sign in sign up
<<
Home , Colonial molasses trade, Molasses Act, Soft drink, Sugar Act

Sweetener usage in America: A brief history and current usage patterns

SD Stellman American Cancer Society , NY

In Williams GM (ed): Sweeteners: Health Effects. Princeton, NJ, Princeton Scientific Publishers, 1988, pp. 1-18. SWEETENER USAGE IN AMERICA: A BRIEF HISTORY AND CURRENT USAGE PATTERNS

Steven D. Stellman American Cancer Society New York, New York

HISTORY

Sweeteners are deeply embedded in human history and culture. The Children of Israel were promised a land flowing with milk () and (invert ), and the Bible abounds with other references to . Figure 1 shows a prehistoric cave-drawing from Valencia, Spain, dating to about 8,000 B.C. which depicts people raiding beehives to obtain honey (McGee, 1984). The ancient Athenians made dumplings from honey, sesame flour, , and oil (Robbins, 1948). Other ancient peoples also made use of fruits and nuts as sources of sweetness. The Romans knew how to press dried currants and almonds to make nougat. Sugar as we know it, however, as sweetening agent (rather than a food component) came relatively late in human history. Said to be native either to India or New Guinea, the plant spread to the South Pacific by 1000 B.C., and was introduced into the Western world by Arabs between the seventh and tenth centuries A.D. In the twelfth century Venice became the Mediterra­ nean center of sugar refining and trade. The skill of Venetian sugar refiners was legendary. Ramazzini wrote, in 1713, "It is the custom to preserve with sugar as table delicacies and for other purposes various kinds of seeds such as almonds, pistachio-nuts, pine-nuts, the seeds of fennel, coriander, and wormwood and also fresh fruits" (Ramazzini, 1940). Ramazzini also devoted an entire chapter of De Morbis Artificum to describing their illnesses. Marco Polo brought reports of sugar refining back from China, where its use is thought to date back to about 100 B.C. It is well known that brought tobacco back to Europe with him on one of his four voyages to the New World. Less well known is that he took sugar cane across with him in 1493, on his second voyage, thereby beginning one of the most interesting- and terrible- sagas in agricultural history. This tropical crop flourished in the West Indian islands. However, cultivation was exceptionally labor-intensive and refining procedures were extremely crude. Sugar was obtained by cutting the canes in the field, crushing i t '

FIGURE I.Reduction and detail from a painting found in the Spider Cave at Valencia, Spain (c. 8000 B.C.). The leader (enlarged at right) may be carrying a basket for the honeycomb. Source: McGee (1984). Reprinted with permission of Harold McGee and Gordon Press Publishers, Inc.

them with heavy weights, and rolling them to press out the , from which the could be boiled off and the sugar solids crystallized out. Figure 2 shows a sixteenth century depiction of this procedure by the Flemish-Italian painter Jan Van der Straet (Stradanus). This was a time-consuming, low-yield process which made sugar a high-priced, near-luxury commodity. The high labor demand was satisfied through importation and exploitation of generations of slaves. Sugar cane growing and harvesting was a pillar of the West Indian and American slave economy for a great portion of our early history. Despite its price, demand for sugar was sufficiently high that the British Parliament, smarting from the effects of competition from French and Spanish Colonies, passed the Act of 1733. This imposed a duty of sixpence per gallon on importation of non-British molasses into the Colonies, which they hoped would force the Americans to buy molasses originating in British . The Act was largely ineffective, due to limited enforcement. Thirty years later, the of 1764 was passed, which cut the import duty to threepence per gallon, but which promised rigorous enforcement. This new British atrocity was seized upon by rebellious colonists as another instance of Taxation without Representation (Schapsmeier and Schapsmeier, 1975). In the late eighteenth century, it was discovered that sugar identical to that from cane could be derived from the sugarbeet plant. During the Napoleonic Wars, all shipments of sugar to France from the Caribbean were cut off. In 1811 Napoleon awarded the Cross of Honor to Benjamin Delessert for developing a method of crystallizing sugar from sugar beets, and decreed that 79,000 acres of French soil be planted, that six experimental stations be established, and that

2 FIGURE 2. Sugar making in the 16th century, as depicted by Jan Van der Straet (Stradanus) in his book Nova Reperta (New Inventions). The sugar cane is shown growing outdoors (right, rear). It is cut (foreground), then chopped into lengths suitable for grinding in the -driven mill (left). is collected, heated in vats (upper right) and is finally poured into "sugar loaf' molds to crystallize into sugar. (Courtesy of the Burndy Library, Norwalk, en.

one million francs be appropriated for commercial development of sugarbeets. Sugarbeets are now the second major commercial source of sugar, next to sugarcane. The modern political economy of sugar probably dates to J 813, when the British inventor Edward Howard developed the vacuum pan, which allowed lower-temperature evaporation of water, and reduced thermal decomposition of the . This also cut the processing time considerably. In 1852, centrifu­ gation was developed by Sir Henry Bessemer, which led to high-volume, high-yield production of sucrose (Hugill, 1979). During the nineteenth century, methods were developed for hydrolyzing starch, mainly from potatoes, to sugar. Towards the end of the century, corn supplanted potatoes as the source of starch, and the was born. During the middle of the present century Japanese chemists developed efficient methods for producing from corn starch, and in 1967 high-fructose corn syrup (HFCS) was introduced into the . Sugar prices and, ultimately, consumption, have been regulated by the Federal Government almost continuously since 1789, when the first tariff of I

3 cent per pound on and 3 cents on loaf sugar were imposed. Tariffs, import taxes and quotas, price supports, and even bounties for domestic growers have been implemented at various times throughout our history. In this century the Depression brought the Sugar Act of 1934, and those of 1937 and 1948 which, with later amendments, sought to preserve domestic ability to supply much of the country's needs (Economic Research Service, l 984a). Though the last Sugar Act expired in 1974, price supports and import duties have been continued administratively. Government protection of sugar prices may yet be sugar's undoing. Since corn sweeteners cost less to produce, high sugar prices allow corn sweetener manufacturers to undercut their sugar competitors. The shift from refined sugar to corn sweeteners documented below is a direct result of this continuing economic warfare.

TYPES OF

All sugars are , that is, organic molecules containing carbon, hydrogen, and oxygen. There are many different families of sugars, distin­ guished by the number of carbon atoms in the basic molecules, and within each family by the stereochemical configuration of the atoms with respect to each other. It is the hexoses (six-carbon sugars) which make foods sweet. The simplest common sugars are the isomeric compounds (or dextrose) and fructose (or levulose ), which are , that is single-molecule enti­ ties. Sucrose is a composed of one molecule of glucose bonded to one molecule of fructose. Two other are , which occurs in many plants (and which is formed in our mouths by the action of ptyalin on the starch we eat), and lactose, or milk sugar. In the body, and in many foods (such as carbonated sodas), sucrose decom­ poses into glucose and fructose (Morrison and Boyd, 1983):

C12H22011 + H20 = C6H1206 + C6H1206 Sucrose+ Water= Dextrose+ Levulose

a process known as inversion because of its effect on plane-polarized light. The resultant mixture of glucose and fructose is known as invert sugar. Most refined sugar products are sucrose-based. Table I shows the composi­ tion of three of the commonest products. from the grocer's shelf is 99.8% sucrose, with trace amounts of moisture and monosaccharides. Brown sugar contains slightly more moisture and about 4% glucose and fructose. Molasses is an aqueous solution of 35% sucrose, 20% invert sugar, and a small quantity of minerals. Honey (Table 2) is an acidic, aqueous solution of varying amounts of mainly monosaccharides (typically about 69% fructose and glu­ cose), along with a small amount of sucrose and other disaccharides. Maple

4 TABLE 1 Composition of Sugar Products Other Compounds Glucose Carbo- (Proteins, Sucrose Fructose hydrates Water Minerals Amino ) Percent White Sugar 99.8 0.05 0 0.1 0.02 Brown Sugar 92 4.0 0 3.5 0.5 Molasses 35 20 6 20 12 5

Source: McGee, 1984.

syrup is an economically minor but highly popular sweetener (Table 3) which contains about 62% sucrose and small amounts of invert sugar (The , 1984).

CONSUMPTION OF SUGARS: DISAPPEARANCE DATA

There are two main approaches to characterization of consumption which are commonly used: estimation of per capita consumption of total or specific sugars in the population as a whole, and estimation of consumption of sugars by individual persons, as determined by food surveys. Each of these methods has its drawbacks, and neither is entirely satisfactory. At best, they provide com­ plementary estimates about where sugar goes and who eats how much of it. We turn first to estimation of per capita consumption. This process is the responsibility of the Economic Research Service of the United States Depart-

TABLE2 Composition of Honey Percent Water 13-23 Fructose 27-44 Glucose 22-41 Sucrose 0.25-7.5 Maltose 3-16 Higher Sugars 0.13-13 Acids, Minerals, Undetermined Substances 0-13 pH= 3.9

Source: The Sugar Association (1984).

5 TABLE 3 Composition of Percent Percent Sucrose 62 Glucose, Fructose Water 35 Malic

ment of Agriculture, which issues a period bulletin called Sugar and Sweetener Report. This complex procedure may be roughly described as the adding up of domestic production, imports, and inventories, followed by deduction of exports, year-end leftovers, military consumption, feed and seed allotments, and other production. Wastage and spoilage are also estimated and deducted. The net result is said to be the quantity which has "disappeared" into the civilian food supply during the year, and, when divided by the mid-year population, yields the "per capita disappearance," which is regarded as an estimate of consumption (USDA, 1972). With these limitations in mind, Figure 3 shows the estimated U.S. per capita consumption of sugar in its major forms: sucrose, dextrose, and corn syrup,

U.S. PER CAPITA CONSUMPTION OF REFINED SUCROSE AND CORN SWEETENERS 1840 -1985

z 0 (J) a:: w a. a:: w a. (J) zCl :::> 0 a.

YEAR

FIGURE 3. Per capita U.S. consumption (pounds) of major refined sugar products (sucrose, dextrose, corn syrup). Minor caloric sweeteners such as honey and are excluded. Source: Economic Research Service (1986); Liebman and Jacobson (1985).

6 from the middle of the nineteenth century to the current decade (Economic Research Service, 1986a; Liebman and Jacobson, 1985). There has been an obvious and quite impressive steady increase up until the height of Prohibition (when sugar was used for making illegal alcoholic beverages), and a more modest increase since the end of the Second World War. Throughout most of the period of increase, the predominant source of sugars was cane and beet. Within the past 20 years, however, new technological developments have brought about a remarkable transformation in the utiliza­ bility of America's highest-production field crop: corn. In addition to its tradi­ tional uses for food and animal feed, over 800 million bushels of the 8.7 billion bushels of corn produced annually are processed into feedstocks for other industrial uses, mainly as grain , corn starch (for textiles and ), glucose and dextrose for the , and high fructose corn syrup (HFCS). Figure 4 shows the number of bushels of corn which ultimately ended up in these four uses since 1972 (Corn Refiners Association, 1986). The specific technological development which has affected the corn industry is the development of high-yield immobilized enzyme systems which permit manufacture of polysaccharide mixtures with highly specific monomer content. The one- and two-unit sugars in these mixtures taste sweet, while the longer polysaccharide chains give the syrup its viscosity. By controlling the relative populations Qf different-sized chains, the manufacturer can tailor the taste and

REFINING USES OF CORN IOOO U.S., 1971-1978

9 00

800 Cl) ...J Alcohol UJ 700 :r Cl) J 600

100 HFCS 0 ~rmll~~-1~ 1971 19.73. 197 5 1977 1979 1981 1983 1985 CROP YEAR

FIGURE 4. Amount of corn (in millions of bushels) put lo refining uses since 1972, by category: alcohol, corn starch, glucose and dextrose, and high fructose corn syrup. Source: Corn Refiners Association (1986).

7 U.S. CONSUMPTION OF CALORIC SWEETENERS REFINED SUGAR vs. CORN SWEETENERS

100

~ 90 LI.Iz LI.I I- 80 II.I II.I ;= l fl) 70 (..) ii: 51r 9 60 c (..) 50 ...J ~ 0 40 I- LI... 0 30 I- l 48.8% z II.I 20 (..) a:: LI.I 0.. 10

0 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 YEAR

FIGURE 5. Percent of total caloric sweeteners from corn vs. percent from refined sugar, U.S. 1975-1985. Minor caloric sweeteners such as honey and maple sugar are excluded. Source: Economic Research Service (1986).

consistency of corn syrup to the particular requirements of a given food or beverage (McGee, 1984). A major historical event flowing from this technology is chronicled in Figure 5, which shows how, since 1975, the percent of total caloric sweeteners has been apportioned between refined sugar on the one hand, and corn-based products on the other. In 1985, for the first time, corn surpassed cane and beet as the principal source of sweetening agents in the U.S. food supply, tipping the scales at 51.2% (compared with about 24% in 1975) (Economic Research Service, 1984b). This relative domination is confirmed on an absolute basis in Figure 6, which shows the estimated pounds per capita consumed of all types of sweeten­ ing agents, including low- or non-caloric, since 1975 (Economic Research Service, 1984b; Glinsmann et al., 1986). Note particularly the emergence of and the impending decline of . One of the most important roles in the preceding shifts is played by soft . Americans seem to be drinking more and more, with no apparent limit to how much liquid we can consume. Figure 7 shows the estimated number of gallons per year we have consumed since 1950 of three popular types of drinks:

8 Total Caloric Sweeteners 130 ~ 120 ;Dextrose +(Honey 110 Edible Syrups 100

90 -HFCS ~ a: 80 <( (.) 70 wIr n. - Refined Sugar tJ) 60 z0 ::::> 0 50 Q.. 40

30

20

10 -Aspartame -saccharin 0

FIGURE 6. Pounds per capita of caloric and low /non-caloric sweeteners, U.S., 1975- 1985, by source (corn vs. refined sugar) or type (aspartame vs. saccharin). Aspartame and saccharin quantities are expressed as sucrose sweetness equivalents. Source: Eco­ nomic Research Service (1984b); Glinsmann et al. (1986).

soft drinks, , and fruit . Each of these has shown an increase, with soft drinks increasing the most (Miller, 1985). Figure 8 shows an almost exponential increase in per capita production of soft drinks, in terms of 12-oz. containers, since the mid-nineteenth century (National Soft Association, 1985). But as startling as this increase appears, equally dramatic is the shift in the way soft drinks have been sweetened which occurred during the course of a single year, from 1984 to 1985. As Figure 9 shows, in that short time period HFCS as the single-item sweetener leaped from 43% to 75% of all beverages, while sugar-even in combination with HFCS-fell to no more than a 3% share. During the same period, aspartame nearly completed its long-anticipated move to edge saccharin out of the low-calorie slot (National Association, 1985; 1986).

CONSUMPTION OF SUGARS: SURVEY DATA

The data in the preceding section represent best-guesses of per capita con­ sumption. This type ofinformation, though valuable, is of extremely limited use

9 PER CAPITA CONSUMPTION OF BEVERAGES U.S., 1950 - 1983

50

a:: - a:: LI.I 0.. z 0 30 CJ) a:: LI.I 0.. a:: LI.I 0.. 20 zCJ) 0 ..J ..J

I I J 1950 1955 1960 1965 1970 1975 1980 1983

YEAR

FIGURE 7. Gallons per capita consumption of soft drinks, fruit juices, and beer, U.S., 1950-1983. Source: Miller (1985).

ESTIMATED ANNUAL PRODUCTION OF SOFT DRINKS U.S., 1840-1985

500

~ 0::

YEAR

FIGURE 8. Estimated annual production of soft drinks, U.S., 1840-1985, expressed as 12-oz containers per person per year. Source: National Soft Drink Association (1985).

10 HOW SOFT DRINKS ARE SWEETENED

1984 1985 3% - Beet I Cane Sugar - 1%

42% High Fructose - Corn Syrup '-...... 75%

HFCS - a Sugar -......

~~B?"~~.:l>l 3 % Aspartame ...... _ =====;';i5% Saccharin 13% - Aspartame a Saccharin 1% - Other, None - 1%

FIGURE 9. Percent of all soft drinks sweetened with various sweeteners or combina­ tions, 1984 and 1985. Source: National Soft Drink Association (1985, 1986).

for making health assessments. For this purpose, nutritional surveys are required. Such surveys, for instance, can provide consumption information for specific age groups, and for special population segments of interest. They are, however, very expensive to perform, and have their own well-known limita­ tions. One important methodological feature is the necessity of distinguishing sugars added during food manufacture and processing from naturally occurring ones. In very few surveys are data on naturally occurring sugars available. Another difficulty is estimation of the exact composition, at time of con­ sumption, of foods in which conversion from one form of sugar to another takes place easily. For example, sucrose is hydrolyzed to glucose and fructose in acidic solutions. Soft drinks often contain carbonic acid, and about 60% of the sucrose initially present is hydrolyzed to these two monosaccharides, thus increasing the actual amount of fructose available for consumption (Glinsmann et al., 1986). Recently a Sugars Task Force of the Food and Drug Administration com­ pleted an analysis of the 1977-78 USDA Nationwide Food Consumption Survey, with specific attention to individual consumption of sugars from differ-

11 AVERAGE DAILY INTAKE OF SUGARS AND CARBOHYDRATES U.S., 1977-1978 300

250

Total Carbohydrates 200 c~ wa: a. 150 ::E "'<( a: (!> 100 Naturally Occuring "71f Je Sugars

50 Added Sugars

0 11-14 15-18 19-22 23-50 51+ Age Age MALES FEMALES

FIGURE 10. Average daily intake of sugars and carbohydrates in Americans age 11-64, included in 1977 NCHS survey. Source: Glinsmann et al. (1986).

ent sources, as well as in relation to total carbohydrates (Glinsmann et al., 1986). Figure 10 shows the number of grams per day of total carbohydrates, naturally occurring sugars (including lactose), and added sugars consumed by subjects in each of five age groups, by sex. In boys there is a peak in the 15-18 age group, followed by a decline in all components at older ages. For females, there is no peak, only a decline from childhood onward. Added and naturally occurring sugars contributed roughly equal amounts to the diets. The Task Force study, while a unique and important source of data, is still lacking in some respects: it is based upon survey data now a decade old, it contains little age-specific datafor older adults, who have the highest morbidity and mortality rates, and it provides no detail on specific food sources of sugars. An example of food-specific sugar intake data is shown in Table 4, which gives the average daily consumption of sugars by a group of school children in one survey (Morgan and Zabik, 1981). About 20% of their total sugar was actually obtained from milk, followed by 13.8% from sweetened beverages. Fruit was in third place (11.5%), followed by other traditional sources of sweetness, including cakes and cookies, jelly, and fruit juice. Food consumption surveys are not necessarily designed, however, for popu­ lation health assessment purposes. In order to study health effects of sweeteners, epidemiological studies must be undertaken. In 1982 The American Cancer

12 TABLE4 Average Daily Consumption of Total Sugars' Mean (g) % of Total Sugar Milk 25.9 20.4 Sweetened Beverages 17.9 13.8 Fruit 17.1 11.5 Cakes, Cookies, Pie 15.3 11.2 Jelly, Syrup, Table Sugar 12.8 9.8 Fruit Juice 12.0 8.8 Cream 7.4 5.5 Bread, Crackers 5.2 4.1 Cereal 4.2 3.3 3.7 2.6 '657 children age 5-12. Source: Morgan and Zabik (1981).

Society inaugurated a six-year follow-up study of 1.2 million men and women in all fifty states, D.C., and Puerto Rico. Over 77,000 ACS volunteers enrolled their relatives, neighbors, and friends over the age of 45 as subjects. Our goal is to study death rates from cancer and other causes in relation to the baseline distribution of various risk factors, including consumption of specific food items (Stellman and Garfinkel, 1986a). As a first step towards such assessments, it is necessary to establish socio­ demographic characteristics of both consumption and illness, in order to design analyses efficiently and to avoid confounding. We have recently published our observations concerning artificial sweetener (AS) consumption (Stellman and Garfinkel, 1986b). Figure 11 shows the percent of women aged 50-69 who were long-time users of AS (for the time period surveyed, mainly saccharin). AS use declined with increasing age, but increased in proportion to relative weight. That is, the most obese women were the highest users. This observation has important implications for health studies. is a risk factor for many chronic diseases, including cancer (Lew and Garfinkel, 1979), so that any behavior (such as AS use) associated with obesity is likely to be a predictor of illness or death, even if not causally linked. Obesity must therefore be regarded as a possible confounding variable in health assessments of AS. There are other food items which may be more important than artificial sweeteners. Two of these are certainly more interesting, if overlooked: chocolate and ice cream, foods of which Americans are especially fond. Just how fond our own study subjects are may be viewed in Table 5, which shows that men are

13 VARIATION OF ARTIFICIAL SWEETENER USE !50 ~ WITH AGE AND WEIGHT ~ 4!5 LI.I I- 11.J 40 LEGE NO 50-54 Ye ~ 10 ~ !5 a..LI.I 0 VERY LOW AVERAGE HIGH VERY LOW HIGH RELATIVE WEIGHT AT START

FIGURE 11. Variation of artificial sweetener use with age and weight. Source: Stellman and Garfinkel (1986b).

more likely to be frequent ice cream consumers than are women, but that the sexes are about equal in consumption of chocolate. We previously observed that frequency of eating ice cream and chocolate were highly correlated, and have developed an "ice-cream-and-chocolate score" for each individual based upon factor analysis of all 28 food frequencies

TABLE 5 Frequency of Consumption of Ice Cream and Chocolate in Male and Female Subjects, Cancer Prevention Study II, by Sex Ice Cream Chocolate Males Females Males Females No. of Subjects 416,778 516,593 351,179 458,712 Days Per Week Percent None 7.8 10.2 21.2 18.5 Occasional 26.2 34.4 31.4 36.8 1-4 53.9 47.6 41.8 38.9 5-7 12.l 7.8 5.6 5.8 100.0 100.0 100.0 100.0

14 TABLE6 Percent of Women in Cancer Prevention Study II Having a Low, Average, or High Factor Score on Dimension Representing Chocolate and Ice Cream Consumption, by Age, and Smoking Habit Low Average High Total Percent Age 40-49 31.3 35.2 33.5 100.0 50-59 34.0 34.2 31.8 100.0 60-69 42.9 30.3 26.7 100.0 70-79 44.3 32.6 23.1 100.0 Education Up to 8th Grade 48.6 28.0 23.4 100.0 Some High School 46.8 31.6 21.6 100.0 High School Graduate 35.0 34.7 30.3 100.0 Some College 36.7 33.6 29.7 100.0 College Graduate 31.4 33.3 35.3 100.0 Smoking Habit Never Smoked 33.2 34.7 32.1 100.0 Current Smoker 43.9 31.1 25.0 100.0 Ex-Smoker 34.2 32.7 33.I 100.0

(Stellman, 1985). CPS-II subjects may be classified as high, medium, or low with respect to this score, indicating their relative frequency for consuming these two foods. The remaining Figures refer to this composite score, confined to women. Analysis for men will be presented in a future report. Table 6 shows the distribution of the ice-cream-and-chocolate factor score by age, education, and smoking habit. Higher levels of consumption are apparent at younger ages, and among more highly educated women. Current smokers are less likely to eat these two foods than are never smokers or former smokers. Since cigarette smoking is a major determinant of present and future health, as well as being associated with specific eating habits, it must be investigated as a potential confounder. Striking regional differences are observed in Table 7, which shows that the North Central Plains states comprised the area of greatest chocolate and ice cream consumption. Finally, the prevalence of present illness is shown in Table 8, within each of the three levels of consumption. This non-specific question ("Are you sick at the present time?'') is an excellent predictor of short-term mortality in our cohort. Ice cream and chocolate were more often consumed by healthy than by ill subjects. This could be due to many possible factors, such as general appetite suppression, lower metabolism rates, or lack of interest in food. Whatever the

15 TABLE7 Percent of Women (Age-Adjusted) Enrolled in Cancer Prevention Study II Who Reported Eating Ice Cream at Least Two Days per Week, by Region of Enrollment• Percent Nebraska 49.2 South Dakota 48.7 Iowa 47.2 Kansas 45.9 Vermont 44.6 North Dakota 42.9 California 34.9 Puerto Rico 34.4 Queens Co. (New York) 33.7 Hawaii 32.5 Nevada 32.2 D.C. 32.1 'The six highest and six lowest study areas are shown.

reason, it is evident that sweetener intake is empirically related to overall health, and that this relationship must be accounted for in epidemiologic studies.

CONCLUSIONS

Consumption of sugars by Americans is at an all-time high. Soft drinks, which comprise one of the largest single sources of sweeteners, have undergone a drastic shift in sweetener content, both caloric (from sucrose to fructose), and non-or low-caloric (from saccharin to aspartame) in an exceptionally brief time. It is too early to predict what health effects, if any, this major change in the American diet may produce. Only limited information is available about dietary intake of specific sugars.

TABLES Percent of Women in Cancer Prevention Study II Who Were Sick at the Time of Enrollment, According to Their Factor Score (Low, Average, or High) on Dimension Representing Chocolate and Ice Cream Consumption Low Average High Sick at Present Percent Yes 12.3 11.8 9.6 No 87.7 88.2 90.4

16 More studies, such as that of the USDA Sugars Task Force (Glinsmann et al., 1986), are needed. New studies of this type should be designed to be of subsequent use in planning, performing, and interpreting epidemiologic studies of health effects of sweeteners. To emphasize the wide variation in patterns of sugar consumption, we have presented some socio-demographic features of ice cream and chocolate con­ sumption among l.2 million men and women enrolled in the American Cancer Society's Cancer Prevention Study II. Consumption was related to age, educa­ tion, smoking habits, and present state of health, all of which are also associated with mortality rates.

ACKNOWLEDGMENTS

I am grateful to the following individuals for providing resource materials: Elizabeth A. Norton, The Sugar Association; Kyd D. Brenner, Corn Refiners Association; Kenneth A. Fried, National Soft Drink Association; Youngmee K. Park, Division of Nutrition, Food and Drug Administration; Robert D. Barry, USDA Economic Research Service; and Michael Jacobson, Center for Science in the Public Interest. I also wish to thank Wendy Hladky, Joe Buardo, and Luz Tejiero for preparation of illustrations, and Sylvia Walker for process­ ing the manuscript.

REFERENCES

CORN REFINERS ASSOCIATION, INC. (1986). Facts on Corn. , D.C.: Corn Refiners Association, Inc. ECONOMIC RESEARCH SERVICE. (1984a). Sugar. Background for 1985 Farm Legislation. Agriculture Information Bulletin No. 478. Washington, D.C.: U.S. Department of Agriculture Economic Research Service. ECONOMIC RESEARCH SERVICE. (1984b). Sugar and Sweetener Outlook and Situation Report, p. 37. Washington, D.C.: USDA Economic Research Service, SSRV9N4 (December). ECONOMIC RESEARCH SERVICE. (I986). Sugar and Sweetener Situation and Outlook Yearbook, p. 24. Washington, D.C.: USDA Economic Research Service, SSR V 11 N2 (June). GLINSMANN, W.H., IRAUSQUIN, H. and PARK, Y.K. (1986). Evaluation of Health Aspects of Sugars Contained in Sweeteners. Report of Sugars Task Force, 1986. Washington, D.C.: Center for Food Safety and Applied Nutrition, Food and Drug Administration. HUG ILL, J. A.C. ( 1979). History of the . In: Sugar: Science and Technol­ ogy (Birch, G.G. and Parker, K.J., eds.). London: Applied Science Publishers. LEW, E.A. and GARFINKEL, L. (1979). Variations in mortality by weight among 750,000 men and women. J. Chron. Dis. 32:563-576. LIEBMAN, B.F. and JACOBSON, M. (1985). How sweet is it? Washington, D.C.: Center for Science in the Public Interest (Poster). McGEE, H. (1984). On Food and Cooking. New York: Charles Scribner's Sons.

17 MILLER, R.W. (1985). Soft drinks and six-packs quench our national thirst. FDA Consumer, October, pp. 23-24. MORGAN, K.J. and ZABIK, M.E. (1981). Amount and food sources of total sugar intake by children ages 5 to 12 years. Am. J. Clin. Nutr. 34:404-413. MORRISON, R.T. and BOYD, R.N. (1983). Organic Chemistry, Fourth Ed. : Allyn and Bacon. NATIONAL SOFT DRINK ASSOCIATION. (1985). Sales Survey of the Soft , 1984. Washington, D.C.: National Soft Drink Association. NATIONAL SOFT DRINK ASSOCIATION. (1986). Sales Survey of the Soft Drink Industry, 1985. Washington, D.C.: National Soft Drink Association. RAMAZZINI, B. (1940). De Morbis Artificum [Diseases of Workers] (W.C. Wright, trans.), p. 427. Chicago: University of Chicago Press. ROBBINS, W.J. (1948). Sugar. New York: Sugar Research Foundation. SCHAPSMEIER, E.L. and SCHAPSMEIER, F.H. (1975). Encyclopedia of American Agricultural History. Westport, CT: Greenwood Press. STELLMAN, S.D. (1985). Chairman's remarks on Session V: Data analysis in cohort studies. In: Selection, Follow-up, and Analysis in Prospective Studies: A Workshop (L. Garfinkel, 0. Ochs and M. Mushinski, eds.), pp. 145-147. National Cancer Institute Monograph No. 67, NIH Publication No. 85-2713. Bethesda, MD: USDHHS, Public Health Service, National Institutes of Health, National Cancer Institute. STELLMAN, S.D. and GARFINKEL, L. (1986a). Smoking habits and tar levels in a new American Cancer Society prospective study of 1.2 million men and women. J. Natl. Cancer Inst. 76:1057-1063. STELLMAN, S.D. and GARFINKEL, L. (1986b). Artificial sweetener use and one­ year weight change among women. Prev. Med. 15:195-202. THE SUGAR ASSOCIATION, INC. (1984). Sugar: A Fact Sheet. Washington, D.C.: The Sugar Association. USDA. (1972). U.S. Department of Agriculture. Major statistical series of the U.S. Department of Agriculture. How They Are Constructed and Used. Vol. 5. Consump­ tion and Utilization of Agricultural Products. Agr. Handbook No. 365.

18