases obtained by passive surveillance from 1988 through 1996 and the cases reported through FoodNet. The estimate also considers that this illness is under reported and under diagnosed and for every reported illness there are assumed to be 20 cases that are not reported (Kennedy, 2000; Mead et al., 1999).

Based on FoodNet data, the yearly estimates of food-related illness attributed to V. parahaemolyticus for 1996, 1997 and 1998 were approximately 2,700, 9,800, and 5,600, respectively (Tauxe, 2000). The 1997 estimate reflects the increased reporting of cases from a large outbreak in the Pacific Northwest. Some variation in estimated cases from year to year is expected, even in the absence of any inter-annual variation attributable to differing environmental conditions.

Specifically for this risk assessment (see Chapter III Hazard Characterization), CDC conducted an in-depth analysis of the available data on the incidence of illness from consumption of raw oysters reported over a 5-year period (1998-2002). CDC estimated there are approximately 2,790 cases of V. parahaemolyticus illness in the United States as result of oyster consumption (Painter, 2003). To obtain this estimate, CDC compared the reported cases from the National Notifiable Diseases Surveillance System (NNDSS) and the Cholera and Other Vibrio Illness Surveillance System (COVISS) because these systems collect reports from all states. Some cases are reported in both systems. A comparison of case information (using "capture-recapture" method for surveillance evaluation) indicated the number of reported cases was 1,125 for the 5-year period (or 225 cases per year). This compares well with FoodNet surveillance data (which represents 13% of the United States population) which indicate there are 300 cases per year in the United States. As noted above, CDC estimates that the number of cases is underestimated by a factor of 1:20 due to underreporting. So the estimated number of cases is 4,500 (225 x 20). Using information relating to V. parahaemolyticus exposure from COVISS, CDC estimates that 62% of all V. parahaemolyticus illness cases are caused by consumption of raw oysters. Therefore, the estimated number of cases of illness from V. parahaemolyticus in raw oysters used in the dose-response modeling was 2,790 (0.62 x 4,500). See Chapter III Hazard Characterization for details.

CDC's Active Surveillance Systems

FoodNet. The Foodborne Diseases Active Surveillance Network (FoodNet) is the principal foodborne disease component of CDC's Emerging Infections Program (EIP). FoodNet is a collaborative project of the CDC, 10 EIP sites (California, Colorado, Connecticut, Georgia, New York, Maryland, Minnesota, Oregon, Tennessee and New Mexico), the United States Department of Agriculture (USDA) and the Food and Drug Administration (FDA).
CDC Gulf Coast Vibrio Surveillance System (GCVSS). The CDC Gulf Coast Vibrio Surveillance System (GCVSS) is a unique regional system that began in 1988 (Levine et al., 1993). Four states initially participated in this program (Alabama, Florida, Texas, and Louisiana). Mississippi was added soon after, and the system has grown to include any and all states that are willing to participate; indeed, in the last few years, the West Coast states have become very active in reporting cases (Crowe, 2002). Investigators in state and county health departments complete standardized Vibrio illness investigation forms on all patients from whom Vibrio isolates are reported. Vibrio reporting comes from individual physicians, hospitals, or laboratories. Illness investigation forms contain clinical data concerning signs and symptoms, underlying illnesses, use of medications, as well as epidemiological information concerning seafood consumption in the week prior to illness. Data from this surveillance system has also been used for case series analysis (see discussion below).

Outbreaks and Sporadic Cases

An outbreak is defined as the occurrence of two or more cases of a similar illness resulting from the ingestion of a common food. The term "sporadic cases" refers to an irregular pattern of occurrence, with occasional cases occurring at irregular intervals. Sporadic cases can be reported as either "case reports" which present pertinent information on individual cases, or as a "case series" which is a study of sporadic cases over a specified period of time.

Outbreaks
The first confirmed case of foodborne illness-associated V. parahaemolyticus infection in the United States occurred in Maryland in 1971 with an outbreak associated with consumption of steamed crabs (Dadisman et al., 1972). Between 1973 and 1998, forty outbreaks were reported to the CDC from 15 states and the Guam Territories (Daniels et al., 2000a). These outbreaks were associated with raw seafood or cooked seafood cross-contaminated with raw or undercooked seafood. Since 1998, there have been three outbreaks caused by V. parahaemolyticus, and all were associated with consumption of oysters (Agasan, 2002; New Jersey Dept of Environmental Protection, 2002; Potempa, 2004).

Table II-2 summarizes the major outbreaks of V. parahaemolyticus gastroenteritis in the United States from 1997 to 2002. In 1997, an outbreak involving 251 cases occurred in the Pacific Northwest (202 in the United States and 49 in British Columbia) (Sample and Swanson, 1997). Of these cases, V. parahaemolyticus infection was confirmed in 209 persons who consumed raw oysters harvested from California, Oregon and Washington and from Canada (CDC, 1998). The most common V. parahaemolyticus serotypes isolated from patients involved in this outbreak were O4:K12 and O1:K56 (Daniels et al., 2000a). In the United States, oyster-associated V. parahaemolyticus outbreaks are more common than other shellfish-associated V. parahaemolyticus outbreaks (Daniels et al., 2000; Agasan, 2002; New Jersey Dept of Environmental Protection, 2002; Potempa, 2004).

Three separate outbreaks occurred in the United States in 1998. In the Pacific Northwest, 48 cases were reported (Therien, 1999). In Texas, a total of 416 V. parahaemolyticus infections were associated with consuming raw oysters harvested from Galveston Bay (Daniels et al., 2000a). Also in 1998, New York reported the first outbreak associated with raw molluscan shellfish harvested from that state and this outbreak included 23 cases, 10 of which were associated with raw oysters (CDC, 1999a).

In the summer of 2002, a cluster of seven cases with V. parahaemolyticus infection appeared to be linked to the consumption of shellfish that was harvested and purchased locally in the Long Island and New York City area (Agasan, 2002). In another outbreak that same year, a total of 11 cases with two fatalities were reported in New Jersey (Mulnick, 2002). These cases were attributed to the above average water temperatures that year and resulted in closing 110 square miles of oyster beds (New Jersey Dept. of Environmental Protection, 2002).

Table II-2. Outbreaks of Illnesses from *Vibrio parahaemolyticus* Associated with Consumption of Raw Oysters in the United States
Year	Location	Number of Cases
1997	Pacific Northwest^a	209^b
1998	Pacific Northwest^a	48
1998	Texas	416^c
1998	Northeast Atlantic	10^b
2002	New York	7
2002	New Jersey	11
2004	Alaska	46 (8^b)

^aThe Pacific Northwest includes California, Oregon, Washington State, and British Columbia.
^bNumber of cases that were culture-confirmed.
^cIncludes 296 cases in Texas and 120 cases in other states traced back to oysters harvested from Texas.

Case Reports
Several case reports have been published that outline clinical presentations and outcomes of patients with V. parahaemolyticus. One such case report describes a 35-year-old woman who sought medical attention for abdominal pain after she had consumed raw fish (Tamura et al., 1993). She presented with gastrointestinal symptoms, redness on lower extremities, fever, polyarthritis and weakness. Vibrio parahaemolyticus was isolated in the stool culture. She was diagnosed as having reactive arthritis induced by V. parahaemolyticus infection. Another clinical case report describes a 31 year-old female with a history of alcohol abuse, hepatitis C virus infection, and cirrhosis (Hally et al., 1995). She presented with diarrhea, weakness, leg pain, and urine retention. The patient had ingested raw oysters and steamed shrimp 72 hours prior to being admitted to the hospital. Vibrio parahaemolyticus was isolated from blood samples. The patient developed cardiac arrest and died six days after presentation.

A suspected case of a laboratory-associated infection was reported in 1973 (Sanyal et al., 1973). One day prior to the development of diarrheal disease the laboratory worker had been handling V. parahaemolyticus strains for the first time. The illness was associated with severe upper abdominal pain, bloody stools, nausea and fever. Weakness and abdominal discomfort continued for two days beyond the onset of illness. No other source of V. parahaemolyticus could be identified, and it was believed that the infection was caused by a relatively small inoculum.

Case Series
Case series data (Angulo and Evans, 1999) was used to analyze the relationship between illness outcomes and pre-existing health conditions. The data were from oyster-related culture-confirmed cases reported to the CDC GCVSS from 1997 to 1998. There were a total of 107 V. parahaemolyticus cases, of which 102 were gastroenteritis only, 5 that progressed to septicemia and 1 death. The overall incidence of septicemia among culture-confirmed V. parahaemolyticus infections was approximately 5% (5 out of 107). Of the cases with information on health conditions, 29% (23 out of 79) of the gastroenteritis illnesses and 75% (3 out of 4) of the septicemia illnesses occurred in individuals with an identified underlying (immunocompromising) health condition. The underlying medical conditions included liver disease, alcoholism, diabetes, malignancy, renal disease, immunodeficiency, hematological disease, and gastric surgery. The data from this case series was used in "Chapter III Hazard Characterization," to estimate the annual number of septicemia cases in susceptible and healthy populations.

Case series have also been reported by others including Bonner et al. (1983), Noland et al. (1984), Kelly and Stroh (1988b), and Levine and Griffin (1993). These studies have also illustrated the association of septicemia with underlying medical conditions. Three case series for illnesses and deaths associated with V. parahaemolyticus infections from consumption of shellfish in Florida from 1981 to 1991 are described below.

A case series of 4 patients who died in Florida due to V. parahaemolyticus infection from 1981 to 1988 was reported by Klontz (1990). All patients were male and all were over the age of 60 years. All died of septicemia. Two of the patients reported eating raw oysters during the week before onset of illness. The median duration of illness was 24 hours. All patients had underlying medical conditions, including cirrhosis, heart disease, prostate cancer and lung cancer.
A case series of 690 Vibrio infections related to raw oyster consumption in Florida during 1981 to 1993 was reported by Hlady and Klontz (1996). There were 355 cases of gastroenteritis, of which 68% were associated with the consumption of raw oysters and 120 (34%) were due to V. parahaemolyticus. Of the 118 cases of septicemia, 83% were associated with raw oyster consumption and 16 (14%) were due to V. parahaemolyticus. Of 467 patients with infections presenting as either gastroenteritis or septicemia, 35% had a preexisting medical condition, such as liver disease, alcoholism, peptic ulcer disease, gastrointestinal surgery, diabetes, antacid medication or immune disorders. While the prevalence of underlying illness was high in the septicemia patients, the majority of patients with raw-oyster associated Vibrio gastroenteritis had no underlying conditions. The reported cases of gastroenteritis caused by V. parahaemolyticus infection were more common during warm weather months.
A case series of 339 Vibrio infections reported in Florida between 1981 and 1994 was reported by Hlady (1997). Culture-confirmed case reports of Vibrio infections, reported to the Florida Department of Health and Rehabilitation Services were investigated. Oyster-associated Vibrio infection was defined as a history of raw oyster consumption in the week prior to onset of gastroenteritis or septicemia. Vibrio parahaemolyticus accounted for 77 of the 339 reported Vibrio infections. Of the 237 raw oyster-associated cases of gastritis, 68 (30%) of the infections were due to V. parahaemolyticus. Of the 193 patients who were hospitalized, 37 (19%) had infection with V. parahaemolyticus. Vibrio parahaemolyticus accounted for 4 (8%) of reported deaths. Patients with septicemia had underlying illness including, but not limited to, cancer, liver disease, alcoholism and diabetes mellitus.

Implicated Foods

Raw oysters are the most common food associated with Vibrio infection in the United States (Hlady, 1997). While thorough cooking destroys Vibrio, oysters are often eaten raw. However, there have been reports of V. parahaemolyticus illnesses associated with other seafood, including crayfish, lobster, shrimp, and crab. In a study from Levine et al. (1993), of 15 patients who ate seafood, the most commonly ingested foods were crabs, shrimp and raw clams. In addition, studies demonstrated the presence of V. parahaemolyticus in fresh fish, mussels and clams (Baffone et al., 2000). In an outbreak of V. parahaemolyticus in the Northeast in 1998, 16 of 23 ill persons ate either raw oysters or raw clams (CDC, 1999a).

Cooked seafood has also caused illnesses. Seafood cooked using seawater from the ships' fire systems caused outbreaks of V. parahaemolyticus gastroenteritis aboard two Caribbean cruise ships in 1974 and 1975 (Lawrence et al., 1979). Half of the 1,200 persons who ate boiled shrimp at a feast in Louisiana became ill with V. parahaemolyticus gastroenteritis in 1972 (Barker et al., 1974). Samples of the uncooked shrimp tested positive, indicating that the shrimp were colonized prior to arrival at the shrimp feast and were not cooked at an adequate temperature to kill V. parahaemolyticus or were re-contaminated after cooking.

Steamed crabs were implicated in two outbreaks in the United States from a cross-contamination with live crabs (Dadisman et al., 1972). In another United States outbreak, crab salad was prepared from packaged processed crab meat, opened the day the meal was served. The crab meat likely became contaminated prior to final packaging (Dadisman et al., 1972). A case-control study of sporadic Vibrio illnesses in two coastal areas of Louisiana and Texas was conducted from 1992-1993. Cooked crayfish consumption was reported by 5 of 10 persons affected with V. parahaemolyticus infection (Bean et al., 1998). In a study by Lowry et al., (1989), the presence of V. parahaemolyticus was surveyed from raw and cooked seafood from New Orleans restaurants. Vibrio parahameolyticus was isolated from all of the raw oysters sampled; the microorganism was isolated in 50% of cooked oyster samples, 67% of boiled shrimp samples, 33% of crab salad samples and in none of the boiled crabs.

Seasonality

The majority of outbreaks of foodborne illnesses associated with V. parahaemolyticus in the United States occur in the warmer months, with 94% occurring between April and October (Daniels et al., 2000a). CDC data (Smith, 2003b) indicates that of the oyster-related, culture-confirmed illnesses due to V. parahaemolyticus from 1988 to 2001, 60% occurred in the summer and only 4% occurred in the winter months. The breakdown of the number of reported cases of illnesses by season is provided in Table II-3. The same associations have been reported in other countries. In India, the monthly isolation of V. parahaemolyticus was more predominant in warmer months (Okuda et al., 1997) and in Japan the monthly outbreaks of food-related V. parahaemolyticus are more prevalent in summer with a peak in August (International Disease Surveillance Center, 1999; IASR, 1998).

Table II-3. Culture-confirmed *Vibrio parahaemolyticus* Illnesses Associated with Consumption of Oysters
Season	2000^a	2001^a	1988 to 2001^a
Winter	1	2	22
Spring	14	17	146
Summer	39	49	354
Fall	8	7	71
TOTAL	62	75	593

^aAnalysis based on oyster- related culture-confirmed V. parahaemolyticus infections reported to the Centers for Disease Control and Prevention (CDC) for which either a date of oyster consumption or a date of illness onset was reported (Smith, 2003b).

Geographic Distribution of Illness

Oysters are harvested in the United States from the Gulf Coast, Mid-Atlantic, Northeast Atlantic, and Pacific Northwest. The climate in these regions is different and there are different harvesting methods and handling practices within the regions that can have an impact on levels of Vibrio in oysters. For example, in the Pacific Northwest, oysters harvested in intertidal areas are typically exposed to higher temperatures longer before refrigeration then those harvested using dredging.

Of the four major oyster-harvest regions in the United States, the majority of oysters (approximately 50%) are harvested from the Gulf Coast and approximately 24% are harvested from the Pacific Northwest (Chapter IV: Exposure Assessment, Table IV-15). During the 1998 outbreaks, the Pacific Northwest shellfish harvested from the Hood Canal area of Washington were responsible for 32 of 48 (67%) of cases in the state of Washington (Therien, 1999). In the Gulf Coast, 20 of 30 harvest sites in Galveston Bay were implicated in the 1998 outbreak. In the Atlantic Northeast region, Oyster Bay Harbor (Area 47) was the only area implicated in the 1998 outbreak of that region (CDC, 1999a).

International Reports of Vibrio parahaemolyticus Cases

Vibrio parahaemolyticus was first identified as a foodborne pathogen in Japan in the 1950s (Fujino et al., 1953). By the late 1960s and early 1970s, V. parahaemolyticus was recognized as a cause of diarrheal disease worldwide. Below is a brief description of recent reports of V. parahaemolyticus illnesses in different parts of the world.

Japan. Prior to 1994, the incidence of V. parahaemolyticus infections in Japan had been declining; however, from 1994 to 1995 there were a total of 1,280 reports of infection due to V. parahaemolyticus (IDSC, 1999). During this time period, the incidents of V. parahaemolyticus food poisoning outnumbered those of Salmonella food poisoning. For both years, the majority of the cases occurred in the summer, with the largest number appearing in August.

Food poisoning due to V. parahaemolyticus in Japan is usually restricted to relatively small-scale outbreaks involving fewer than 10 cases. From 1996 to 1998, there were 1,710 incidents, including 496 outbreaks, with 24,373 cases of V. parahaemolyticus reported. The number of cases of V. parahaemolyticus food poisoning doubled in 1998 as compared to 1997 and again exceeded the number of Salmonella cases (IDSC, 1999). Similar to the 1994 to 1995 period, outbreaks were more prevalent in the summer with a peak in August and relatively few outbreaks occurred during winter months. Boiled crabs caused one large-scale outbreak, involving 691 cases. However, the majority of outbreaks were small in scale, but occurred frequently. There were 292 outbreaks and sporadic reports of V. parahaemolyticus involving 5,241 cases in 1996. In 1997, the incidence increased to 568 outbreaks and sporadic reports, with 6,786 cases, and in 1998, there were 850 outbreaks and sporadic reports (IDSC, 1999). The increased incidence during 1997 to 1998 has been attributed to an increased incidence of serovar O3:K6.

India. A hospital-based active surveillance of V. parahaemolyticus infections in Calcutta, India, conducted from 1994 to 1996, identified 146 patients (Okuda et al., 1997b). The incidence suddenly increased in February of 1996 and remained elevated until August of that year when surveillance ended. The increased incidence of V. parahaemolyticus infections was associated with an increased prevalence of O3:K6 strains. This serovar had not been isolated in Calcutta prior to February of 1996. The incidence of diarrhea due to V. parahaemolyticus strain O3:K6 accounted for 63% of the strains isolated from patients in Calcutta between September 1996 and April 1997. The virulant O3:K6 strains isolated from travelers arriving in Japan from Southeast Asian countries was indistinguishable from O3:K6 strains found in Calcutta, India (Matsumoto et al., 1999).

Vietnam. Five hundred forty eight cases of V. parahaemolyticus infection were detected between 1997 and 1999 in the Khanh Hoa province of Vietnam (Tuyet et al., 2002). Of these, 90% occurred in persons over 5 years of age, 421 (77%) reported vomiting, 258 (53%) presented with watery stools, 34 (6%) reported bloody stools. None of the patients died at the time of discharge from the health care service. A risk factor for infection was high socioeconomic status, which led the authors to believe that the source of infection was fresh seafood since only the most affluent members of the community can afford this delicacy. There was no definitive information on consumption.

Chile. Between November 1997 and April 1998, several gastroenteritis cases were reported in Antofagasta, a city in northern Chile (Cordova et al., 2002). The outbreak was associated with consumption of shellfish. This was the first report of V. parahaemolyticus causing an outbreak in Chile. Isolates were obtained from patient stool specimens and fresh shellfish. It was speculated that the exceptionally warm seawater caused by "El Nino" may have favored a bacterial bloom.

Spain. Between August and September 1999, an outbreak with 3 clusters of illness occurred in Galicia, Northwest Spain (Lozano-Leon et al., 2003). Sixty four persons were ill, 9 case patients were hospitalized. The most common symptom was diarrhea; other symptoms included abdominal cramps, nausea, headache, fever and vomiting. The median duration of illness was 3 days, and onset was within 12 to 24 hours after consumption of raw oysters in a typical outdoor street market. Vibrio parahaemolyticus was isolated in stool of all case patients. All patients resided in one of 2 cities near the outbreak site.

Taiwan. Vibrio parahaemolyticus has become a leading cause of foodborne disease outbreaks in Taiwan (Chiou et al., 2000). Vibrio parahaemolyticus accounted for 64% (542/850) of the food-associated outbreaks in Taiwan between 1995 and 1999. The O3:K6 serovar accounted for 0.6% of V. parahaemolyticus infections in Taiwan in 1995. This increased to 50% in 1996 and reached a peak of 84% in 1997. Comparison of outbreak data indicates that the high incidence of foodborne V. parahaemolyticus outbreaks from 1996 to 1999 can be attributed to the increase in O3:K6 infections.

III. HAZARD CHARACTERIZATION/DOSE-RESPONSE

The Hazard Characterization component of a risk assessment describes the adverse effects on the host of a particular substance, organism, or other hazard. In the current risk assessment, a quantitative evaluation was conducted of the dose-response relationship between the levels of V. parahaemolyticus ingested and the frequency and severity of illness. The dose-response relationship for V. parahaemolyticus was derived using human clinical feeding trial studies and epidemiological surveillance data. The probability of illnesses (gastroenteritis and septicemia) and the incidence of severe disease (septicemia) were evaluated.

Factors Influencing the Dose-Response Relationship

Dose-response relationships are influenced by three factors: the pathogen (e.g., virulence characteristics), the environment (e.g., the food matrix), and the host (e.g., susceptibility and immune status). These factors are described below.

Virulence Characteristics of Vibrio parahaemolyticus

Several different virulence traits have been associated with the pathogenesis of V. parahaemolyticus strains. These include their ability to:

produce a thermostable direct hemolysin (TDH) (Miyamoto et al., 1969);
produce a thermostable-related hemolysin (TRH) (Okuda et al., 1997a);
produce urease (Kelly and Stroh, 1988a);
invade the enterocytes (Akeda et al., 1997);
produce an enterotoxin (Honda et al., 1976b); and
produce pili as possible attachment/colonization factors (Nakasone and Iwanaga, 1990).

Currently, the only trait that has definitively been demonstrated to reliably distinguish pathogenic from non-pathogenic V. parahaemolyticus is the production of TDH. The tdh gene was first cloned from a Kanagawa-positive strain by Kaper et al. (1984). The so-called, Kanagawa Phenomenon (KP) is the exhibition of β-hemolysis induced by this haemolysin on a special blood agar (Wagatsuma) medium. This phenotype is strongly associated with clinical strains (Miyamoto et al., 1969). Pathogenic strains possess a tdh gene and produce TDH, whereas non-pathogenic strains lack the gene and the trait. For the purpose of this risk assessment, pathogenic V. parahaemolyticus are defined as those strains that produce TDH.

Food Matrix Factors

Food matrix factors such as fat levels, acidity, salt content, and other characteristics can have a significant impact on the ability of a pathogen to cause disease (Foegeding, 1997). For example, gastrin, the most potent stimulant of gastric acid secretion, is released after eating a protein-rich meal, such as oysters (West, 1985). Because most enteric pathogens, including V. parahaemolyticus, are sensitive to acids, the increased production of gastric acid actually provides a protection against infection. On the other hand, consumption of highly buffered foods (such as cooked rice) or antacids may decrease the number of microorganisms needed to cause illness because of their effects on gastric pH. For example, the ID₅₀ (the dose at which 50% of infected subjects become ill) observed in feeding trials with V. cholerae O1 was substantially lower when the microorganism was ingested with antacid vs. no antacids (Levine et al., 1981).

Host Factors

Host factors such as the general health status, presence of underlying disease, nutritional status, or physical stress can play an important role in an individual's response to infections. The immune status, especially of those individuals who are immunocompromised due to disease or medical treatments can influence occurrence and/or severity of foodborne diseases. Intrinsic factors such as age, sex, and genetics further influence the immune system, and thus the susceptibility of an individual to disease. For illness associated with V. parahaemolyticus infection, the severity of the disease is strongly associated with the presence of underlying medical conditions. The impact of immune status on the initial colonization and infection of the gastrointestinal tract is less clear-cut.

Human Clinical Feeding Studies

Several human clinical feeding trials were conducted prior to 1974 using pathogenic V. parahaemolyticus. The available data from these studies are briefly summarized here. Information on non-O1 V. cholerae is also provided as this represents a possible surrogate microorganism with respect to future investigations.

Feeding Trials with Vibrio parahaemolyticus

Takikawa (1958) used a Kanagawa-positive strain in a human volunteer study and showed that V. parahaemolyticus caused diarrhea in 1 of 2 individuals fed a dose of approximately 10⁶ cells. Diarrhea occurred in both individuals fed approximately 10⁷ cells. The ingested doses were not directly determined, but were instead estimated assuming that V. parahaemolyticus cultures can reach maximum growth densities of approximately 10¹⁰ cells per milliliter. These data were selected for the dose-response model.

In a study by Aiso and Fujiwara (1963), three clinical isolates (2 Kanagawa-negative strains and 1 Kanagawa-positive strain) and one shell fish isolate (Kanagawa-negative strain) were tested. The cultures were suspended in salted milk and were fed just prior to eating a normal meal. Illness only occurred with the Kanagawa-positive strain fed at a dose of 10⁹ organisms. Symptoms developed 5 to 11 hours after challenge. Typical symptoms included violent abdominal pain, diarrhea and vomiting in each of the 4 volunteers. The data for the Kanagawa-positive strain were selected for the dose-response model.

In a third study (Sanyal and Sen, 1974), three Kanagawa-negative strains isolated from cases of gastroenteritis were fed to groups of four volunteers each. No illness was observed in any of the volunteers at doses as high as 2 x 10¹⁰ cells. A Kanagawa-positive strain also isolated from a gastroenteritis case produced no symptoms at a low dose of 200 viable cells; however, abdominal discomfort was reported by 1 of 4 volunteers at a dose of 2 x 10⁵ viable cells, and 2 of 4 volunteers experienced abdominal discomfort and diarrhea at 3 x 10⁷ viable cells. All volunteers received antacid tablets prior to challenge with cultures suspended in gelatin. Only the data from the Kanagawa-positive strains were used in the dose-response model.

In another study, human exposure to 15 Kanagawa-negative strains isolated from fish produced no illnesses when doses as high as 10⁹ viable cells were used (Sakazaki et al., 1968). It was not reported how many volunteers were challenged in this study. These data were not used in the dose-response model.

A personal communication from Kasai (1971) reports that it took 6 to 8 hours incubation for a V. parahaemolyticus Kanagawa-positive strain to cause disease whereas a Kanagawa-negative strain required approximately 18 hours to cause disease after challenge. The infecting dose was reported to be approximately 10⁶ organisms. No information was provided in the communication about the dose level or number of volunteers in the study. These data were not used in the dose-response model.

Feeding Trials with non-O1 Vibrio cholerae

Two human clinical feeding studies have been conducted with non-O1 Vibrio cholerae, a potential surrogate for Vibrio parahaemolyticus. In one study, healthy volunteers were fed 10⁵ to 10⁹ levels of non-O1 V. cholerae. One of the three strains caused no diarrhea in 2 volunteers fed 10⁵ cells, 2 of 3 fed 10⁶, 1 of 2 fed 10⁷ and 3 of 3 fed 10⁹. Two other strains produced no disease at doses as high as 10⁹ cells(Morris et al., 1990). In a second study, Vibrio cholerae O139 Bengal fed to volunteers caused diarrhea in 2 of 4 fed 10⁴ cells and in 7 of 9 fed 10⁶ cells (Morris et al., 1995). The pathogenicity of this serotype more closely resembles Vibrio cholerae O1, and as such may be less useful as a potential surrogate.

Animal Studies

Animal studies using V. parahaemolyticus or a surrogate microorganism are potentially useful as a basis for extrapolating dose-response estimates for humans. Animal studies can also be useful for assessing the virulence potential of different strains and serotypes, susceptibility of sensitive subpopulations (i.e., immunocompromised), and the role of specific virulence determinants. Several V. parahaemolyticus animal studies have shown the virulence potential of TDH-negative strains. However, it remains to be determined whether the virulence potential of these strains also applies to humans. The effect of food matrices and other environmental factors on virulence and the dose-response relationship can be evaluated more readily in animal studies than in human studies. Potentially relevant animal dose-response data and identified factors influencing the infectivity of V. parahaemolyticus in animal models are described in this section. Although potentially informative, animal data were not utilized in the dose-response model for this risk assessment because the measures of the severity of illness in relevant animal studies did not correspond with definitions of human illness on which reporting statistics are based and therefore provided little additional information with respect to quantitative risk prediction/characterization of human illness.

A limited number of animal studies have been conducted using V. parahaemolyticus. In one study, suckling rabbits infected orally with a Kanagawa-positive strain at doses of 10⁹ to 10¹⁰ had positive blood cultures in 9 of 36 tested, positive spleen cultures in 11 of 21 tested and positive liver cultures in 14 of 21 tested (Calia and Johnson, 1975). Similar doses of a Kanagawa-negative crab isolate were negative for bacteremia, liver or spleen invasion in all 12 animals challenged (Calia and Johnson, 1975).

Hoashi et al. (1990) conducted 7 experiments in which mice were challenged intraperitoneally with 4 TDH⁺and 3 TDH^- strains. In the combined results of all 7 experiments, no deaths were reported with a dose of 10⁵ cells; 4% deaths with a dose of 10⁶; 61% deaths with a dose of 10⁷, and 90% deaths with a dose of 10⁸ cells. Combined results of 2 experiments in which mice were challenged orally with TDH-positive strains resulted in 38% deaths with a dose of 10⁷ cells, 57% deaths with a dose of 10⁸ and 80% deaths with a dose of 10⁹ cells (Hoashi et al., 1990). There were no significant differences in mortality between the TDH⁺ and TDH^- strains at any of the doses.

In rabbit ileal loop model the effective dose required to produce ileal loop dilation in 50% of rabbits for three Kanagawa-positive strains ranged from 2.6 x 10⁵ to 7.7 x 10⁶ cells (Twedt et al., 1980). It was estimated that the initiation of positive loops occurred with doses from 10² to 10⁵ cells (Twedt et al., 1980). Seven clinical isolates were tested belonging to four different serotypes that possess one or more virulence factors: TDH, TRH, and urease, in relation to the ability to cause diarrhea (Kothary et al., 2000). All strains were found to induce fluid accumulation in suckling mice and diarrhea in a ferret model after oral inoculation in a dose-dependent manner. The relationship between clinical and environmental origins of these strains was not evaluated.

Epidemiological Data

Epidemiological investigations of V. parahaemolyticus provide directly relevant information on the dose-response in humans. These data may be somewhat limited if there is a lack of information for the ingested dose associated with reported cases of illness. However, even when epidemiological data is not informative as to dose-response, such data often provide valuable information on the likelihood of illness (gastroenteritis) progressing to more severe outcomes (i.e., septicemia, death) in susceptible versus otherwise healthy populations. Information on the annual incidence of illness from surveillance data and outbreak investigations is provided in "Chapter II. Hazard Identification."

CDC estimated the annual illness burden from pathogenic V. parahaemolyticus associated with the consumption of raw oysters as 2,790 cases of illness per year (Painter, 2003). For additional information, see Chapter II: Hazard Identification.

Data Selection and Criteria for the Dose-Response Model

The selection of data for use in the Dose-Response model considered the availability of the data and limitations of data sources. Consideration was given to using the dose-response of an appropriate surrogate bacteria and/or host (i.e., animal model), which could provide a more suitable basis for risk prediction/characterization if uncertainties such as immune status and food matrix effects were substantially reduced. If a surrogate dose-response is to be more informative than the available feeding trials data, then better information is needed with respect to response rates associated with low dose exposure (including knowledge of relevant biomarkers) and the effect of the (oyster) food matrix on the dose-response relationship. However, the potential difference between a surrogate dose-response and that of V. parahaemolyticus adds an additional uncertainty with respect to risk prediction/characterization. For the purpose of this risk assessment, human clinical feeding studies with pathogenic V. parahaemolyticus were used. A summary of the selection criteria and evaluation of each identified human clinical feeding study is provided in Table III-1.

Table III-1. Summary of Criteria and Selection of Human Clinical Feeding Studies for Dose-Response Modeling
Study	Selection Criteria			Used in Dose-Response Model?
Study	Dosed with Vibrio parahaemolyticus	Pathogenic strains?^a	Dose Level Reported?	Used in Dose-Response Model?
Aiso and Fujiwara, 1963	Yes	Yes	Yes	Yes
Takikawa, 1958	Yes	Yes	Yes	Yes
Sanyal and Sen, 1974	Yes	Yes	Yes	Yes
Sakazaki et al, 1968	Yes	No	Yes	No
Kasai, 1971	Yes	Yes	No	No
Morris et al., 1990	No (V. cholerae)	Not applicable	Yes	No
Morris et al., 1995	No (V. cholerae)	Not applicable	Yes	No

^a For the purpose of this risk assessment, pathogenic Vibrio parahaemolyticus strains are those characterized as Kanagawa Phenomenon-positive.

Limitations of the Available Human Feeding Trial

The limitations of the available human feeding trial and surrogate studies for use in dose-response modeling are summarized below. Some of the studies were performed using uncharacterized strains.

No information was available on the immune status of the volunteers. Previous exposure of the volunteer to V. parahaemolyticus could provide some immunity to infection.
A dose range limited to relatively high doses of V. parahaemolyticus was used.
The V. parahaemolyticus dose was not administered with a food matrix; except for one study, which used salted milk (Aiso and Fujiwara, 1963). This is problematic because a food matrix can either increase or decrease stomach acidity. Protein-rich meals, such as oysters, would increase stomach acidity. Because V. parahaemolyticus is sensitive to stomach acids, the presence of oysters may increase the infective dose.
In most cases, antacids were administered with the V. parahaemolyticus dose. It is common to administer oral challenge dose either in or in conjunction with an alkaline solution or a fat emulsion (e.g., cream) in order to neutralize or minimize the impact of stomach acidity. This practice attempts to create less variability in stomach acidity among volunteers. The practice also effectively mimics achlorhydric (e.g., low stomach acid) conditions, which are common in a significant portion of the United States population, particularly in the elderly. While this helps to control the dose in the experimental context, it introduces an uncertainty with respect to inferring the dose that causes infection when V. parahaemolyticus is consumed with oysters. The magnitude of the difference between an infectious dose administered in an antacid, in comparison to that ingested in food, is generally unknown.
The number of volunteer subjects is small in each study. Most studies do not provide information on the volunteers such as gender, age, and health status. In general when information was provided, the majority of the volunteer subjects were male and relatively young (aged 25 to 40).

The human feeding studies were performed prior to 1974 and it is unlikely that any future human feeding studies with V. parahaemolyticus will be undertaken to resolve these issues due to an apparent cardiotoxicity of TDH in animal models (Honda et al., 1976a; Seyama et al., 1977).

Assumptions Made for the Dose-Response Model

All individuals are equally susceptible to probability of gastroenteritis.
Septicemia may only occur subsequent to gastroenteritis.
The likelihood that an infection will lead to more severe symptoms varies depending on pre-existing health conditions.
Approximately 7% of the population has underlying medical conditions and are at higher risk of V. parahaemolyticus septicemia once the gastrointestinal tract is infected.
Only 1 in 20 cases of V. parahaemolyticus illness is culture-confirmed.
The Kanagawa Phenomenon-positive strains used in the human volunteer studies are representative of pathogenic V. parahaemolyticus with respect to estimation of the steepness of the dose-response curve.
The slope of the dose-response curve was assumed to be the same for both the controlled feeding trials and oyster-related exposure situations.

Modeling the Dose-Response Relationship

The structure of the dose-response model is shown in Figure III-1. The V. parahaemolyticus dose-response model was developed by fitting a distribution to the selected human feeding trial data. The resulting estimate of the shape of the dose-response relationship was then modified by "anchoring" the mean risk predictions to be consistent with epidemiological surveillance data. The probability of cases of gastroenteritis progressing to septicemia was also calculated.

Figure III-1. Schematic Representation of the Development of the Vibrio parahaemolyticus Dose-Response Model

Studies and Data Sources Used for Dose-Response

Aiso and Fujiwara, 1963. Data from human clinical trial used to fit dose-response model.
Sanyal and Sen, 1974. Data from human clinical trial used to fit dose-response model.
Takikawa, 1958. Data from human clinical trial used to fit dose-response model.
Painter, 2003. Estimate of annual incidence of V. parahaemolyticus illness. Data used to 'anchor' dose-response model and adjust for limitations of the human clinical trial data.
Angulo and Evans, 1999. Data on culture-confirmed cases with medical history used to estimate the probability of septicemia.
Klontz, 1997. Estimate of percentage of United States population with underlying chronic medical conditions used to calculate probability of septicemia cases in this subpopulation.

Fitting Three Dose-Response Functions to Data

First, the available human feeding trial data for the incidence of gastrointestinal illness from the three selected studies [Takikawa (1958), Aiso and Fujiwara (1963), and Sanyal and Sen (1974)] were pooled. Collectively, a total of 20 healthy volunteers were administered pathogenic V. parahaemolyticus at doses ranging from 2.3 to 9-log₁₀ cfu in a bicarbonate buffer. In these three studies, 9 of 20 subjects developed symptoms of gastroenteritis. No illnesses were reported for the lower doses of 2x10² and 2x10⁵ cfu of V. parahaemolyticus. However, at higher doses (>1x10⁶ V. parahaemolyticus organisms) between 50% and 100% of the human subjects became ill. A summary of the dose levels, number of subjects, and number that develop illness is provided in Table III-2.

Table III-2. Summary of Data from the Human Feeding Trial Studies Used for the *Vibrio parahaemolyticus* Dose-Response Model
Dose (cfu)	Number of Subjects	Number of Illnesses	Rate of Observed Illness	Reference
2 x 10²	4	0	0	Sanyal and Sen (1974)
2 x 10⁵	4	0	0	Sanyal and Sen (1974)
1 x 10⁶	2	1	0.5	Takikawa (1958)
1 x 10⁷	4	2	0.5	Takikawa (1958)
3 x 10⁷	2	2	1.0	Sanyal and Sen (1974)
1 x 10⁹	4	4	1.0	Aiso and Fujiwara (1963)
Total Subjects = 20		Total Illnesses = 9

Secondly, the dose-response models were selected. Dose-response models are used to define the shape of the dose-response curves, allowing the extrapolation from the observed data from the human feeding trials to other (lower) dose levels. Three dose-response models, Beta-Poisson, Gompertz, and Probit, were evaluated. These models exhibit different behaviors at low dose levels; that is they would predict different probability of illness for the same exposure levels. These models are parametric, meaning that they can be described by a mathematical (i.e., algebraic) equation. The mathematical equations for these three models are shown in Table III-3. Additional details about the model selection are provided in Appendix 4.

Table III-3. Dose-Response Model Equations for the Probability of Illness as a Function of Ingested Dose
Dose-Response Model	Equation ^a
Beta-Poisson	Pr(ill \| d) = 1 - (1 + d/β)^-α
Probit	Pr(ill \| d) = Φ(α + β * log₁₀(d))
Gompertz	Pr(ill \| d} = 1 - exp[-exp[α + β * log₁₀(d)]]

^a For the Beta-Poisson, α and β are the shape (steepness) and location parameters, respectively. The approximation used for the Beta-Poisson dose-response function applies when α << β (and β>>1). For the Probit and Gompertz models, α and β are the location and shape (steepness) parameters, respectively. For all three models, d denotes the dose. For the Probit model Φ denotes the cumulative distribution function of a standard normal random variable.

Next, the dose response models were fit to the observed feeding trial data as shown in Figure III-2. The models were fit to the data by the maximum likelihood criteria; that is, the values chosen for the model equation parameters shown in Table III-3 were the values which maximized the likelihood of the model predicting data similar to the observed data. The adequacy of model fits to the data was evaluated using a likelihood ratio based goodness-of-fit measure. All of the models provided an adequate statistical fit to the data. For more information about estimated model parameters and the statistical evaluation of the model fits, see Appendix 4.

The Maximum Likelihood Estimate (MLE) is the most likely value of all possible outcomes (i.e., the best estimate of the probability of illness). The best estimates of the dose corresponding to a 50% probability of illness (i.e., the MLE of the ID₅₀) were determined to be 2.8×10⁶, 4.0×10⁶, and 3.2×10⁶ organisms/serving for the Beta-Poisson, Gompertz and Probit dose-response models, respectively. Although these estimates are not substantially different at the ID₅₀, the differences are much more substantial at low dose levels as can be seen in Figure III-2. For example, the estimated risk of illness is approximately 5 cases per 10,000 servings for the Beta-Poisson model at a dose of 1,000 V. parahaemolyticus organisms/ serving. However, at the same dose, the estimated risk is approximately 10-fold higher based on the Gompertz and approximately 10-fold lower based on the Probit. The differences between these models are less substantial for high doses that exceed 100,000 organisms per serving.

Selection of the Beta-Poisson Dose-Response Model

An evaluation of the uncertainty distributions of the risk predications for the three dose-response models was conducted (Appendix 4). This comparison indicated that considering the residual predictions of uncertainty, the three models were comparable. Therefore, for simplicity, one model was chosen to use in the risk characterization. Of the three models evaluated, the Beta-Poisson model is the only one that meets the mechanistic criteria identified by FAO/WHO (2003). The criteria include consideration that there is no threshold level (i.e., a single cell can cause illness). The Beta-Poisson model was therefore considered the most appropriate model to use for this risk assessment.

Figure III-2. Comparison of the Beta-Poisson, Gompertz, and Probit Dose-Response Models Fit to Data from Human Feeding Studies

Dose-Response Adjustment Factor

The V. parahaemolyticus human feeding trial data is the most complete data set available to describe the relationship between dose and the probability of illness. However, there are apparent biases in these data relative to what may be expected from exposure to V. parahaemolyticus by a diverse population consuming raw oysters. For example, the human feeding trials included concurrent antacid administration and no concurrent administration of oysters (food matrix) with the V. parahaemolyticus dose, which potentially changes the infective dose. Thus, the ID₅₀ observed in feeding trials would be expected to be lower than that of the general population based on effect of the food matrix vs. buffer on the infective dose.

Figure III-2 shows the relationship between dose and the probability of illness. Using the Beta-Poisson curve and the predicted exposure levels (see Chapter IV Exposure Assessment), the model would predict too many illnesses in comparison to epidemiological data. For example, using the Gulf Coast summer harvest, the mean exposure to pathogenic V. parahaemolyticus from oysters is predicted to be 20,000 organisms per serving (~100 cells per gram) (see Chapter IV: Exposure Assessment). At this level of exposure, the risk of illness would be predicted to be substantially greater than 0.001 (i.e., >1 illness in 1,000 servings). Accounting for the number of servings per year, this rate of illness would be approximately equivalent to 4,000 illnesses/year associated with the Gulf Coast summer harvest. This predicted rate is too high, considering that CDC estimates there are only 2,790 cases/year (Painter, 2003) for the entire United States population.

Based on the above considerations, the dose-response model was adjusted or "anchored' to be consistent with both the CDC's estimate of the average annual number of cases occurring per year and the estimated number of servings consumed (Chapter IV: Exposure Assessment). This adjustment factor represents the effect of the apparent differences between the dose-response observed in human volunteers under controlled conditions versus that in the general population when exposure is associated with the oyster food matrix.

The shape of the dose-response curve (i..e., the slope or steepness) was assumed to be the same for both the controlled feeding trials and oyster-related exposure situations. However, the location of the curve was shifted, using the adjustment factor. For the Beta-Poisson model, the resulting expression used for risk prediction was taken to be:

Pr(ill | d) = 1 - (1 + (d/(γ*β)))^-α

where γ is the dose-response adjustment factor.

The magnitude of the adjustment factor was estimated by iteratively running the risk characterization model and adjusting the location of the curve to be consistent with CDC's estimated average annual illness burden of approximately 2,800 cases (Painter, 2003). For the Beta-Poisson model, the resulting dose-response adjustment factor was estimated to be 27, which corresponds to a difference of 1.4-log₁₀ between the ID₅₀ under the controlled versus oyster-related exposure scenarios. The difference between the adjusted and unadjusted curves is shown in Figure III-3.

The solid line shown in Figure III-3 is the MLE of the Beta-Poisson model fit to the pooled human feeding studies data and the dashed line shows the shift adjustment (location) made so that the model predictions agree with the epidemiological surveillance data. From Figure III-3, it can be seen that the dose corresponding to a 50% probability of illness (ID₅₀) for the unadjusted curve is approximately 3 million and that of the adjusted curve is approximately 80 million.

Figure III-3. The Beta-Poisson Dose-Response Model for Vibrio parahaemolyticus Fit to Human Feeding Trials and Adjusted Using Epidemiological Surveillance Data

Figure III-3. The Beta-Poisson Dose-Response Model for Vibrio parahaemolyticus Fit to Human Feeding Trials and Adjusted Using Epidemiological Surveillance Data
[The solid line is the best estimate of the Beta-Poisson Model fit to pooled human feeding studies. The dashed line shows the shift adjustment so that the model predictions agree with epidemiological surveillance data. MLE denotes the maximum likelihood estimate. ID₅₀ is the dose corresponding to a 50% probability of illness.]

Uncertainty Characterization of the Dose-Response Relationship

Uncertainty in the dose-response relationship was characterized by performing a procedure called non-parametric bootstrapping. This procedure involves hypothetical replication of the observed human feeding study. However, given the limited number of possible outcomes (illness rates), the procedure was conducted as follows. For each possible outcome, the model was refit by the maximum likelihood criteria to obtain a set of parameter estimates, one corresponding to each possible (but unobserved) outcome. Weighting was assigned based on the probabilities of the outcomes. An uncertainty distribution was derived based on the parameter estimates and the weighting. The details of these calculations are provided in Appendix 4.

Figure III-4 shows a graphical representation of the weighted set of dose-response curves from the bootstrapping procedure. The 21 curves in this set were used in the Risk Characterization model. For each simulation (run of the model), a single curve was randomly selected, based on the assigned weight for that curve (the uncertainty distribution). The thick black curve shown in Figure III-4 is the curve that received the most weight (i.e., had the highest probability and would be selected most frequently). The weights for each curve and other supporting information are provided in Appendix 4.

Figure III-4. Vibrio parahaemolyticus Dose-Response Curve and Uncertainty

Figure III-4. Vibrio parahaemolyticus Dose-Response Curve and Uncertainty
[The dark line indicates the dose-response curve with the highest weighting (16.5%) and the 20 gray lines represent the dose-response curves with lower weightings (<1% to 13%).]

We did not apply uncertainty to the dose-response adjustment factor used to bring the model-predicted illnesses in alignment with the reported epidemiological illnesses (i.e., the shift shown in Figure III-3). To incorporate uncertainty in the dose-response shift an effort to assess the uncertainty in the number of illnesses occurring annually (i.e., uncertainty in the number of underreported illnesses) would need to be undertaken. See Appendix 4 for additional information regarding uncertainty in the dose-response model.

Predicted Probability of Illness

The Beta-Poisson Dose-Response model shown in Figure III-4 estimates the probability of the total V. parahaemolyticus risk per serving (gastroenteritis alone and gastroenteritis followed by septicemia) as a function of dose. For example, using the curve with the highest weight (the dark line in Figure III-4), the probability of illness is approximately 0.5 for a dose of approximately 100 million cfu. This means that for every 100 servings at that dose level, approximately 50 individuals will become ill. At exposure levels of approximately 1,000 cfu, the probability of illness is relatively low (<0.001). The probability of illness approaches 1.0 (i.e., 100% certainty of illness) at exposure levels around 1x10⁹ cfu.

Severity of Illness

For the purpose of this risk assessment, it was assumed that there is no sensitive subpopulation with respect to the occurrence of an infection leading to gastroenteritis. However, given the occurrence of illness, it was estimated that it was more likely that the infection leads to a severe outcome (e.g., septicemia or death) among individuals with an underlying chronic medical condition.

The probability of gastroenteritis progressing to septicemia in healthy and immunocompromised individuals was estimated using an application of Bayes' Theorem (see for example, Fleiss, 1973). The equation below illustrates the relationship between the frequency of a given outcome, health status, and the probability of the outcome.

Pr(illness outcome | health status)
= (Pr(health status | illness outcome) * Pr(illness outcome))/Pr(health status)

where, Pr(illness outcome | health status) denotes the frequency or probability of an illness outcome type within a subpopulation of individuals defined by the existence of a common predisposing health condition ("health status").

All factors on the right hand side of the equation are identifiable based on a set of CDC's epidemiological case series data reported by Angulo and Evans (1999). The statistics of the case series were:

107 cases of gastroenteritis
5 cases of septicemia
1 death

Of the cases with available information:

23 of 79 (29%) cases occurred in individuals with underlying chronic conditions
3 of 4 (74%) septicemia cases had an underlying chronic condition

Substituting the observed data into the above equation provides an estimate of the probability of septicemia occurring. Thus, for the subpopulation identified as having an immunocompromised chronic health condition, the probability of septicemia (given that illness occurs) was estimated as follows:

Pr(septicemia | immunocompromised)

= Pr(immunocompromised | septicemia) * Pr(septicemia)/Pr(immunocompromised)

= ((3/4) * (5/107))/(23/79) = 0.12

The probability of septicemia occurring consequent to culture-confirmed illness in healthy individuals and the total United States population was estimated in a similar fashion (see Appendix 4).

It is important to recognize that the estimated probabilities based on the CDC data pertain to culture-confirmed illnesses; i.e., these are probabilities conditional on both the occurrence of illness and the identification of that illness by a confirmed culture. Analysis of the cases series data (Angula and Evans, 1999) indicates that the rate of reported illnesses that are culture confirmed is higher in individuals with an immunocompromising health condition compared to individuals with no pre-existing health condition. It was assumed that approximately 7% of the United States population has an underlying medical condition (Klontz, 1997). Therefore, the equation was modified to account for the differential reporting rates for culture-confirmed illness for immunocompromised versus healthy subpopulations. For details of this analysis, see Appendix 4.

As shown in Table III-4, the overall estimated risk of progression to septicemia occurring subsequent to V. parahaemolyticus illness is 0.0023, or approximately 2 cases of septicemia per 1,000 illnesses. For immunocompromised individuals, however, the probability of gastroenteritis progressing to septicemia is approximately 10-fold higher, with approximately 25 cases per 1,000 illnesses. This translates to a mean of approximately 7 cases per year of septicemia for the total population, 2 cases per year for the healthy population, and 5 cases per year for the immunocompromised population.

Table III-4. Probability of Septicemia in Patients with Gastroenteritis from *V. parahaemolyticus* Infection
Population	Probability of Septicemia	Mean Number of Cases (per 1000 Illnesses)	Mean Number of Cases (per Year)^a
Total	0.0023	2	7
Healthy Individuals	0.00063	<1	2
Immunocompromised Individuals	0.025	25	5

^a Number of Cases per Year = (total illness/year) × (probability of septicemia) × (percentage of population). Total illness/year assumed to be 2,800 (Painter, 2003); 7% of the population assumed immune compromised (Klontz, 1997) and 93% assumed healthy.

Food

Section Contents Menu

Pathogenic Vibrio parahaemolyticus in Raw Oysters

Quantitative Risk Assessment on the Public Health Impact of Pathogenic Vibrio parahaemolyticus in Raw Oysters

Table of Contents

RESPONSE TO PUBLIC COMMENTS

CONTRIBUTORS (2004 Version)

ACKNOWLEDGEMENTS (2004 Version)

CONTRIBUTORS (2001 Version)

ACKNOWLEDGEMENTS (2001 Version)

Interpretive Summary

PREFACE

INTRODUCTION

SCOPE AND GENERAL APPROACH

HAZARD IDENTIFICATION

HAZARD CHARACTERIZATION

EXPOSURE ASSESSMENT

RISK CHARACTERIZATION

Predicted Illness Burden

Sensitivity Analysis

Model Validation

WHAT-IF SCEENARIOS

Reducing Levels of Vibrio parahaemolyticus in Oysters.

Reducing the Time-to-Refrigeration

Re-submersion of Intertidally Harvested Oysters

Sample-Based Control Plans

CONCLUSIONS

EXECUTIVE SUMMARY

Background

Scope and General Approach

Results

Conclusions

I. INTRODUCTION

Background

Scope

Risk Assessment Overview

Hazard Characterization/Dose Response/Severity Assessment

Exposure Assessment

Risk Characterization

Using the Model as a Tool: "What-If" Scenarios

Chapter II: Hazard Identification

Chapter III: Hazard Characterization

Chapter IV: Exposure Assessment (Harvest, Post-Harvest, Consumption)

Data:

Modeling:

Data:

Modeling:

Chapter V: Risk Characterization

Chapter VI: 'What-If' Scenario

II. HAZARD IDENTIFICATION

Vibrio parahaemolyticus

Illnesses Caused by Vibrio parahaemolyticus

At-Risk Populations

Annual Incidence

CDC's Active Surveillance Systems

Outbreaks and Sporadic Cases

Implicated Foods

Seasonality

Geographic Distribution of Illness

International Reports of Vibrio parahaemolyticus Cases

III. HAZARD CHARACTERIZATION/DOSE-RESPONSE

Factors Influencing the Dose-Response Relationship

Virulence Characteristics of Vibrio parahaemolyticus

Food Matrix Factors

Host Factors

Human Clinical Feeding Studies

Feeding Trials with Vibrio parahaemolyticus

Feeding Trials with non-O1 Vibrio cholerae

Animal Studies

Epidemiological Data

Data Selection and Criteria for the Dose-Response Model

Limitations of the Available Human Feeding Trial

Assumptions Made for the Dose-Response Model

Modeling the Dose-Response Relationship

Studies and Data Sources Used for Dose-Response

Fitting Three Dose-Response Functions to Data

Selection of the Beta-Poisson Dose-Response Model

Dose-Response Adjustment Factor

Uncertainty Characterization of the Dose-Response Relationship

Predicted Probability of Illness

Chapter IV: Exposure Assessment
(Harvest, Post-Harvest, Consumption)