Chairman - Dr. Fred Shank, FDA

Subcommittee Members:

Mr. Charles Bartleson

Dr. Robert L. Buchanan

Dr. Robert Gravani

Dr. Joseph M. Madden

Dr. Marguerite A. Neill

Dr. Michael T. Osterholm

Dr. Morris Potter

Dr. William H. Sperber

Dr. Don Vesley

Dr. Kaye Wachsmuth

FDA

Mr. Phil Derfler, General Counsel's Office

Mr. Jerold Mande, Executive Assistant to the Commissioner

AGENDA ITEM PAGE

Overview Presentations:

Introduction - Dr. Michael Friedman, FDA

Report of O157:H7 Outbreaks Caused by Juices -

Dr. Patricia Griffin, Centers for Disease Control

and Prevention

Concepts in Emerging Food-Borne Pathogens -

Dr. Robert L. Buchanan, Agricultural Research

Services (ARS)

Specific Microbial Concerns for Fresh and Minimally

Processed Foods - Dr. Douglas Archer, University of

Florida

Questions of Clarification

Odwalla Situation - Stephen C. Williamson

Greg Steltenpohl

FDA Findings

Facility Inspection and Follow-up - Dr. Frank

Scholl, San Francisco District Office

Laboratory Analyses - Dr. Richard Rude, Seattle

District Office

Findings in Washington State -

Dr. Russell Alexander, King County

Dr. Phil Tarr, Associate Professor of Pediatrics,

Children's Hospital - Seattle

California Findings, Packers & Growers Inspections -

Mr. Stuart Richardson, Department of Health Services,

California

Questions of Clarification

Labeling Issues - Dr. Edward Scarbrough, Center for

Food Safety and Applied Nutrition

Juice Industry Practices - Mr. Rick A. Kress,

President, Juice Processing Division, Seneca

Foods Corp.

Growing and Harvesting Practices -

Dr. Richard Dougherty, Washington State University

Dr. Eugene M. Kupferman

Mr. Paul Baker, Baker Farms

Organic Production - Dr. Harold Ricker, Agricultural

Marketing Service, Organic Standards

Questions of Clarification

Identify appropriate GMPs for the production and

harvesting of fruits and vegetables

Appropriate GMPs - Dr. Barry Swanson, Washington

State University

Future GMPs for the Apple Industry - Franklyn

Carlson, Carlson Orchards

Public Commentors

Adjournment

DR. SHANK: We have a full schedule, for those of you who have taken a few minutes to look at the agenda. It's going to be two full days, a lot of public participants, and that's exactly why we're holding this meeting, to get all the information on the record that is humanly possible.

Let us get started. I'll have some more introductory remarks to make about the operation of the meeting and those types of issues, but at this point I'd like to call on Dr. Michael Friedman, who is the Deputy Commissioner for Operations with the Food and Drug Administration, who is going to give us the introduction to the meeting. So let's go ahead and get started. Dr. Michael Friedman from FDA.

DR. FRIEDMAN: Thank you very much, Fred.

I want to begin by welcoming you all and thanking you for your interest in this public meeting on the safety of fresh juices. I recognize that at this time of year there are many demands that are placed upon your schedule, and I'm very gratified that you share our concerns, a fact indicated by your presence here in such numbers today.

The topic of today's meeting is the safety of fresh juices. The purpose of the meeting will be to provide an opportunity to evaluate the current science and to review the technologic and safety factors associated with the movement of fresh juices from producer to consumer.

Within this context, we will also be considering specific pathogens, such as E. coli O157:H7, which have caused a recent outbreak of food-borne illness, but in addition, also salmonella and other species of note.

This meeting is an important component of the consideration of whether and how FDA's regulatory program for fresh juice and juice products should be revised, to discuss and exchange information on relevant safety issues, to identify research needs where that's appropriate, to consider whether additional consumer education is necessary, and to consider other measures to reduce the risk of future outbreaks.

We are not here specifically to focus on one outbreak such as Odwalla. We must look to the broader context, the safety of a whole line of products; and while E. coli O157:H7 is a particularly virulent organism, it is not confined to one product category. It has also been found, as you know, in many products, unfortunately often after it has caused substantial harm.

But, again, this is not the only pathogen that is of concern to us, and we are asking you today to help us consider in this meeting the broader context of food safety. We would like to take a more global view. It is certainly not our intention to point fingers, assign blame, look at specific cases, only to the extent that it helps inform in a more general way.

Excuse me just a second.

 Pause

DR. FRIEDMAN: You may not be having trouble hearing me, but I was having trouble thinking. Thank you. Sorry.

Speaking for FDA, we are holding this public meeting to learn as much as possible from recent food-borne illness outbreaks involving fresh juices. As you can see from the agenda, we have invited a number of speakers who can provide a comprehensive overview of the current picture. Not only do we wish to learn about the recent outbreak involving apple juice and other apple juice-containing products produced by Odwalla, we're also interested in the entire area of food-borne pathogens, both emerging and those that we know about that may be associated with fresh or minimally processed juices. These issues must be factored into any agency action to enhance the safety of these products.

We are most interested in getting firsthand information from those in the apple and fresh juice industries, as well as those from the public health and academic communities. FDA wants to know about current manufacturing processes and how these practices may or may not contribute to microbial contamination of the product. We wish to hear from those who have real-world experience because you can provide us with important information on how vulnerabilities in these practices can be corrected without severely restricting consumer choice.

Not only is this meeting about the current manufacturing processes used by industry in order to minimize microbial hazards, we want to become more knowledgeable about promising new technologies and interventions coming on line that could enhance tomorrow's product safety.

During the course of the next two days, we will explore the science that is the essential foundation of any future action, and we will be learning what the current practices are, looking at new technologies that are coming, hearing opinion regarding optimal solutions. We will also be receiving comments from various individuals who have requested an opportunity to make a public statement, and that certainly will be offered.

Viewpoints will be presented on current juice industry practices from the perspective of both fresh and pasteurized juice producers. The producers of fresh juices and certain other products such as cider are generally small and local. Some of those producers are with us today, and we appreciate their participation.

We also have producers of pasteurized juices, and we expect a broad-ranging discussion of the techniques, such as processing temperature and the relationship to product stability and safety associated with these sorts of considerations. We will also address the question of what should we tell the consumer. The American consumer has a spectrum of perceptions with respect to juice products. Some believe that fresh and minimally processed juices are the healthiest and that organic production is preferable. Others prefer processed juices, and we recognize that are inherent risks associated with all practices.

Let me reiterate the purpose of the meeting: to establish a record of the current industry practices for the production of juice products and the science underlying the production of safe juices. FDA will use this record as the foundation of its critical review of our current practices, and it will hopefully form the basis of future actions.

I would like to remind you that the record of this meeting will remain open through January 3, 1997. There are copies of the Federal Register notice announcing this meeting available to you, and this provides specific information that you may wish to have for submitting additional written comments.

In closing, we're here to listen and to learn. We're optimistic that this exchange of information and this dialogue will facilitate the full and more complete review of our current policies. And while there will be consensus on some matters, there will undoubtedly be strongly held divergent viewpoints expressed also. I am confident, however, that everyone in this room shares the mutual goal of assuring, first, that products are safe, that consumers are provided with the number and variety of choices which they desire, and, third, that the number of outbreaks in the future should be kept to an absolute minimum.

We're very much looking forward to hearing from you and, again, I thank you for your participation and attendance.

On a personal note, I regret that I will not be able to be here for the meeting because other unavoidable activities are calling me away. I find that somewhat frustrating personally, but I'm very glad that there will be large representation from our Food Center to monitor, evaluate, and participate in the meeting. Certainly people from CFSAN are essential in this process.

We appreciate your consideration, we appreciate your participation in developing this important body of information over the next two days.

Thank you very much.

DR. SHANK: Thanks, Mike.

At this time let me take a little bit of your time and possibly put the meeting a little more in perspective and to introduce some people that we feel you need to know about.

On the design of the meeting, as you look through the agenda, you will see that we're starting off with a background on illnesses associated with juices. We'll then get into some more general information on pathogens in fresh or minimally processed foods. We will review the Odwalla situation. Then we will have a brief review of the labeling requirements that may have a bearing on fresh or minimally processed juices. We'll change our thrust then to looking at some of the practices, the current practices in the juice industry, as well as the current practices in the growing and harvesting of produce.

The program to this point is to provide a background for our further deliberations. It is to set the scene, so to speak, so that we can consider if changes are necessary and to start to think about how any necessary changes might be accomplished.

The remainder of the meeting then is to consider where we go from here. We will address appropriate GMPs for the production of fresh juice. This is to include from the farm to the table. It will include GMPs for growing and harvesting. We will give consideration to the identification of critical control points in the Hazard Analysis and Critical Control Point System.

We will consider whether or not there are new technologies or intervention strategies to control pathogens of concern and are there sanitizers or food additives available to ensure safe products and what are the appropriate uses of sanitizers and food additives that are available in their use in the production of juices.

Then we come to that ever important question: Is pasteurization of juices appropriate and necessary? What advice, if any, should be given to the consumer, and how should this advice be communicated? As an example, should this be through labeling? That will be another topic of consideration. And, finally, attention will be given to future research needs.

Let me make a few observations about our participants. As we consider what needs to be done in the future, that is, that portion of the program that I mentioned following the background, we have generally identified two speakers for each topic. The desire was to have presentations from different perspectives. We wanted to have a balance of academia, of industry, trade associations, and the public. And as Mike said earlier, this is a public meeting. We want to hear from all of you that have ideas on how to produce safe juices in the future.

More than three dozen so-called public participants have indicated their desire to make presentations, and if anyone desires to present that has not come forward to this point, please see the people where you signed in and get your name on the agenda.

We are requesting that you keep your presentations to preferably five minutes, and I recognize that some have asked for ten minutes. These requests and additional speakers from the floor will be accommodated as time permits.

At this point let me introduce the head table here from--the majority are from FDA, but to my immediate left is Dr. John Vanderveen, who is Director of the Office of Plant and Dairy Foods and Beverages with the Center for Food Safety and Applied Nutrition. Next to him is Mr. Phil Derfler who is from our general counsel's office. Dr. John Kvenberg is to his immediate left, who is the Strategic Manager for HACCP.

Morris Potter, Dr. Potter, is the Assistant Director for Food-Borne Disease, the Division of Bacterial and Mycotic Diseases, from the National Center for Infectious Disease, Centers for Disease Control and Prevention. And on his left is Dr. Joe Madden, Strategic Manager for Microbiology.

We have a group here in the front that are facing us. This is the Fresh Produce Subcommittee of the National Advisory Committee on Microbiological Criteria for Foods. I've got that list someplace.

 Pause

DR. SHANK: On this side over here is Mr. Charles Bartleson from the Washington State Department of Health. To his right is Dr. Bob Buchanan, Microbiologist, from the Agricultural Research Service, USDA, in Philadelphia. Dr. Kaye Wachsmuth, who is Acting Deputy Administrator for the Office of Public Health, the Food Safety and Inspection Service of USDA. Dr. Bob Gravani, Professor of Food Sciences from Cornell University. Dr. Neill, who is Professor of Medicine from the Brown University School of Medicine. Dr. Mike Osterholm, who is State Epidemiologist from the Minnesota Department of Health. Dr. Don Vesley, who is Professor, Environmental and Occupational Health, University of Minnesota. Dr. Bill Sperber, Senior Corporate Microbiologist from Cargill, Incorporated.

This subcommittee has been convened to look at fresh produce as an official function of the National Advisory Committee for Microbiological Criteria for Foods. They are here today in order to consider the deliberations of this meeting. They will consider the deliberations of this meeting in the future work of their committee, but I would underscore that FDA is holding this meeting for the agency's benefit to get that full comprehensive record of issues related to fresh juices.

I should also mention that there will be a full transcript of this meeting available. For this reason, and so that everyone can hear the full proceedings, we are asking that anyone desiring to speak to do so from a microphone.

I think that you all will agree with me that we have a very aggressive agenda for the next two days, and at this time I'm going to proceed with the agenda.

Our first presenter is Dr. Patty Griffin. Dr. Griffin is Chief of the Food-Borne Diseases Epidemiology Branch of the Division of Bacterial and Mycotic Diseases from the Centers for Disease Control and Prevention. Patty, do you want to come forward, please?

DR. GRIFFIN: Good morning. I'd really like to commend FDA for putting this meeting together. It's a busy time, and I know it took them a lot of time and resources to do it at a busy time of year. So I'd like to thank FDA from myself and from others at CDC.

In November 1991, four children from the Fall River area were admitted to the Massachusetts Children's Hospital, the Boston Children's Hospital, with HUS, a form of kidney failure that's caused by E. coli O157:H7. Their families spent a lot of time waiting while the kids were on dialysis, and they got together and decided that it was fish sticks that had made them all sick. The Health Department heard about the cluster and invited CDC to join the investigation of the fish sticks and other possible causes.

Before I tell you what the investigation found, I'm going to start my slides. So I'm going to give you an overview of E. coli O157:H7, which I'll call O157. I'll then talk about the Fall River outbreak and then about apple cider and juice outbreaks, both before and after 1991. I'll discuss an orange juice outbreak at a theme park and end with a summary and recommendations.

E. coli O157:H7 is so named because it has the 157th somatic or O antigen ever identified and the 7th flagellar or H antigen. This slide shows the sequence of events in infection. After O157 is ingested, usually three or four days pass while the organism colonizes the large bowel and multiplies. The incubation period can vary between 1 and 10 days. Illness then begins with non-bloody diarrhea and abdominal cramps. In many, but not all persons, the illness then progresses to bloody diarrhea, and in most persons, the illness resolves within a week. But in a few, about 5 percent, it progresses to HUS.

This slide lists features of patients with O157 infection that was severe enough that they sought medical attention and had their stools cultured. Over 90 percent had diarrhea, bloody stools, and abdominal cramps, but fewer than half had vomiting or fever, and this is typical.

Most O157 infections are not detected, and this is mainly because stools are still not cultured on the special sorbitol MacConkey medium needed to screen for O157, but also not all persons with O157 infection see a doctor or, if they do, even have a stool culture.

While most of our knowledge about O157 comes from outbreaks, this slide shows the complications of O157 in U.S. outbreaks. Overall, 22 percent of patients who are hospitalized, 5 percent developed HUS, and 1 percent died.

Now, the severity of illness varies widely with each outbreak. For example, in a nursing home outbreak in Canada, the death rate was 35 percent, and that's typical that you get more severe illness at the extremes of age, the elderly and the very young.

Hemolytic uremic syndrome was first described in Switzerland in 1955. It's characterized by low blood count and kidney failure. It's caused by O157 and other Shiga toxin-producing E. coli. It affects persons of all ages, but it's most common in children under 5 years old.

I'll just stop for a minute. Is there anyone having trouble hearing me? Raise your hand if you are. Okay.

HUS is a severe illness. Most of the affected children need a blood transfusion, 4 percent have a stroke, and 4 percent die during the acute illness. Half need a kidney machine in the hospital, and 4 percent still need it years later.

This little girl, like the little boy before her, survived HUS, but this little girl was in a coma for several weeks and now requires institutional care.

While I mentioned earlier that many O157 infections are missed, this slide shows the percent of U.S. laboratories that screened either all stools or all bloody stools for O157 by year. You can see at the end that even by 1994, only about half of labs were screening all stools or all bloody stools for O157. So even if a person developed O157 diarrhea and went to see their doctor and the doctor decided to have a stool culture, only about 50 percent of these people would have that stool culture looked for O157.

Now, you can see in the last two bars that screening increased in 1993 and 1994, and that was after publicity from the large outbreak in January of '93.

This slide shows the number of U.S. outbreaks by year, and you can see that it also increased markedly in 1993 and 1994, parallel to the increase in labs culturing for O157. So these data suggest that O157 infections have been grossly under-recognized for many years, and that part of the reason we're finding more outbreaks is that more labs are looking for O157.

This map shows in yellow states that have ever reported an O157 outbreak. Outbreaks are really just the tip of the iceberg. Most cases occur sporadically, so clearly this infection occurs nationwide.

This is a dairy farm, one of many dairy and beef farms, where O157 lives when it's not causing illness.

And this is a CDC epidemiologist doing one of the first investigations that linked O157 to cattle. O157 lives in the intestines of healthy cattle, deer, and sheep.

How is O157 infection acquired? This table shows the modes of transmission for the 75 U.S. outbreaks for which transmission was known. Foods from cattle accounted for half, with ground beef responsible for 43 percent, other beef for 5 percent, and raw milk for 3 percent. Non-bovine foods caused 19 percent of outbreaks, water 12 percent, and this was mostly lake swimming water that people swallowed; person-to-person spread caused 17 percent, and one outbreak was caused by an ill food handler.

The most famous ground beef outbreak was the multi-state outbreak in January 1993 in Washington, Idaho, Nevada, and California. Over 700 ill persons were reported, mostly children; 178 were hospitalized, 58 developed HUS, and 4 children died. The recall of hamburgers was based on epidemiologic data and is estimated to have prevented 800 more cases.

This slide lists non-bovine food products linked to O157 outbreaks. They include salad items like lettuce, ranch dressing, mayonnaise, pea salad, potato salad, fruit salad, sandwiches, apple cider and juice, punch, turkey, and cantaloupe--quite a varied list. I think the message here is that anything that could be cross-contaminated from beef or from cattle manure or from deer or sheep manure could transmit O157.

I want to return to that outbreak that I mentioned at the beginning, the HUS outbreak in Massachusetts in the fall of 1991. As I mentioned, there were four children with HUS from the Fall River area who were all admitted to the same hospital. O157 was isolated from the stools of three children, and there were reports of bloody diarrhea in family members.

When you have only four children ill, it's almost impossible to prove whether the cause was fish sticks or something else. So we set about case finding. We interviewed their ill friends and relatives. We reviewed charts in emergency rooms and physicians' offices looking for cases. We sent letters to physicians and labs asking them to look for O157 cases. And, finally, after we'd figured out the cause, we issued a press release.

We identified 23 persons with O157 infection. Now we had enough to do a case control study to determine the source. We used a standard questionnaire and compared items eaten by each ill person with items eaten by healthy control persons of similar age from the same neighborhood. Surprisingly, we found that ill persons were less likely than controls to have eaten ground beef. However, most ill persons, but few controls, drank cider from one cider mill, called Cider Mill A here, and all could have drunk cider produced on the same day.

Cider Mill A, which is shown here, used mostly drops, and that's apples collected from the ground. They weren't washed or brushed. A preservative wasn't added, and the cider wasn't pasteurized. Its pH was between 3.6 and 3.8. We wondered whether these practices were typical for the industry, so we conducted an anonymous survey of apple cider manufacturers in January 1992 at a meeting of the New England Small Fruit Growers Association. Forty-five growers were present, 80 percent of them completed the survey, and we found that all of them used drop apples, only 33 percent routinely washed and brushed their apples, and only 6 percent always used a preservative. So the practices at Cider Mill A were typical of other producers in the area.

To learn whether O157 was a plausible vehicle, we worked with Dr. Doyle's group at the University of Georgia. They inoculated cider with 10⁵ organisms of the outbreak strain. You can see that O157 survived in unpreserved cider for over 20 days. However, when we added 0.1 percent sodium benzoate, survival was less than 7 days, which was encouraging except that a lot of people consume cider quite soon after it's made.

During this outbreak, we reviewed the literature for other reports of outbreaks of cider and apple juice before 1991, and we found that in 1922 in France, 24 persons developed probable typhoid fever after drinking cider. The apples used in the cider had been rinsed with river water that had probably been contaminated with sewage. In 1974, in New Jersey, 296 persons developed salmonella diarrhea after drinking apple cider. The apples had dropped onto a manured field. And in 1980, in Ontario, 14 children developed HUS and one died after drinking apple juice at a fair.

We wondered after our outbreak in Massachusetts how frequently apple cider and juice caused illness, so we've been watching for other outbreaks, and we know of five apple cider and juice outbreaks after 1991. In 1993, in Maine, 213 persons developed Cryptosporidia diarrhea after drinking cider made from apples that had dropped from trees on the edge of a cow pasture. This cider had been prepared by school children. And this fall, in Connecticut, 10 persons developed O157 diarrhea, 2 had HUS, after drinking commercial cider. Well water used for rinsing the apples had coliforms.

This fall, in New York, 32 persons developed Cryptosporidia diarrhea after drinking commercial cider. Well water used for rinsing the apples had coliforms. And this fall, in Washington, six persons developed O157 diarrhea after drinking cider made at a church event. The apples had been washed in a chlorine solution.

And the last outbreak, the one that got us all together, was in western North America last month. I'll give you a brief overview, and others, I know, will be filling in details.

O157 was linked to consumption of Brand A unpasteurized apple juice. Most cases had consumed pure apple juice; 6 percent consumed blends of apple juice and other juices. All Brand A apple juice was recalled on October 30th based on epidemiologic data. It's important to note that the recall was five days before O157 was cultured from juice.

Cases occurred in British Columbia, Washington, California, and Colorado. Sixty-six juice-associated cases were identified, of which 14 had HUS and one child died. Fifty-eight percent of cases were children 5 years old or less. The transmission route for 61 was drinking juice, for one was spread from a person who drank juice, and for four, either of these routes was possible.

This graph shows the dates of illness onset for cases. The latest case shown on the far right was probably acquired from another person who drank the juice. The latest "enjoy by" date for the juice was October 24th. That was for the suspect juice. The recall was on October 30th, and the report of O157 in juice was on November 4th.

The FDA lab in Seattle helped the traceback immensely by culturing O157 from apple juice produced on October 7th. Subtyping demonstrated the same O157 strain in patients and juice. The juice produced on October 7th was made from apple lots X and Y. Many visibly poor-quality apples from one of these lots were discarded. Others may have gotten into the juice. Lots X and Y apples were used in juice blends produced between October 5th and 9th. Juice produced between October 5th and 9th could account for 97 percent of cases.

The production practices used for apple juice in this outbreak were more stringent than in previous outbreaks. Drop apples may have been used, although contracts with suppliers specified no drops. Apples were washed with a phosphoric acid solution, agitated, and brushed. To monitor juice quality, juice was tested for bacteria and yeast after distribution.

To summarize the nine outbreaks from apple cider and juice, of those with information available, 33 percent did not use drop apples, 88 percent rinsed apples, 60 percent brushed apples, 22 percent disinfected apples, and 11 percent added a preservative. Clearly, these practices have not been sufficient to prevent contamination and subsequent illness.

I'd like to change gears now and speak briefly about another juice that concerns us. I want to tell you about an outbreak of salmonella infections from unpasteurized orange juice in Florida in 1995. We identified 62 cases of salmonella in visitors to Theme Park A in Orlando. They had traveled there from 21 states. Their average age was 10 years, and 22 percent were hospitalized. Our case control study showed that illness was associated with drinking unpasteurized orange juice at character breakfasts. That's breakfast at which cartoon characters are present.

Plant A orange juice was shipped interstate. We and others cultured unopened containers weeks after the outbreak and found between 220 to 1,700 E. coli per 100 mls of orange juice. Eleven containers yielded salmonella of three different serotypes. The pH of the juice was 4.

The oranges came from many groves, but a major grove used surface water for irrigation. The oranges were often knocked from the trees onto the ground, and later, cultures of both soil and the surfaces of oranges yielded salmonella.

The plant investigation showed that the plant used a phosphoric acid rinse to clean the oranges. Salmonella strains were isolated from a toad and a frog outside the plant, and animal droppings were found inside the plant. This environmental work was done by Mickey Parrish of the University of Florida.

So, in summary, outbreaks from cider and juice have been recognized for decades. Sporadic cases are probably common but unrecognized. Unpasteurized cider and juice, even at pH 3.6 to 4, can transmit pathogens. Current production practices do not guarantee the safety of apple juice, apple cider, and orange juice. And product recalls based on epidemiologic data can prevent illness.

The challenges to control this problem include that deer, which can run through apple orchards, can carry both O157 and Cryptosporidia; that only a few organisms of O157 and Cryptosporidia are needed to cause human illness; that O157 and Cryptosporidia are acid tolerant and can survive in cider for as many as four weeks. Data on orange juice is similar. And adding preservatives to cider has had mixed results, at best.

One more. Good.

Our recommendations for apple cider and juice are to strengthen industry practices and regulations to assure that apple cider and juice shipped interstate is safe, which means an additional barrier to microbiologic growth, such as pasteurization, and to label unpasteurized juice and cider to indicate the risk of illness if these products are still permitted. We must also determine the practices and regulations needed to improve the safety of other juices.

My last slide shows the many collaborating agencies in the apple juice outbreak and in the orange juice outbreak.

Thanks for your attention.

 Applause

DR. SHANK: I'm going to ask you to hold your questions for Dr. Griffin until the question-and-answer period a little bit later this morning.

Our next speaker will be Dr. Bob Buchanan, the research microbiologist of Eastern Regional Research Center of the Agricultural Research Service. Bob is going to address some concepts in emerging food-borne pathogens. Bob?

DR. BUCHANAN: Thank you, Fred. And I've bypassed the need to use the slide changer. I have a more primitive form of slide changer.

 Laughter.

DR. BUCHANAN: What I've been asked to do for the next few minutes is to talk about some of the concepts that go into considering the emergence of a new pathogen, and it can be a rather complex process. I'm not going to be talking just about juices or just about E. coli O157, but I have tried to select some examples of different issues or different factors that influence it that are related to the area that we're talking about today.

Could I have the next slide?

It becomes apparent as you go back and look at the records that it's a changing world. If you go back 50 years ago, to a food microbiologist there were only basically five organisms that caused a lot of food-borne disease. They were sort of classic pathogens.

Next slide.

However, that's changed dramatically in the past 50 years, and now we have a list that these are just a few of the new ones, and, in fact, each year I add a new organism to the list as we have what are called emerging pathogens, new organisms that previously had not been considered to be a cause of disease associated with foods that have cropped up. This year we've added a new one, that second item, cyclospora. E. coli O157, of course, is one that we're focusing on today. But there has been a continuing emergence of new pathogens, and there has also been the re-emergence of classic pathogens in new situations. Both of them are of interest when we're trying to determine the public health impact in terms of food safety.

Next slide.

Now, when considering food safety issues and the emergence of pathogens, we always have a balance that you have to deal with, and you can't consider the organism by itself, but you always must consider it in balance with the people that are going to be affected and also the food. The pathogen will have different capabilities of causing disease which may change over time. The host may have different susceptibilities to disease. And, likewise, the food may either enhance or decrease the risk associated with an emergence of a pathogen.

When you change the balance, that's when you get emergence or re-emergence of a disease. When you change the genetics of the organism or you change susceptibility of the host or you change the way you manufacture and distribute foods, that is when we get the emergence of a new disease or the re-emergence of a known pathogen.

Next slide.

The factors that can contribute to the emergence of a new pathogen are very complex, and I don't expect anyone to read this. Actually, I'm going to go through this in a little bit more detail, giving specific examples, and this is just a small number of the factors that can contribute to the emergence of a new microbiological problem.

Basically, we can divide these down into three areas, representing those three I talked about: changes in the organism, changes in the host, or changes in the food.

Next slide.

Let's consider first examples of factors associated with the host, and the four I have highlighted up here are the ability to cause disease, that is, the pathogenicity of the organism; the geographical and temporal distribution and density of the pathogen in the environment; the ability of that pathogen to either grow or survive in the food; and then the ability to adapt to new niches and reservoirs. Each of these can contribute to the overall disease process.

Next slide.

Using the example of one of the organisms we're concerned about today is one that we have some really definitive, well-established proof in terms of its genetics. E. coli dates back to the dawn of food microbiology, and it goes back to the 1890s when it was first proposed as a non-pathogenic indicator of fecal contaminati on. And it rests as a non-pathogenic organism, and, in fact, still most E. coli that you isolate are non-pathogenic. However, in the 1940s, we started to have reports of infantile diarrhea associated with this organism. In the 1960s, we began to active food-borne cases associated with helabel(?), a cholera-like toxin, and then finally in the 1980s, we've had the emergence of the enterohemorrhagic E. coli.

Each one of these different classes of pathogens within the species represents the acquisition of a gene or genes that cause a virulence factor that induces a disease. So, for example, the 1960s represent the acquisition of a cholera-like toxin gene by the organism on a plasmid. In the 1980s, we see the acquisition of genetic material for Shiga-type toxin, for a number of other virulence factors that the organism has picked up from a bacteriophage. So we have the evolution of the organism, in this case subspecies that have become more pathogenic.

Next slide.

In addition to the genetic material, we also have the fact that the organism can be selected for to adapt to conditions, and certainly one of the things that has become known about the E-hax(?) is that they have a number of different systems within them that allow them to become resistant to things like acid and heat and a number of other factors. And, in fact, what happens is that if you don't kill them, you may actually induce them into a state of being more resistant to other factors. So this is an area that's fairly new. We're only starting to understand the physiological basis for it, but certainly we can demonstrate that we can induce these organisms to be quite acid tolerant.

Next slide.

The other thing that we have to consider when we're talking about the organism is not just the organism itself, but its reservoir. Where does it exist in nature? And how has that reservoir changed? Probably the best known that I hear that is an indirect effect is Lyme disease. Lyme disease is a disease that emerged in 1975 in Lyme, Connecticut, and it took awhile to figure out what it was, but it became apparent that in Lyme disease the natural reservoir for this organism is the deer population. And as the deer population in that region increased, the number of deer then increased the number of deer ticks, and the deer ticks are the vector that take the organism of concern, which is a spirochete, and transmits it from the deer population into the humans. So, obviously, here one of the ways of controlling this disease is not only through the control of the organism, but control of the deer population.

I think this one is a very relevant example to some of the issues that we're facing today because certainly, for O157, deer populations have been implicated as one of the potential reservoirs for the organism.

Next slide.

The other thing when we're dealing with emergence that we have to look at is where is its progression. We typically have the initial emergence of the organism, and there is some question about what the initial emergence was with this organism. But once you have it emerge and you have it passing through a population, it becomes biologically amplified. You have a much greater load within the environment, and then you'll have the organism, particularly if it's fairly adaptive, it will start adapting itself to new niches and new reservoirs. And I think that's one of the things we're seeing with O157 now. As the load of it in the environment increases, it's going to be spreading out to different places. It's really quite a competitive organism and can move and survive and actually thrive in a number of environments.

So we're seeing new products, new reservoirs, the area of potential person-to-person spread, and the humans acting as a reservoir is something that we have to consider now.

Next slide.

That brings us into the area of examples of host-associated factors, and there are a number of t hese. I'm not going to spend a lot of time on them this morning, but you can have differences in the susceptibility to disease, you can have the acquisition of immunity, you have effects of demographics. What is your population density? What is the population age? How does that match with susceptibility to disease? General socioeconomic factors can, in fact, impact the overall susceptibility to disease. And then one of the factors that is also sometimes underrated is the host manner in which they handle the food, prepare the food, their food preferences, et cetera. Certainly in the initial emergence of E. coli O157, the fact that a substantial portion of the population likes ground beef in a rare state, this certainly contributed to the initial amplification.

Next slide.

We also have the food. If the vehicle that we're moving this organism around is the food, then we have to consider how changes in the food will affect the potential emergence of a pathogenic organism. And this can either be because of the composition or other inherent characteristics of the food itself, or it can be changes in the technologies, either to produce, to process, to distribute, and even to cook the product.

Next slide.

I've just picked a couple of very simple examples here that might be pertinent. Certainly there was concern a number of years ago that as we become more proficient in making varieties of tomatoes, the pH of the tomato in certain varieties has risen up above the magic 4.5 that we currently use for calculating processes, thermal processes. And I know there was some concern that we would get the potential for bot outgrowth in these products if we made products from them, and then treated them as if they were the same tomato. So here changes in varieties can have an impact on changes in the composition of the product, which in turn affects its ability to resist pathogens.

Next slide.

Changes in technologies can be very simple, and this is one that came out of FDA's history. It's well documented that there were concerns with mushrooms because of the introduction of plastic wraps. Plastic wraps represent a barrier to oxygen. Mushrooms are a very rapidly respiring product that restriction of oxygen getting into it actually allowed the environment to become anaerobic to the point where it could support the growth of clostridium botulinum or also staph aureus. Here even a small change in technology, something as simple as what we put over the package, had an impact on the capability--the potential for the emergence of a pathogen, or in this case the re-emergence of a pathogen. The solution was equal ly as simple. They poked holes in the plastic wrap to let some oxygen in.

Next slide.

I also want to indicate that there is an inherent qualitative difference when we have to deal with high-dose and low-dose organisms. When you become--when you're working with pathogens that have to reach relatively high numbers, your basic thrust of food processing is to prevent the growth of the organism because you need a relatively high level before you're going to have a problem. When you're dealing with high infectivity organisms, particularly those that have severe symptoms, we get into a different mode where you have to not only control their growth but you also have to just control their presence. So you have to be able to control their survival or you have to be able to control just their mere presence. And as we get an organism that is more and more infective, that is, the lower and lower the number that you have to have someone consume to get a disease, the more and more we have to move to controlling just the mere presence of the organism.

Next slide.

The final concept I'd like to talk about is we also have in the emergence of disease geographical barriers. If you're someone looking at emerging pathogens and not doing it in food but in general, the place you look for is an isolated area that they're just building a road into. Because when you get a mixing of a previously isolated area, that's when you're going to find new pathogens. One of the classic places to see disease problems is in boot camp in the military when you mix people from all different regions, and these will lead to the emergence of disease outbreaks. Or probably the best example is they estimate that the HIV virus has been around for about 10,000 years, but it wasn't until roads went into that area of Africa where it was originally isolated from that it then became dispersed.

It's important to note here--could I have the next slide? It's a very different world. There's a tremendous globalization and regionalization of the food market, and basically we can disseminate an organism around the world in 24 hours, either in people or in products, and we ship food around the world.

And the last slide.

So there's two take-home messages when we're working with this. Any change in the biological agent, the host, or the food has the potential of resulting in the emergence or re-emergence of a food-borne pathogen. Then the other is nothing is constant. Change is something that we deal with all the time. So it's something that we have to consider. As you make changes, what is the potential impact on the emergence of a pathogen?

 Applause

Thank you.

DR. SHANK: The next presenter this morning is Dr. Douglas Archer. Doug is Professor and Chair of the Department of Food Science and Human Nutrition at the University of Florida. He's going to address some specific microbial concerns of fresh and minimally processed foods. Doug?

DR. ARCHER: Thanks, Fred. It's good to be here. It's given me an excuse why I don't have my Christmas shopping done.

I don't have slides this morning because I was waiting for Bob Buchanan to send me his talk so that we wouldn't overlap, and Bob didn't do that. So I don't have anything to show you. But I do have a few things I do want to say.

The topic that I was asked to talk about are specific microbiological concerns of fresh and minimally processed foods, and you've heard some of those concerns already. But I think what I'd rather do a little bit is stick with some basic truths, and they all are relevant to the issue of microbiological safety of foods.

The first one is that any potential pathogenic microorganism on a food that will receive no further processing is of concern because it is one step from the host. And we saw that a little bit graphically before from Dr. Griffin's presentation. The salmonella out in the middle of an orchard has a long way to go before it finds a human being, and it becomes a game of chance. Simple mathematics: Will it make it to the host, and will that host become ill?

We've been reminded lately that some microorganisms are inherently more virulent than others, ev en within their own genus or within their own species. And we're learning that the environment of food itself may modulate virulence. Dr. Buchanan alluded to that. An organism that has been exposed to an acidic environment or other kind of a stressful environment may well be more virulent for human beings than one that has been relatively pampered.

We are concerned with those organisms that Dr. Buchanan referred to as the very virulent microorganisms. Some of those are becoming fairly well known. Among the bacteria, those of most concern would be organisms such as Shigella, the causative organism of dysentery; salmonella, certain species or certain subtypes of salmonella; and certainly E. coli O157 and the other so-called virocytotoxigenic E. coli.

We've also become concerned with certain protozoans and microsporidiums such as Cryptosporidia, Giardia, and cyclospora most recently. And the viruses are always of concern because of their very, very low infectious dose for humans. These include hepatitis A and the Norwalk agent.

Now, when these are present in or on a ready-to-eat food, illness may well result. But we do ha ve to consider something, and that is that these organisms are becoming more widespread in our environment. In fact, salmonella, once referred to as a zoonotic disease, a disease associated with animals, has been declared a geonosis by the World Health Organization. That means it's everywhere in the environment, and we need to recognize that and deal with it.

The protozoa and the viruses and certain bacteria underscore how normally water-borne agents may contaminate food. Water frequently becomes a component of food, or water is used to sanitize food. And if that water is contaminated, the food will become contaminated as well.

The more widespread these organisms become in the environment, the more likely it is that humans will encounter them and illness may result.

Now, there's an old saying when people try to compare two things that are not similar, and that is, it's like comparing apples and oranges. Now, that axiom recognizes that both are fruits, but they are different from one another. One could also say it's like comparing grapes and strawberries. They, too, are different, and I would hope that FDA and the Advisory Committee would recognize that juices are very different from intact fruit or vegetables from which they are made, and they're very different one from the other. They should not be compared, and the potential risks should not be assumed by comparison without data in hand.

There have been incidences of illness associated with fresh juices, and that is, juices which are not pasteurized. And that's why we're here today. We're here to discuss the problem and potential solutions.

There is the reality also that has to be recognized that there are a substantial number of individuals who prefer fresh over pasteurized juice products, whether these are fruit juices or vegetable juices. And we can debate pasteurization's effects on flavor, but even with hard data in hand, there are certain consumers that would not change their minds and would still prefer a fresh product.

Some consumers would forego juices entirely if they were processed in any way. So the issue is: Can they be made safe? And a sub-issue of that: Should all juices be pasteurized?

I can only offer my opinion as a former public health official and regulator, and now as an academic. I'll answer the questions as follows: Yes, unquestionably they can be made safe. And, no, mandatory pasteurization is not necessary.

We really need to define how safe is safe, and consider the denominator of the total volume of product that's consumed--meaning fresh juice--then consider the illness incidents that have occurred.

A frequent argument is, despite the large denominator of consumption, some illnesses may be occurring but not being detected. And that argument may be valid, but I think that argument could also be made for pasteurized or other processed foods as well. And I was reminded when I was thinking about this issue that the largest outbreak of urciniosis(?) ever in the United States was caused by a pasteurized product, and the largest outbreak of salmonellosis ever in the United States was also caused by a pasteurized product.

So pasteurization itself is not a panacea. Pasteurization can be defeated, and it can be defeated either by a high microbial load or by post-process contamination.

So can fresh juices be made safe? I believe they can. New technologies are available now, or will be shortly, and I think what we need to consider is how to fast-track some of these into being. Many of them, many of the processes of which I'm aware, will be classified in some way, shape, or form as a food additive; that is, they may become a component of the food itself. Where safety is an issue, if possible, those should be given priority and fast-track approval.

Later you will hear how Florida dealt with its problem. Now, Dr. Griffin alluded to Theme Park A outbreak, and we were painfully aware of that occurrence. But the Florida industry and the Florida state regulatory authorities got together and in a very excellent, cooperative environment, have worked through a series of regulations that everyone can live with and we believe goes a long way toward solving the problem.

Now, there is another basic truth, and that was alluded to in Dr. Griffin's talk. One of the problems is that we don't understand the ecology of the microorganisms with which we are concerned. It's only a very, very recent finding that deer and sheep are carriers of E. coli O157, and I would hazard a guess that that's not the end of the list of species that do. Clearly, we have to understand where these organisms are occurring before we can do anything about keeping them out of food.

There's another basic truth as well. Fresh juices can be made safe if the pathogens are removed before either the intact fruit or vegetable is crushed or squeezed or otherwise turned into juice. They can be made safe when attention is paid to basic sanitation principles. There is no intrinsic reason that fresh juices cannot be produced, distributed, and consumed as safe and wholesome products. So I would urge the Advisory Committee of FDA to go slow in this regard. I would define the concerns very specifically before I acted on any, and don't compare things that are not alike.

It's tempting to apply what may seem to be a quick fix, pasteurization, but it may also be unnecessary. There is no such thing as absolute safety in any product. Absolute safety, unfortunately, is a myth, and we've come to recognize that time after time.

If we can define a specific problem in a specific commodity, we can usually fix it. That's one of the basic principles of what has come to be known as HACCP, the Hazard Analysis and Critical Control Point System of assuring food safety. And few people would argue that manure and human food need to be kept apart. That's a given.

I think it would be very easy to look at this problem and say it's simple, we have a process, we can kill the microorganisms and everything will be fine. We could do that. We could easily commercially sterilize everything, and that would add even a greater margin of safety. But I don't think it would be very desirable.

I think one thing that we always need to remember is we're dealing with a very clever enemy. Dr. Buchanan alluded to that in terms of how these organisms adapt, evolve, and that's a constant process. It's been a constant process since the Earth was created. But one thing we should keep in mind is they don't have a brain and we do, and in most cases, we can figure out ways to deal with them effectively once we know the facts.

Thank you for the opportunity of addressing this group, and I'll put you back on schedule.

 Applause

DR. SHANK: Thanks, Doug.

This brings us to our first opportunity for questions. We will take questions from the floor for the speakers this morning as well as the questions from the head table or the subcommittee. Down here. Would you use the mike and would you identify the speaker to whom you're addressing the question, please?

MR. CHITTENDEN: My name is Eric Chittenden. I own the Cold Hollow Cider Mill in Waterbury Center, Vermont. I'd like to ask Dr. Archer, just for my own curiosity, he alluded to a couple of situations where in pasteurized products salmonella and other pathogens were found, if he could for me, and perhaps for others, give me that information. That was kind of interesting to me.

DR. SHANK: Patty, do you just want to come on down, too? I'm sure you're going to get questions.

DR. ARCHER: Well, in one of those situations in 1982, it was called the tristate outbreak, and it involved pasteurized milk. And the milk was fine, and the pasteurization process had worked and done exactly what it was supposed to do. But the milk bottles or the containers were put in cartons that had previously been on a pig farm, and some of the pig manure managed to drip down on the caps, by capillary action got up under, and once the milk carton was opened, it was instantly inoculated with ursinia aneracolita(?).

In the larger outbreak of salmonellosis, again, pasteurized milk was the product. This was post-pasteurization contamination within the pasteurizat ion plant itself. Salmonella had become lodged in a pipe that was very remote in the plant, and this was a plant with virtually hundreds of miles of piping. And something went wrong, and a slug of salmonella got through, survived, and caused, conservatively, about 400,000 illnesses. It can happen.

DR. OSTERHOLM: Fred?

DR. SHANK: Yes.

DR. OSTERHOLM: Mike Osterholm. Just as a point of clarification, I think it's very important for the audience to understand that we're not talking about a failure of pasteurization. We're talking about a failure of pasteurization system, and that the pasteurization based on time and temperature worked. It's just that any system--in fact, just not to correct our speaker, but the largest outbreak of salmonellosis is actually the Swan's Ice Cream outbreak, larger than the Jewel Dairy outbreak that you're referring to. In that outbreak, again, the product was pasteurized but then was subsequently hauled in trucks which also hauled raw egg product. And so, again, you could say that was a failure of pasteurization, but it was not. It was a failure of a system post-pasteurization. And I think just for the audience that's an important distinction.

DR. ARCHER: It is an important distinction, Mike, and it should be made. But it's also, I think, an important distinction to point out that when we were investigating the possible pasteurized milk outbreak involved Listeria, and when the actual laboratory studies were done, there was an incredibly narrow margin of safety within the process itself. And as you know, it's a logarithmic death of any organism when heat is applied, and it was calculated that had the load of Listeria or other heat-resistant organism been large enough to start with, pasteurization would not work.

DR. YURSTROBUS: My name is Dr. Yurstrobus  ph . I'm a consultant here in Washington. I had a question for the CDC--two questions, actually, points of clarification. It seems to me there are many different types of fruit juices--grapes, oranges, apples, grapefruit. The CDC presentation was largely on apples and oranges, but it itself noted that there were differences in the number of outbreaks, the sources of contamination, the organisms, the particular use of sanitary practices. And I guess I'd have two questions. One is I wanted to clarify whether the recommendations apply equally to all of these juices and whether the CDC has done a comparable spiking study to the Doyle study on orange juice that has been produced under the new State of Florida regulations that were installed after the Theme Park A episode.

And my second question is whether we can get a sense of what the denominator is, given the fact that a large number of people may drink juice on a daily basis in the United States, which would give us some sense of what the risk is.

DR. GRIFFIN: The first question I think was, Are the recommendations the same for other juices as what I made? I made specific recommendations for apple juice and cider, and that was to strengthen practices to assure that it was safe. And what I said for other juices is that we need to improve the safety of other juices, and I didn't recommend how that should be done. So I think that was a pretty broad recommendation.

The next question is, Did we do similar spiking studies on orange juice formulated under the new guidelines? In general, CDC doesn't do these sorts of studied unless we can't find anyone else in food microbiology who will do them. And we were fortunate in having Dr. Doyle's group, Dr. Larry Beauchat, working with us on the apple cider. And I think that people at the University of Florida, Mickey Parrish, may have done some work on orange juice, but I'm not familiar with any new studies on orange produced under the new regulations. Perhaps that will come up later in the conference.

Then as far as the denominator of people who drink juice, I don't have that sort of data. The only data that we do have is that some of you may have heard that CDC, with monetary assistance from USDA and FDA, has begun active surveillance for food-borne disease in five sites, and we have about 5 percent of the U.S. population under surveillance. And to try to understand more about food-borne illness and food consumption practices , we surveyed a random sample of people in those areas, and I looked--one of the questions we ask them is: Have you drunk apple cider in the past five days? And our data from this past March through September showed that a range of between five to nine respondents said they had drunk cider in the previous nine days. And that doesn't include data from October and November when we might expect it to be higher than 9 percent.

MR. SAPERS: My name is Gerald Sapers. I'm a food technologist with the Agricultural Research Service in Philadelphia. I'm a colleague of Bob Buchanan's. I'd like to make a comment about one remark that Dr. Buchanan made concerning low-acid tomatoes.

In the 1970s and early 1980s, we did a lot of research in this area, and one of the points that we made was that not only must we consider the inherent low acidity of some varieties, but also the fact that as the fruit became senescent, the acidity decreased. And more to the point, in the presence of certain fungal infections and areas of decay, the acidity decreased greatly so that the pH could reach neutrality in some cases.

This is relevant, I think, to the apple situation since in the micro-environment of decay areas in apples, it's quite possible that the pH may be greatly elevated. And, in fact, we do have some preliminary data showing that this is the case. And so this micro-environment might be an appropriate area where O157 might not only be able to survive, but to grow.

DR. SHANK: Thank you.

Caroline?

MS. DeWAAL: Caroline Smith DeWaal, Center for Science in the Public Interest. I have a question actually to all your panelists, but I'd be very interested in information from CDC specifically. You've said that we believe that O157:H7 survives in the manure or in the intestinal tract of cattle, deer, and sheep. I want to know specifically what is the data on survival in the live animal for deer and sheep. I'd also like to know whether CDC has done any data evaluation showing whether fields have been treated, fields which are associated with outbreaks, from O157 have been treated with cow manure or whether there's any traceback to the actual manure of deer or sheep populations. But I have heard that a lot of this data on deer specifically may be from small slaughter operations where you have cattle being slaughtered and deer being slaughtered and there may be cross-contamination issues involved.

DR. SHANK: Who wants to be first on that one? Come on up, Patty.

I would remind you, Caroline, that some of those issues may be addressed later this afternoon as we talk about the practices.

DR. GRIFFIN: You mentioned survival of O157 in cattle, deer, and sheep. I'm not sure that's what I would call it. It can colonize these animals, and when we go onto farms, we can culture the organism from the feces of these animals. We're much more likely to culture it from the young animals, the post-weaned heifers, than from the older animals, the cows. And when we go back to the same animals, we often find that their feces are negative for the organism, and when we go back again, they may be positive again. So the more you culture, the more positive animals you're likely to find on a particular farm.

The data on deer and sheep is very new data. We at CDC haven't done this ourselves, and it is in the medical literature. You can find that there are published articles on this, and I can help you find them if you have difficulty.

As far as fields associated with outbreaks being treated with cow manure, I don't think we've been successful enough in our tracebacks that we've found outbreaks and been able to trace an outbreak from a food to a particular field. However, some of the pioneering work on O157 in cattle has been done by Dale Hancock and Tom Besser in Washington State. And they've found some evidence suggesting that treating fields with a slurry of cow manure and the letting cattle go out and graze on those fields may be associated with cattle herds that are more likely to carry O157. We need a lot more work on the ecology of O157 in cattle and other animals.

DR. WACHSMUTH: I know during the orange juice outbreak we learned from the industry in Florida that this type of orange juice was growing in popularity and that that industry was expanding. It had been very small, and then it increased substantially in a very short period of time. So that's a little troublesome.

But what I'd like to ask Dr. Archer is: What is the bottom line of the Florida regulation? What was done to ensure safety? And when was that effective?

DR. SHANK: Before we take that question, Kaye, I think Bob has a follow-up to Caroline's question.

DR. BUCHANAN: Caroline, I think at this point the information on the reservoirs for the organism is one of the areas that is--we don't have enough data on it. There has been an assumption that the natural reservoir for this organism is cattle. This is based on what I would consider a very, you know, limited viewpoint that we haven't really gone out into the environment. My feeling is that cattle is either a secondary or a transitory reservoir. I don't think we've found the real one yet.

I assume that people here will be able to give more information, but with the limited areas that I've looked at in terms of apple orchards in the last few weeks, the raising of cattle and the use of cattle manure is not something that occurs in those areas. But the deer populations, everyone that I've queried about the deer populations has indicated a tremendous upswing in deer populations in the last ten years that they just run through the orchards.

Now, this again may not be the reservoir. The point is we can't look at these problems knowing that we have the solution ahead of time, because we'll be disappointed.

DR. SHANK: Doug, you had at least one or two questions there for you.

DR. ARCHER: Yes. I'd start with a comment. That is, I wouldn't be concerned about the expansion of an industry unless they're presupposing a problem with the industry. I think expanding an industry is a good thing.

The other thing I'd say is I think the regulations in Florida were enacted on--is Betsy Woodward here?--February 6, 1996? February 20th. Okay.

The bottom line of those regulations, I think it's a lot of good, common-sense sanitation. General sanitation principles applied rigorously. That's the bottom line. And I think that's really--going back to a lot of what we've heard, we're hearing about deer and cow and, you know, in the case of Cryptosporidium and Giradia, you might throw in beaver and elk and a bunch of other animals. Well, it just makes good sanitation sense. Manure and food need to be kept apart, and if it's unavoidable that they come in contact with one another, it has to be removed from the food before the food encounters human beings. And that's basic sanitation.

MR. MARTINELLI: Good morning. I'm Frank Martinelli from Orchid Island Juice Company. I'm the plant manager there, and I'd like to apologize in advance with regards to microbiological contamination and microbiological information. I am a layman. But I have a question for Dr. Griffin.

At Orchid Island Juice Company, we produced over 100 million servings of fresh squeezed orange juice and grapefruit juice and without any reported incident of food-borne pathogenic illness. And my questio n is: What level of safety are you trying to achieve? And could you put in perspective for the layman the risks that we're dealing with with regards to fresh squeezed juice as opposed to the risks a person takes when they get in their car to drive to the grocery store to get that juice?

DR. GRIFFIN: That's a good question. I wish I had the answer to it. I think I alluded to the problem when I said that we're recognizing more outbreaks from juice, but we think that sporadic cases are more common. It's very difficult for us to measure the number of illnesses from juice because it tends to be a quite widespread product and outbreaks tend not to be recognized. One of the reasons we have the new food-borne disease act of surveillance system with the five sentinel surveillance sites is to try to measure some of this sporadic disease.

So we're detecting more outbreaks from juice, but we think that those are just an indication that there are also sporadic cases. And when I said that we were looking for more outbreaks after 1991, when we had our first recognize outbreaks, it's because we've had the same questions that you have. How big a problem is this? And it's going to be difficult to measure.

When we had 55 cases of E. coli O157 infection in the city of Las Vegas associated with the 1993 big hamburger outbreak, that entire outbreak was missed in the State of Nevada. The only reason it was found is because the mother of a child who was hospitalized with kidney failure read in the newspaper about the outbreak in Washington State and said: Oh, isn't that funny? My child ate at one of those restaurants before she got sick.

So not only do we miss sporadic cases, we miss even big outbreaks. So it's very hard to measure the impact.

I think that right now it's clear that we have a problem with unpasteurized apple juice. We've had a problem with some unpasteurized orange juice made by particular methods. And I think that we need to look at the critical control processes that go into the making of juice to try to determine if juice is safe.

DR. SHANK: Burt, does your issue still hold?

MR. BARTLESON: I think the point that I wanted to make is that when we start looking at the ecology of E. coli, the more we look, the more we find. When we started looking, we looked at cows, and we found it in cows. And we discovered that a single cow could be a carrier and then not be a carrier and then carry it again later, and that you had a prevalence in herds of 1 percent and it was different cows at different times.

But when we started looking around the farm, we found that not only was it in the cows but it was in the water that the cows drank. It was in the flies that were flying around. It was in the dogs that were running around on the farm, in the deer, and it's all over the place. It's also in birds. So we can't just assume that if we control cow manure, deer manure, and sheep manure that we've got this problem licked. It's more complex than that, and there are other sources of contamination beyond just those three.

DR. OSTERHOLM: I think this question regarding what is the risk is a very important, relevant question, and there's really two parts to that. One is the numerator and one is the denominator, how many people are drinking the product, but how many cases are actually occurring. And, Fred, if you could indulge me, I just have two slides. Could I show them? You have the changer up there.

In Minnesota, where we have probably the most active--one of the more active statewide programs for E. coli O157:H7 infection--and I say that because this past two years we reported out more cases in Minnesota than in any other region of the United States combined to the CDC, and I do not believe it's because we have more incidence of disease or higher incidence. It's because really we have our laboratories looking for it.

The only point I want to make here is you can see 345 cases reported, which are really the tip of the iceberg. Most of them were summertime cases. Twenty-four cases of HUS were reported.

Fred, if you would go to the next slide, actually we'll do this quickly.

If you can focus that a little bit, you can see seasonally in 1994 we have this increase in the summer. Each red box is an isolate that our laboratory received a case report, 135.

Next slide.

In 1995, there was 182 with a big peak in the summertime. You say, well, what's going on out there? What we've actually done is taken this a step further and done a further characterization of these isolates, a clonotype analysis, where it's just not O157:H7, but now we can tell a John T. Smith from a John P. Smith from a John Q. Smith. And all E. coli O157:H7 are not the same.

Next slide, please.

If you look in 1994, that's that same graph you saw in all red, but each colored box reflects one particular strain that more than one isolate came through that was exactly that clonotype. And in those instances, you see there were 20. In 52 instances, a single white box represents a single unrelated molecular subtype, so in total here we really dealt with 72 different E. coli O157:H7s.

Next slide, for my last one.

In 1995, it was even more dramatic. What if you had looked at that big peak and you had assumed that that was all E. coli O157:H7 from one source? You have the lavender boxes; you have the pink boxes; you have the yellow boxes; you have the red boxes.

The point I want to make here is we're missing outbreaks all the time, because unless you're doing this clonotype-type work, you are truly mixing apples and oranges in your analysis. And this past summer we had an opportunity to have other states in the Midwest supply their E. coli isolates from their public health laboratory. The same pulses that went through Minnesota in the same week went through the entire upper Midwest. These are perishable food items that are coming through that are pulsing through the system, creating what we call a series of mini-epidemics, m-i-n-i epidemics, m-a-n-y epidemics. Processed juices, et cetera, clearly could be part of this.

Just as a final note to add here, in the outbreak with Odwalla, note that northern California was one of the five centers that Patty referred to for food-borne disease surveillance, yet that was not picked up in northern California despite the distribution of the product there in cases. It was picked up in Seattle-King County because they are one of the few places that have the combination, clinical expertise combined with this type of further identification.

So I think we have to, as Patty said, which very much supports her point, we're missing outbreaks all the time. Most of these did not come up to a recognize outbreak because we didn't have the resources to go and do additional follow-up, we had so few cases, that we could be having juice outbreaks or fresh fruit and vegetable outbreaks, or whatever, all the time. And these are not long-term kind of products in the market, as you can see. Many of these come through literally in a one- or two-week period, and they're gone.

Just as a final note to make on that, in Minnesota, just to give you a sense, the Swan's Ice Cream outbreak, which I referred to earlier, because of the fact that this ice cream was a home-delivered product, we had a very good opportunity to actually learn how many people had consumed it and got sick, because we could go back home by home by home. In Minnesota, based on the consumption and the number of illnesses, we estimate 30,000 cases conservatively occurred in Minnesota. In 1994, this was the number one Minnesota-based news story as determined by AP. So it was not that people were not aware of this. Yet out of those 30,000, only 151 cases were medically detected and reported to us, for less than 0.5 percent. So even in these boxes up here, every time you hear about 100 cases or 200 cases or 300 cases, remember the multiplier effect and realize what's coming through our system.

DR. SHANK: Thanks, Mike.

Yes?

MR. WOOD: I'm Richard Wood with Food Animal Concerns Trust. I was wondering if some of our presenters might give a rather subjective assessment in terms of how they view the level of research that's focused on O157:H7 in determining its reservoir and outbreak situations and related to some of the other questions. In terms of dollars and the amount of scientific literature that's out there, how does it compare to research on some of the other food-borne pathogens? It's my understanding th at the dollars haven't been there as they should be, and I'd like some comment on that, if possible.

DR. SHANK: Dr. Buchanan, do you want to take a stab at that one?

DR. BUCHANAN: Well, I can give you the party line. It's never enough.

We see two things going on here. The amount of information we need about O157 is very large, and it goes from very basic molecular biology all the way up through the ecology of it in the real environment. We have people doing little pieces of it all over the country. There are funds available for O157. It's never enough. I mean, the speed at which you achieve is directly dependent on the dollars you put in it. It is also one of the more important problems, but it's not the only microbiological concern. There are a number of them that are working through.

We also have the reality that there aren't many research food microbiologists in the country, and so there's only so much you can do. And, finally, field work is a real pain. Going around collecting deer feces is not the most exciting research, nor is it the most pleasant to do. But it's still one of those things that needs to be done.

DR. ARCHER: I'll take a crack at that from a different aspect. Being with a university that is looking for funding, having people out in the field scooping up feces, I think that's why the Lord in his wisdom invented graduate students.

 Laughter.

DR. ARCHER: So the labor force is there. But I would hazard a guess that if I applied for a grant from the Federal Government to go and try and methodically look at what the fecal carriage rate of E. coli was in various species, and that was next to a grant that was looking at the basic molecular biology of E. coli O157, I think I could hazard a guess which one would get funded and which one wouldn't. And that's just the reality of the times. The glitzier stuff seems to attract the money. Unfortunately, it's the grunt work that probably has more importance.

DR. BUCHANAN: And I would concur with that. If it's not sexy, it doesn't get funded, even though it may be needed.

DR. GRIFFIN: I agree with both of the previous speakers. In 1987, five years after the first O157 outbreak--and in that first outbreak O157 was cultured from the ground beef that caused the outbreak. In 1987, CDC sent EIS officers to do the first survey of cattle herds for O157 because it hadn't been done yet. And they were investigating an outbreak, and they really felt that the work needed to be done, and that was the study that showed that O157 can colonize many cattle and is more easily isolated from young calves than from older animals.

I think that the work being done now in Washington State, as I mentioned before, is very important, and we would very much like to see more epidemiologic studies on farms of the risk factors for some animals carrying O157 and other animals not carrying it, for some farms having animals with O157 and other farms not having animals with O157. We'd like to see epidemiologic studies on slaughter plants that look at the risk of contamination of meat during slaughter and processing, comparing the practices in plants that produce meat that's contaminated with microorganisms to those of plants that produce meat that is less heavily contaminated.

DR. SHANK: Dr. Potter?

DR. POTTER: Thank you, Fred.

A question for Dr. Archer. You stressed the differences between juices and fruits and vegetables from which they're derived and the differences among the various kinds of fruits and vegetables. What do you think the important characteristics are that would let us group those into relevant risk factors for focused control strategies? And are there control strategies that transcend those differences that could be generally applied? You seem to have a dim view of the public health benefits of pasteurization, and I wonder if you have other suggestions.

DR. ARCHER: I don't have a dim view of pasteurization. Pasteurization has saved countless lives over the past decades. But I do have a dim view of applying something if it's not needed. And I go back to the gentleman who asked the question before, producing 100 million units of the product and we've never heard a complaint, and the answer was, well, it's probably sporadic cases. I mean, I'd worry a whole lot more about the status quo of the water supply in the United States and the sporadic illnesses caused by that than anything else.

Now, to get back to your question, Morry, I think it, again, just becomes common sense, and in the terms of microbiology, I'd be worried about juices in which organisms could grow. I mean, all of life, microbiology included, is a numbers game. Even here where we're working with very, very infectious organisms, it's still a matter of a sufficient number of them have to get into the human being to cause an illness. That number may be very low in some individuals, but, nevertheless, if the population can expand, the risk increases. And so those juices--well, first of all, the epidemiology ought to go some distance in telling the story. If there's a documented problem, yes, attention should be focused on that.

Where there is no problem, and where I get a little nervous, is when we extrapolate between the juices and, okay, we've seen some information that fresh apple cider there have been some problems, and now we're jumping off into, say, grape juice because it's a juice.

I don't think that quite follows. I think we need to know a little bit more before we propose a panacea solution to all juices.

DR. POTTER: Doug, before you leave, could you expand on that just a little bit? If the fact that they're both fruit juices isn't the relevant characteristic, what might we look for so that we don't have to wait for an outbreak per product? Is there some way that we can say these juices have relevant characteristics that are similar and, therefore, we can view them as a group? Or do, in fact, we have to wait for an outbreak per product?

DR. ARCHER: Historically, that's been how it's been done, if I'm not mistaken. Unfortunately, we have had to wait for the outbreak and then figure out why it happened.

Can we do a little bit better job of prospective thinking? Probably we can. Again, going back to a common theme here, it seems to be manure. If you keep the product away from manure, it seems that it will be a little bit safer. Well, then, possibly food products that aren't going to come into contact with manure or have a very low probability might be ones you might not want to focus on.

But it's still--I mean, look at the number of causative organisms, and there's going to be some kind of a chain of events where they get into a food regardless. I mean, we've had hepatitis A outbreaks on fruits and vegetables, and, you know, it happens sometimes. Those are things that should be avoidable, but when people are involved, mistakes get made on occasion.

DR. SHANK: Bob?

DR. BUCHANAN: There's a variety of factors that have the potential for influencing it, but if we're dealing with a product that is relying on acidity to inactivate the pathogen, as in an acidic fruit juice, the key issues are: What is the pH, and what is the acid that makes up the major component within that fruit juice? So, for example, comparing apples and oranges is not appropriate is in reality true because they have two different acidulants. One is malic acid and one is citric acid. You also have to look at the characteristic of the fruit itself, its potential for internalizing a pathogen, the potential, for example, of taking an organism in the blossom end in the two respective fruits.

So there's a variety of things. I think we can do it in a more systematic manner, but that's just a couple of the areas.

DR. OSTERHOLM: This is a question really for Doug or anyone else in the group here, but we're concentrating on manure, but I think we also had an experience this past year or past two years with cyclospora and the issue there. And for those of us that have the belief that eimeria of birds and cyclospora in humans are the same bug, clearly bird droppings become a very important role. Anytime you're dealing with fruits or vegetables, you've got bird droppings also involved and the potential for contamination of the product that might go into juices. Could somebody comment on the issue of in addition to cow manure or any other animal manure but also bird droppings, which would not be a necessarily a ground-related issue?

DR. SHANK: I don't see any volunteers at this point. Kaye?

DR. WACHSMUTH: This goes back to the research. Just for public information, we have a multi-agency food safety research group formed, mainly at the behest of Mike Taylor, who is the Acting Under Secretary for Food Safety at USDA. That group did identify areas that we're raising here, the ecology and other things for all of the enteric pathogens, but also put some emphasis on the other enterohemorrhagic E. coli other than O157:H7 since we have no idea about the infectious dose. We do know there have been some outbreaks in other countries. So we're certainly not ignoring those things, and FSIS will take this agenda to ARS, which is a very large research service, with millions of dollars, and we hope to put some more of that money towards these very things, the ecology and some of the human incidence, even, if we can influence institutions like CDC and NIH.

DR. SHANK: Burt? And then our public commentor.

MR. BARTLESON: I think that what Dr. Osterholm showed us is happening all the time. We're seeing little tiny pulses of disease that are so small that because we only have one state doing the work, we can't get enough cases to do a case control study. And, you know, we looked at Odwalla, the outbreak, pretty heavily and what we discovered was that we had another little six-person outbreak from 300 gallons of juice that was pressed at a church dinner.

Now, if we hadn't been enhancing our surveillance and collecting more stool cultures, we wouldn't have found that at all. And I think that's what's happening all the time. If you put the money in the laboratory and you put the money in the research, you're going to find there are more problems.

The second thing I wanted to mention is I don't think that as epidemiologists in local health departments sometimes we ask the right questions. We look at two days of food history, and with low-dose organisms like E. coli, the incubation period may be seven days or longer. And so we may be totally missing the sources.

The other thing is, of course, people's memory isn't very good, so you've got to have enough people to actually figure out the outbreak.

MR. WILSON: My name is Dan Wilson. I run an apple orchard and a small-scale fresh cider operation in upstate New York. I have a question for Dr. Osterholm. I was interested to see your slides, and I wanted to find out a little bit more of the background information on what caused these outbreaks, because I understand--or I saw the peak, the very strong peak in July and August for outbreaks, but I know that in the upper Midwest apple season and, therefore, the bulk of the processing of apples doesn't occur until mid-September through November.

DR. OSTERHOLM: First of al, to answer your question, it's unknown. That's a lack of resources. We're fortunate to have this much done. We're able to pick up some of the bigger outbreaks, and I think what you heard just now from Burt was a good example. If we have three or four or five cases, often we don't have enough information to be able to pursue this further in a way that would allow us to understand that. So that's an area that's very deficit in this country and one that we all need to invest in, because without good surveillance data, you know, it's kind of like trying to run the O'Hare air traffic control tower with tin cans and string. You know, we're not very effective. And I think that's what we're kind of doing right now in this country, and we need better data that way to answer your question very specifically.

Seasonality-wise, we agree with you. There are some very interesting patterns there, but I think again the point is that just as those may not be fruit and vegetable outbreaks associated with your product or whatever, clearly they are occurring in the fall and we're missing even those. So as you just heard, the Washington outbreak with the church group in the pressed apples would never even have been picked up if the Odwalla outbreak hadn't occurred. We think we're missing them big-time in Minnesota, and if we're missing them, I can tell you the rest of the country is really missing them. And that I think is the key issue. Please don't come in here and say we've got nine outbreaks on record and that's all we have. That is true, but that would really defy the point, I think, of what is the true burden of disease and what's associated with what, and that's what we really have an obligation to find out.

DR. SHANK: Thanks, Mike.

Let me ask you to hold until a little bit later. We're eating into our break.

Let's take a quick break at this point and be back in 15 minutes, please.

 Recess.

DR. SHANK: According to my watch, it's 10:30. So if you could come on back in and take your seats, we'll get started.

Let me have your attention. I think that in view of the time we need to get started. We do have a full schedule, and we need to keep as close to schedule as we can.

You will notice that we have another person up front here with us since the break. Mr. Jerry Mande has joined us. Jerry is the Executive Assistant to the Commissioner, and so we welcome him to the front.

At this point we're on the program to review what has been called the Odwalla situation. We have two presenters. Your program indicates one. But we have Stephen Williamson, who is the chief executive officer for Odwalla. We also have Greg Steltenpohl, who is the founder and chairman of Odwalla. And so I would ask both of those gentlemen to come forward, and we'll proceed with those presentations at this time.

MR. WILLIAMSON: Hello. My name is Stephen Williamson, and I am the CEO of Odwalla. I would like to thank the FDA, the CDC, and all the state and local authorities for helping us through this crisis. I am very happy to be here today to share our story with you, and there are five things I want to talk about today: first, who we are, give you a quick background of who we are; two, how we responded to this crisis; three, how this experience has changed us; four, give you an overview of the HACCP plan we want to talk about; and, five, talk about consumer choice.

To start with, who we are. Slide No. 1, if we could. Thank you.

To understand us, you actually need to understand our vision statement. Key words in here are soil to soul, people to planet, nourishing the body whole--probably not a typical Washington, D.C. vision statement, but this is--

 Laughter.

MR. WILLIAMSON: --the fundamentals of our company. We believe we are a nourishment company, and that makes what we've gone through that much more difficult.

Okay. Patty, you were better at this than me. How do you work this thing?

 Pause

MR. WILLIAMSON: Core values, a couple things are important to understand here. Honesty, integrity, respect, leadership through creativity, innovation, example. These are key to us, and they were very important for us in handling this crisis.

Next slide, please.

A couple key facts about us. Last year we did $60 million in sales, $633,000 in net income. We have just under 600 employees, just under 200 routes, 4,600 accounts. We do all our own DSD. We did produce 37 million units last year. Since the inception of the company, we estimate we've had several hundred million servings. We have 24 different flavors and 80 different units we make. Our distribution system is shown here through seven states. We have 22 distribution centers.

Next slide, please.

All our own fleet, and we make deliveries seven days a week to our accounts. So I've given you a brief sense of who we are. What I'd like to do now is talk about the recall and how we handled that.

Next slide, please.

As I mentioned earlier, our crisis communication plan is totally based off of our vision and our core values. When this call came from Carl Osaki  ph in Washington State, we simply asked them, What do you need to know and what can we do? And what they said, very clearly, was it was an apple-related problem, apple juice, apple juice. And we said, well, do you understand that we also have apple juice in a lot of our blends? Immediately we said we'd recall all our apple juice, all our apple-related--even though epidemiologically they had said to us at this point it was only apple juice, we were not comfortable just pulling apple juice. If there's a problem in our apple juice, we thought it would probably be in our blends as well. So we immediately pulled everything off the shelf.

A couple of things to think about is how far do you go, and I think that we're an exception in that we went much further than people suggested. Not only did we take apple off, we took apple-related, and one day later we took all our carrot juice off, even though there was no epidemiological evidence with carrot juice. It was made on the same line, and consumer safety is our number one priority, and, therefore, it came off.

Immediately we also reached out to all the families through the media saying we will pay all medical expenses. If you were made ill from our juice, we'll cover the costs. And we I think very quickly established a very good working relationship with the FDA and the CDC and all the state and local authorities. Randy Wykoff  ph from the FDA set up daily conference calls with us which were really important for us because we didn't know what was going on, and having this one-hour conference call every day for the first ten days was extremely helpful. We needed to know information that people wanted, like our distribution records, the accounts, anything. We were there trying to supply the information to be responsive.

Then we had investigators in our plant from 11 days from the FDA and the CDC--I guess the CDC and the state and local authorities, and our attitude there I think is important to point out, which was, hey, nothing to hide, our records are open, you're going to find something with us or before us, that's okay, let's find out the issue here. And we were totally open. Anything you want, it's there.

Next slide, please.

Obviously the impact on our company was pretty substantial. In the Pacific Northwest where most of the press focus happened, our sales dropped almost to zero. That was augmented by some miscommunications in terms of what was being recalled. But the impact was pretty dramatic. It also instantly changed our distribution system from one that puts it out to one that brings it in, which is no small task. We made 4,200 stops in a 24-hour period. We had our total recall completed in 48 hours. Eventually we took over 40 forty-eight-foot trucks full of our product and had it destroyed. It was an incredibly tough period to go through.

I also want to point out Commissioner Kessler's comments in a press conference on the Friday after the incident. He was patched into a press conference in Seattle, and he said that this is not just an Odwalla problem, this is a fresh apple juice industry problem. And we appreciate that comment, but we also felt that was a very difficult comment because we know what we do and I've been around to many, many different fresh juice plants in the country. We didn't want to put the industry on notice, but we were also in a situation where--I have three children. I have a daughter who's 20 months. She drinks mango-tango every day, fresh apple juice every day. My two other children drink tons of apple juice. I'm a father. What do you do? I mean, this obviously freaked us out, and we felt that we needed to reach out and let people know this was an issue.

We've tried to conduct ourselves in a way that we think is both full of integrity and also trying to be sympathetic for all the other people in the business. And it's been no easy path. It seems either way we go we end up in a tough situation.

Next slide, please.

What I'd like to do now is take you through a little bit of how this experience has changed us, and what I want to do is start to give you a little bit of background about our plant.

We're really proud of our production facility. We're proud of what we've accomplished, and this has clearly been an incredible blow to us. A couple of facts here.

This year we were budgeted to process 155 million pounds of fruit, over 50 different ingredients. Orange, carrot, and apple are our largest poundage. It's approximately 10 million gallons of finished product budgeted for this year. Our production facility opened in August of '93. It's a 95,000-square-foot facility on 50 acres located in the Central Valley in California. We have approximately 175 employees, and we run six days a week.

Next slide, please.

This gives you a very simple line drawing of how our plant works. We have two primary juice-making operations: the extraction, which is for citrus, and the press, for carrot and apple. From there it goes into storage tanks where it's held for a matter of minutes or hours, and it goes into the mix and blend room where we add our purees, our water, our dried powders. It's mixed, it's blended, it's bottled, and it's put into our finished goods holding place where it's picked by route, shipped out. Approximately 60 forty-eight-foot trucks a week go out to our 22 distribution centers.

Next slide, please.

This is, again, a simple line drawing of our old press line. When fruit comes in, it's inspected. If it makes it past the inspection, it goes into storage where it's held for a matter of hours or days, depending on our volume. It's dumped on to a table. It's culled. After culling, it goes through a Van Mark scrubber where it is scrubbed, and it is sprayed with acid detergent. It's rinsed. It was ground, pumped, and filtered, and bottled, and we can go through--the next slide, please--to give you a little overview of the operation as it was prior to this.

It doesn't seem like that slide is focused. Can you see that? Is that focused?

VOICES: No.

MR. WILLIAMSON: Could we focus back there?

Okay. Well, anyway, I will keep going. What I'm trying to point out here is this is the old facility, and you'll see the sorting going on there, two people sorting. After that it goes into--this is a Van Mark scrubber. It's rated for approximately, I think, 24,000 pounds per hour. We were running it less than 50 percent of that, and in there was where the acid detergent was sprayed. It goes through this wall, and on the other side of this wall is our extraction operation.

Next slide, please.

Apples going through the sorting.

Next slide, please.

And here inside is the extraction room, our Hammermill grinder, and then the press.

Next slide, please.

Here is a recent survey from the U.S. Apple Associate that came out post this incident, and I think it's important to point out here, if you look at this, we think we answered this to the highest order. We do wash, we do brush. We use a Van Mark peeler-scrubber rated for apples at well below its capacity. We do not recirculate water. We cull before. We visually inspect, and we attempt to use no bruised and decay cuts or punctured fruit.

Next slide, please.

Just a quick overview of some of the sanitation and clean-in-place systems. Our CIP system is a six-step process. It's very thorough. Our equipment surface cleaning operation is a four-step process, and it has real-time verification. I'm going to talk mor e about that in a moment. And our in-process sanitation happens if our lines stop for more than ten minutes, and it's a three-step process.

Next slide, please.

One of the things that we use and we're very happy with and proud of--and, again, I want to make sure people understand I'm not a scientist, nor am I a lawyer, so I'm not necessarily good at answering lots of questions. But we used an IDEXX luminometer every day to help us, because one of the things we think in sanitation is real time is very important. If you do your sanitization process and you don't get your results for 24 hours, well, you've already run another day of production. So we used this tool to get instantaneous feedback on were we clean or were we dirty, and if we're dirty, we go through the process again. And we found it be helpful.

Next slide, please.

Our process controls and GMPs included these categories for pathogens, product less than 4.3. We had a rodent plan, an insect plan, spoilage and careful control of the cold chain for materials and finished goods, sanitation plan, and our product quality controls included--every batch was Aston Brix(?) pH tested. We did yeast and mold and aerobic plate counts on all our single-strength products and a rotation of our blends. We tracked consumer complaints very carefully, and our low-acid product is where we focused.

What we did is we looked at what we thought was the most dangerous product, which was--or the most potential for concern, which was vegetable cocktail. It was 70 percent carrot juice and the other 30 percent things like beets and celery, where we thought we had the highest risk. So we did testing for coliforms and E. coli on that every time it was made.

Next slide, please.

This is a slide showing our results over the last three years of our aerobic plate counts in apple juice. I guess the thing that's been so difficult for us about this is we really thought we had made a tremendous amount of progress, and we've taken our plate counts down from over 100,000 to 12,000, and we thought we were getting better. We had been continually investing in our plant. We put over $10 million into our plant. We've continually invested in a management team, and when this hit, it was clearly an incredible shock.

Next slide, please.

When we got the call, we reached out for help, and I really want to tell people, not that you ever want to go through this experience, but the FDA was incredibly helpful in putting us in contact with people. So was the CDC. Patricia Griffin was very helpful in getting me to Mike Doyle. John Vanderveen was very helpful in getting me to people that were helpful to try to help us understand what was going on here. We clearly needed experts.

Michael Doyle has been to our plant, Dr. Michael Doyle, and so was Dr. Nick Nickelson. They've been very involved in helping us re-emerge. The National Food Lab, and Sherry McIntyre from the National Food Lab is here, and we have results we want to share with you. I'm not going to put them on the screen, but anyone who wants them, we did do a series of thermal death time studies on apple juice. We took five strains of E. coli. We looked for the most heat-resistant and the one that thrived most i n apple juice, and we have the equation here. We'd like to share it with anyone who wants it. We were surprised that there is actually--that was the only research--it didn't exist. We had to go do it. And we want to share that with you today. Anyone who wants that, please come up and see me or Sherry McIntyre. We said that's public information and it should be out there.

Next slide, please.

I think a number of people have mentioned this, and I think this is important to point out. Apples, carrots, and oranges are different, and they have different things you need to consider: the sourcing, the shape, the ease of cleaning, and the method of processing. Each one needs to be considered. When we went about trying to emerge from this, they were very influential in how we emerged. Carrots are a fundamentally different industry than apples. With the advent of the baby peeled, the carrot industry has changed a tremendous amount. And I think you've got to consider all these things when you're trying to figure out how to make fresh juice safer, which I think this is what this is about.

Next slide, please.

I'm going to give you a quick overview of some of the changes we've made, some of the learnings from this experience. We were actually in the process of putting in an apple and citrus receiving station that's outside of the plant, the physical building, prior to this happening, but it went on line right after this. And what we're trying to do here is build more lines of defense, receive fruit outside the plant, wash it with a neutral wash, a neutral detergent, rinse it, do a primary sorting, put it into cleaned intra-company bins, plastic bins that haven't been outside the plant, and from there go into fruit storage.

Next slide, please.

This is a little overview of it.

Next slide.

An orange bulk truck being unloaded.

Next slide.

Next slide.

Sorting operation.

Next slide.

The loading into the bins.

Next slide, please.

So our new apple press line. When this happened, we did determine that we should put in more lines of defense, and we did make some changes quickly. The major changes are we switched from a Van Mark brush-washer--Van Mark peeler-scrubber to a flat table brush-washer. We spray a sanitizer now of chlorine solution, and we have put in a flash pasteurization system. So let's go through that quickly. I can give you an overview.

Next slide, please.

So this is a new table for brush-washing the apples, and you can see the Van Mark scrubbers in the background. Here's our sorting going. Remember, this is the second time it's been sorted at this point.

Next slide, please.

Another thing to point out, I'm sure that many of these people know about this, many don't, but, again, our belief is let's share standards to make things work better. These are undulating rolls where the fruit goes over, and what happens here is, if you look at fruit when it goes through a roller, it tends to just go end over end and continue. You don't get the top and the bottom, and this system here allows for a lot of rotation so you clean the top and the bottom, which tends to be where the most dirt ends up. That's an innovation that we came up with. I don't know how many other people use it, but it's been helpful for us in trying to get as much agitation in the process.

Next slide, please.

This is our flash pasteurization system we put in. Again, I referenced the thermal death time studies that we did with the National Food Lab. We've come up with a process where we extremely quickly raise the temperature, hold it for a very short period of time. Right now we're running it with many, many D-kill. There's my first use of some scientific words. And we drop it down extremely quickly.

Now, we went through a long series of taste testings to understand, you know, why do this. It's expensive, it's a process that we think retains a lot of the flavor. Curiously, there's not a lot of research that's been done on the effects of heat and the benefits of freshness. That's something my partner will talk more about.

Next slide, please.

Just a quick overview of our citrus operation. Again, it's received outside. It goes through one washing, one cull. It's put into our bins. It comes into the storage. When it's taken out of storage and put on to the operation, it goes through two scrubbings, two separate scrubbings with an acid detergent, a secondary sorting, a rinse, a sanitizing, and I'll take you through the slides.

Next slide, please.

So here we are, there are two different tables here for the sorting, Table 1, Table 2, and then you've got--for scrubbing, excuse me, washing and s crubbing where the acid detergent is sprayed. And then you've got your culling, and then it goes through the wall into the extraction room.

Next slide, please.

Inside our extraction room, we use FMC machines. We recently switched during the summer from another extraction method to the FMC method because we find this method gets much less juice contact with the skin. That's a very important consideration. And we have found our plate counts dropped accordingly.

Next slide, please.

So I said I would talk about five things: who we are, how we handled the crisis, how this experience has changed us, and I want to now talk about the fourth thing, which is HACCP. Pretty early on in this process it became clear to us in working with Dr. Doyle, Dr. Nickelson, talking to John Kvenberg from the FDA and John Vanderveen, that HACCP was going to need to be part of our operation. And so we are in the process of developing a complete HACCP plan. We're initially focusing on the microbial hazards. I'm going to show you in a moment our apple and citrus microbial HACCP plan containing the four critical control points.

I want to reiterate I think you need to consider each fruit and each vegetable separately. They all have their own characteristics. You know, when you make apple juice, you take the skin, you pulverize it, and that's part of the juice. It's hard to separate the skin from the apple. That's not the case in a carrot. A carrot is much more scrubbable. You can polish it, you can scrub it. The baby peeled actually is big carrots that are scrubbed down. That's where our primary source of ingredients comes from for carrots. And then I think it's also important HACCP is not the salvation in and of itself. It's got to be built off a system of good sanitation SOPs, good manufacturing processes, and process controls.

Next slide, please.

Here is our HACCP plan for apple juice. There are four points. We are in the process of getting all points in place. I think we're learning a lot. Point No. 2 on fruit hold and test, one of the things that we've learned which is very difficult is it is hard to test for this pathogen, and there's a lot of debate on whether Point No. 2 is going to be a worthwhile step. But we feel we should at least try it and pursue it, and we are working with Dupont and a few others from Michael Doyle's group on what actual testing methods we should use. We're in the process of experimenting with two at this point in time.

You can see in Point No. 3, the critical control Point No. 3, we believe it's not just flash pasteurization here. It is washing and scrubbing, sanitizing, the culling. All of that plays into this, and then finally, flash pasteurization. Then, of course, downstream from that, as a number of people have mentioned, you've got to make sure you don't reinfect.

Next slide, please.

I think the thing to point out here is on citrus you'll see on Point 3 we think fruit processing is fundamentally different in citrus versus apple, and, therefore, we do not at this point in time think at all that flash pasteurization is right for citrus. And I want to just go back and make it clear on apple. We don't think flash pasteurization is the answer for apple. It's what we have found so far to be a line of defense we're comfortable with, but there are other promising technologies. We've looked at pressure systems. We've looked at light systems. And we believe we will find something that works better and is fundamentally in line with our view of soil to soul and being a nourishment company. But at this point in time, flash pasteurization is in there. But if we find something else and we don't want to use it, we'll use something else. It's part of, again, the HACCP system.

The last slide, please.

So I'm going to turn it over in a moment to my partner, Greg Steltenpohl, the founder and chairman. What I do want to just close on are a couple of things I think are important. Safe, fresh juice is the issue. We need to create standards to ensure safety. We have been most impressed with Florida in how they deal with the citrus industry. We have looked to them to get good information and help us in our progress. Florida has been very responsive in moving forward.

We want to create a forum to share learnings, and as I said, we'll share our HACCP plan with anyone. We'll share all of our research on apple that we did at the National Food Lab. We think more research needs to be conducted, and again, I want to reiterate. You've got to understand the difference between a carrot, an apple, and an orange, and all fresh fruits and vegetables. We embrace HACCP. We're going to complete the chemical and the foreign object or physical side of the HACCP shortly, and we want to reiterate we have only adopted flash pasteurization as one of the lines of defense in our HACCP, and we absolutely believe in consumer choice.

At this point I'd like to turn it over to my partner Greg.

 Applause

MR. STELTENPOHL: Thank you, Stephen. I also would like to thank the FDA for allowing us to come and present, and to members of the CDC and state and local health agencies that have really been extremely respectful of our company during this whole crisis, and we sincerely appreciate that.

When our company was founded 16 years ago, I had the ideals of consumer health and wellness foremost in mind. And so as you can probably imagine, the irony of having this crisis affect a company such as Odwalla has made a deep and lasting impression. And we continue at this point to feel greatly saddened by the fact that there have been young children affected, and our thoughts and hearts continue to go out to those people and their families.

But in moving forward, we have to understand what we can do, and I come here today to really pose some questions to this group. And so in the spirit of inquiry and sharing, I'd like to share with you a few major dilemmas that we think are facing all of us.

Along with the emotions that we felt towards the victims, a lot of our organization has also felt proud, and I think it's important to mention that even before this incident our company had a history of cooperation, working with the FDA, the FDA had been in our plant as recently as two months before the outbreak, and we have always had an open record policy.

When this happened, we did several things that I would just like to add to that Stephen mentioned. In addition to immediately executing the recall prior to the bacteriological evidence, we also stopped making apple juice until just a little over two weeks ago when we introduced the flash pasteurization. We also to this date are not continuing fresh pressed carrot juice, and that's because we feel not all the information is in and that we hope to gain knowledge from this conference, in particular to help us put a better HACCP in for carrot juice.

By putting the public health first, we think that there's an important lesson for all companies here and that we can testify to the amount of overwhelming consumer response that has happened as a result of the way our company handled this crisis, and we think it's important to note literally thousands of contacts have been made to our company by consumers reinforcing the way it was conducted, our early recall, our proactivity, and, in fact, in a consumer study that contacted over 2,500 consumers conducted by Werthlin worldwide, we received a 96 percent approval rating of the company, and in many cases an increase in consumer perception about the brand value.

But this whole incident because a real life expression for what happens to a company with embedded core values, and we believe that this should be hopefully a strong signal to other companies that may encounter this type of situation in the future.

While we can be proud of our actions, you have heard from Stephen's presentation that we focused very strongly on internal activities that could eliminate or reduce this risk in the future. But these fundamental questions that have been raised by the introduction of HACCP and by the question of pasteurization is something I'd like to raise to all of you.

The first question I pose is: Can this group in this room, which is representing all the different factions from consumers to industry, the state and local agencies, the Federal Government, can we act much more proactively and not reactively to situations such as outbreaks? For example, today in Patty Griffin's talk, our company was interested to note that there were several other outbreaks occurring concurrently with ours that we were unaware of, and even though our company has been in the spotlight of the media, we were not informed of these outbreaks. We were informed of ones that happened prior, but this industry can benefit greatly from a synchronization and a connectivity of its operating groups. This way we can help cut off emerging hazards.

As one avenue, we would propose to catalyze the formation of a fresh juice industry trade association. We feel this could streamline the communications and create regular industry review of new technologies, best practices, and successful hazard reduction programs, possibly look at labeling issues as well.

We believe that HACCP programs must be individualized by those fruit industries that know how to handle and know the particular idiosyncratic features of those fruits or vegetables.

A clear precedent has been set by the seafood industry for successful proactive HACCP implementation together with the FDA. Based on what we're learning, we feel that mandatory pasteurization would be a premature and unnecessary step in light of the vast new technologies that are emerging. And I can assure you, since our company is located next to Silicon Valley, the amount of contact we've had from universities and from inventors and from research organizations has literally been overwhelming.

We believe there are a few promising technologies; in particular, the University of California at Berkeley and Lawrence Livermore Laboratories have several technologies that are ready for industrial application, and there are many more.

Another challenge and a secondary question is: How can we best understand the public health benefits of this industry that we are engaged in? Before pasteurization is considered as a final step, we believe that the nutritional benefits and, thus, the public health benefits of fresh juices are poorly understood. We believe more research is necessary. What are their specific nutritional attributes? Can we develop standards of nutritional identity? Early studies are promising. Our work with Corning Hazleton and National Food Laboratories leads in this direction, but we think clinical studies are the next step. There are several major universities that are interested in working with our company on large-scale population studies as well.

Additionally, because of the quantifiable nutrient levels present in fresh juices, they represent a unique public health opportunity to measure the effect of fresh foods because the dosages are clear and because you can have quantifiable dosages in a bottle that's definable.

Finally, and most important, is the question of balance. With public health foremost in mind, can we provide increasing degrees of food safety while protecting the nutritional and value and integrity of that same food supply? We must respond to the nation's changing health needs, to be sure, and as Dr. Kessler has pointed out, there is no better strategy than prevention. The number one cause of preventable death in the United States is poor diet.

We are hearing from thousands of Americans ranging from young children to consuming adults, and we would like to say that the fact is the story of Odwalla's history and, in fact, the outcome of our company was not determined by marketing calculation or advertising expenditures. The hundreds of millions of units that Stephen mentioned and the 37 million units that we achieved in distribution last year has been achieved by supplying a product that the public wants. That fact cannot be understated.

So I'd like to propose to this group that we could face the challenge for a win-win public health situation. So let's not remove the very same nutrients that are most needed to protect the immune systems of healthy people in this country in today's environment. So then we can move forward and hopefully have a healthy informed choice in the hands of the public.

Thank you.

 Applause

DR. SHANK: Thank you, Greg.

This brings us to the portion of our program where we're going to start looking at FDA findings. The first topic is the facility inspection and follow-up. Mr. Frank Scholl, who is Director of Investigations from our San Francisco District Office, will make that presentation. Frank?

DR. SCHOLL: I'll make this relatively brief since we're running behind schedule.

We first found out about this problem on October 30th. I got a call from our people in Seattle District in the late morning sometime. At that point there was the epidemiological link to apple juice, Odwalla apple juice, but we had no codes. We had a set of onset times, and we were starting from that point. We got somebody into the plant that afternoon. The state food and drug branch also got a couple people in that afternoon. At that point we were concentrating on collecting samples.

Because of the nature of this product, as someone mentioned earlier, there isn't a lot of--most apple juice is consumed fairly rapidly from the point it's purchased. We had no product containers or product left at that point. We collected about 11 samples. At the same time Seattle District went out to the Odwalla distribution centers and collected about a dozen samples. Those were put under analysis immediately.

That night I went home and thought I was done for the day, but Commissioner Kessler set up a conference call that started at 6:30 Pacific time and ran until past midnight. I don't have any idea how many people were involved in that call. They came and went. At one point about a dozen people from Odwalla were tied into it.

Initially, the health people from King County in Washington, CDC, Patty Griffin, the Commissioner, and our people from CFSAN discussed the epi information, made a determination that it was sound, and that we should go with the assumption that the apple juice was the causative agent, brought the firm on at about that point. I think it was an hour-and-a-half into the call, or two, explained the situation to them. They responded by agreeing to recall their apple products. The apple juices and their blended products, all in all I believe it's 13--13 different of their 24 items they have.

Subsequently, as Stephen mentioned, they pulled the carrot juice off