A. Larry Branen P. Michael Davidson Seppo SalmiBookFi.org FOOD ADDITIVES

Food Additives

Second Edition Revised and Expanded

edited by

A. Larry Branen

University of Idaho Moscow, Idaho

P. Michael Davidson

University of Tennessee Knoxville, Tennessee

Seppo Salminen

University of Turku Turku, Finland, and Royal Melbourne Institute of Technology Melbourne, Australia

John H. Thorngate III

University of California Davis, California

Marcel Dekker, Inc.

New York • Basel

iii v

xiii

809

897 899 913

Index

915

Introduction to Food Additives

A. LARRY BRANEN and R. J. HAGGERTY

University of Idaho, Moscow, Idaho

I. INTRODUCTION

According to the Food Protection Committee of the Food and Nutrition Board, food addi- tives may be defined as follows:

a substance or mixture of substances, other than a basic foodstuff, which is present in a food as a result of any aspect of production, processing, storage, or packaging. The term does not include chance contaminants.

Since prehistoric times, chemicals have been added to foods to perform special functions. Although basic foods contain no additives, as foods are processed for conversion into a variety of products, an increasing number of additives are generally used. Technological advances in food processing have increased the variety and use of these additives. Today, more than 2500 different additives are intentionally added to foods to produce a desired effect. The use of these additives is a well-accepted practice but is not without controversy. In this chapter, we explore some of the major benefits and risks of using additives. In subsequent chapters, each category of additives and some of the benefits and potential risks are explored in more detail.

II. TYPES OF ADDITIVES

Additives can be divided into six major categories: preservatives, nutritional additives, flavoring agents, coloring agents, texturizing agents, and miscellaneous additives. Several lists of these additives are available and as will be noted throughout this book, several additives commonly serve more than one function in foods. In Europe and other parts of Additives can be divided into six major categories: preservatives, nutritional additives, flavoring agents, coloring agents, texturizing agents, and miscellaneous additives. Several lists of these additives are available and as will be noted throughout this book, several additives commonly serve more than one function in foods. In Europe and other parts of

A. Preservatives

There are basically three types of preservatives used in foods: antimicrobials, antioxidants, and antibrowning agents. These additives are grouped under the category of preservatives in the INS system. The antimicrobials, with E and INS numbers ranging from 200 to 290, are used to check or prevent the growth of microorganisms. Antimicrobials are discussed in further detail in Chapter 20 . In addition, the book Antimicrobials in Foods (Davidson and Branen, 1993) gives a complete treatment of these additives. Antimicrobials play a major role in extending the shelf-life of numerous snack and convenience foods and have come into even greater use in recent years as microbial food safety concerns have in- creased.

The antioxidants (INS 300–326 and E300–E326), discussed in further detail in Chap- ter 18 , are used to prevent lipid and/or vitamin oxidation in food products. They are used primarily to prevent autoxidation and subsequent development of rancidity and off-flavor. They vary from natural substances such as vitamins C and E to synthetic chemicals such as butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT). The antioxidants are especially useful in preserving dry and frozen foods for an extended period of time.

Antibrowning agents are chemicals used to prevent both enzymatic and nonenzy- matic browning in food products, especially dried fruits or vegetables and are discussed in Chapter 19 of this book. Vitamin C (E300), citric acid (E330), and sodium sulfite (E221) are the most commonly used additives in this category. These additives are classified as either antioxidants or preservatives in the INS system, but retain the same numbers as in the E system without the E.

B. Nutritional Additives

Nutritional additives have increased in use in recent years as consumers have become more concerned about and interested in nutrition. Because of this increased interest, this Nutritional additives have increased in use in recent years as consumers have become more concerned about and interested in nutrition. Because of this increased interest, this

Vitamins, which as indicated above are also used in some cases as preservatives, are commonly added to cereals and cereal products to restore nutrients lost in processing or to enhance the overall nutritive value of the food. The addition of vitamin D to milk and of B vitamins to bread has been associated with the prevention of major nutritional deficiencies in the United States. Minerals such as iron and iodine have also been of extreme value in preventing nutritional deficiencies. Like vitamins, the primary use of minerals is in cereal products.

Amino acids and other proteinaceous materials are not commonly used in foods. However, lysine is sometimes added to cereals to enhance protein quality. Proteins or proteinaceous materials such as soya protein are also sometimes used as nutritional addi- tives, although they are most commonly used as texturizing agents.

Fiber additives have seen increased popularity in recent years with the increase in consumer interest in dietary fiber. Various cellulose, pectin, and starch derivatives have been used for this purpose. Recently, naturally derived fiber from apples and other fruits as well as sugarbeets has been introduced as a fiber additive. Fiber additives are not well defined and in reality have little or no direct nutritional value, although they do have indirect nutritional benefits. In some cases, fiber additives also provide improved texture to food products and are categorized in the INS and E system as bulking agents, thickeners, or stabilizers.

The number of food additives used for special dietary purposes has increased sig- nificantly in recent years with an emphasis on the replacement of fat to reduce calories. United States food processors use more than 16 billion pounds of fat each year (Anony- mous, 1995), and the fat replacement industry grew rapidly in the 1990s, although it appears to have peaked in recent years, and the growth of this sector has subsided (Sloan, 1997). Fat replacers include many texturizing agents and, as indicated in Chapter 11, include carbohydrate-, protein-, and fat-based systems.

The increased interest in nutrition has also led to the rapid growth of the functional food or nutraceutical industry with the development of several additives for the purpose of enhancing overall health. Chapter 10 describes some of the current interest in essential fatty acids as additives, while Chapter 12 describes some of the major additives used in engineering clinical nutritional products.

C. Coloring Agents

Most coloring agents are used to improve the overall attractiveness of the food. A number of natural and synthetic additives are used to color foods. In addition, sodium nitrite is used not only as an antimicrobial, but also to fix the color of meat by interaction with meat pigments. The colors are included in the E system as E100–E180 and in the INS

as 100–182. As indicated in the review of coloring agents in Chapters 16 and 17 , there has been much controversy regarding their use. Although synthetic coloring agents continue to

be used extensively, there has been significant increased interest in natural colorants as

D. Flavoring Agents

Flavoring agents comprise the greatest number of additives used in foods. There are three major types of flavoring additives: sweeteners, natural and synthetic flavors, and flavor enhancers.

The most commonly used sweeteners are sucrose, glucose, fructose, and lactose, with sucrose being the most popular. These substances, however, are commonly classified as foods rather than as additives. The most common additives used as sweeteners are low- calorie or noncaloric sweeteners such as saccharin and aspartame. These sweeteners, as discussed in Chapter 15 , have had a major impact on the development of new foods.

In addition to sweeteners, there are more than 1700 natural and synthetic substances used to flavor foods. These additives are, in most cases, mixtures of several chemicals and are used to substitute for natural flavors. In most cases, flavoring agents are the same chemical mixtures that would naturally provide the flavor. These flavoring substances are discussed in detail in Chapter 13 . The acidulants (see Chapter 21 ), which add a sour taste, often serve other purposes, including preservation.

Flavor enhancers (INS 620–642 and E620–E640) magnify or modify the flavor of foods and do not contribute any flavor of their own. Flavor enhancers, which include chemicals such as monosodium glutamate (E621) and various nucleotides (E626–E635), are often used in Asian foods or in soups to enhance the perception of other tastes. These chemicals are covered in detail in Chapter 14 .

E. Texturizing Agents

Although flavoring agents comprise the greatest number of chemicals, texturizing agents are used in the greatest total quantity. These agents are used to add to or modify the overall texture or mouthfeel of food products. Emulsifiers and stabilizers are the primary additives

in this category and are discussed in more detail in Chapters 23 and 24 . Phosphates and dough conditioners are other chemicals that play a major role in modifying food texture. Phosphates are some of the most widely used and serve a number of functions in foods, as discussed in Chapter 25 .

Emulsifiers (INS 429–496 and, primarily, E431 and E495) include natural sub- stances such as lecithin (INS 322 and E322) and mono- and diglycerides as well as several synthetic derivatives. The primary role of these agents is to allow flavors and oils to be dispersed throughout a food product.

Stabilizers include several natural gums such as carrageenan as well as natural and modified starches. These additives have been used for several years to provide the desired texture in products such as ice cream and are now also finding use in both dry and liquid products. They also are used to prevent evaporation and deterioration of volatile flavor oils.

Phosphates (E338–E343) are often used to modify the texture of foods containing protein or starch. These chemicals are especially useful in stabilizing various dairy and meat products. The phosphates apparently react with protein and/or starch and modify the water-holding capacity of these natural food components.

Dough conditioners such as steroyl-2-lactylate and various humectants such as so- dium silicoaluminate are also used as texturizing agents under very specific conditions.

F. Miscellaneous Additives F. Miscellaneous Additives

III. BENEFITS OF ADDITIVES

There are obviously many recognized benefits to be derived from additives. Some of the major benefits are a safer and more nutritious food supply, a greater choice of food prod- ucts, and a lower-priced food supply.

A. Safer and More Nutritious Foods

There is no question that the preservative and nutritional additives used in foods increase the safety and overall value of many food products. The use of several antimicrobials is known to prevent food poisoning from various bacteria and molds. Antioxidants, used to prevent the development of off-flavors, also prevent the formation of potentially toxic autoxidation products and maintain the nutritional value of vitamins and lipids. As indi- cated in Chapter 9 , the use of various nutritional additives such as vitamins is also of proven value in preventing nutritional deficiencies.

B. Greater Choice of Foods

Most major supermarkets today carry more than 20,000 food items, providing the con- sumer a wide choice of food products. The availability of additives has allowed the produc- tion of numerous out-of-season foods and a variety of new food products. Additives have increased the development of convenience foods, snack foods, low-calorie and health pro- moting (functional) foods, exotic foods, and a variety of food substitutes. Convenience has been built into TV dinners and breakfast cereals as well as several microwave products. Additives allow these foods to be pre-prepared and still maintain acceptable flavor, texture, and nutritional value. Although many of these foods can have added convenience through the use of new packaging approaches or other processing methods, most depend on preser- vatives and texturizing agents. It is estimated that the shelf life of cereal products can be increased over 200% by the use of antioxidants (Branen, 1975).

The snack food industry has continued to be successful because the use of coloring and flavoring additives make available a wide array of snack items. These items, which are commonly subjected to high-temperature processing and are expected to have an ex- tended shelf-life, also contain preservatives.

The increased interest by consumers in dieting has resulted in a proliferation of low- calorie food items. The use of saccharin and cyclamates opened the market for various food products with reduced calories, and by 1996 these and other calorie reduction agents became the highest selling category of food additives at $1.3 billion annually (Anonymous, 1995). These sweeteners are now being phased out, being at least partially replaced with aspartame, yet remain the primary additives used in low-calorie foods. However, many emulsifiers and stabilizers have allowed a reduction in the lipid content of foods, thus also lowering calories. As is noted in Chapter 11 , fatty acid esters of sucrose have allowed an even greater reduction in the lipid content of foods. Coloring and flavoring agents have enhanced the appeal of these foods to consumers.

Stabilizers, emulsifiers, and coloring and flavoring additives have also allowed de- velopment of a number of food substitutes, especially dairy and meat substitutes. Marga- Stabilizers, emulsifiers, and coloring and flavoring additives have also allowed de- velopment of a number of food substitutes, especially dairy and meat substitutes. Marga-

The greatest increase in food additive use in the next several years is likely to be in the functional food and nutraceutical industry. Several recent publications have noted the rapid growth of this industry both in the United States and in Europe (Hollingsworth, 1999; Sloan, 1997, 1998, 1999). Although definitions vary, Sloan (2000) defines a func- tional food as a food or beverage that imparts a physiological benefit that enhances overall health, helps prevent or treat a disease/condition, or improves physical or mental perfor- mance via an added functional ingredient, processing modification, or biotechnology. When low-calorie and fatfree foods are included, the total market for functional foods exceeds $92 billion in the U.S. and is expected to grow annually 6–10% (Sloan, 1999, 2000). Although lower fat content continues to be a major factor that motivates food purchase decisions, overall good health is now the major motivating factor (Sloan, 2000).

C. Lower-Priced Foods

Although there have been few recent studies to indicate that additives reduce the overall price of foods, a study reported in 1973 (Angeline and Leonardos, 1973) indicated that, at least for some processed foods, total removal of additives would result in higher prices. This study was based on the premise that the consumer would still desire the same type of foods in the absence of additives. The researchers reported that if, for example, additives were removed from margarine, consumers would have no alternative but to purchase a higher-priced spread such as butter, which usually contains few or no additives. They also reported that if additives were removed from bread, franks, wieners, and processed cheese, new processing procedures, increased refrigeration, and improved packaging would be required, at a higher cost, to keep the same type of products available. In 1973 prices, it was estimated that a consumer of sandwich fixings, including bread, margarine, franks or wieners, and processed cheese, would pay an additional $9.65 per year if additives were not available.

Although packaging or processing procedures could be developed to replace the need for additives, in most cases processing or packaging alternatives are not as cost- effective as the use of additives. However, it must be recognized that some of the additive- containing foods could be replaced in the diet with foods free of additives. It is also important to realize that the assumption that food additives lower the price of foods is based on maintaining the same type and quality of foods that we currently have available. Without additives, we could still have an excellent food supply at a reasonable cost. How- ever, to provide consumers with the variety of foods along with the other benefits men- tioned, would cost more without additives.

IV. RISKS OF ADDITIVES

Despite the benefits attributed to food additives, for several years there have also been a number of concerns regarding the potential short- and long-term risks of consuming these substances. Critics of additives are concerned with both indirect and direct impacts of using additives. As for many of the benefits mentioned, there is not always adequate scientific proof of whether or not a particular additive is safe. Little or no data are available concerning the health risks or joint effects of the additive cocktail each of us consumes

The indirect risks that have been described for additives are the converse of some of the benefits attributed to their use. While it is accepted that through additives a greater choice and variety of foods have been made available, there is no question that additives have also resulted in the increased availability of food products with a low density of nutrients. These so-called junk foods, which include many snack items, can in fact be used as substitutes in the diet for more nutritious foods. Recently the food industry has attempted to address this criticism by adding nutritional additives to snack items so that these foods are a source of selected vitamins and minerals. The long-term effectiveness of this is questionable. Obviously, educational programs are needed to ensure that consum- ers select nutritious foods. Some scientists, however, feel that there is a place in the diet for foods that provide pleasure even if no direct nutritional benefit can be ascribed to their consumption.

Of greater concern than the indirect risks are the potential direct toxicological effects of additives. Short-term acute effects from additives are unlikely. Few additives are used at levels that will cause a direct toxicological impact, although there have been incidents where this has happened. Of particular concern are the hypersensitivity reactions to some additives that can have a direct and severe impact on sensitive individuals even when the chemicals are used at legally acceptable levels. The reactions to sulfites and other

additives, as described in Chapters 3 – 5 of this book, are examples of such a problem. With proper labeling, however, sensitive individuals should be able to avoid potential allergens.

Toxicological problems resulting from the long-term consumption of additives are not well documented. Cancer and reproductive problems are of primary concern, although there is no direct evidence linking additive consumption with their occurrence in humans. There are, however, animal studies that have indicated potential problems with some addi- tives. Although most of these additives have been banned, some continue to be used, the most notable being saccharin.

Most existing additives and all new ones must undergo extensive toxicological eval- uation to be approved for use. Although questions continue to be asked regarding the validity of animal studies, there is a consensus among scientists that animal testing does provide the information needed to make safety decisions. The procedures used for this evaluation and the current philosophy regarding safety testing are outlined in Chapters 2 and 3.

V. BALANCING RISKS AND BENEFITS

Due to the difficulties in precisely defining the risks and benefits of individual additives,

a legal rather than a scientific decision is commonly made regarding the safety of a food additive. In such a decision, the potential risks must be weighed against the potential benefits. A common example of this balance is saccharin. Although there is no direct evidence that saccharin, in the low amounts consumed in foods, causes cancer in humans, risk evaluation in rats indicates a potential for cancer in humans. On the benefit side, saccharin is an excellent noncaloric sweetener that is useful for diabetics and those inter- ested in reducing consumption of calories. Many consumers feel that the benefits of having saccharin available as a sweetening agent outweigh the risks. On the basis of available risk information, however, the FDA initially issued a ban on saccharin in the early 1970s. The U.S. Senate, recognizing the consumer demand for low-calorie foods, subsequently a legal rather than a scientific decision is commonly made regarding the safety of a food additive. In such a decision, the potential risks must be weighed against the potential benefits. A common example of this balance is saccharin. Although there is no direct evidence that saccharin, in the low amounts consumed in foods, causes cancer in humans, risk evaluation in rats indicates a potential for cancer in humans. On the benefit side, saccharin is an excellent noncaloric sweetener that is useful for diabetics and those inter- ested in reducing consumption of calories. Many consumers feel that the benefits of having saccharin available as a sweetening agent outweigh the risks. On the basis of available risk information, however, the FDA initially issued a ban on saccharin in the early 1970s. The U.S. Senate, recognizing the consumer demand for low-calorie foods, subsequently

Concern regarding the safety of additives has declined in the United States since the enactment of the saccharin moratorium. As noted in Chapter 6 , a 1997 study indicates that only 21% of supermarket shoppers were concerned about additives and preservatives,

a significant decline from a 1987 study. Reduced consumer concerns plus a changing political environment away from consumerism and a move toward more responsible use of additives by manufacturers have lessened the controversy over additive use. The poten- tial risks of additives are well recognized, but the beneficial role these additives play in food production, processing, and utilization are also felt to be essential to the maintenance of our current food systems. With the convenient, tasty, and nutritious foods demanded, or at least desired, by consumers and the increasing overall demand for foods as populations increase, food additives will continue to play an important and essential role in food pro- duction. There will, however, continue to be concern regarding the potential risks associ- ated with long-term consumption of small amounts of these chemicals and the possible interactive toxicological effects. As methods improve for evaluating these toxicological effects, some additives may be banned. At the same time, the same information may be used to develop safer new additives or techniques for using existing additives in a way that will lessen risk.

New technology is likely to have a profound impact on the use of food additives in the future. Of these, recombinant DNA biotechnology may have the greatest effect on the future development and use of food additives. Recombinant DNA biotechnology is already routinely used for production of additives through bioprocessing, including or- ganic acids, bacteriocin preservatives, enzymes, microorganisms, vitamins, and minerals (Institute of Food Technologists, 2000). Biotechnology may also decrease the need for food additives. Plants have been produced through recombinant DNA with increased shelf- life and nutritional value, thus decreasing the need for a variety of additives. Although it is expected these recombinant DNA methods will be accepted in the future, there are currently several questions being raised regarding the risks and benefits of these products as well.

VI. LEGAL QUESTIONS

The final decision regarding additive use will most likely fall on governmental agencies that will evaluate the available information of potential risks and benefits to reach informed decisions. There is a need to harmonize these legal decisions on a worldwide basis, espe- cially with the continued increase in movement of processed foods between countries. Most likely these decisions will be specific to each country and depend on the perceived benefits, which may vary from country to country. It is hoped that research can continue to better define safety evaluation procedures and their interpretation. Research will also

be needed to better define the use and benefits of additives and identify possible alterna- tives. Existing laws in most countries appear to reflect consumer concerns and provide adequate protection. Chapters 7 and 8 provide excellent overviews of major laws govern- ing use of additives in the United States and Europe. As research results become available,

Informed scientists, food producers and consumers, and legal authorities need to continue to meet on a worldwide basis to develop strategies for addressing concerns re- garding additive use. It is doubtful that the interest in the wide variety of foods made available with additives will decline in the future. With the expected continued increase in per capita income on a worldwide basis, the demand for a variety of convenient foods will also continue to increase. At the same time, as consumers become better educated, they may also want less risk. In rare cases, decisions will be left to consumers, but most will be made by legal authorities. We undoubtedly will continue to live in a society in which additives are a way of life.

REFERENCES

Angeline, J. F., Leonardos, G. P. 1973. Food additives—some economic considerations. Food Tech- nol . April:40–50. Anonymous, 1995. Food Additives: U.S. Products, Applications, Markets. Technomic Publishing, Lancaster, PA. Branen, A. L. 1975. Toxicology and biochemistry of butylated hydroxyanisole and butylated hy- droxytoluene. J. Am. Oil Chem. Soc. 52:59. Codex Alimentarius Commission. 2001. Class names and the international numbering system for food additives. Codex Alimentarius: Vol. 1A—General Requirements. www.fao.org/es/esn/ codex/standard/volume 1a/vol la 1a_e.htm .

Davidson, P. M., Branen, A. L. 1993. Antimicrobials in Foods. Marcel Dekker, New York. Hanssen, M. 1984. E for Additives: The Complete ‘‘E’’ Number Guide. Thorsons Publishers,

Wellingborough, England. Hollingsworth, P. 1999. Keys to Euro-U.S. food product marketing. Food Technol. 53(1):24. Institute of Food Technologists. 2000. IFT expert report on biotechnology and foods: benefits and

concerns associated with recombinant DNA biotechnology-derived Foods. Food Technol. 54(10):61.

Jukes, D. 2001. Food Additives in the European Union. 2000. Food Law. www.fst.rdg.ac.uk/ foodlaw/additive.htm . Sloan, A. E. 1997. Fats and oils slip and slide. Food Technol. 51(1):30. Sloan, A. E. 1998. Food industry forecast: consumer trends to 2020 and beyond. Food Technol.

52(1):37. Sloan, A. E. 1999. Top ten trends to watch and work on for the millennium. Food Technol. 53(8): 40. Sloan, A. E. 2000. The top ten functional food trends. Food Technol. 54(4):33.

Food Additive Intake Assessment

SEPPO SALMINEN

University of Turku, Turku, Finland, and Royal Melbourne Institute of Technology, Melbourne, Australia

RAIJA TAHVONEN

University of Turku, Turku, Finland

I. INTRODUCTION

Food additives are special chemicals that are added into our food supply on purpose, and they are meant to be in the food at the time of consumption. Both international organiza- tions and local governments generally evaluate the safety of food additives. The goal of local assessment is to take into account local food supply and cultural differences in dietary habits that may influence the intake of food additives.

Food additive intake assessment has three major goals:

1. Monitoring the intake of chemicals and relating it to the acceptable daily intake (ADI) values

2. Identifying consumer groups that may be at risk for food additive intake close to or higher than the ADI values

3. Provide information for the regulatory bodies for reassessing the food additive regulations in case of high intake in all or some consumer groups

The major aim of the intake assessment is to protect consumer health and to assist in developing food additive regulations.

II. SCOPE AND PURPOSE OF FOOD ADDITIVE INTAKE ASSESSMENT

The safety evaluation of food additives is based on the assessment of toxicity of the chemi-

Table 1 Toxicological Classification of Food Additives Based on Available Safety Data Group A

Substances with an established ADI value Group B

Substances generally regarded as safe Group C

Substances with inadequate data Group D

Flavoring components Group E

Natural components used as additives without any scientific safety data or with very limited data

extent to which ADI values may be exceeded. The ADI value is determined by the Joint FAO/WHO Expert Committee on Food Additives (JECFA). The ADI value defines an estimate of the amount of food additives, expressed on a body weight basis, that can be ingested over a lifetime without appreciable health risk (WHO, 1987). This sets the limits on food additive intake assessment and also the goals for the assessment procedures. Sev- eral approaches can be taken, and food additives can be classified based on the safety data (Table 1). However, it is important to assess the use of specific components, whether additive or ingredient, and to define the safety data needed for each purpose. This then forms the basis for intake assessment and risk evaluation.

III. REGULATION OF MAXIMUM LEVELS OF FOOD ADDITIVES

In the United States, the principle of certain food components being generally recognized as safe (GRAS) was established in early legislation and later rigorously defined to include scientific evidence (Wodicka, 1980). Other regulations are included in the Code of Federal Regulations on specific food additives. In the European Union, three major directives regulate the use of food additives in member countries. In Australia and New Zealand, the regulation is similar to that in Europe. In Japan, differences exist in the regulation of all additives from natural sources, while strict regulation concerns chemical additives.

Usually the regulation is based on the acceptable daily intake values determined by the JECFA. This committee started establishing the ADI values as early as 1956. The ADI value is not an exact figure or a mathematical value, but it gives an estimate of how much of a chemical can be relatively safely ingested daily by normal consumers. The ADI value is not related to the intake over a few days or even a few months, but rather the daily intake over a lifetime (Fondu, 1992). It has also been pointed out that the ADI value, which depends on a series of factors, is not a constant regulatory number. Rather it is a guide serving to calculate the acceptable limits of different chemical agents incorpo- rated into our food supply (Truhaut, 1992; Fondu, 1992). As the use of food additives is generally regulated on the basis of the ADI values, the applicability of the ADI value has been discussed especially in terms of infants and children. These studies were brought together by ILSI Europe in a consensus meeting (Clayton et al., 1998). It was recom- mended that special ADIs should not be created for infants and children, but rather that special sensitivity should be taken into account when assessing and defining individual ADI values (Clayton et al., 1998; Larsen and Pascal, 1998).

IV. EUROPEAN REGULATORY SITUATION

The assessment of food additive intakes has become increasingly important in Europe The assessment of food additive intakes has become increasingly important in Europe

The European approach was extensively discussed in a workshop organized by ILSI Europe concerning the scientific assessment of the regulatory requirements in Europe (Howlett, 1996; Wagstaffe, 1996). This meeting also summarized the different approaches taken by many European countries to fulfill the commission requirements on monitoring food additive intakes (Penttila¨, 1996; Cadby, 1996; Lawrie and Rees, 1996; Verger, 1996). Also, the differences between various additive groups were discussed, and special attention was given to flavoring substances. As there are only a few hundred additives, there are over 2000 flavoring substances and this poses special problems for intake assessment (Cadby, 1996). Methods for assessing the intake of flavor componds have been compared by Hall and Ford (1999).

V. METHODS OF ESTIMATING DIETARY INTAKE OF ADDITIVES

Methods of estimating dietary food additive intake can be classified as either one-phase or two-phase. A one-phase method uses information from one data source, usually con- cerning food additive production and usage. A two-phase method combines information from two data sources; these usually concern additive concentrations in foods and food consumption. In the latter case the investigator is required to decide how to combine the two different types of data to estimate the food additive intake. The sources of information and methods available to the investigator of dietary food additive intakes are summarized.

Figure 1 shows a diagram of information sources concerning food additives and food consumption. Various methods of additive intake estimation are based on the data sources written in capital letters. One-phase methods require information about food addi- tive production and usage. Two-phase methods require information about food additive concentrations in foods and food consumption. In duplicate meal studies, both these types of information are collected from one source, which increases the accuracy of this method of intake estimation over the others.

There is an inherent inaccuracy in all the methods of estimating food additive in- takes, and this varies in degree between methods of calculation. Each method is based on different assumptions. For example, in one method all food items are assumed to con- tain the maximum permitted food additive concentrations, which results in overestimated intake values. Several methods employ two different data files, one on food consumption levels and the other on food additive concentrations in various food items. Usually investi- gators do not collect all the information they require, but use other food surveys to supply the missing information. In combining data from different sources, they incorporate the

Figure 1 The methods used for assessing the intakes of food components and food additives. The combination of the methods from top down indicate increasing cost and accuracy in intake assessment.

carefully assessed. A critique of dietary survey methodology has shown that large random and systematic errors can be introduced into inadequately planned experimental protocols (Bingham, 1985). For example, when participating subjects estimate the weight of foods consumed by methods other than by weighing, the errors involved can vary from 20 to

50. Also the responses of subjects asked to recall their food intake during the previous

24 hours have been shown to vary by between 4 and 400% of observed intakes. This means that experimental protocols need to be carefully designed to provide the most accurate information from the data to be collected. The methods applied to assess the intake of contaminants, additives, and nutrients have also been reviewed by Petersen and Barraj (1996). Dietary assessment methodologies have also been critically reviewed and recom- mendations made by a Nordic compendium (Lennerna¨s, 1998; Berglund, 1998; Goldberg and Black, 1998; Hambraeus, 1998).

A. One-Phase Methods

1. Estimation Based on Production and Foreign Trade One estimation of food additive usage in a given country can be calculated from the 1. Estimation Based on Production and Foreign Trade One estimation of food additive usage in a given country can be calculated from the

Examples of this type of calculation have recently been published (Conning, 1986). Estimates of the annual utilization of additives in the United Kingdom were combined with the assumptions that 32% of the UK production is exported and that 20% of food purchased is wasted. The calculation resulted in an estimated intake of 0.53 g of synthetic additives per person per day, or 440 ppm in a daily diet of 1200 g. It was suggested that this food additive intake would comprise preservatives, antioxidants, colors, emulsifiers, and phosphates. A similar calculation was used to estimate the dietary concentration of flavors to be of the order of 0.003 ppm.

2. Surveys of Food Industry Usage Records of the purchase or use of food additives by the food industry can be collected

to estimate the total usage of additives. Dividing this amount by the number of consumers yields an estimate of the average intake of that food additive. The U.S. Food and Drug Administration (FDA) has sponsored a number of surveys on the use of food additives; the third was completed in 1977. The National Research Council undertook this survey to elicit information concerning the concentrations of additives while determining which additives were used in the products available to consumers. The potential food additive intakes were then calculated using this data file in combination with estimates of food consumption (National Research Council, 1979). Furthermore, the survey committee di- vided each major category into approximately 200 subcategories, which were in turn sub- divided into food classes. This classification system enabled the surveyors to ensure that

a sufficient number of responses were collected for each category. About 2500 companies received the survey questionnaires, and about 500 returned them. It was clear from the information obtained that the 500 respondents included 71 of the ‘‘Top 100’’ food compa- nies in the United States. The investigators assumed that companies generally use similar concentrations of food additives in the same types of food. Thus the results of the survey were considered to represent the national diet adequately and therefore to provide reasonably accurate and useful estimates of food additive levels entering the U.S. food supply. The calculated intake estimates for each additive were derived using the following data: the average concentrations (weighted mean use levels) of additives in food, based on the re- sponses from the manufacturers, the frequency with which foods containing each additive were eaten each day, and the average amount of foods (weight) consumed on each occasion.

The individual average daily intake of each additive was estimated by multiplying the total substance concentration used in foods of each category by the average consump- tion of such foods on each eating occasion. This figure was in turn multiplied by the frequency with which the foods were consumed over the 14-day survey period, and the result was subsequently divided by 14 to obtain the average daily intake. The investigators commented that it was likely that almost all the additive intakes calculated in this way would be overestimated. Therefore the results should be regarded as the upper limits of

a potential range of intakes. In Finland, questionnaires were circulated to the members of the Association of Food Industry, who were asked to record both their use of food additives

Table 2 Food Additives Intakes: Comparison of Two Analysis Methods in Finland

Use of additives by the food industry

Food control analysis Additive

(mg/person/day) Sodium benzoate

(mg/person/day)

46 40 Sorbic acid

2.3 6.6 Source : Penttila¨, 1996.

Government authorities conducted another study (Penttila¨ 1996) during the same period. Although the two studies reported variations in the estimated intakes (Table 2), explana- tions for these differences could be suggested without invalidating the experimental method of either report. However, in most countries, no valid information on food additive intake is available.

In a method described by Buchet and Lauwerys (1983), the housewife was the refer- ence person who stored duplicate meals of all the foods she ate. The sample meals were stored in the domestic refrigerator until they were transported to the laboratory for analysis. The use of duplicate meal samples from large catering establishments, such as hospitals, may also be considered because of the ease with which the preparation and collection of meal samples can be organized. The number of analyses required for each food additive is equal to the number of duplicate meals collected and increases when the experimental design includes representative sampling of population subgroups (socioeconomic groups, ethnic minorities, etc.) across seasons and different geographical areas. Although difficul- ties are encountered when adopting the analysis of duplicate meals to estimate the additive intakes, there are several advantages to using this experimental protocol. In the first in- stance, it provides accurate estimates of food additive intakes of individuals and realistic variations of these intakes. Also, food consumption data and analytical data on single food items are unnecessary for such a study. This experimental procedure could therefore be used to assess the accuracy of the other methods of estimating food additive intakes and thus be used to validate their results. Furthermore, a duplicate meal study can be used to ascertain whether any significant losses or gains in additive content of foods occur during the preparation of those foods in the home.

B. Two-Phase Methods

1. Assumption of Maximum Permitted Levels This method is based on the assumption that the concentration of the food additive in each

food item is the permitted maximum. The estimated daily intake is therefore calculated by multiplying the maximum permitted level (mg/kg) of additive in the food item by the average consumption of various food items (g/day). This calculation is carried out for all

(mg/day) for that food additive. Information on food consumption data of population subgroups of special concern, such as children or an ethnic minority group can be added. This enables the estimation of potential additive intakes of special population subgroups. This survey method was used by WHO to investigate the potential intakes of 54 additives; only three were found to exceed their corresponding ADIs. The use of a different list of maximum permitted additive concentrations resulted in nine additives exceeding their ADIs (Lu, 1973). Subsequent studies (Toyoda et al., 1983; van Dokkum et al., 1982; Niemi et al., 1982; Gergely, 1980) using analytical methods revealed that actual consumption of various food additives rarely exceeds their ADIs. This leads to the conclusion that the use of maximum permitted additive concentrations in food overestimates the intakes because additives are seldom used in all the foods for which they have been approved. Even when they are used, their concentrations are usually below the maximum permitted levels. How- ever, improvements in analytical techniques and analytical capacity now enable the deter- mination of food additive concentrations in duplicate meals or food samples. Therefore the maximum permitted additive concentrations method should be used for the estimation of additive intakes only when analyzed additive concentrations are not available. Recently, in Italy the maximum theoretical intake estimates of antioxidants showed a potential excess intake of BHT, but not other antioxidants (Leclerq et al., 2000).

2. Market Basket Method The market basket method, also referred to as a total diet study, uses selection of food

items representing the typical pattern of food consumption. The foods are purchased from retail shops and are then prepared and cooked (with herbs, spices, dressings, etc.) to incor- porate the usual gains and losses of food additives. The food items are then sorted into food groups (cereals, fats, fish, fruits, meats, oils, vegetables, etc.), and the items of each group are combined according to the proportions in which they occur in the diet as indi- cated by food consumption statistics. The foods in each group are then homogenized, and the concentrations of food additives in the homogenate are determined. The daily food additive intake for the food group is estimated by multiplying the measured concentrations of additives by the average consumption of foods in that food group. An estimate of the total intake is the subsequent sum of the intakes calculated for each food group. It is not appropriate to use the market basket study for estimating the intake of food additives with very limited use because the additive concentrations in food groups are diluted and may fall below the analytical detection limits. The use of this method could therefore underesti- mate the food additive intakes. The accuracy of the information obtained in this type of study is dependent on the initial construction of the typical diet. In addition, the accuracy of these estimates of additive intakes for the population of a country can be increased by taking into account any regional differences in food consumption (Toyoda et al., 1985), by using the different types of foods available in different shops and by incorporating seasonal variations of food consumption into the study (van Dokkum et al., 1982). These studies are conducted in several places around a country and several food items and brands are included. A good case example of these studies has been conducted in the United States for the Food and Drug Administration. The so-called FDA total diet studies have been reported from 1989 to 1991 assessing dietary intakes of pesticides, selected elements, and other industrial chemicals (Gunderson, 1995).

The market basket method is a broad concept that can be applied to many different approaches to the study of food additive intakes. The intake estimates may concern the The market basket method is a broad concept that can be applied to many different approaches to the study of food additive intakes. The intake estimates may concern the

VI. FOOD CONSUMPTION DATA

A. Food Balance Sheets

National food balance sheet studies coordinated by the United Nations Food and Agricul- ture Organization (FAO) include figures concerning the national production, import, ex- port, and net consumption of food and the contribution of each food to the intakes of energy, protein, and fat (FAO, 1984). The per capita consumption is calculated by simply dividing the net consumption of food by the total population. The use of food balance sheets in studies estimating food additive intakes is inadequate for a number of reasons. The balance sheet studies do not provide information about the variations in food con- sumption caused by different dietary patterns. The list of food commodities (50–100 items) is limited (for instance, household surveys contain 250–300 items) and does not cover all the items containing food additives; the list basically contains raw food materials. For example, the consumption of cereals such as wheat and barley is presented, whereas the consumption of bread, cakes, and biscuits is not. Processed foods are included only when their foreign trade is of particular interest; for example, processed fruits, cheese, and butter are included in the balance sheets for Finland (FAO, 1984).

The use of food balance sheets in the estimation of food additive intakes should therefore be restricted to studies of food items that are included in the national balance sheets. If crude estimates of the intakes of food contaminants (such as heavy metals and pesticides) are required, then food balance sheets could be used.

B. Household Surveys

Household surveys are conducted in most western countries primarily for estimating ex- penditures on food and certain other commodities to enable the calculation of economic indices. This survey requires the householder to keep a diary on food purchases, but these records do not provide information about the foods actually consumed. It is assumed that the members of the household subsequently consume all the foods that are purchased. The number of households participating in the surveys varies; in Finland the number is approximately 10,000 for a population of 5 million (Central Statistical Office of Finland), whereas in Great Britain the sample is approximately 7500 (MAFF, 1986) for a population Household surveys are conducted in most western countries primarily for estimating ex- penditures on food and certain other commodities to enable the calculation of economic indices. This survey requires the householder to keep a diary on food purchases, but these records do not provide information about the foods actually consumed. It is assumed that the members of the household subsequently consume all the foods that are purchased. The number of households participating in the surveys varies; in Finland the number is approximately 10,000 for a population of 5 million (Central Statistical Office of Finland), whereas in Great Britain the sample is approximately 7500 (MAFF, 1986) for a population