Click for next page ( 127


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 126
Variety in Foods HELEN SMICIKLAS-WRIGHT, SUSAN M. KREBS-SMITH, and JAMES K~BS-SMITH Nutritionists generally accept the idea that eating a variety of foods ensures the selection of a nutritionally adequate diet. In the United States, variety is a fundamental tenet of dietary guidance, representing the first of seven guidelines issued by the U.S. Department of Agriculture and the U.S. Department of Heals and Human Services (USDA and DHHS, 19801. Furthermore, Recommended Dietary Allowances (RDAs) issued by the Food and Nutrition Board of the National Research Council are "intended to be met by a diet of a wide varied of foods rather than by supplementation or by extensive fortification of single foods" (NRC, 1980, p. 1~. Indeed, from the 1920s to the present, the idea of variety has continued, although changes have occurred in the suggested kinds and amounts of food to eat (Wolf and Peterkin, 19841. The notion that a variety of food choices enhances the likelihood of selecting a nutritionally adequate diet is not limited to food guidance in the United States. Canada's Food Guide recommends variety in food choices and eating patterns as a basic nutrition principle (Department of National Health and Welfare, 19791. Food variety is also recommended for people in developing countries, such as rural populations with limited- food patterns (Robson and Wadsworth, 19771. This paper addresses two questions: (1) What is food variety? (2) What benefits does an increasingly varied diet provide? Pertinent literature is reviewed, and some preliminary data are presented on variety in American diets and the relationship between variety and dietary quality. 126

OCR for page 126
VARIETY IN FOODS 127 REVIEW OF LITERATURE Food variety has been described as a simple guideline, but one that represents a complex food selection technique (Wolf and Peterkin, 1984~. The complexity is illustrated by the diversity and proliferation of food products on supermarket shelves. The average supermarket may contain 15,000 items (Connor, 1980), and thousands of new products are intro- duced each year (Solomon, 1983~. Although many new products may be no more than minor alterations in flavor or package size, the general picture is one of immense choice. This vast selection of food items may be responsible for consumers' difficulty in understanding the concept of variety. A Canadian study (De- partment of National Health and Welfare, 1979) examined consumer un- derstanding of such nutrition concepts as adequate diet, balanced diet, and food variety. When asked to interpret the advice, "Eat a variety of foods each day," many respondents simply listed different kinds of foods. Some stated that it meant eating different kinds of meat or balanced meals. About one-third of respondents reported that variety meant a little of evening, and about 15% described it as not eating the same food every day. Consumer confusion about variety may arise not only from the plentitude of food products but also from the professional community's venous interpretations of, and recommendations for, variety. One interpretation of variety is that it represents the selection of foods from among major food groups. For example, Nutrition and Your Health: Dietary Guidelines for Americans advises consumers to include selections from six food groups: fruits; vegetables; whole grain and enriched breads, cereals, and grain products; milk, cheese, and yogurt; meats, poultry, fish, and eggs; and legumes (USDA and DHHS, 1980~. A second interpretation of variety is the selection of foods from within food groups. Ideas for Better Eating: Menus and Recipes to Make Use of the Dietary Guidelines recommends choosing different foods within each group of foods and emphasizes more servings of fruits, vegetables, and grain products- especially whole grains and frequent consumption of dark-green vegetables, dry bean dishes, and starchy vegetables (USDA, 19811. In a third interpretation of variety, Canada's Food Guide recommends varying food preparation methods for example, using raw, cooked, canned, and frozen fruits and vegetables (Department of National Health and Wel- fare, 19821. It suggests that even more variety may be achieved by altering the size of meals and the time and location in which meals are eaten. Perhaps the various interpretations stem from the many expected ben-

OCR for page 126
28 EATING TRENDS AND NUTRlTlONAL CONSEQUENCES efits of food variety. Probably the most frequently stated benefit is the increased assurance of adequate nutrient intakes (Guthrie, 19771. Other benefits from variety include avoiding either deficiencies or excesses of single nutrients (USDA and DHHS, 1980), ensuring an appropriate bal- ance of micronutrients (Mertz, 1984), and reducing the likelihood of exposure to contaminants in any single food item (USDA and DHHS, 19801. Variety has also been recommended as a way of increasing eating enjoyment (Department of National Health and Welfare, 19791. Few of these specific benefits of variety have been closely examined, although several researchers have studied the importance of food variety in general (Black and Sanjur, 1980; Caliendo et al., 1977; deGwynn and Sanjur, 1974; Duyff et at., 1975; Krondl et al., 19821. Krond! and coworkers (1982) measured food variety in the diets of elderly Canadians by asking how many of 181 commonly available foods were eaten in the previous year. Participants were classified as having diets of limited variety or greater variety. The mean, less one standard deviation, was selected as the cutoff point separating limited from greater vanety. Thus, 49 or fewer items represented limited variety, and 50 or more represented greater variety. Variety of food use was determined for 90.~% of sampled women and 69.4% of sampled men. Greater variety was associated with higher educational achievement, higher health rating, and a stronger effort to maintain health. However, the study did not measure nutrient intakes, which would have allowed comparisons between variety and nutritional adequacy. Sanjur and coworkers used food diversity scores to study dietary patterns in venous populations (Black and Sanjur, 1980; Caliendo et al., 1977; deGwynn and Sanjuu, 1974; Duyff et al., 1975~. Duyff and coworkers (1975) examined the diets of Puerto Rican-American teenagers by mea- suring food diversity and nutritional adequacy. Diversity was defined as the average number of different food items eaten by a respondent per day. Intakes of calcium, iron, vitamin A, and vitamin C were calculated from a 3-day record. The mean diversity per day 12.5 itemswas described as low. More diverse diets were reported to be more nutritionally adequate; however, the data on this relationship were not presented. Another study examined food diversity among Puerto Rican women attending a maternal and infant care clinic (Black and Sanjur, 1980~. The diversity score was constructed as follows. From a frequency distribution of all food items consumed in a 24-hour period, the 16 food items most frequently consumed were selected. Women who consumed fewer than 8 of the selected items received a score of 1; women who consumed from 8 to 11 selected items received a score of 2; and women who consumed more than 11 of them received a score of 3. Results showed that women

OCR for page 126
VARIETY IN FOODS 129 who were migrants to the community had higher diversity, nutrition knowl- edge, and food preference scores. Nutritional adequacy was not measured in the study. Yet another food diversity score was computed by Caliendo and co- workers (1977) in Heir study of 113 preschool children residing ~ New York State. The score was based on Be number of foods consumed by 20% or more of the children during a 24-hour period; the reported number was 20. Absent from the literature is information on He extent to which Amer- ican diets exhibit variety and how variety improves the quality of those diets. Such information could provide a more precise interpretation of variety for consumers. To meet this need, we measured variety in venous sex and age groups and examined the relationships of dietary variety to both nutritional adequacy and macronutrient balance. METHODS Studly Population Data for this study were obtained from the basic survey portion of USDA's 1977-1978 Nationwide Food Consumption Survey (NFCS) (USDA, 1980). In that survey, individual household members recalled 1 day's food intake and kept a diary for 2 additional days. Information was gathered from a stratified area probability sample of 15,000 households in the 48 conterminous states and the District of Columbia (USDA, 1980~. Deter- m~ning which persons in each household would be asked to participate differed by season. In the spring, all persons in each household were asked to provide food intake information. In the fall, winter, and summer, only one-half of those persons 19 years of age and older were asked to partic- ipate, except for those persons in one-member households who were asked to participate regardless of age. Proportional representation was ma~n- tained in these other seasons by double counting each record for respon- dents 19 years of age and older, except for respondents from one-member households. Additional weighting factors were applied to all persons in the survey to account for nonrespondent households. Three-day food rec- ords were obtained from 2S,030 persons (36,255 weighted). In order to save costly computer time, only a subsample of that population was ana- lyzed in He present study. The NFCS sample population used for this study was selected as follows. From the unweighted spring portion of the NFCS, a straight 10% random sample was selected. For the other seasons, in order to achieve an appropriate age distribution, a 10% random sample of all

OCR for page 126
130 EATING TRENDS AND NUTRITIONAL CONSEQUENCES persons 19 years or older, regardless of household size, was selected. After the sample was drawn, pregnant and lactating females and children under 1 year of age were excluded. Although the study population was not a representative sample of the U.S. population, it represented a large group of people more than 1 year of age from all regions of the United States. Variety Measure Vanety was defined as Me number of unique food items, characterized by distinct NFCS code numbers, that were consumed and reported in 3 days. Such a scoring system is simple yet accounts for as much variation as possible within the NFCS data. The individual food codes differentiated items on the basis of such factors as added ingredients, preparation method, and extent of fat removal. Table 1 presents some examples of NFCS food codes and their descriptions. Overcounting and undercounting of unique food items are possible with the NFCS food codes. For example, some identical foods, variously de- scribed, have different codes. If a person reported the same food in dif- ferent ways, overcounting of unique food items would occur, and results would erroneously show more variety in food consumption. Similarly, some NFCS food codes refer to food mixtures (e.g., sandwiches, casse- roles). If one person reported eating a tuna salad sandwich and another person, consuming the same food, reported eating bread, tuna, mayon- naise, and celery, undercounting of unique food items would occur for the former person, and results would erroneously show less variety in food consumption. Overcounting, however, was determined to have a negligible effect on results, and adjustment for undercounting could not be made because TABLE ~ 1977-1978 Nationwide Food Consumption Survey Examples of Food Codesa Food Code Description Milk, cow's, fluid, whole Milk, not furler specified Milk, cow's, fluid, low fat (2%) Milk, chocolate . 111-1100 1 1 1-0000 111-1211 115-1100 581-0503 S81-OS19 581-21 13 Macaroni and cheese Cheese ravioli, no sauce Rice casserole, no cheese aFrom USDA, 1980.

OCR for page 126
VARIETY IN FOODS 131 (1) the NFCS code book does not include all ingredients of mixtures and (2) counting ingredients could invalidate a mixture as a unique food item if the ingredients were also reported elsewhere as individual food items. Because a simple measure of variety was desired for this study and because correction for all overcounting and undercounting was impossible, the study proceeded with the number of unique food codes on each person's record as the measure of vanety; this measure was labeled unique foods. Dietary Quality Measures Dietary quality was assessed by two measures of nutrient adequacy (nutrient adequacy ratios and mean adequacy ratios) and by the percentage of calories from fat, protein, and carbohydrate. Nutrient adequacy ratios (NARs) were calculated for 10 nutrients (protein, calcium, iron, mag- nesium, phosphorus, vitamin A, thiamin, riboflavin, vitamin Bit, and vitamin C) according to the following equation: NAR = person s 3-day average intake of nutrient A second NAR was calculated for vitamin B6, but the value listed in the RDA table (NRC, 1980) was not used as the denominator. Instead, we used 0.02 mg of vitamin B6 per gram of protein intake. Vitamin B6 values in the RDA table were based on this formula but were also based on an assumption that many adults consume an average of 100 g to 1 10 g of protein. Many adults, however, do not consume such high levels of protein; therefore, the RDA table values may overestimate the need for vitamin B6 (Guthrie and Crocetti, 1983~. A mean adequacy ratio for 11 nutrients (MARll) was calculated by adding the percent RDAs met (truncated at 100%) for each of 11 nutrients and dividing the sum by 11. Therefore, an MARll of 100 represents 100% of the RDAs for all nutrients included in the score. RESULTS Extent of Variety The number of unique foods reported by this study population varied from 4 to 63 (Table 2), and all age and sex groups showed a wide range of unique food consumption. Children as young as 1 to 3 years of age displayed vanety, ranging from 12 to 45 different items. The mean number

OCR for page 126
32 EATING TRENDS AND NUTRITIONAL CONSEQUENCES TABLE 2 Mean Number and Range for Unique Food Items and for Total Number of Foods Reported in 3 Days by Sex and Age of Respondents Number of Unique Age Number of Foods Toml Number of Foods Sex (years) Penons Mean + SD Range Mean + SD Range Male 1-3 82 24.9 + 7.0 12-43 40.9 + 11.1 19-70 4-6 86 27.1 + 6.5 7-44 41.5 + 9.8 20-64 7-10 136 28.1 + 6.7 15-46 41.7 + 9.2 23-70 11-14 135 27.8 + 7.2 6-49 43.2 + 12.3 17-79 15-18 157 27.4 + 7.1 10-46 42.4 + 12.4 15-76 19-22 115 24.6 + 7.8 9-46 38.3 + 11.4 16-67 23-50 566 27.5 + 7.9 4-56 45.0 + 14.6 9-109 51-69 269 27.1 + 8.3 7-56 48.1 ~ 16.5 8-112 ~70 117 25.7 + 7.9 4-50 50.1 + 14.4 12-93 Female 1-3 69 24.1 + 6.3 12-45 39.5 + 9.4 24-70 4-6 93 26.2 + 6.0 8-40 40.4 + 8.5 22-64 7-10 124 28.5 + 7.2 11-48 42.4 + 9.9 20-70 11-14 137 27.4 + 7.1 5-47 40.0 + 10.3 7-66 15-18 138 24.2 + 7.5 7-50 36.7 + 12.0 9-76 19-22 118 23.7 + 8.2 7-51 35.9 + 12.2 12-71 23-50 751 25.3 + 8.1 4-63 40.4 + 13.9 7-119 51-69 405 26.6 + 8.2 6-61 45.5 + 13.6 9-98 ~70 203 24.2 + 6.8 9-50 44.2 + 12.1 16-93 Total 3,701 26.2 + 7.8 4-63 42.7 + 13.5 7-119 of unique foods for the total sample of 3,701 persons was 26.2 + 7.8. The mean number of unique foods for each age and sex group was sim- ilar approximately 25 foods. The number of total food items reported (total foods), also shown in Table 2, ranged from 7 to 119. As with unique foods, all age and sex groups exhibited wide ranges. However, there are some sex and age differences in the mean values for total foods repotted. Males aged 23 years and older had the highest mean values females aged 1 ~ to 99 very had the lowest mean value. , _O ~ _ ~ _ ~ ~ To simplify reporting of the results, the population was divided into seven groups based on the number of unique foods reported (1-15, 16- 20, 21-25, 26-30, 31-35, 36-40, and 41 or more) and eight groups based on total foods reported (1-10, 1 1-20, 21-30, 31-40, 41-50, 51-60, 61-70, and 71 or more) (Tables 3-71. Table 3 displays, for the total group, the mean number of unique foods and total foods and the percentage of persons in each unique foods and total foods category. Relatively few persons, 7.3% of the total sample,

OCR for page 126
VARIETY IN FOODS TABLE 3 Mean Number of Unique Foods and Total Foods Reported by Sample Population and Percentage Distribution of Persons Who Reported Consuming Various Numbers of Unique Foods and Total Foods in 3 Days (N = 3,701) . . . 133 Measure Mean + SD Number of Foods Percentage of Persons Reporting Unique foods 26.2 + 7.8 Total foods 42.7 + 13.5 1-15 16-20 21-25 26-30 31-35 36-40 ~41 1-10 11-20 21-30 31-40 41-50 51-60 61-70 271 7.3 16.2 24.4 24.6 15.8 7.4 4.3 0.3 3.0 14.5 29.4 27.2 15.9 6.8 2.9 consumed fewer than 16 unique foods, and an even smaller number, 4.3%, consumed 41 or more unique foods. lThe mean total foods score was about 43, most persons reporting between 20 and 60 total food items. Relationship Between Variety and Various Quality Measures Table 4 presents the bivariate relationships between unique foods and each of the 11 NARs. All the NARs showed a positive relationship with the measure of variety used here. Variation in the number of unique foods appeared to account for more variation in the magnesium and thiamin NARs than in other nutrient NARs. In contrast, relatively little NAR variation for protein, iron, and vitamins B6 and BE was explained by variations in unique foods. Except for the lowest unique foods category, the NARs for protein were high. Table 5 presents the distribution of persons in each unique foods cat- egory by MAR11 score. With each increase in number of unique foods, the proportion of persons with MAR11 scores of less than 60 dropped precipitously, whereas the percentage of persons with scores of 80 or higher multiplied several times. Less than 1% of those with 20 or fewer unique foods achieved an MAR11 score of 100, and almost none with more than 30 unique foods had a score of less than 60.

OCR for page 126
134 e~ C~ ;^ Ct _ o U. o 3 ._ o z ~D _` U. CK z C~ o _ C) et 3 - ~ C~ c ~ _ ~ ~ ._ -, ~ 5) ~ o ._ et ~0 ~ _ o c _ . ~, _ ,_ gt)= o == o Z a3 Z oo V) - _ 11 A\ Z ol il Z _ - oo ~n t ~ Z; ~ _ o o~ o t1 o 11 Z ~ _ - ~o o 11 Z - o U) 11 - 1 Z _ } oo ~ o~ oo 4~~ ~h _ ~ ~ _ mm - ~=m" . . . . . . . . . . o o o o o o o o o o o Q ~ ~c~ ~ oo oo _ ~ oo 0 oo o~ 0\ c~ ~ ~ ~ ~ ~ cr~ . . . . . . . . _ o o o o o o o o o o o oo ~ o o~ ~ ~ ~ o ~ ~ 0 ~ 0\ 0\ 0\ cr ~ ~ c~ ~ ~ . . . . . . . . . . . oooooooooo ~ oo In 00 0` U) 0 oo oo oo CM ~ cr' . . . . . . . _ O 0 0 0 0 0 o~ U oo ~ cr~ . . . O 0 0 0 ~ ~ ~ ~ 00 ~ 00 oo oO o~ ~ oO o~ 00 . . . 0 0 0 0 0 0 0 0 0 0 0 oo X . . . . . O 0 0 0 0 oo ~D O oo ~ oo o~ 00 . . . . . O 0 0 0 0 C~ 0 ~ ~ ~ oO ~ 00 oo oo oo ~ . . . . . . . . . . . O 0 0 0 0 0 0 0 0 0 0 ~ ~ ~ ~ oo oo ~ ~ ~ ~ I . . . ~ . . . ooooooo ~ 00 . . . O 0 0 ~ e~ .~o '5= ~ eO o eU 0,:t ,~,D,~,:5.= Z ~ ~ ~> - Q o o o W - C) ._ 04 ._ C) ns U) 1:: o ._ - o = Z so~ . . &~ _ o ~Ct -, Ko o o 9D --~5 E ~ ~ ~Z V Z

OCR for page 126
VARIETY IN FOODS TABLE 5 Percentage Distribution of Mean Adequacy Ratios for 11 Nutrients (MARll) Obtained from Various Numbers of Unique Foods in 3 Days 135 Number of Distribution of MAR11 Scores (%) Unique Foods Scores Scores Scores Scores Scores Reported Na 100 1 - 15 270 9.3b 26.3 37.4 26.7 0.4 16-20 601 1.5 1 1.3 38.8 47.4 1.0 21-25 905 0.1 3.9 24.5 70.3 1.2 26-30 910 0.0 0.4 16.4 79.1 4.1 31-35 584 0.0 0.2 17.2 87.5 5.1 36-40 273 0.0 0.0 4.0 86.5 9.5 241 158 0.0 0.0 2.S 81.0 16.5 Total 3,701 aNumber of persons reporting. bPercentages may not total 100 due to rounding. Table 6 shows He mean percentage of caloric intake from protein, fat, and carbohydrate for each unique foods category. As the number of re- ported unique foods increased, there was a slight decrease in mean per- centage of caloric intake from protein and a corresponding increase in mean percentage of caloric intake from carbohydrate. Interestingly, the mean percentage of caloric intake from fat was virtually the same for all unique food categories. Another study (Guthrie and Wright, 1984) in which a different subset of Me NFCS population was used showed that as unique foods increased, the percentage of reported meats and meat alternatives decreased slightly, whereas reported fruits and vegetables increased. TABLE 6 Mean Percentage of Caloric Intake Obtained from Protein, Fat, and Carbohydrate by Persons Who Reported Consuming Various Numbers of Unique Foods in 3 Days Calonc Intake . Number of From From From Unique Foods Protein (%), Fat (%), Carbohydrate Reported Na Mean + SD Mean + SD Mean + SD 1-15 270 18.4 + 5.1 39.1 + 9.8 41.7 + 12.5 16-20 601 16.9 + 3.5 40.2 + 7.8 42.5 + 10.1 21-25 905 16.9 + 3.2 40.5 + 7.1 42.3 + 8.9 26-30 910 16.2 + 3.2 40.8 + 6.6 42.8 + 8.3 31-35 584 16.0 + 2.8 40.5 + 6.4 43.6 + 8.3 36-40 273 15.5 + 2.8 40.6 + 5.8 43.9 + 7.8 241 158 15.4 + 2.5 1 40.9 + 5.2 43.7 + 6.9 Total 3,701 . aNumber of persons reporting.

OCR for page 126
36 EATING TRENDS AND NUTRITIONAL CONSEQUENCES TABLE 7 Mean Number of Kilocalories Obtained per Day by Males and Females (Aged 23 to 50 Years) Who Reported Various Numbers of Unique Foods in 3 Days (N = 1,317) Number of Males Unique Foods Reported Females kcal, kcal, Na Mean + SD N Mean + SD 1-15 36 1,719 + 500 80 858 + 357 16-20 68 2,069 + 634 134 1,234 + 392 21-25 136 2,158 + 568 185 1,464 + 424 26-30 133 2,407 ~ 608 179 1,704 + 495 31-35 -107 2,684 + 773 98 1,853 + 371 36-40 56 2,806 + 651 41 1,940 + 430 ~41 30 2,880 ~ 611 34 2,199 + 665 Total 566 751 aNumber of persons reporting. Th,e relationship between, the mean number of kilocalories (kcal) ob- tained per day and the number of unique foods reported was also examined in this study. Table 7 shows values for males and females aged 23 to 50 years. For both sexes, mean caloric values increased as unique foods increased. For males who reported the least variety' the mean daily caloric intake was 1,719 kcal; for those males who reported 41 or more different items, mean daily caloric intake was 2,880 kcal. Women who reported 1 to 15 items had a mean calonc intake of 858 kcal, whereas those who reported 41 or more foods obtained 2,199 kcal. All age and sex groups showed similar relationships. , . Because there seemed to be no reason that variety per se would increase caloric intake to such an extent, another question emerged. Perhaps va- riety, as defined by the unique foods measured here, was interrelated with Me absolute number of foods eaten (total foods). If the unique foods measure was highly correlated with total foods, this might explain the positive relationship between variety and caloric intake. Indeed, the cor- relation between unique foods and total foods for this population was determined to be 0.76. In other words, 58% of the variation in unique foods is explained by total food scores. This interrelationship prompted the reexamination of the relationship between unique foods and MAR 1 1 this time controlling for total foods reported. The relationship was examined by a correlation-regression analysis, using MARll as the dependent variable and unique foods, total foods, and the interaction of unique and total foods as independent variables. Table 8 shows the equation obtained by regression analysis. This can be used as a model to predict MAR11. There is a significant interaction between unique foods and total foods in their relationship

OCR for page 126
VARIETY lN FOODS Parameter TABLE 8 Equation Obtained by Regression Analysis to Predict MAR11 Score by Number of Unique Foods and Total Foods Reported in 3 Daysa Regression Coefficient (b value) Number of unique foods Total number of foods Interaction Intercept , aMAR1 1 = 28.07 + 1.68 (unique foods) + 0.95 (total foods) b _ 0.02 (unique foods x total foods). bp ~ 0.0001. 1 .68b 0.9sb 0.026 28.0~7 137 to MAR11 scores. This interaction suggests that the effect of variety on overall nutritional adequacy is different for various numbers of foods eaten. Table 9 shows He estimated MART 1 scores that can be predicted from the model for various numbers of total foods and unique foods. Although these are not actual sample means, these data fit the model reasonably well. Regardless of the total number of foods reported, MAR11 scores increased win increasing variety, but the amount of change associated with an increase in unique foods was greater when total foods were low Man when total foods were high. Therefore, when only 20 foods were reported in 3 days for every ~ unit increase in unique foods, a 1.3 unit increase in MAR11 scores was expected. However, when 60 food items were eaten in 3 days, only about a 0.5 unit increase in MAR11 for such an increase in unique foods was expected. TABLE 9 Estimated Mean Adequacy Ratios for 11 Nutrients (MART is), by Total Number of Foods and Vanous Numbers of Unique Foods Reported in 3 Days . . Number of Estimated MARlls Unique Foods 20 Total 30 Total 40 Total 50 Total 60 Total Reported Foods Foods Foods Foods Foods 15 66.3 72.8 79.3 85.8 92.3 20 72.7 78.2 83.7 89.2 94.7 25 83.6 88.1 92.6 97.1 30 89.0 92.5 96.0 99.5 35 96.9 99.4 lOl.9a aMARI Is are truncated at 100, but the linear relationship suggests that at high levels of both unique foods and total foods, average un~ncated MART Is would exceed 100.

OCR for page 126
138 EATING TRENDS AND NUTRITIONAL CONSEQUENCES SUMMARY AND DISCUSSION The goal of nutrition education is to provide simple and reliable advice to consumers that will guide them in selecting a healthful diet. Such a diet must not only provide adequate amounts of the essential nutrients but also must guard against excessive consumption of kilocalories and harmful constituents. Although the advice to "eat a variety of foods" sounds like a simple axiom, consumers may be confused about implementing the recommen- dation. Even within the nutrition community, there are various interpre- tations of how variety is to be achieved. Nutntionists expect many benefits from variety, although few of these benefits have been tested. This study examined the extent of variety in American diets and the effect of variety on dietary quality. A variety measure was selected to account for as much total variation in food choice as possible- including variation among and within major food groups and variation from altered cooking methods. Use of this simplified measure to examine bivariate relationships initially suggested that increased variety was associated with improved nutritional adequacy. This result seemed to confirm one major stated purpose of food variety the selection of a nutritionally adequate diet. However, furler analysis determined that variety was also related to the total number of foods that a person ate. In this study population, persons with increased variety consumed more foods. The accompanying high NARs and caloric intakes may have resulted more from increased total foods than from increased variety. Therefore, nutritionists must be cognizant of their implicit advice to increase the number of foods when they suggest eating a variety of foods. Furthermore, when the number of foods was controlled and variety examined, earlier conclusions about the relationship between variety and overall nutritional adequacy (MAR11) required alteration. That is, variety affected overall nutritional adequacy less when the number of foods was high and more when the number of foods was low. This suggests that dietary variety is more important for persons consuming a limited number of foods, such as females aged 15 to 22 years. Further research on food variety is necessary so that nutritionists can provide the best dietary guidance to consumers. Our ongoing research examines the relationships between variety (controlling for the total num- ber of foods) and the other measures of dietary quality the individual NARs, caloric intake, and percentage of caloric intake from each of the macronutrients. We are also studying the separate effects of dietary variety within and among major food groups, e.g., whether food variety within

OCR for page 126
VARIETY IN FOODS 139 some groups is more important to dietary quality than in other groups, or how dietary quality is affected when a person simply chooses foods by food group without regard to choices within groups. This research needs to be accompanied by studies Cat identify how consumers interpret variety when they are faced win a complex food supply. Such research will contribute significantly to dietary guidance. ACKNOWLEDGMENT This research was partially supported by Contract 58-3198-2-57 from Be Human Nutrition Information Service, Consumer Nu~ihon Center, U.S. Department of Agriculture. REFERENCES Black, S. J., and D. Sanjur. 1980. Nutrition in Rio Piedras: A study of internal migration and maternal diets. Ecol. Food Nutr. 10:25-33. Caliendo, M. A., D. Sanjur, J. Wright, and G. Cummings. 1977. Nutntional status of preschool children: An ecological analysis. J. Am. Diet. Assoc. 71:20-26. Connor, J. 1980. Food product proliferation: Part II. Natl. Food Rev. Summer:10-13. deGwynn, R. E., and D. Sanjur. 1974. Nutritional anthropometry, diet and health-related cor- relates among preschool children in Bogota, Colombia. Ecol. Food Nutr. 3:273-281. Department of National Health and Welfare. 1979. Nutntion Concepts Evaluation Study. Nu- tntion Education Unit, Health Promotion Directorate, Health Services and Promotion Branch, Ottawa, Ontano, Canada. Department of National Health and Welfare. 1982. Canada's Food Guide: Handbook. The Ministry of Health and Welfare, Health Promotion Directorate, Ottawa, Ontario, Canada. Duyff, R. L., D. Sanjur, and H. Y. Nelson. 1975. Food behavior and related factors of Puerto Rican-American teenagers. J. Nutr. Educ. 7:99-103. Guthrie, H. A. 1977. Concept of a nutritious food. J. Am. Diet. Assoc. 71:14-19. Guthrie, H. A., and A. F. Crocetti. 1983. Implications of a protein-based standard for vitamin B6 Nutr. Rep. Int. 28:133-138. Guthrie, H. A., and H. S. Wright. 1984. Assessing Dietary Intake. Second Report for Contract 58-3198-2-57. The Human Nutrition Information Service, Consumer Nutntion Center, U.S. Department of Agriculture, Washington, D.C. Krondl, M., D. Law, M. A. Yurkin, and P. H. Coleman. 1982. Food use and perceived food meanings of the elderly. J. Am. Diet. Assoc. 80:523-529. Mertz, W. 1984. The essential elements: Nutritional aspects. Nutr. Today 19:22-30. NRC (National Research Council). 1980. Recommended Dietary Allowances, 9th ed. A report of the Food and Nutrition Board, Assembly of Life Sciences. National Academy of Sciences, Washington, D.C. Robson, J. R. K., and G. R. Wadsworth. 1977. The health and nutritional status of primitive populations. Ecol. Food Nutr. 6:187-202. Solomon, G. 1983. New foods proliferate without high technology. Nutr. Week 13(26):4-6. USDA (U.S. Department of Agriculture). 1980. Food and Nutrient Intakes of Individuals in One Day in the United States, Spring, 1977. Nationwide Food, Consumption Survey 1977- 78, Preliminary Report No. 2. U.S. Department of Agnculture, Washington, D.C.

OCR for page 126
{40 EATING TRENDS ED NOON CONSEQUENCES USDA (U.S. Department of Agnculh~e). 1981. Ideas for Better Eating: Menus and Recipes to Make Use of the Dietary Guidelines. U.S. Government Printing Office, Washington, D.C. USDA and DHHS (U.S. Department of Agriculture and U.S. Department of Health and Human Services). 1980. Nutrition and Your Health: Dietary Guidelines for Americans. HG232. U.S. Government Printing Office, Washington, D.C. 20 pp. Wolf, J. D., and B. B. Petedcin. 1984. Dietary guidelines: The USDA perspective. Food Technol. 38:80-86.