NUTRINET October 1999 Volume 13 Number 5 A bulletin of nutrition, food, and health information Editor: Sybil Woutat 145 FScN, 1334 Eckles Avenue St. Paul, MN 55108 A service of the Department of Food Science and Nutrition University of Minnesota Extension Service WEIGHTY ISSUES Craig Hassel APPROPRIATE USE OF MEDICINAL HERBS Craig Hassel IMPROVING CALCIUM INTAKE - FSNEP1998 PROJECT REPORT Marla Reicks ABOUT MEAT: COST OF MEAT R.J. Epley CENTENNIAL CUISINE: Mary Darling Regulating cooking temperature; an essential talent World War I food rationing: win the war by saving food! World War II rationing: food for defense! Food for freedom! EVALUATION OF RISKS RELATED TO MICROBIOLOGICAL CONTAMINATION OF READY-TO-EAT FOOD BY FOOD PREPARATION WORKERS AND THE EFFECTIVENESS OF INTERVENTIONS TO MINIMIZE THOSE RISKS Bill Schafer White paper, section one: a literature review pertain ing to foodborne disease outbreaks caused by food work ers, 1975-1998 White paper, section two: interventions to prevent or minimize risks associated with bare-hand contact with ready-to-eat foods WEIGHTY ISSUES Craig Hassel Earlier this month, two well-publicized reports on obesity came out in the October 7 issue of New England Journal of Medicine. Both found that heavier women and men had roughly 50 to 100 percent higher mortality than lighter individuals. In this case, "heavy" means a body-mass index (the weight in kilograms divided by the square of the height in meters) of about 30 or more (greater than or equal to 174 lb for a woman 5'4" tall and greater than or equal to 208 lb for a man 5'10"tall), and lighter means a body-mass index below about 25 (145 lb for a woman 5'4" tall and 174 lb for a man 5'10" tall). Currently, about one fifth of U.S. adults have a body-mass index of 30 or more. Both studies used self-reported, rather than measured, height and weight. One view among scientists maintains that this method introduces no meaningful bias into the findings, whereas other scientists see studies using self-reported weight as flawed. Regardless, these studies do highlight the need to prevent obesity in the population. However, they do not address the issue of treating obesity. That is, they do not add to our information on the benefits or consequences of weight loss among overweight individuals. Also this month, the Star Tribune featured an article on the weight loss product "Metabolife." This product has been around for a while as you probably know. It contains guarana seed as a source of caffeine (guarana tea is drunk in Brazil as a stimulant), Ma Huang as a source of ephedrine (a potent stimulant ), ginseng (seen generally as a tonifying herb, ginger (used often to prevent motion-sickness and nausea), vitamin E, magnesium, zinc, chromium, lecithin, and a few other odds and ends. Basically what we have here is an herbal upper. Most of the concern focuses on whether the stimulant activity provided by the Ma Huang and guarana taken in combination may be ill-advised, especially for those who may have heart problems or high blood pressure. Here is another example of how a romanticized embracing of herbal products can lead to problems. Ma Huang is traditionally used in Asian cultures for asthma and other respiratory difficulties, not as a weight loss aid. Research about 10 years ago in Europe began to look at ephedrine as a metabolic stimulant to help overweight subjects lose weight. While there are success stories with "Metabolife," any individual considering or using this product ought to proceed with caution and the guidance of a knowledgable health care professional. It seems to me to be marketed as a quick fix, which always poses the threat of inappropriate use, particularly among those who most urgently want the quick fix. The message of "healthy at every size" may be something to offer people considering this product. APPROPRIATE USE OF MEDICINAL HERBS Craig Hassel The previous example showcases what many would consider an inappropriate use of herbs, especially without taking into consideration a person's health status or history, diet or lifestyle and the many other factors that go into wise decisions around health, weight and lifestyle. Many complementary health care professionals would say that "Metabolife" is not complementary medicine at all, but rather a substitution of an herb concoction for amphetamine pharmaceuticals in a biomedical context. They would say that you cannot define complementary medicine by the tools or modalities used, but rather by the underlying conceptual framework or map which guides the use of these tools or modalities to improve health. Most often, we hear that herbs are used within a biomedical context (St. John's wort for depression, ginger for motion-sickness, etc). This represents the same substitution of an herb for a pharmaceutical product within a biomedical framework, or a biomedical use of herbs and would not be considered complementary medicine. So, by this perspective, one needs to shift the focus away from the tools or modalities and toward the underlying conceptual framework guiding the use of these tools. I think this is an extremely important point to consider as one tries to evaluate the confusing array of herbs, health care products and modalities now available in the marketplace. One of the main difficulties we face is knowledge of how to use medicinal herbs in an appropriate way. The challenge within biomedical science is to characterize appropriate use of herbs based upon the chemical composition of herbs. A brief look at the chart below gives us some insights into the distinction between pharmaceutical drug preparations and medicinal herb products. CHARACTERISTIC DRUG HERB Composition Precisely defined and Poorly defined; controlled; dozens of chemical few constituents constituents Active ingredient Usually one or two; Poorly defined; many known precisely; constituents dosage controlled, may contribute to ac concentrated source tive effects, less concentrated Whereas pharmaceutical drugs are well-defined preparations where the precise chemical composition of the product is known, medicinal herbs are complex plant systems or extracts comprised of tens or hundreds of chemical constituents. Many or most of these constituents are undefined or ill defined, making it difficult to understand how herbs work in the same way a well-defined drug works. Because of this inherent complexity, science is not yet able to attribute overall physiologic effects to a precise chemical compound or group of compounds. Other systems of medicine and health care, for example traditional Chinese medicine, do not use chemical analysis as a means to define appropriate use. Rather, appropriate use of herbs is defined through sensory qualities of herbs (organoleptic analysis) combined with careful and elaborate patient observation and diagnosis. Although many of our great-great grandparents may have had significant knowledge of plant medicinal qualities and remedies, much of this knowledge has been lost over the past century with the explosion of pharmaceutical products available. While we have gained a great deal, it seems we have also lost some knowledge and perspective. IMPROVING CALCIUM INTAKE - FSNEP1998 PROJECT REPORT Identification of learning objectives for youth for improving calcium intake. Overview: The purpose of this study was to identify learning objectives in the behavioral, cognitive, affective and skill domains for 4th-5th grade low-income youth for improving calcium intake. Surveys were developed which measured reported frequency of behaviors related to intake of calcium-rich foods, skills practiced to enable intake, attitudes and intentions, and knowledge. Surveys were administered in schools by Nutrition Education Assistants (NEAs) during the months of April, May and June 1998. Food record forms were also collected from youth to provide estimates of intake from the food groups for the previous 24 hours. The results of the surveys and food record forms were used to suggest possible learning objectives which enable or predict desired behaviors related to getting enough calcium. Staff involved in the completion of project: Campus Faculty and Graduate Student: Marla Reicks, Extension Nutritionist, Department of Food Science and Nutrition Corrie Sandeno, Nutrition Graduate Student, Department of Food Science and Nutrition University of Minnesota Extension Service Nutrition Education Assistants: Gloria Wolf, Scott County Terry Drake, St. Louis County Shirley Nelsen, Mille Lacs County Leslie Downs, Beltrami County Kelly Hall, White Earth RTC Kathy Bussert-Pleschourt, Rice County Mary Wint, Kanabec County RESULTS: Surveys were administered in schools by Nutrition Education Assistants (NEAs) during the months of April, May and June 1998. Instructions were provided and suggested procedures were explained to NEAs in person to standardize the data collection process. The survey questions were incorporated into a machine scannable form by the Office of Measurement Services, University of Minnesota. Survey data and accompanying food record forms were collected from students by 4 NEAs in schools in Beltrami, Mille Lacs, Carver and Scott Counties as well as the White Earth Reservation. The number of servings of food were coded onto the machine scannable food intakeform by the graduate student using agreed upon criteria to classify foods and servings sizes to maintain consistency and accuracy of food intake data. The Office of Measurement Services scanned the forms and data were translated into Excel files. The data were analyzed using the Statistical Analysis System (SAS) program. About half of the total number of students (N=256) were boys and half were girls. The majority were 4th grade students (64%) and 36% were 5th grade students. Most were Caucasian (85%) with some classifying themselves as multi-racial (7%) or Native American (7%). Food intake data were collected for 104 students. These children consumed an average of 3 servings from the milk, yogurt and cheese group per day. Most indicated drinking milk for breakfast, but fewer indicated drinking milk at lunch and their evening meal. Total fruit and vegetable intake averaged about 3 servings per day, well below the recommended 5 servings of fruits and vegetables per day, but within the range reported in nationally representative samples of children this age. Fruits and vegetables as part of mixed dishes and fried vegetables were included in the estimated intake. Children averaged about 5 servings per day of foods from the grain group (bread, cereal, rice, noodles, pasta, tortillas, etc.). This is also less than the recommended 6-11 servings per day. For foods that are mainly fats, oils, and sweets, children's average intake was about 2.4 servings per day which could be replacing fruits, vegetables, and grains in the daily diet. Correlation coefficients for the relationship between the responses to survey statements about calcium and intake from the milk, yogurt and cheese group were determined. Students were also divided into low and high consumers (low = those consuming < 3 servings from the milk, yogurt and cheese group per day and high = those consuming > 3 servings a day). About half of the children were in each of the low and high consumer groups. T-tests were used to determine significant differences between low and high consumers. There were only two behaviors for which there was a significant relationship between reported frequency of the behavior and intake of dairy products (choosing pudding, frozen yogurt or ice cream for dessert and snacks, and having milk with snacks). Those children who were high consumers were also more likely to report more frequent selection of dairy products for dessert or snacks and having milk with snacks. Most children reported eating breakfast and there was a strong correlation between eating breakfast and eating cereal and milk with breakfast. Many children (71%) indicated they drink milk for breakfast every day, 69% indicated they drink milk for lunch every day, while only 52% reported drinking milk for their evening meal on a daily basis. About 25% report taking a vitamin pill with calcium three or more times per week. There was a negative correlation between thinking it is cool to order milk in a restaurant and intake from the dairy group. Only five percent of the children reported trying to lose weight by not drinking milk and while the relationship between this response and intake was negative, it was not significant. A large percentage of children indicated they like milk (93.4%) while only 77.4% indicated they like yogurt. Overall, the intake of dairy foods was not affected by food preferences since most children indicated they liked these foods. Most children also indicated that their parents thought it was important for them to drink milk (91.4%) and few reported getting sick from dairy products (4%-8%). Skills related to intake from the milk, yogurt and cheese food group significantly associated with the actual intake were figuring out if they were getting the recommended number of servings and ordering milk when eating at restaurants. Frequency of practicing these skills was also significantly different between high and low consumers. For most skills, about one-third of the children reported practicing these skills most of the time or always. About 87 percent reported storing milk on the front shelf in the refrigerator and 58 percent reported asking their parents to buy milk most of the time or always. The five items that assessed knowledge related to calcium intake were considered separately and as an aggregate score (Calcium knowledge). The score for one single item was negatively related to intake (Period of life when need people need to get enough calcium). Children were not sure whether you need calcium your whole life or only when they are young and growing. Most children could accurately identify good sources of calcium. The behaviors, affective items, skills and knowledge positively related to intake of foods from the milk, yogurt and cheese group are listed below as recommendations for learning objectives for the curriculum topic area of Increasing Calcium Consumption. LEARNING OBJECTIVES: Behaviors: Choose pudding, frozen yogurt or light ice cream for dessert and snacks. Have milk with snacks. Drink milk when eating at restaurants. Affective items: Think it is cool to order milk in a restaurant. Skills: Figure out if getting the recommended number of servings each day from the milk, yogurt and cheese food group. Order milk when eating at restaurants. Knowledge: Know that it is important to get enough calcium throughout life Marla Reicks, Extension Nutritionist ABOUT MEAT COST OF MEAT In 1970, the disposable personal income was $3,564 per person. The meat and poultry expenditures were $148.59 per person. Thus, meat and poultry expenditures as a percent of disposable income were 4.2 percent, or 30.1 percent of the total food expenditures. However, in 1997, the disposable personal income was $21,969 per person. The meat and poultry expenditures were $428.01 per person. Thus, meat and poultry expenditures as a percent of disposable income were 1.9 percent, or 18.2 percent of the total food expenditures. Yes, the cost of meat has increased, but disposable income has increased even more. Consumer complaints about the price of meat are almost non-existent because the price has increased more slowly than disposable income has increased. In my opinion, the development of new meat products is trying to focus on consumers who are curious about new product and don't mind paying more per pound compared to traditional meat products. What is being developed? First, convenience is increasing in popularity. Precooked meat products are appealing more and more to consumers. However, precooked meat products must also contain antioxidants as processors continue to incorporate "natural"-based antioxidants. New species of meat products are being produced and offered to clientele. Buffalo and red deer fairly expensive, but many consumers are trying them from time to time. New production methods are starting to surface more and more. For example, hogs are being raised in other than completely enclosed confinement areas. Organic meat will really expand when standards are finalized by USDA. Organic meat will sell. The questions are: how much will be produced, and at what price? Finally, branded meat will be slowly oriented towards local producers. Small plants that are USDA inspected or Minnesota inspected (equal to) are becoming more and more interested. Richard Epley Extension Animal Scientist, Meats CENTENNIAL CUISINE: REGULATING COOKING TEMPERATURE; AN ESSENTIAL TALENT In "Food on the Frontier," author Marjorie Kreidberg wrote, "The joy felt by a housewife as she gazed upon her newly acquired cookstove was usually abruptly terminated by her first attempts to use it." The successful use of a cook stove required understanding how to vent the heat through the parts of the stove, as well as understanding what kind of fuel was needed for a hot or a moderate fire. Acquiring this ability to adjust heat while cooking brought about a revolution in food preparation, according to Reay Tannahil in her book, "Food in History." A 1905 "Good Housekeeping " article, "The Kitchen Fire," explained how to prepare and regulate the cookstove fire, as well as how to polish the stove, so the stove would look "proud and happy" at the end of the day. When some women mastered their cookstoves, they would not part with them, and they moved those heavy stoves across the prairies in covered wagons! Then the stoves were moved from lean-tos, to shacks, to cabins, and finally into fine frame homes. Although some of those cookstoves may have been cherished, they were difficult to clean; especially cleaning the soot-filled the stovepipes. And accidents did happen. For example, in Mankato it was reported that "Mrs. Harry Footner . . . was badly burned while shaking down the ashes in her coal stove. The coffee pot on the stove was overturned, the hot contents running over her head and shoulders." "The Homesteader's Cookbook," 1971, printed in honor of the history of the Minnesota River Valley, contains an article from the "Shakopee Weekly Argus," dated February 13, 1873: "Never fill a stove more than half to two-thirds full of hard coal, even in coldest weather. When the fire is low, never shake the grate or disturb the ashes, but add from ten to 15 lumps of coal and set the draft on. When these are heated through and somewhat ignited, add the amount necessary for a new fire, but do not disturb the ashes yet. Let the draft be open half an hour. Then shake out ashes. The coal has thoroughly ignited, and will keep the stove at high heat from 6 to 12 hours, according to the coldness of the weather. In very cold weather, after the fire is made, add coal every hour." While maintaining the fire in the cookstove and regulating the temperature may have been a talent mostly developed by women, there were some men who were famous for their baking skills. In the winter lumber camps in northeastern Minnesota, good cooks were essential for good morale among the loggers, and the cooks had community-wide reputations to uphold, even in the summer. For example, at a Fourth of July celebration, 1899, at Sandy River Crossing near Aitkin, the Neils Neilson family celebrated with some eight or ten families. According to the new book, "A Minnesota Remembrance: Making A Life in The Land God Forgot" by Robert O. Harder, the baked goods served at that picnic made for a wonderful feast. One brother wrote to his sibling in Minneapolis, "Jimmy Green, Murphy's camp cook that everybody likes so well, baked up a storm-- bread, donuts, cakes, and pies. We ate till we nearly busted." When fuel for the cookstove was needed the most, it was not always readily available. During February of 1907 the "Norman County Index" printed that coal was in short supply and there was some available at Perley, Minnesota, a town on the North Dakota border about 20 miles from Ada, the county seat. The article also encouraged people to use peat and weed seeds as alternative fuel. Access to electric and gas appliances increased in the early part of this century. In the summer of 1915 the Agricultural Experiment Station at the University of Minnesota in Crookston was the site of a demonstration of cooking with gas appliances. The demonstration took place outdoors in a large tent and was conducted by the Boys and Girls Club and an appliance dealer. Cooks also needed to know whether hot, moderate, or slow oven heat was needed for various foods. In the 1923 edition of her cookbook, Fannie Farmer wrote, "Experience is the best guide for testing temperature of oven. Various oven thermometers have been made, but none have proved practical." She gave the following rules for baking in a gas oven: Hot oven requires 2 burners lighted. Moderate oven requires 2 burners, halfway on. Slow oven requires 1 burner halfway on. For baking a loaf of bread, Mrs. Farmer wrote the following directions: Light both burners five minutes. Put loaves on upper shelf. After ten minutes turn off back burner and turn front burner down one-fourth. Gradually reduce heat until it is only one-third on. Turn [flame] entirely off five minutes before bread is done. Small loaves bake in 40 minutes. Large loaves bake in 60 minutes. How complicated! It is a wonder that any bread got baked! About the same time Fannie Farmer's book was published, the Good Housekeeping Institute published recipes with very exact temperatures for baking. Examples of oven temperatures were 275 for a fruitcake, 320 for Company Cake, 325 for Almond Loaf, 360 for Southern Chocolate Cake, and 375 for cup cakes. The Institute staff was perhaps overly optimistic; first, about the availability of ovens with thermostats and, secondly, about the ability of the thermostats to precisely control oven heat. Another suggestion for estimating the temperature of an oven was found in a 1932 "Household Handbook" published by Clarkson-Rishoff Company in St. Paul. According to the "Handbook," one of the simplest ways for an experienced baker to tell the temperature was with his or her bare hand. A hot oven is one in which you can hold the hand until you have counted 30. If you cannot count to thirty the oven is very hot, too hot for most baking. A moderate oven permits you to hold your hand to count to 40 . A slow oven permits you to count to 45. An oven in which you can hold your hand much beyond the count of 45 is really too cool for even the slowest baking. Certainly opening the oven door to insert one's hand while counting to 30 or 45 reduced the oven's heat immediately. And one needed to know how fast to count. A third method, the "paper method" of testing oven temperature was described in a 1937 booklet prepared by the Royal Baking Powder Company: Although this test is not as accurate as either the oven control or thermometer, it will prove helpful when they are not available. Place a piece of white unglazed paper in center of oven after it has been lighted ten minutes. The heat is correct when the paper browns in time specified: Hot or 475 to 500 in « minute. Moderate Hot or 400 to 425 in 1 minute. Moderate or 350 to 375 in 1« minutes. Slow or 325 in 2 minutes. Obviously it did take experience to gain the skill needed to judge the temperature of an oven. And if it was too hot or too cold for the item to be baked, what was the next step? Change the dampers? Change the fuel? For many of us today, we can say, "Hats off!" to our predecessors. Regulating the oven heat was a matter of intuition, art, and perhaps some luck. By 1935-1940 thermostats were built into both ovens and surface units. The Robertshaw Thermostat Company of Pennsylvania promoted the assets of automatic, constant temperatures: " . . . no more twiddling with the gas cock and peeking at the oven thermometers." Controlling heat for electrical surface units depended upon developing the technology that made flexible tubular heating units possible. The tubes could be shaped into circles that fit under pans or shaped for grills and ovens. Switches also became more sophisticated. But in Goodhue County's booklet celebrating the 75th anniversary of Extension Home Study Groups, the Leonites Group, about 25 neighboring farm women, recorded that in 1941-42 they still did their "cooking on wood ranges, hardly anyone had electricity or indoor plumbing." Almost all Minnesota homes had electrical power by the 1960s. "Forecast for Home Economists" published an article about the revolution in household appliances that was written by Eileen Burke. The title was "Appliances: A Women's Work is Never Done." According to her article, by the 1930s electric cooking had increased power and people were talking about the necessity of 220-volt lines. In addition to many other power plants, the locks and dams on the upper Mississippi River were producing cheaper hydroelectric power. This included the Ford Dam on the St. Paul-Minneapolis city limits. The speed of electrical units was advertised in the 1950 issues of "Better Homes and Gardens." The Hotpoint ad claimed that "In six minutes flat this amazing unit brews 6 cups of coffee, using cold water." The General Electric range ad boasted that the extra high-speed unit was 20 percent faster and "boils a pint of water in four minutes or less." Burke summarized the changes in range temperature control, saying " . . . in both built-in and free-standing units, thermostatic controls now bring the same accuracy to surface cookery that we've enjoyed in oven cooking for many years. And, in effect, convert any frying pan into an automatic skillet." Cooks have made amazing changes over the generations, first learning how to use cookstoves with a variety of fuels, then thermostatically controlled ranges, and, more recently microwave ovens --which is another story to tell about family food preparation during this century. Today it seems that whenever there is a major power failure during a winter storm, news reporters find stories to tell about families that had minimal problems surviving the cold, thanks to their wonderful cookstoves! So much for a century of progress! WORLD WAR I FOOD RATIONING: WIN THE WAR BY SAVING FOOD! It was a boom time for agriculture at the turn of the century when the United States was fighting the Spanish American War in Cuba and the Philippines, but by the time the United States entered World War I, grain reserves were depleted. In his book about the first century of the Minnesota Extension Service, former director Roland Abraham pointed out that the world supply of wheat was low and cargo ships of grain had been sunk at sea, losing grain meant to feed Europe. Crop yields for both wheat and potatoes had been low in 1915 and 1916 and the supply of seed corn was short. When the U.S. declared war, the citizens of this country were united under the motto, "Food will win the War! Win the War by Saving Food!" "The Farmer" magazine reported that Congress had passed the Food Control Bill to prevent hoarding and extortion, reduce waste, and ensure fair prices. People were asked to voluntarily limit or ration their use of wheat, meat, fat, and sugar. Meatless Tuesdays and Wheatless Wednesdays were encouraged. In the June 1917 issue of the "Farm Journal," President Wilson was quoted as saying, "Upon the farmers of this country, in large measure, rests the fate of the war and the fate of the nations." The Secretary of Agriculture, David J. Houston, added, "Upon an ample food supply with careful conservation and economic distribution may depend the safety of this country and the preservation of democracy." Even an advertisement of a sugar company promoting its product, a scarce commodity, read, "We must feed not only our own people but also millions in Europe." Countless ways of conserving food were identified in columns with titles like "What YOU can do now!" Family meals should be as largely vegetarian as father and the boys will stand for;' Save seeds for next year's garden; Use barley flour instead of wheat flour for making bread; Don't eat bread! Have another potato instead; Save leftover cereal, stale bread, or bread crumbs from your cutting board to thicken soup, stewed vegetables, or hash; Use the tough outer leaves and stalks of vegetables like cel ery and cabbage. And former President Theodore Roosevelt added, "Cut out Booze! Save the Grain!" The Royal Baking Powder Company prepared a booklet of non-wheat bread recipes that was endorsed by the U.S. Food Administration. The wheat could be saved for export and at the same time Americans would have healthful food in the form of cornmeal and other coarse flours, whether they liked to eat it or not. Alice Ames Winter, Chair of the National Defense Council Women's Committee wrote, "Women must keep in mind the world necessity and the world program while they are getting dinner, but the rest of the family must -cheerfully- (emphasis added) accept the changed menu and learn to like war breads" However, Alan Darling, a former resident of Montrose, South Dakota, remembered: Back in the winter or spring of 1918, my mother couldn't get wheat flour. She baked our bread on the farm. So she bought barley flour and rye, and followed some government recipes to make rye and barley bread--using very little wheat flour. I was six years old and remember the result as heavy, almost unchewable--and dark colored! She never tried again. We ate a great deal of corn bread, hominy, cornmeal mush, and oatmeal. No Cream of Wheat. Catherine Swift from South Dakota State College prepared an article on "Patriotic Meals" for the November 1917 issue of the "Farm Journal" to help people plan meals with minimal amounts of wheat and meat. The meals for one day were the following: corn flakes, buckwheat cakes and sausages for breakfast; tomato soup, cheese fondue, creamed potatoes, and nut bread with butter for lunch; codfish cakes, Johnny cake [cornbread], rhubarb sauce, and oatmeal cookies for supper. Note the lack of red meat and minimal use of wheat. WORLD WAR II RATIONING: FOOD FOR DEFENSE! FOOD FOR FREEDOM! For people in the United States, World War I lasted about 18 months, but World War II lasted for four long years. And rationing was not voluntary. Starting in May 1942 the government issued ration books each month with little red and blue coupons of different "point" values. There were coupons for sugar, meat, coffee, canned foods, fish, butter and cheese, as well as books with coupons for gasoline, rubber tires, and shoes. Anna Schumann Darling, a retired Nebraska school teacher, recently wrote: When WWII began I was teaching in Hastings. I lived with another teacher in an apartment. We had ration books. We had trouble getting sugar, coffee, etc., because we never got home from school early enough to be able to buy things at the store. They were usually sold out of sugar, etc. by the time we got home in late afternoon. We ate lots of vegetables and not much meat. In a 1943 booklet called "Your Share," Betty Crocker of General Mills fame explained how homemakers should watch their "P's (points) and "Q's" (quantities) in their ration books. For example, one pound of steak was assigned 12 points; one pound of ground beef was assigned 7 points, and one pound of liver, kidney, or flank meat was assigned even fewer points. So people were supposed to buy less popular cuts of meat to make their points go farther. But, somehow, liver and other organ meats, nourishing though they might be, never became popular. The foreword in the same booklet lent moral support: Hail to the women of America! You have taken up your heritage from the brave women of the past. Just as did the women of other wars, you have taken your positions as soldiers on the Home Front. . . . you are, first and foremost, women with the welfare of your families deepest in your hearts . . . . You must make a little do where there was abundance before. . . . you must prepare satisfying meals out of your share of what there is. Your must heed the government request to increase the use of available foods and save those that are scarce. . . .And we salute you all! (Betty Crocker, Home Service Staff, General Mills, Inc.) The motto or theme, "Feeding the Family for Health, Morale and Victory," encouraged statewide food conservation. The University of Minnesota Extension Service taught programs on gardening, caring for fruit orchards, and canning and storage of food, including butchering meat for home use. According to Katherine Jellison's book, "Entitled to Power," in 1942 the Minnesota members of the Extension Homemaker Groups pledged to raise 75 percent of their food at home. In 1941-42 Blue Earth County Home Demonstration Agent, Viola Mach Dietz, collaborated with the Red Cross, Civilian Defense, and the Parent Teachers Association. Over 2000 people, both rural and town residents, attended the first of a series of five meetings; volunteer leaders conducted 96 sessions for neighborhood groups. The topics included "Home Grown Food Supply in Wartime," "Right Foods Put More Life in Living," and "Three Meals a Day the Minnesota Way." Goodhue County Extension Homemaker Group leaders taught lessons planned around the consumer's pledge: "I will buy carefully, I will waste nothing, and I will take care of the things I have." In their 75thAnniversary booklet, a group called Happy Circle wrote: The World War was still being fought and food shortages were everywhere. Conservation of our resources were stressed a lot. [Those] were the days of sugar rationing, gasoline rationing and also rationing of fruits, meats, and canned vegetables. Each family was issued rationing stamps for each member of the family. Farmers had big gardens and raised their own animals for eating. Sometimes rationing stamps were traded with another family. Lack of meat, or at least the feeling of not having enough meat, was a major problem. Alan Darling, a retired minister, spoke of rationing meat in northern Illinois during the war: "Some how I was appointed to the County Meat Ration Board with authority to have black market meat handlers arrested and tried. Thank the Lord, we didn't have many cases. Just one I can think of now where a couple kept on after being fined several times and the Board closed their meat business. I think they also served a jail sentence." "The American Woman's 3-Way Meat Stretcher Cook Book" from the Culinary Arts Institute in Chicago contained the following statement about the black market and the availability of meat: "The vital necessity of meat for everybody ought to keep all of us on the alert against Black Market operations. Bootlegging meat is a crime against every citizen, not just a slick way to secure more meat." Another booklet or planning guide prepared by the Home Economics Institute of the Westinghouse Corporation in 1943 encouraged people to share meat to be sure everyone had enough, and not to waste even a spoonful of gravy. The booklet contained directions on how to extend meat with cereal and how to store it so it didn't spoil. Also people were encouraged to voluntarily, ". . . set aside one day a week besides Friday to serve meat alternates in place of meat. It's little enough for those of us on the home front, when you consider the way our boys on the fighting front gladly volunteer for dangerous missions that may cost them their lives." This was a heavy burden, but there were also efforts to boost morale. Speaking as Betty Crocker, the home economists of the General Mills Test Kitchens had a lot of ideas to help people be more positive about the food shortages. Locally, General Mills cooperated with the Minneapolis Municipal Market in sponsoring a "Vitamins for Victory Party" in August 1942 to increase awareness of locally grown health-giving produce. Betty Crocker also suggested ideas for inexpensive neighborhood or family parties so people would relax and enjoy each other's company. A waffle supper or a garden salad picnic were among the ideas; the whole group would share ration points so everyone could enjoy rationed items like coffee while having a good time. General Mills also declared that cereal foods were the "official extender" of meat, stews, and soups during the wartime rationing. If anyone was going to promote cereal foods for extending meat and stews, General Mills certainly had the right and responsibility to do so. They pointed out that most nourishing foods were scarce or restricted, but "cereal foods are still available in many -delightful- [emphasis added] forms." The Frigidaire Division of General Motors prepared a booklet on how to clean and defrost a refrigerator to maximize its ability to circulate cold air to chill the food and reduce waste. Frigidaire acknowledged that it was difficult to get refrigerators repaired but its entire resources were pledged to an all out war effort until victory was won. The company assured people that when the war was over there would be many wonderful new products, resulting from the research and technology developed during the war. Ultimately the message was a patriotic message: Be patient, do your part, postpone having your needs met, and you will be rewarded when we win the war! There was a shortage of labor to help with the planting and harvesting of crops on farms. Local businessmen and professional men put on their gloves and coveralls and volunteered six, nine, fifteen hours a week to help harvest the corn and hay. Young women from the Walker area traveled by bus to southern Minnesota to help in the cornfields. Often Extension Agents coordinated those labor pools. Laborers were needed in agriculture, and laborers in manufacturing plants had a change in lifestyle, too. Men and women were working swing shifts and night shifts, so families were less likely to have meals together. Today's memories of rationing food during the war remain distinct.. One gentleman wrote, " I remember using honey and white Karo syrup for sweetening. Gooseberry jam made with honey was good. It was okay for me on breakfast food when we were short of sugar, but I couldn't take it in my coffee for breakfast!" The effort to conserve food continued after World War II was over. Rationing and standing in line for a beef roast may have ended, but there was still a great need to feed the people in Europe and other parts of the world who had not been producing food for several years. Both lack of food and bad weather plagued European countries. In April 1948, the Marshall Plan was put into effect to feed and rebuild Europe and, it was hoped, to minimize the growth of communism as European workers struggled to find answers to solve their problems. American churches, granges, schools, unions, and other groups, organized "Friendship Trains" of corn, soy, hogs and dairy products that rolled across mid-America to the East Coast for shipping to Europe. Displaced persons came to the Midwest, war brides from Europe joined their soldier-husbands, and people learned new ways of living together in crowded housing and college classrooms. Employed women were told to stay home and let returning soldiers have their jobs. Families were reunited and babies were born. Not only was there a baby boom, everything else was booming, too. Food for the family would no longer be thought of in terms of produce from the family garden or farm. Frozen foods, cake mixes, ready-to-eat cereals, and countless new grocery items would soon line the supermarket shelves. People in the armed services had been around the world, and while they were hungry for home cooking, they were also ready to enjoy the foods they had learned about elsewhere, like pizza and tempura. Hopes were high for a better, modern world for all. Mary Darling, Extension Nutrition Specialist EVALUATION OF RISKS RELATED TO MICROBIOLOGICAL CONTAMINATION OF READY-TO-EAT FOOD BY FOOD PREPARATION WORKERS AND THE EFFECTIVENESS OF INTERVENTIONS TO MINIMIZE THOSE RISKS September 1999, U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition Jack Guzewich, RS, MPH; and Marianne P. Ross, DVM, MPH http://vm.cfsan.fda.gov/~ear/rterisk.html INTRODUCTION: The Food and Drug Administration (FDA) publishes the Food Code which provides guidance on food safety, sanitation and fair dealing that can be uniformly adopted by jurisdictions for regulating the retail segment of the food industry. The model Food Code is the cumulative result of the efforts and recommendations of many contributing individuals, agencies, and organizations. Section 3-301.11 of the 1999 Food Code, entitled "Preventing Contamination from Hands" was added to the code in response to outbreaks of foodborne illness caused by food that had been contaminated with pathogens transmitted by food preparation workers. FDA believes that the considerable number of illnesses transmitted by food worker contamination of food demands rigorous intervention measures. The following is a summary of current information from scientific Literature or provided to FDA that evaluates the factors related to contamination of foods by food workers and the effectiveness of interventions to prevent or minimize contamination of ready-to-eat food by food workers. Three major intervention areas are addressed: exclusion of ill food workers from the workplace, removal of pathogens from the hands of food workers, and the use of barriers to prevent bare-hand contact with ready-to-eat foods. Information provided in this review includes all applicable submissions that were received in response to Federal Register Notice, Vol. 64, No. 63, Friday, April 2, 1999. On September 16, 1999, CDC released data on the incidence of foodborne disease in the United States. --------------------------------------------------------- WHITE PAPER, SECTION ONE A literature review pertaining to foodborne disease outbreaks caused by food workers, 1975-1998. Jack Guzewich, RS, MPH Marianne P. Ross, DVM, MPH ABSTRACT: A search was conducted of the published scientific literature for the period 1975-1998 to identify articles that described outbreaks of foodborne disease that were believed to have resulted from contamination of food by food workers. A total of 72 articles that described 81 outbreaks involving 16 different pathogens were identified. Viral agents, specifically hepatitis A and Norwalk-like virus, accounted for 60% (49) of the outbreaks in this review. Ninety-three percent (75) of the outbreaks involved food workers who were ill either prior to or at the time of the outbreak, depending on the organism involved. In most of the remaining outbreaks, an asymptomatic food worker was believed to be the source of the infections. Eighty-nine percent of the outbreaks (72) occurred in food service establishments as compared to 11% (9) that were attributed to foods prepared in domestic settings. Sandwiches, salads, and miscellaneous hot food items that required extensive hand contact during preparation accounted for the majority of foods involved in the outbreaks. This review provides evidence that food workers, particularly ill food workers, can serve as the source of infection in foodborne outbreaks and that hand contact with foods represents a mode by which contamination may occur. INTRODUCTION: Outbreaks of foodborne disease are caused by foods that are contaminated intrinsically or that become contaminated during harvesting, processing, or preparation (Torok et al., 1997). It has been estimated that seven pathogens found in animal products (E. coli O157:H7, Listeria monocytogenes, Campylobacter jejuni, Clostridium perfringens, Toxoplasma gondii, Salmonella spp., and Staphylococcus aureus) account for approximately 3.3-12.3 million cases of illness and 3,900 deaths in the United States each year (Buzby & Roberts, 1997). The annual cost of these foodborne illnesses, which includes costs to individuals, industry, and the public health sector, is estimated to be $6.5-$35 billion (Buzby & Roberts, 1997). Foodborne diseases are known to contribute to both human morbidity and mortality as well as to health care costs (Campbell et al., 1998). These costs include the expenses entailed in controlling the disease, medical treatment costs, business losses, and losses in productivity. A recent study estimated the total societal cost of one particular foodborne outbreak of hepatitis A to be $809,706 (Dalton et al., 1996). This included the costs associated with the infected food worker believed to be the source case, 43 secondary cases, and the potential exposure to 5000 patrons. The study suggested that the cost of outbreaks due to food workers can in some instances far exceed the costs associated with outbreaks due to person-to-person transmission of infectious agents. According to a report by the Centers for Disease Control and Prevention (CDC), hands may be the most important means by which enteric viruses are transmitted (LeBaron et al., 1990). Further, contamination of food by an infected food worker is the most common mode of transmission of hepatitis A in foodborne disease outbreaks. In these outbreaks, the vehicles involved have most often been foods that were not cooked or that were handled improperly after cooking (Centers for Disease Control, 1990, 39(14):228-32). Food workers may transmit pathogens to food from a contaminated surface, from another food, or from hands contaminated with organisms from their gastrointestinal tract (British Medical Journal, 1990). Therefore, hand contact with ready-to-eat foods (defined as food that is edible without washing, cooking, or additional preparation by the consumer or by the food establishment and that is reasonably expected to be consumed in that manner U.S. Public Health Service, 1999|), represents a potentially important mechanism by which pathogens may enter the food supply. During the five-year period from 1988-1992, 2,423 outbreaks of foodborne disease were reported to the CDC (Bean et al., 1996). Of these outbreaks, 1,435 had information reported concerning contributing factors. During this period, the two most commonly reported practices that contributed to foodborne disease were improper holding temperatures of foods and poor personal hygiene of food workers, reported in 59% and 36% of outbreaks, respectively. Delicatessens, cafeterias, and restaurants were the most common places where contaminated food was reportedly eaten (Bean et al.,1996). For the period 1983-1987, there were 2,397 outbreaks reported with 1,257 of these outbreaks reporting contributing factors (Bean et al., 1990). Improper holding temperatures were reported in 63% of the outbreaks and Poor personal hygiene were reported in 28% of the outbreaks. In New York State during the period 1980-1993, among the outbreaks where contributing factors were reported, the most commonly reported contributing factor was contaminated ingredients (23%) (Guzewich, 1995). Inadequate refrigeration was reported 20% of the time and infected food workers were reported 17.6% of the time (Guzewich, 1995). Enteric pathogens that are believed to be capable of being transmitted by food workers include, but are not limited to: E. coli, hepatitis A virus, Salmonella spp., Shigella spp., and Clostridium perfringens (Paulson, 1994; Restaino & Wind, 1990; Snyder, 1997). In addition, pathogens such as Yersinia, Proteus, Campylobacter, and Klebsiella, originating from raw animal products, can contaminate hands and then be transferred to foods, equipment, and other workers (Paulson, 1994; Restaino & Wind, 1990; Snyder, 1997). The purpose of this review was to search the published scientific literature for examples of foodborne disease outbreaks believed to have resulted from the introduction of pathogens by food workers. METHODS: We conducted a search of the published English-language scientific literature for the period 1975-1998 for articles that described foodborne disease outbreaks believed to have resulted from the introduction of pathogens into food by food workers. In order for an outbreak of foodborne disease to be classified as one resulting from the contamination of food by a food worker, we required that at least one of the following 2 criteria be presented convincingly in the published report. The first criteria: sufficient epidemiologic evidence was presented to link the food worker with the outbreak. Factors considered here included whether the putative time of the employee's contact with the implicated food was consistent with the incubation period for the illness experienced by those persons who ate the food; the strength of the association between the food(s) prepared by the employee and the illness that followed consumption of the food(s); the biologic plausibility of the food(s) serving as the vehicle; whether the cessation of contact of the food worker with the food(s) in question resulted in a reduction in illness; consistency of findings with previous reports; and, finally, the likelihood that alternative explanations could account for the illnesses. The second criteria: laboratory evidence implicating a food worker as the source of contamination had to include either the identification of the etiologic agent from an anatomical site of the food worker suggesting colonization or infection or, in the case of hepatitis A infection or Norwalk-like virus infection, serologic evidence had to suggest acute infection at the time of food preparation. The following electronic databases were searched: PubMed, Grateful Med (MEDLINE), Educational Resources Information Centre (ERIC), Agriculture Online Access (Agricola), Food Science Technology Abstracts (FSTA), Biological Abstracts (BIOSIS PREVIEWS), and the Centers for Disease Control and Prevention (CDC) publication website. Keywords used for the search included terms such as 'foodborne disease', 'food handler', 'foodborne outbreak' and 'food preparer'. Outbreaks of foodborne disease that occurred in countries other than the United States and that met the above criteria were included to emphasize the global importance of sanitary food preparation practices. Viruses of the Norwalk and Norwalk-like group are in the family of viruses known as Caliciviridae. For simplicity, we categorized all such viruses (such as Snow Mountian Agent and Small Round Structured Virus) as Norwalk-like viruses. RESULTS: Using the criteria described in the Methods, we identified a total of 72 articles for review. Several articles described more than one outbreak; consequently, a total of 81 foodborne outbreaks were incorporated in our review. In the 72 articles reviewed, 16 organisms were identified as the etiologic agents. Hepatitis A and Norwalk-like viruses were the most frequently reported organisms (Table 1). Outbreaks ranged in size from five persons to 3,175 persons. A total of 14,712 persons were affected in these 81 outbreaks. Three percent (440) of the case-patients required hospitalization. Two case-patients died, both as a result of infection with Salmonella enteritidis. One of the fatalities was a 70-year old man who had had cardiac surgery while the other patient recently underwent a laparotomy. Seventy-eight percent (63) of the outbreaks occurred in the United States, 11% (9) in the United Kingdom, and 4% (3) in Canada. One outbreak took place in Singapore (Goh et al., 1984) and one in Jordan (Khuri-Bulos et al., 1994). One outbreak occurred on a cruise ship sailing from the U.S. to the Bahamas (Dalton et al., 1996). Three outbreaks occurred on international flights: one originating in the U.S. (Hedberg et al., 1992), one originating in the U.K.(Burslem et al., 1990), and one originating in Tokyo (Centers for Disease Control, 1975). The two flights originating abroad had connections in the U.S. and one of these outbreaks was traced back to a food worker in the U.S. Several large outbreaks where food workers were implicated were captured in our search. These include an outbreak of Shigella sonnei in 1991 that involved approximately 3,175 persons (Lee et al., 1991); an outbreak of Norwalk-like virus in 1984 in which approximately 3,000 persons became ill (Kurtisky et al., 1984); and an outbreak of Salmonella enteritidis in 1980 involving 866 persons (Burslem et al., 1990). Specific food items were implicated as vehicles for transmission in 89% (72) of the outbreaks (Table 2). Sandwiches, salads, and miscellaneous hot food items, such as mashed potatoes and ham, accounted for the majority of foods involved in the outbreaks. In cases where multiple foods were implicated, these items were counted separately. Other foods implicated in outbreaks included baked goods, beverages, fruit salads, and miscellaneous cold foods such as aspic glaze (obtained from powdered gelatin, used to retain the fresh look of food products while extending shelf life), rice dressing, and canned salmon. Of the nine remaining outbreaks where a specific food item was not identified, five were hepatitis outbreaks that were associated with eating at the establishment where the food worker was ill with hepatitis A during the appropriate time frame. The remaining 4 outbreaks involved bacterial illnesses ( 2 Salmonella typhii, 1 Salmonella enteritidis, and 1 Shigella sonnei) where a food worker was ill with the outbreak bacteria just prior to the outbreak and acquisition of the illness was associated with eating at the establishment at times when the implicated food worker was on duty. Eighty-two percent (66) of outbreaks provided sufficient epidemiologic and laboratory evidence to implicate a food worker as the source of infection; 17% (14) relied on epidemiologic evidence alone whereas for the remaining 1% (1) of outbreaks, laboratory findings alone contributed the evidence. Ninety-three percent (75) of the outbreaks involved food workers who were infectious either prior to or at the time of the outbreak, depending on the organism (i.e., Norwalk-like virus is thought to be shed primarily at the time of symptoms whereas hepatitis A, with an incubation period of 15-50 days, had maximum infectivity during the latter half of the incubation period). In the remaining 6 outbreaks, 1 food preparer denied illness and refused testing, 1 food preparer was not clinically ill but was responsible for contaminating foods with hands after changing diapers, 4 articles provided insufficient information concerning laboratory testing and the implicated food workers denied illness. The majority of the outbreaks associated with food workers involved transmission of the pathogen to food by the food worker's hands. In the description of the findings from the outbreak investigations, authors specifically listed hand contact as a factor in the transmission of the pathogen in 34 outbreaks (and specifically mentioned that the food worker was not wearing gloves in 14 of these outbreaks) or implied that hand contact was a factor in the transmission of the pathogen by discussing one or more of the following factors in an additional 38 outbreaks: handwashing, poor personal hygiene, food workers "handling" ready-to-eat foods such as adding parsley to dishes, open sores on hands or arms, and food workers snacking on food. The remaining 9 outbreaks associated with food workers did not specifically mention how the food worker transmitted the organism to the food. In this latter group, all food workers were ill. Four of the 72 articles specifically mentioned that food workers were wearing gloves; however, gloves were worn improperly (i.e., not worn consistently or not worn to cover lesions) or were not worn during food preparation and handling. Several scenarios by which food contamination occurred from bare-hand contact were described in the 72 articles. In an outbreak of Norwalk-like virus gastroenteritis that infected 67 persons, the food vehicle was food prepared by an asymptomatic food worker who reported recovering from mild gastrointestinal illness 48 hours prior to the outbreak. The worker used bare hands to slice meats and de-bone cooked chicken (Patterson, et al., 1993). Staff members involved in an outbreak of Shigella sonnei gastroenteritis in the week prior to an outdoor festival prepared cold ready-to-eat foods that led to an outbreak involving 3,175 persons. A few food preparers were still symptomatically ill with febrile diarrheal illness during food preparation for the festival. An uncooked tofu salad had been thoroughly mixed by hand by these staff members and there was limited access to proper handwashing facilites (Lee et al., 1991). Four employees of a hotel bar developed hepatitis A due to mixed drinks Prepared by a bartender who was asymptomatic at the time he prepared the drinks; in addition to mixing drinks, he placed garnishes as well. This bartender reported experiencing cough, nausea, and vomiting 3 weeks prior to the outbreak and was subsequently diagnosed with hepatitis A infection (Kosatsky & Middaugh, 1986). A bakery worker experienced vomiting and diarrhea while preparing butter cream frosting. The employee prepared frosting by submerging his bare arm up to the elbow in the frosting as it was being mixed in order to scrape the sides of the vat. A subsequent outbreak of Norwalk-like virus infection involving 129 ill persons resulted from ingestion of frosted items that were contaminated by the suspected source case (Kurtisky et al., 1986). In addition to the hands of food workers being a source of contamination of the implicated foods, other contributing factors were identified during the investigations. Inadequate food storage or other problems with preparation practices were described, including unsanitary food-contact surfaces, cross-contamination, and improper holding temperatures for foods. After contamination of foods by the hands of food workers, temperature abuse may have enhanced the growth and multiplication of bacteria in the implicated products. DISCUSSION: In our search of the published scientific literature for reports of foodborne disease believed to have resulted from contamination of food by food workers, we found more instances of disease caused by viral agents than bacterial agents. Specifically, hepatitis A and Norwalk-like viruses accounted for 60% (49) of all the outbreaks included in this review. During the 81 outbreaks, 14,712 persons were infected. Ninety-three percent of these outbreaks involved food workers who were ill either prior to or at the time of the outbreak. Eighty-nine percent (72) of the outbreaks occurred at food service establishments, such as restaurants, cafeterias and catered functions as compared to 11% (9) of the outbreaks which were attributed to foods prepared in domestic settings. The majority of foods involved in the outbreaks included sandwiches, salads, and miscellaneous hot items. Foods such as sandwiches and salads often involve intensive hand contact during preparation and are not cooked prior to consumption. Contaminated hot foods suggest that hand contact occurred after cooking. There were several limitations inherent in this review: first, the reports of foodborne disease outbreaks identified in the published literature represent only a small fraction of foodborne disease outbreaks and an even smaller fraction of all foodborne disease. The likelihood of an outbreak being brought to the attention of health authorities depends on consumers' and physicians' awareness, interest and motivation to report the incident, and the surveillance activities of state and local health agencies. Diseases with short incubation periods are more likely to be recognized as foodborne disease outbreaks than those with longer incubation periods. Pathogens that cause mild illness may be underreported. A second limitation of the review is that the studies included often differed in the extent to which the outbreaks of disease were investigated or reported. As a result, uniform data were not presented regarding the role of hand contact in all of the outbreaks. For example, some factors that make foodborne disease detection inherently difficult are the fact that foods are often not available for testing, food workers may refuse to submit stool or blood samples for analysis, and samples may not be collected until days or weeks have elapsed following contamination of the implicated food. Food consumption histories are often impracticable; therefore the link between illness and the consumption of contaminated food may not be ascertained. Viral foodborne diseases are exceptionally challenging. Viruses have small infective doses and infection, especially of immunocompromised individuals, can occur with low-dose contamination (Stolle & Sperner, 1997). Virus excretion is of short duration, which may make diagnosis difficult. Because hepatitis A has a mean incubation period of 30 days, poor recall and recall bias can be substantial problems in investigations (Warburton at al., 1991). Foodborne outbreaks due to hepatitis A are most often caused by food contaminated during preparation by an infected food worker (Centers for Disease Control, 1993). Therefore, food workers in retail settings with acute hepatitis A infection constitute a great risk for consumers since one person has the potential to transmit the virus to a large number of people (Angelillo et al., 1996). Since thorough cooking inactivates hepatitis A virus, outbreaks almost always involve foods that are not cooked between contamination and consumption (Centers for Disease Control, 1983). Because of the limitations described above, it is most likely that this review markedly under-represents the true number of foodborne outbreaks related to food workers. Additionally, one cannot establish from this review what the true role of hand contact of food by food workers is in the total burden of foodborne disease. Nevertheless, despite these limitations, several themes became apparent: first, numerous examples of foodborne disease outbreaks were identified in which hand contact of foods by food workers was believed to be the source of infection. Second, both viral agents as well as bacterial pathogens were involved, with parasitic agents being less common. Third, in addition to contamination of foods by food workers, a number of other substandard food practices were often involved. In sum, our review of the literature provides evidence that ill food workers can be the source of infection in foodborne outbreaks and that hands of food workers can transmit pathogenic organisms to foods. It has been suggested that food establishments that mass-produce cold, ready-to-eat food items should exclude ill workers from food preparation responsibilities and provide strict oversight of food preparation and handwashing practices (Hedberg et al., 1992). Exclusion of food handlers from duties for 48-72 hours after termination of diarrhea and vomiting may be adequate to prevent and control Norwalk-like virus outbreaks (Parasharet al., 1998). Table 1:Organisms identified in foodborne outbreaks, 1975-1998. AGENT, # (OUTBREAKS) Hepatitis A, 28 (34.57%) Norwalk-like virus, 21 (25.93%) Staphylococcus aureus, 6 (7.41%) Shigella sonnei, 5 (6.17%) Salmonella typhimurium, 5 (6.17%) Salmonella enteritidis, 4 (4.94%) Group A streptococcus, 4 (4.94%) Giardia, 2 (2.47%) Salmonella paratyphi, 1 (1.23%) Salmonella javiana, 1 (1.23%) Vibrio cholera, 1 (1.23%) Shigella flexneri, 1 (1.23%) Cryptosporidium parvum, 1 (1.23%) Yersinia enterocolitica, 1 (1.23%) Total: 81 (100%) === Table 2: Foods identified in foodborne outbreaks, 1975-1998. FOOD ITEM(S) IMPLICATED, # OF TIMES THIS ITEM WAS IMPLICATED IN AN OUTBREAK Hot food items such as mashed potatoes, ham and turkey, 24 Cold salads, 19 Cold sandwiches, 16 Other cold foods such as canned salmon and rice dressing, 9 Bakery goods such as glazed products, 6 Beverages such as mixed drinks and ice slush beverages, 6 Fruit including fruit salads, 6 Combination of salad and sandwich, 4 == REFERENCES: For a complete list of the references cited in the above white paper, contact Bill Schafer, Extension Food Safety Specialist, University of Minnesota Department of Food Science and Nutrition, at == --------------------------------------------------- WHITE PAPER, SECTION TWO Interventions to prevent or minimize risks associated with bare-hand contact with ready-to-eat foods. Jack Guzewich, RS, MPH Marianne P. Ross, DVM, MPH INTRODUCTION: In the food industry, contamination from microorganisms can be responsible for infectious disease outbreaks passed from food employees to consumers via food (Paulson, April1996). In a report by the Hospitality Institute of Technology and Management (Docket C-3), it was stated that, "due to lack of adequate handwashing by employees who prepare, process, and handle food in a retail setting, the potential for foodborne illness of fecal-oral nature continues to be problematic". The origins of microbial contaminants in food service facilities include the environment, the food worker, the source of the food, and the food itself (Docket C-8). Food may also be contaminated by food employees via exposure to raw animal products during processing (Paulson,1994; Coates, et al. 1987; Docket C-8). The number of pathogenic cells on a food employee's hands is presumed to be directly related to the probability of transfer of microorganisms from hands to cooked food products (Restaino and Wind, 1990). It has been stated that handling cooked products with bare hands is one of the major hazards of cooked foods (Bryan, 1995). This review will address the many different interventions that can be used to minimize or eliminate the contamination of ready-to-eat foods by food workers. Two major areas have been presented in this section: removal of pathogens from the hands of food workers, and barriers to bare-hand contact with ready-to-eat foods. METHODS: The information presented here is the result of a literature search of the past five years on PubMed by using keywords 'handwashing', 'hand washing', and 'handwash products'. Other reference items included in this document were submitted to the FDA before and during the development of this paper, and in response to a Federal Register Notice. The information presented in this paper forms the basis for current opinion and judgements made on these topics. Value judgements were not made concerning the quality of the data or the methods used to generate the data. REMOVAL OF PATHOGENS: Background and Definitions: Handwashing is the removal of soil and transient microorganisms from the hands (Larson, 1995). Hand antisepsis is the removal or destruction of transient microorganisms (Larson, 1995). Degerm- ing, or hygienic hand disinfection, refers to the reduction of predominantly transient microorganisms with the use of germicidal agents or antiseptic detergent formulations (Sheena and Stiles, 1982; Ayliffe et al., 1987; Nicoletti et al., 1990). Hygienic hand disinfection is also called healthcare personnel wash, which is a term that describes the washing of hands by food employees and healthcare personnel in order to eliminate transient microorganisms. (Bartzokas et al., 1987; Sattar and Springthorpe, Cambridge Univ Press,1996). Plain or nonanti- microbial, nonantiseptic soaps are detergent-based cleansers that have no bactericidal activity and, by mechanical action, are used for physical removal of dirt (Larson, 1995). Antimicrobial or antiseptic soaps, on the other hand, contain ingredients with in vitro and in vivo activity against microorganisms on the skin (Larson, 1995). Antimicrobial soaps are considered drugs and are regulated by the Food and Drug Administration since they are intended to inhibit or kill certain skin flora. Microflora of skin are categorized into two types: resident and transient. Resident bacteria are those organisms that normally reside on the skin, in this case, the skin of the hands. Ninety percent of resident flora of the hands are coryneform and coagulase negative staphylococci (Miller, 1994). Among the resident microflora, Staphylococcus aureus is the only true pathogenic organism of food safety concern (Docket C-3; Miller, 1994; Lowbury et al., 1964). Resident flora are not easily removed by mechanical friction (Docket C-3; Larson, 1995) since they are buried deep within the pores where they are protected by sebaceous gland secretions (Restaino and Wind, 1990; Miller, 1994). The variety of resident organisms is significantly less than the kinds of microorganisms that can serve as transient organisms (Restaino and Wind, 1990). Transient organisms are of concern because they are readily transmitted by hands unless removed by the mechanical friction of washing with soap and water, or destroyed by the use of an antiseptic solution (Larson, 1995). Transient organisms can be considered skin contaminants that are acquired from environmental sources and become attached to the outer epidermal skin layer (Docket C-3; Restaino and Wind, 1990). Hands, as well as contaminated gloves, serve as vectors for transmission of transient microorganisms (Fendler at al, Part I, 1998; Docket RPT-1). According to Miller (1994), transient bacteria cause great concern to the food service industry because these organisms are loosely attached to the surface of the skin and can easily contaminate food products if employees do not wash their hands adequately. Hands, arms and fingers of food employees may become contaminated with fecal microorganisms after using the toilet. These organisms include salmonellae, E. coli, Staphylococcus aureus (Snelling, 1991), Clostridium perfringens, shigellae, and hepatitis A virus (Docket C-3; Restaino and Wind, 1990). Organisms from animal sources such as Yersinia, Proteus, Campylobacter, and Klebsiella, can be transmitted to hands and thus transferred to foods, equipment, and other workers (Paulson, 1994). Transfer of Organisms: Hand transfer can be a significant mode of transmission of bacteria and viruses from person to person, from person to surface or vice-versa, and from person to food (Docket C-8). Touch-contact-associated bacterial transfer is facilitated by wet hands as compared to dry hands (Docket C-8). Transfer of organisms and viruses occurs less frequently when the contaminated material or hands are dry (Larson, 1985; Ansari et al., 1988). Residual moisture on hands after handwashing has been found to play an important role in the transfer of bacteria and viruses (Springthorpe and Sattar, 1998). The longer the duration of hand drying, the fewer bacteria were transferred to other surfaces (Springthorpe and Sattar, 1998). Viral Transfer: Frequent and thorough handwashing has been found to help prevent and control the transfer of viruses and other infections from food employees in field settings (Sattar and Springthorpe, Cambridge Univ Press,1996). However, handwashing agents vary in their ability to eliminate viruses on hands; they tend to eliminate bacteria more effectively than they do viruses (Sattar and Springthorpe, Cambridge Univ Press 1996). Non-enveloped viruses, such as rotavirus, calicivirus and hepatitis A virus, survive better on skin than enveloped ones, such as herpes virus and influenza virus (Sattar and Springthorpe, Infect Control & Steriliz Technique,1996; Joklik, 1988). Most of the in vivo evaluations of handwashing products have been conducted With bacteria; whereas, very few published reports evaluate the agents against viruses. Rotaviruses are transmitted by the fecal/oral route and usually cause gastroenteritis in infants and small children. They can also be problematic for the elderly and immunocompromised individuals. Caretakers of infants must practice proper handwashing techniques after changing diapers. Rotaviruses survive for long periods in contaminated water, on hard surfaces, and on hands (Benenson, 1995). Rotaviruses are capable of surviving on human hands for up to 4 hours (Sattar and Springthorpe, Infect Control & Steriliz Tech, 1996; Ansari et al., 1989; Sattar et al., 1994) and can be readily transferred between hands and inanimate objects (Ansari et al., 1989). Infectious rotavirus on hands can be readily transferred to other surfaces and vice versa. The amount of virus transferred is related to the duration of contact and the amount of pressure applied during contact (Sattar and Springthorpe, Infect Control & Steriliz Tech, 1996). Contamination can occur from the preparation of various types of foods by persons with rotavirus-contaminated hands (Ansari et al., Rev Infect Dis, 1991). Lack of data on the role of foods as a vehicle may be due to investigation insufficiencies following foodborne outbreaks of gastroenteritis rather than the inability of rotaviruses to be distributed via food (Ansari et al., Rev Infect Dis, 1991). In a study by Ansari et al. (1988), hands contaminated with rotavirus were then allowed to dry for 20 or 60 minutes and then touched to clean metal surfaces and vice versa, for 10 seconds at a pressure of 1 kg/cm2. Virus transfer from hands to surfaces and from surfaces to hands was similar. After 20 minutes the mean percent of transfer ranged from 16.1-16.8%. At 60 minutes, the mean percent of transfer was 1.6-1.8%. Transfer from contaminated hands to clean hands was 6.6% at 20 minutes and 2.8% at 60 minutes. The author concluded that virus transfer after 20 minutes was quite high compared to that at 60 minutes. This suggests that the inocula were not completely dry after 20 minutes. The author concludes that hands may play a vehicular role in the transfer of rotavirus. Repeated experiments did not show detectable human parainfluenza virus transfer from fingers to inanimate objects when using pressure of 1 kg/cm2 for 5 seconds, although the virus could be transferred from inanimate objects to fingers (Ansari et al., J Clin Microbiol, 1991). Hepatitis A virus survives well on environmental surfaces and on human Hands for up to 7 hours (Sattar and Springthorpe, Infec Control & Steriliz Tech, 1996), can be easily transferred to and from hands and surfaces, and is resistant to many disinfectants used in food establishments (Mbithi et al., 1993). In a study of 10 handwashing agents (Alcare, Aquaress, Bacti-stat soap, Dettol, Ethanol 70%, Savlon, Scrub Stat IV, Septisol soap, Triclosan hand soap), alcohol solutions were effective in preventing the transfer of hepatitis A virus from washed hands to metal surfaces (Mbithi et al., 1993). In a study by Bardell (1995), it was found that the presence of Herpes Simplex Virus-1 (HSV-1) on gloved hands led to the contamination of lettuce and ham. This might not be expected with virus in a dry state that was adhering to the fingertips. Since the transfer of HSV-1 to the foods occurred when the virus inoculum was dry and the surfaces of the foods were moist, the moist surface of the foods apparently enhanced viral transfer. Bacterial Transfer: According to a study by Scott and Bloomfield (1990), if surfaces are contaminated with low numbers of organisms (120 organisms/cm2) such as E.coli, Salmonella, and S. aureus, contact with fingers can transfer organisms in sufficient numbers to pose a potential infection hazard. The ability to transfer staphylococci to cooked potatoes during handling, peeling, cutting and garnishing has been recognized (Bryan, 1995). Listeria spp. Have been found to survive on the fingertips for over 11 hours, thus food may be contaminated with Listeria via the fingers of food employees (Snelling et al, 1991). Since L. monocytogenes has a long survival time, this may have serious implications for food processing establishments (Snelling et al, 1991). Chlorhexidine gluconate in alcohol was found to be effective in reducing Listeria from fingertips (Snelling et al, 1991). Cooked foods are vulnerable if touched by Salmonella-contaminated fingers that have been contaminated by low numbers of the bacteria (Pether and Gilbert, 1971). Salmonella spp., as well as E. coli, can be transferred between hands, raw foods, and cooked or processed foods (Pether and Gilbert, 1971). In a study by Pether and Gilbert (1971), it was found that washing with soap and water, followed by drying with paper towels, reduces Transient carriage of Salmonella spp. on hands unless the bacterial contamination is very high. This study showed that standard handwashing techniques did not remove inocula of 10 6 organisms but did remove some organisms with inocula of 10 3-10 4. All Salmonella organisms were removed after handwashing with inocula below 10 3. Transmission via hands contaminated with Camplyobacter can be a major Route of infection. Transmission to hands may take place through cross-contamination of foods and utensils when food workers process raw foods (Coates et al., 1987). In a study by Coates et al. (1987), Camplyobacter spp. were effectively removed by handwashing with either Soap and water or water alone, followed by drying with paper towels, yet the organism existed on wet hands. Campylobacter was also eliminated by 70% isopropyl alcohol. This study demonstrates the importance of hand drying to reduce the incidence of sporadic Campylobacter infections. Parasite Transfer: The literature search produced only one article that briefly mentioned parasitic infections. Detergent and soap products can inactivate parasites, such as lice and pinworms. The ova are more resistant but can be easily removed by mechanical action of handwashing (Borgatta, 1989). Handwash Technique: According to Springthorpe and Sattar (1998), proper hand hygiene requires three components: a proper protocol, an appropriate handwashing/cleansing agent, and compliance. Fingers are thought to be the most important part of the hand in terms of the transfer and spread of pathogenic microflora (Ansari et al., J Clin Microbol,1991). An evaluation of handwashing techniques reveals many different schools of thought. Discrepancy in duration of the procedure, type of product used, and temperature and pressure of the water are just a few of the variations presented in the literature. Quantity of soap varied from study to study. It was found that 3-5 ml. was sufficient (Larson, 1987; Ayliffe,1988) when using antiseptic soap. Larson states that there is no advantage of using more than 1 ml. when using nonantiseptic soap (Larson, 1987). The methods and duration of handwashing were variable among the many articles reviewed. A range of 5-30 seconds was noted (Paulson,1994; Nicoletti et al, 1990;Miller, 1994;Butz et al, 1990; Larson, 1987; Larson et al, 1989; Paulson, 1992; Namura et al., 1994, Docket C13). Duration of handwashing is important for mechanical action as well as to allow sufficient contact time with antimicrobial products (Larson, 1995). Increased friction by rubbing hands together or using a scrub brush allows for greater reduction of transient bacteria even with the use of plain soaps or detergents (Restaino and Wind, 1990; Docket C-3). Butz, et al. (1990) performed a comparative study of the immediate antimicrobial effectiveness of four handwash products for health care personnel. Immediate effectiveness was determined by sampling hands immediately after handwash compared with samples taken prior to handwash. In this study, they adapted the handwashing protocol set forth by the American Society for Testing and Materials (ASTM). The ASTM protocol is as follows: wet hands under warm water (100 -108 F); apply 3 ml. handwash product; rub vigorously over all hand surfaces, concentrating on interdigital spaces and nailbeds; apply a small amount of water and lather for 15 seconds; rinse for 30 seconds; dry with a clean paper towel. The 1986 CDC Guidelines for Hand Washing and Hospital Environmental Control recommended that, for routine handwashing, plain soap can be used with a vigorous rubbing of all surfaces of lathered hands for at least 10 seconds, followed by thorough rinsing under a stream of water (Docket C-3). The American Society for Microbiology (ASM) recommends vigorous scrubbing for 10-15 seconds and the Association for Professionals in Infection Control and Epidemiology (APIC) recommends 15-20 seconds of vigorous hand washing (Docket C-13). In a study by Ojajarvi (1979), it was found that a handwash of 2 minutes' duration removed only 3% more transient microorganisms than did a 15-second wash (Docket C-13). Handwashing with warm water is thought to exacerbate the damage done to The skin's barrier function (Springthorpe and Sattar, 1998). However, it has been suggested that warm water (110< F-120< F) at a water flow of 2 gallons/minute is sufficient to wash off the pathogens that have been loosened by handwashing with plain soap or detergents (Docket C-3). The activation energy of antimicrobial agents is easier to achieve at higher temperatures, thus surfactants and other antimicrobial components in handwashes work more efficiently (Docket C-13). In a study of handwashing at various water temperatures, a significant difference in resident microflora removal was seen between washing and rinsing with 70< F and 120< F water (Docket C-8). There were no resident microflora removed at 40 F, despite the use of soap and manual handwashing.Washing and rinsing with warm water brings resident flora from deep skin layers to the surface where they are removed with washing or drying (Docket C-8). Since transient microorganisms are not normally located in deep skin layers and are more easily removed by routine hand washing, water temperature may not play a role in the removal of these organisms from the skin (Docket C-8). In any case, water temperatures must be within a comfortable range to the user in order to be effective and practical. Numbers of organisms decrease as the duration of handwashing increases but this occurs only to a certain point. Chamberlain et al. (1997) demonstrated increased bacterial counts of both naturally-occurring and artificially-inoculated organisms after a 3-minute handwash as compared to a 10-second regimen. Mean bacterial counts for unwashed hands was significantly greater (p=0.04) than for hands washed for 10 seconds. Mean data for hands washed at 3 minutes was not significantly different from unwashed hands (p=.30). Mean data for hands washed at 3 minutes was also not significantly different from hands washed for 10 seconds (p=.22) (Chamberlain et al., 1997). Washing for 10 seconds removes transient bacteria from hands and results in decreased recovered organisms. Washing for 3 minutes removes transient bacteria but brings residual flora to the surface, thus increasing recovered organisms from hands. Very high frequency (>25 times/day) of handwashing shows increased skin irritation (Bartzokas et al., 1987) and increased bacterial counts, possibly due to the defatting of skin, which has been shown to increase the survival of Staphylococcus arueus on hands (Larson, 1985). Excessive handwashing can interrupt the skin's normal protective barrier function by cracking or damaging the skin, altering the skin's pH, removing skin lipids, decreasing moisture, or changing its normal flora (Larson, 1995; Springthorpe and Sattar, 1998; Larson, 1985; Larson et al.,1989; Borgatta, 1989). As handwashing frequency, duration and aggressiveness increases, damage to the stratum corneum layer can cause dry skin, chapping, pain, cracking and fissures. Dry skin then causes increased shedding of both skin cells and skin microflora (Docket C-8). One study found that a double handwash of 15 seconds provided a 20% Greater reduction in resident bacteria than a single wash; whereas, the second wash provided 0.25%-2.25% greater reduction in transient organisms (Docket C-13). Transient organisms, rather than resident microorganisms, and cross contamination via hands are of greatest concern in food establishments. In a study by Stiles and Sheena (1987) in a meat processing plant, levels of hand contamination varied between work stations. For workers in the kitchen and packaging areas, no differences in reduction of bacteria could be noted among handwash agents. Transient bacteria were found after handwashing during in-use operations, indicating that handwash techniques were not effective. However, the visible contamination of hands in the meat cutting areas may have prompted these workers to perform a more extensive handwash, which then resulted in significant differences in reduction of bacterial counts. Rings and Fingernails: In a hospital study by Jacobson et al., a one-minute wash with Ivory soap and a surgical scrub brush, followed by a one-minute rinse, was used to determine bacterial counts on subjects with rings versus those without rings (1985). It was found that, although wearing rings increased the number of microbes found on the skin, thorough handwashing produced no statistically significant difference in bacterial counts for subjects with rings as compared to subjects without rings (Jacobson et al., 1985). In another study in the healthcare setting, a 10-second handwash and 10-second rinse were used to demonstrate differences in microbial counts in subjects with and without rings (Salisbury, 1998). The authors concluded that there was a greater reduction in microbial counts after handwashing for those persons without rings (Salisbury, 1998). This effect was more apparent when the initial microbial load was greater than 1000 cfu/ml before handwashing. One article from the healthcare setting determined that nurses with artificial nails concealed greater numbers of gram-negative rods before and after handwashing as compared to nurses without artificial nails (Pottinger et al., 1989). The authors suggested that a 10-second handwash alone may be inadequate to prevent bacterial shedding. Rings and jewelry are thought to harbor food debris, microbial contaminants, food allergens, and caustic sanitizers or disinfectants; all of these items may be irritating and react with ring metals, which may result in decreased or inadequate handwashing (Docket C-8). Methods of Evaluation: This literature search provided studies on various handwash products. However, it should be noted that no standard, approved method exists to evaluate and analyze handwash products. Methods used to analyze products vary in relation to: setting, technique and duration of handwash, number of handwashes, water temperature and pressure, use of a prewash, amount and type of product tested, in vivo versus in vitro studies, microbiological analysis, and microorganisms used for testing purposes. This makes it difficult to compare the various products. The proposed Health- care Continuum Model addresses the need for a uniform means of evaluation and classification of topical antimicrobial formulations (Jones, 1998). For the category of food employee handwash products, this model suggests specific in vitro and in vivo testing methods for efficacy, performance, and use pattern. An approach to testing the efficacy of food employee handwash products, based on the Healthcare Personnel Handwash Evaluation, has recently been developed (Docket RPT-1). Paulson believes that topical antimicrobial handwash products used for healthcare personnel are chemically and antimicrobially similar to those used in food service; however, food service products should also effectively remove the organic load of food ingredients and fat (Docket RPT-1). There are currently no chemicals or formulations capable of reducing virus infectivity by at least 99%, which is the standard measure of effectiveness of chemical disinfectants. To date, there is no such criterion for the efficacy of in vivo testing of antiseptics for viruses (Sattar and Springthorpe, Cambridge Univ Press1996; Sattar and Springthorpe, Infect Control & Steriliz Technique,1996). Handwash Products: 1. Soaps and Detergents: Different conclusions have been drawn concerning the effectiveness of plain soaps. Handwashing with plain detergent soap and water can physically remove microbes, but antiseptic agents are necessary to kill or inhibit microorganisms (Larson, 1995; Ehrenkranz, 1992). Plain soap is used primarily in the mechanical removal of transient microorganisms whereas antimicrobial products are used for the mechanical removal and killing or inhibition of both resident and transient flora (Larson, 1995). Handwashing with plain soap should be sufficient to remove transient microflora from the hands of food service employees (Docket C-3; Paulson, 1994). However, antimicrobial soap is statistically more effective in both immediate and residual properties (Paulson,1994). One study demonstrated that plain soap was as effective as alcohol or chlorhexidine, but only in terms of activity against Gram-negative bacilli (Ayliffe et al., 1987). It is unclear if plain soaps are more effective than antimicrobial soaps, since there is literature to support both types of products. However, the review of the literature seems to indicate that plain soaps are effective for the physical removal of transient organisms and that antimicrobial products are needed for the inhibition of transient and resident flora. Apparent increases in bacterial counts associated with sanitizers as well as handwashing alone may be due to methodology accuracy and precision, inherent variability in individual dermal parameters, and data scatter due to insufficient controls (Docket C-13). When handwashing frequency is low (<6 washes per day), there is less advantage of using antimicrobial soaps compared to nonantimicrobial soaps (Larson et al., 1989). Antimicrobial soaps are recommended at higher frequency handwashes when long-term reduction in colonizing microflora is needed (Larson et al., 1989). Debate has ensued concerning the presence of bacteria on bar soaps. Bar soaps were found to have higher bacterial cultures after use than liquid soaps (McBride, 1984) but several studies found that the bacteria were not transferred to hands on subsequent use (Heinze, 1985; Bannan and Judge, 1965; Heinze and Yackovich, 1988). 2. Chlorhexidine Chlorhexidine gluconate (CHG) was found to be an effective product in terms of residual effects against bacteria (Docket C-3; Larson, 1995; Bartzokas, et al., 1987; Ayliffe at al., 1988). Bartzokas et al. (1987) found that CHG had a residual efficacy of log 10 reduction factor of 4.15. Chlorhexidine gluconate was found to be effective against nosocomial infections (Sheena and Stiles, 1982; Doebbeling et al., 1992; Aly and Maibach, 1979) and fungi (Nicoletti at al., 1990; Namura et al., 1994; Stiles and Sheena, 1985; Larson et al., 1986), although it is not effective against viruses (Mbithi et al.,1993; Sheena and Stiles, 1983). A study by Stiles and Sheena (1985) found that 4% CHG had a 99% reduction in transient organisms. Aly and Maibach (1979) determined that 4% CHG had immediate activity (85% reduction)and residual activity (98.2% reduction) of resident bacteria. In a study by Ayliffe et al.(1988), it was found that CHG had a log 10 reduction factor of 4.92. Therefore, CHG was more effective in terms of residual activity than 7 other products tested. A test configuration using 0.75% CHG formulation demonstrated a 3 log 10 difference between the treatment group and a control group (untreated) at the immediate and one hour sample (Docket RPT1). Recovery populations for the group using 0.75% CHG were 4.45 log 10 for the immediate sample and 3.37 log 10 for the 1-hour sample. Recovery Populations for the untreated group were 7.49 log 10 for the immediate sample and 6.16 log for the 1-hour sample. In this study, CHG demonstrated statistically significant immediate and persistent antimicrobial activity (Docket RPT1). 3. Alcohols Alcohol was the most immediately effective product against bacteria but had limited residual activity (Docket C-3; Paulson, 1994; Coates et al., 1987; Larson, 1995; Butz et al., 1990; Ayliffe et al., 1988; Aly and Maibach, 1979; Larson et al., 1986; Paulson,1994; Ly et al.,1997). Alcohols and formulations containing 70% alcohols were most effective in reducing the numbers of E. coli and rotavirus (Ansari et al., 1989). In contrast, they were not as effective against viruses such as hepatitis A (Mbithi et al., 1993). Alcohol applied to hands for as short as 15 seconds has been found to be effective in preventing transmission of Gram-negative bacteria (Larson, 1995). Products containing isopropanol or ethanol are very effective in decreasing bacteria in areas around and under the fingernails (Mahl, 1989). Bacterial counts were found to increase after very frequent washing and with the use of alcohol sanitizers (Miller, 1994). Alcohol gel sanitizers that do not require rinsing may be ineffective on their own due to the fact that there is no mechanical action to wash away bacteria (Paulson, 1994; Miller, 1994;Docket C-8). Thus the end result may be increased resident bacteria, including pathogenic S. aureus, on the hands. As they dry, alcohol products may pull resident bacteria from deeper skin layers, thus an increase in resident bacterial counts may be noticed (Docket C-8). Antiseptic handrubs, such as alcohol gel sanitizers, can be used only to inhibit microorganisms, without any mechanical effect on soil removal (Larson, 1995). It has been reported that the detergent base in these handrubs, and not necessarily the antiseptic product, is the cause of greater damage to the skin when compared to plain soaps (Larson, 1995). These detergents can irritate the skin by removing normal skin oils and impairing the skin's normal barrier function (Paulson, 1998). Aly and Maibach (1979) found that alcohol had immediate activity (84% reduction) but had limited residual activity (90% reduction) as compared to other products. A study by Ayliffe et al. (1988) determined that the most immediately effective products for hygienic hand disinfection were those containing alcohols (log 10 reduction factors of 2.5-3.8) but alcohols showed little or no residual effect. In a study to demonstrate immediate and persistent antimicrobial activity of a commercially available alcohol gel product, recovery populations were 7.85 log 10 for the immediate sample and 6.81 log 10 for the 1-hour sample (Docket RPT1). Recovery populations for the control group (untreated) were 7.49 log 10 for the immediate sample and 6.16 log 10 for the 1-hour sample. There was no significant difference in antimicrobial activity between the group that used the alcohol gel and the control group (Docket RPT1). Alcohols are not cleaning agents; therefore, they are not recommended for use in the presence of physical dirt (Larson, 1995; Docket C-8). Due to the shortcomings of alcohol sanitizers in the presence of soil, the build-up of emollients after repeated use, and the lack of effectiveness against certain viruses, it is recommended that hands be washed before alcohol application (Docket C-3; Restaino and Wind, 1990;Docket C-8). A combination of the rapid effects of alcohol and the persistent effect of chlorhexidine gluconate could serve as a template for a desirable antiseptic product (Paulson, 1994; Larson, 1995; Snyder, 1993). 4. Iodophor Stiles and Sheena (1985) determined that iodophor was effective for reduction of several transient bacteria (99.2-99.5% reduction). Using a procedure involving 6 handwash treatments over a period of 2 successive days, Sheena and Stiles (1983) determined that chlorhexidine gluconate had a residual effect whereas iodophor (0.75% available iodine) did not indicate a residual effect. Iodophors are generally accepted as antibacterial agents for hand hygiene but products containing high iodine concentrations (i.e. >0.75%) create some hesitation among users due to product odor and staining of skin (Stiles and Sheena, 1985). In addition to its broad-spectrum activity against bacteria, viruses, spores, and protozoa cysts, 0.5% active iodine has a low toxicity to humans (Docket C-4). A handwash towel that is comprised of a 5% Povidine-Iodine solution (PVP) represents new technology for application to the hands of food preparation/food processing personnel (Docket C-4). The antimicrobial properties of PVP, along with its low toxicity to humans, suggests that is compound will contribute to the reduction of microbial levels in food processing facilities (Docket C-4). 5. Triclosan Triclosan has a broad spectrum of activity against gram-positive and most gram-negative bacteria. It has immediate antibacterial effects as well as persistent activity on the skin and is only minimally affected by organic matter (Larson, 1995). Triclosan 1.5% was found to have residual efficacy (3.78 log 10 reduction factor) against transient bacteria (Bartzokas et al, 1987). A study using 0.3% triclosan found that after 6 washes per day for 5 days, there was no significant difference in mean log 10 colony-forming units compared to other products. After 18 washes per day for 5 days, the control group (nonantimicrobial soap) mean count decreased 0.30 logs while the tricolsan group decreased 1.55 logs. In this study, tricolsan performed significantly better than the control soap after high-frequency handwashing (Larson et al., 1989). 6. Para-chloro-meta-xylenol (PCMX) PCMX is considered to have good activity against gram-positive organisms, is less active against gram-negative organisms, and has fair activity against viruses, some fungi and against the tubercle bacillus (Larson, 1995). Its activity is minimally affected by organic matter and it has a persistent effect over several hours (Larson, 1995). In a study comparing handwash products at two different handwashing frequencies, 0.6% PCMX demonstrated no difference in mean log 10 colony-forming units when compared to control (nonantimicrobial) soap after 6 washes per day for 5 days. After 18 washes per day for 5 days, the control group mean count decreased 0.30 logs while the mean count for PCMX decreased 1.74 logs. After 18 washes, the effectiveness of PCMX was significantly better than that of the control soap; however, tricolsan and PCMX were not significantly different from each other (Larson et al., 1989). 7. United States Department of Agriculture (USDA) List of Proprietary Substances and Nonfood Compounds Employee, or simply "E", classifications were used to describe and categorize antimicrobial hand soaps and sanitizing compounds. This classification system applied to USDA-approved handwash products to be used in meat processing plants. This system was discontinued by the USDA in 1998.The USDA classification is mentioned in this review since this system may still be familiar to the intended audience. When in use, this classification system was based on manufacturer's research claims and recommendations rather than on research conducted by USDA. E2 compounds were those used for handwashing and sanitizing. For purposes of E2 classification, hands did not need to be washed prior to the use of such compounds but must have been thoroughly rinsed after use (Paulson, 1994; Miller, 1994). E2 soaps usually contain triclosan or parachlorometaxylenol (PCMX) and had immediate antimicrobial effects primarily due to mechanical action (Paulson, 1994). Examples of these products included Clean & Smooth, Purell Antibacterial Lotion Soap, and Derma Klenz (Miller, 1994). E2 products were effective in terms of immediate and persistent antimicrobial effects but were very irritating to the skin (Paulson, 1994). E2 products showed a more significant reduction in resident and transient (Miller, 1994) organisms compared to antibacterial soaps. 8. Fingernail brushes: From this literature search, only one author published information concerning the use of fingernail brushes (Docket C-3). The author concludes that fingernail brushes are necessary in order to remove debris from fingertips and under and around nails, since the subungual area of the hand (area under fingernail) contains the highest numbers of microorganisms. However, the use of a brush that is too stiff or excessive use of fingernail brushes may damage the epidermal layer of the fingertips. The author recommends the double hand wash method when food service employees begin a shift and then again after using the restroom. A single hand wash, without the use of a nailbrush, is adequate for the removal of most transient bacteria during routine food preparation operations. The outline of the double hand wash is as follows: 1) using warm water, and with water flowing over fingertips, brush the fingertips and beneath nails for 10-12 seconds; 2) after using brush, store brush side up, without the use of sanitizing solutions; 3) apply soap to hands, lathering all surfaces for 5-7 seconds; 4) rinse; 5) dry with paper towel. Studies were conducted with an inoculated finger washing experiment using a 0.1 ml. solution of Serratia marcescens, containing 20,000,000-100,000,000 organisms/ml. (Docket C-3). These studies have shown a 1,000:1 reduction in bacterial counts after the first wash, with an additional 100:1 reduction after the second wash and then even further reductions after paper towel use. After use of the nail brush, it was determined that fewer residual organisms remained on the brush as compared to fingertips. The residual microorganisms that remain on the fingernail brush could be transferred to the next person using the brush; however, there would be another 99.98% reduction once the subsequent user begins washing. The author believes that door knobs and other fomites present a greater risk for cross-contamination than the common use of a nail brush; however, no scientific data were presented to support this statement. Hand Drying: 1. Hot Air Dryers Measures of hand drying effectiveness include such things as speed of drying, effective removal of microorganisms, degree of dryness, and prevention of cross-contamination (Docket C-8). One study of hot air dryers revealed an increase in the number of bacteria on hands after dryer use, as well as the presence of bacteria in the nozzles of hot air dryers themselves (Blackmore, 1989). A comparison of three models of dryers demonstrated an increase in numbers of bacteria remaining on hands after drying from 136% to 187% (Blackmore, 1989). After using hot air dryers with insufficient cycle length or heat, there may be a tendency for persons to finish drying hands on clothes (Blackmore, 1989; Matthews & Newsom, 1987). This could serve to increase the chances of organism transfer and cross-contamination (Blackmore, 1989). Matthews and Newsom (1987) found that a 30-second cycle drying time was insufficient to dry hands thoroughly. They also concluded that, of the four commercial dryers that were tested, hot air dryers appear to be safe from a microbiological standpoint. A study by Ansari et al. (Am J Infect Control, 1991) found that, irrespective of the handwashing agent used, electric air drying produced the highest reduction in numbers of E. coli and rotavirus when compared to either paper towels or cloth towels. For example, after washing with soap and water with no drying, there was a 77% reduction of rotavirus on hands; whereas, a reduction of 91.74% was noted with warm air drying, 86.8% with paper towels, and 80.4% with cloth drying. It was also noted that the use of 70% isopropanol can reduce the transient microflora to undetectable levels and in such instances, any of the methods of hand drying could be used. The author concluded that, in terms of washed hands, the reduction in contamination during drying is more critical when less effective hand washing agents are used. 2. Paper Towels Several studies demonstrated a significant reduction in bacterial counts when hands were dried with paper towels (Blackmore, 1989; Georgia-Pacific, 1996) and with cloth towels (Blackmore, 1989) since the friction effect physically removes bacteria from hands. A comparison of calculated mean values of bacteria from hands before and after drying revealed a 29% reduction when hands were dried with paper towels and a 26% reduction when dried with a continuous cotton towel (Blackmore, 1989). However, the bacteriologic quality of continuous cloth towels is inferior to that of paper towels (Blackmore, 1989). This can be due to the laundering process as well as the fact that bacteria can be transferred from one user to the next as the towel is rotated and pulled in order to obtain a clean area. The paper making process leads to substantial reduction, if not elimination, of microorganisms in paper towels (Georgia-Pacific, 1996). Paper towels are considered to be the most sanitary hand drying method (Georgia-Pacific, 1996). It was noted that the number of bacteria on paper towels was low before use but markedly increased after hands were dried indicating physical removal of microbes from the hands (Georgia-Pacific, 1996). The friction applied when drying with paper towels further reduces bacterial counts. Under experimental conditions, an average of 95% reduction in contamination with S. aureus was obtained after rinsing with tap water alone and drying hands with paper towels (Georgia-Pacific, 1996). This frictional removal of transient bacteria from hands takes on greater importance if handwashing is not performed properly (Georgia-Pacific, 1996). Paper towels can also be useful in situations where foot pedals are not available such that faucets and door handles can avoid being touched with cleaned hands (Georgia-Pacific, 1996). Although there are advantages offered by paper towels, there are issues of hand contamination with pathogens from paper towel exit areas as well as dispensers that utilize cranks, buttons, and levers (Docket C-8). 3. Handwash Machines: The major advantage of handwash machines is consistency of the handwash procedure (Vesley et al., 1985; Paulson, 1993). Another side benefit may include the monitoring of employees' handwash frequency. The antimicrobial efficacy of handwash machines is equivalent to that of manual handwashing procedures (Pauslon, 1992; Vesley et al., 1985). A study of the efficacy of antiseptic handrub lotions with handwashing machines revealed that a 30-second soap wash prior to use of handwashing machines with a chlorhexidine/alcohol combination (0.5% CHG and 77% ethyl alcohol) resulted in 90% reduction of bacteria from hands (Namura et al., J Derm, 199421:481-485). In a study comparing a plain soap manual handwash with chlor- hexidine gluconate machine wash, it was found that automated handwashing has more standardized and consistent wash results than did the manual wash (Paulson, 1993). After the first handwash, the variability of wash results was .719 logs for the manual wash and .53 logs for the automated wash. The manual wash variability was not significantly different from baseline measurements (0.749 logs) while the automated wash had less variability than baseline measurements (1.225 logs). After 5 washes, the manual wash variability was .74 logs which was not statistically different from baseline measurements. However, the automated wash variability was .505 logs, which was statistically significantly less than baseline. A standard handwashing machine comparing manual handwash with plain soap and 0.3% triclosan with machine wash using 2% chlorhexidine gluconate, it was noted that both the manual wash and chlorhexidine gluconate wash were statistically equivalent in degerming effectiveness (Docket RPT1). After the first manual wash there was a log 10 reduction of 2.07 from baseline and a 2.03 log 10 reduction after the fifth manual wash. The chlor- hexidine gluconate machine wash revealed a 1.84 log 10 reduction from baseline after the first wash and 1.88 log 10 reduction after the fifth wash. The machine wash was effective in degerm- ing the skin while having the advantage of increased wash control (Docket RPT1). This study also determined that users need not have any specific training in handwashing techniques to properly use the automated handwash system. A system employing brushes with the automated wash was not more effective than systems without brushes. This may be due to the fact that fingers were not placed far enough into the machine to make sufficient physical contact and that brushes are unable to clean between fingers (Docket RPT1). BARRIERS: Barriers to bare-hand contact with ready-to-eat foods include such things as gloves, deli wraps and utensils. In this review, gloves were the only barriers that were included due to lack of available data regarding other barrier methods. Gloves: Most glove studies have been conducted in the healthcare setting. In terms of food establishments, the main purpose of wearing gloves is to prevent pathogenic organisms from being transmitted to foods via hand contact from food workers (Paulson, Food Quality, 1996). An intact vinyl or latex glove (i.e., one with no punctures, tears, or holes) will provide protection from transmission of contaminating microorganisms from hands (Paulson, Food Quality, 1996). Ehrenkranz believes that glove use promotes a false sense of security Among healthcare workers since contaminated gloves have led to patient-to-patient spread of nosocomial infections (Ehrenkranz, 1992). Considering the glove to be protective can promote poor handwashing practices and increased microbial growth on the hands (Fendler et al., Part I, 1998). According to Bardell (1995), it is not uncommon for gloves to be worn for long periods of time without being changed and it is not unusual for food employees to put gloved hands to their mouths or noses without changing their gloves. It is the opinion of one author that the wearing of gloves to prepare and serve food does not prevent cross-contamination since glove wearers continue to touch contaminated surfaces or raw foods, thereby inoculating the glove surfaces with microorganisms (Docket C-3). The use of gloves alone does not provide a sufficient barrier against transmission of pathogenic microorganisms from food employees to consumers (Fendler et al., Part II, 1998). Handwashing was strongly encouraged prior to gloving (Snyder, 1997; Fendler et al., Part I, 1998; Docket RPT-1; Paulson, April1996) and after removal of gloves (Larson, 1995; Doebleling et al., 1988; Olson et al., 1993). E. coli counts increased on hands that were not washed prior to gloving (Paulson, June/July, 1996). This occurred after glove changes at one-hour and three-hour intervals. No significant growth of contaminating microorganisms was found on hand surfaces after 3 hours of consecutive glove wearing when hands were effectively washed prior to gloving (Paulson, June/July 1996). It has been demonstrated that both the interior and exterior of gloves can become contaminated with surface hand microorganisms if the hands are not washed prior to gloving (Docket C-3). Hands themselves can also be contaminated with organisms found on the glove surface. Microbial contamination of hands occurred more frequently with vinyl than with latex gloves (Olsen et al., 1993). According to Paulson (June/July, 1996), wearing gloves can present an even greater potential for transmission of disease. The author feels that microorganisms residing on skin are provided a more favorable environment for growth on gloved hands as compared to ungloved hands due to increased levels of moisture and nutrients. The author recommends that when gloves are worn, hands must be washed with an effective product prior to donning the gloves. The author also suggests that both handwashing with an antimicrobial product and gloving will provide more protection to those performing high-risk tasks (e.g., preparing, cooking, or wrapping food) than either method used alone (Paulson, 1997). According to Larson et al. (1989), handwashing is often omitted when gloves are used and organisms on the hands can multiply rapidly inside the moist and warm environment of the gloves. The use of gloves does not replace handwashing, especially since bacteria and viruses can leak through gloves (Larson et al., 1989). It has been shown that up to 18,000 Staphylococci organisms can pass through a single glove hole during a 20-minute period despite the fact that hands were washed for 10 minutes prior to gloving (Docket C-8). When a glove break occurs, a liquid bridge of microbial contamination can flow from hands to surfaces and foods (Docket C-8). Loose-fitting gloves may increase the risk of microbial contamination and transfer, as well as rendering them cumbersome. Gloves that are too tight can cause discomfort and may result in multiplication of microorganisms due to incubation and sweating inside the gloves (Docket C-8). Vinyl and latex gloves were tested in the hospital environment by mimicking patient care (Korniewicz et al., 1990). At "lower use levels", which included donning and removing gloves or rubbing the gloved hand with a washcloth, there was no statistically significant difference found between latex and vinyl gloves in terms of leakage. However, at "higher use levels", which included attaching capped needles to syringes and removing needles several times, or wrapping and taping blunt objects, 63% of vinyl gloves leaked compared to 7% of latex gloves. Glove leaks were more frequent with vinyl than with latex gloves (Olsen et al., 1993; Korniewicz et al., 1990; Kotilainen et al., 1989). In a hospital study, 43% of unused vinyl gloves had perforations (Best, 1992). In a hospital study using both vinyl and latex gloves, leaks occurred more frequently at the tip of the index and middle fingers (Kotilainen et al., 1989). This same study found that vinyl gloves were more likely to have multiple leaks as compared to latex gloves. One author stated that packaging for nonsterile use was the major factor associated with an increased glove leakage rate, implying that less care is taken in the molding of gloves for nonsterile use or that such gloves are not adequately tested for leakage (DeGroot-Kosolcharoen and Jones, 1989). Another potential problem with latex gloves is that some persons exhibit an allergic reaction to latex (Muller et al., 1998; Schwartz, 1995). Several studies reported the effectiveness of various products in removing bacteria from latex and vinyl gloves (Doebbling et al., 1988; Newson and Roland, 1989; McCarthy, 1996; Best, 1992). Best (1992) concluded that latex contributes to the trapping of microorganisms, possibly due to the three-dimensional lattice structure of latex, which allows for elasticity. It was found that after washing with several different commercial products, recovery of Staphylococcus aureus was minimal or negative (ranging from 0 cfu/ml to 39 cfu/ml). In a hospital study by Doebbeling et al., it was found that microorganisms adhere to latex gloves and are not easily washed off despite friction and cleaners (1988). A 3x2 factorial design tested a standard concentration of one of four nosocomial pathogens and one of three different hand cleansing agents to cleanse gloves. The agents reduced the median log 10 counts of organisms on gloves to 2.1 to 3.9 after an initial inoculation of 107 colony forming units. The proportion of positive glove cultures was as follows: Staphylococcus aureus, 8%-100%; Serratia marcescens, 16%-100%, and Candida albicans, 4%-60%. These results varied greatly after use of different handcleansers (P <0.001). There was considerable variability for Psuedomonas aeruginosa, 20%-48% (P=0.085). After glove removal, the observed proportions of hands contaminated with the nosocomial organisms varied from 5%-50% with variability depending on the particular handwashing agent used (P<0.001) (Doebbeling et al., 1988). The authors recommended handwashing after glove removal and suggested that it may not be beneficial to wash and reuse gloves in a hospital setting. SUMMARY: This review has addressed the many different interventions that can be used to minimize or eliminate the contamination of ready-to-eat foods by food workers. Three major areas have been presented: exclusion of ill food workers from the workplace, removal of pathogens from the hands of food workers, and barriers to bare-hand contact with ready-to-eat foods. Part One of the White Paper discussed the transmission of pathogens from food workers to foods. Exclusion of ill or infected food workers from the workplace is one intervention that can be applied in response to the information presented. Removal of pathogens from the hands of food workers can be accomplished by various modalities. Handwashing technique, including duration of handwash, water temperature, hand drying method, and frequency of handwash, play an important role in pathogen removal. Handwashing agents, such as detergents, soaps, sanitizers, and antimicrobial agents, vary in their ability to remove pathogens. Factors such as type of pathogen, duration of contact with hands, and characteristics of organic material present on hands must be considered when selecting an appropriate handwashing agent. Hand drying methods range from hot air dryers to cloth and paper towels. Factors such as cycle length of air drying, friction used with towel drying, and type of towel used can all influence the removal of pathogens from hands. Handwashing machines are also used to remove pathogens from hands. They are found to offer consistency and compliance monitoring capabilities but must be evaluated based on their mechanism of action, such as cycle length, water pressure and quantity, as well as the products utilized in the handwash procedure. Barriers to bare-hand contact with ready-to-eat foods include such things as gloves, deli wraps and utensils. In this review, gloves were the only barriers that were included due to lack of available data regarding other barrier methods. Issues related to the use of gloves as barriers include the glove material, glove permeability, duration of wearing, and handwashing techniques prior to and after wearing. == REFERENCES: For a complete list of references cited in this white paper, contact Bill Schafer, Extension Food Safety Specialist, Department of Food Science and Nutrition, University of Minnesota, at .