Click for next page ( 14


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 13
2 TI1E NATIONWIDE FOOD CONSUMPTION SURVEYS Background information on the Nationwide Food Consumption Survey presented in this chapter was provided by U.S. Department of Agriculture Human Nutrition Information Service specialists at a Committee-sponsored June 1983 workshop and October 1983 symposium. This information is presented without Committee comment, assessment, or evaluation. HISTORY Each year, the U.S. Department of Agriculture (USDA) provides per capita estimates of the nutrient content of the civilian food supply. These estimates are derived from food balance sheets of supply and wholesale distribution tables. They estimate food disappearance by subtraction of wholesale stocks and exports from production and imports. Estimates of the total food disappearance into civilian distribution are divided by the total civilian population. Because these data take into account farm and wholesale stocks but not retail stocks, they may be viewed as representing the wholesale level. The civilian consumption or disappearance data include consumption of food in homes, in eating places, and other institutions away from home, as snacks, by pets, and through waste and in loss .3 This data set, which dates back to 1909-1913, is USDA's longest historical series related to nutrition. It allows USDA to follow trends in the per capita availability of food and nutrients over a 70-year period. It provides no information about the dis tribution of food at the household and individual level . The first USDA survey of national scope was conducted to fill this gap in 1936-1937. Since then, such surveys have been conducted at roughly 10-year intervals -- 1935-1935, 1942, 1948, 1955, 1965-1966, and 1977-1978. The first four surveys included only information on the use of food in households, whereas the more recent surveys (spring of 1965 and four seasons of 1977-1978) obtained information both on household food use and on the dietary intake of individual household members. It is the 1977-1978 survey that is examined in this report. Data from the 1977-1978 Nationwide Food Consumption Survey (NFCS) provided information on food consumption at two levels: food used by entire households at home and food intake of individual household members . . . both at home and away from home. From these data and data on the com- position of foods, estimates were made of the-nutrient content of house- hold food supplies and of the nutrient intake of individual household members.

OCR for page 13
14 PURPOS ES The fundamental purpose o f the NFCS is "to measure the food and nutrient content of the diet and the money value of food used by U.S. households and the food and nutrient intakes at home and away from home of individuals" (R. Rizek, Director, Consumer Nutrition Division, Human Nutrition Information Service, USDA, personal communication) . Other4 ma jar purposes of the NFCS are as follows: ~ To measure the kind, amount, nutrient content, and monetary value of foods consumed by the U. S. population and specific groups . foods . To describe the practices of families in the use of specific Thus, the NFCS provides both measures of the amounts of food and estimates of their nutrient content in the United States. The estimates of nutrient content in U.S. diets are used to estimate dietary adequacy. Repeated surveys permit measures of U.S. diets and estimates of dietary adequacy over time, thus providing a data base for determining dietary changes and trends in such changes. The survey obtains information both on the kinds and on the quan- tities of food. The nutritive value of diets ingested by adults and children -- classified by sex, age, region of residence, income, and race -- can then be calculated. The survey also obtains information on various household characteristics. DES ION GENERAL FEATIJR F..R The most recent NFCS was conducted from April 1977 through March 1978. The 1977-1978 NFCS sample was a sample of approximately 15,000 U. S. households . The household survey component collects information on the household, and the individual survey component collects information on individual household members. The data collection method used in the household component of the NFCS was a 7-day recall of food used by the household. Although the data collection method was standardized, the information reported varies with the household being sampled, e.g., with the characteristics of the house- hold, the type and extent of participation in food assistance programs, the type and quantity of food used, and the cost of the food. With respect to food cost, for example, information was obtained on the price of the food purchased; on the additional monetary value of other food

OCR for page 13
- 15 - used at home that may have been a gift, a payment, or produced in the home; and on the expense incurred for food obtained and eaten away from home. For the individual component, individual household members provided an interviewer with recall of 1 day's food intake and then maintained a diary of food intake for 2 additional days. thus, the individual com- ponent of the NFCS collected a total of 3 consecutive data-days of information on one person's dietary intake. STATISTICAL DESIGN The universe for the 1977-1978 basic survey consisted of households in the 48 conterminous states. Special surveys were conducted in Alaska, Hawaii, and Puerto Rico and for the elderly and for low-income households in the 48 conterminous states. The design of the household food use phase of the basic survey was that of a stratified probability sample for the conterminous United States. The nation was divided according to the nine Department of Commerce census regions and by three urban categor ies: central city, suburban, and nonmetropolitan. For reporting purposes, the information from the original nine census regions was "collapsed" and reported as information from four geograph ic regions . The survey covered a full 12 month period, beginning in April 1977 and ending in March 19 78. Sample households were drawn for each of the four seasons, and information was collected from approximately 3,750 households in each quarter. Almost 15,000 households completed the household food use phase of the NFCS questionnaire. The national sample of households was des igned to b e s e 1 f -weighting, i.e., sample households were selected for interview in a manner that would be representative over time (by year, quarter, month, week, or day). Because completion rates were slightly below target in the first two quarters, household statistics were adjusted for nonresponse by quarter, region, and urbanization. The universe for the individual dietary intake phase of the basic survey consisted of members of households in the 48 conterminous states. The sample consisted of members of sample households participating in the household food use phase of the survey. During the spring quarter, the Survey subjects consisted of all household members. In the remaining three quarters, all household members 18 years old and under were included in the sample, but only half those over 19 and over. Persons in one-person households were included regardless of age. As a result, the intake of approximately 34,000 persons was reported.

OCR for page 13
- 16 - Statistical adjustments were made to account for nonresponse by house- holds. In addition, for the three seasons other than spring, proportional representation among sex-age groups was maintained by treating the record for each person 19 and over as two records. DATA COLLECTION METHODS Both household and individual data were collected primarily by interview in the household. Interviews were conducted by trained and experienced inter- viewers in an effort to ensure consistency in data collection. Each household was contacted at least a week before the interview, in part to schedule an appropriate interview time. At this initial contact, the interviewer asked that the person most knowledgeable and responsible for meal planning and preparations (the "household food manager") keep memory aids on food use, such as store receipts or menus. During the 2- to 2.5-hour interview, the house- hold food manager was asked to recall, with the aid of a food list and other memory aids, the kinds and amounts of food that had disappeared from the home food supplies over the previous 7 days. This recall included food that had been eaten, discarded, and fed to household pets as leftovers. The household food manager was also asked about several household characteristics. If participating in the individual dietary intake phase, as well as the household food use phase of the survey, the household food manager was also asked to recall the food eaten on the previous day -- both at home and away from home. At this time, the household food manager also provided information on dietary intake of children under 12 in the household. The household food manager was shown how to keep food diaries for both manager and children for 2 additional days. Teen-agers and adults in the household recalled their own food intake for the interviewer. The interviewer returned after 2 days to review and collect the food records kept by household members. HOUSEHOLD SURVEY COMPONENT The household questionnaire had four major sections: household character- istics, foods used from home food supplies, home food production, and house- hold income. Household characteris tics included: Information on general food shopping practices. A list of household members with information on sex, age, and whether pregnant or nursing. o The number of meals and snacks from household food suppliese

OCR for page 13
Expense for food bought and eaten away from home. De s Or ipt ion o f the dwe 11 ing . Educational level and employment status for heads of household. Participation in food assistance programs. The interviewer asked the respondent to recall, from a detailed list, foods used from home food supplies during the previous 7 days. When a food was mentioned that did not appear on this food list, the interviewer sought a detailed description of that food. For each food item reported, information was sought on: ~ Quantities -- in pounds or units convertible to pounds. F_ -- including fresh, frozen, canned, or dried . Source -- purchased, home-produced, or received as gift or pay. Purchase and price -- for food reported as purchased, information on the unit of purchase and the pr ice . From this information and ache information about quantity of food used, the monetary value of consumption for each food could be calculated. Where food was reported as having been produced at home, information was sought on the categories of food produced and on home freezing and canning. The last questions asked were related to household income . For each household member 14 years old or older, information was sought on the amount and sources of income during the previous full month. In addition, infor- mation was sought on the amount and sources of income for the entire household during the previous calendar year. Detailed questions were asked about par- ticipation in the food stamp program and the face value of any food stamps received. Information on participation in other food assistance programs was also requested. In handling data obtained on household food use, several computations were made, and the results were entered on a tape. Each food was assigned one of ate-out 3,000 food codes. .These codes identified foods both by major food marketing group and subgroup and by food processing form and its variations. For example, fresh chicken breasts would be coded as 4511 201 16 10 11 1l,. with the coding sys tem indicating:

OCR for page 13
- 18 - 4511 -- chicken. 4511 _ -- chicken breasts, fresh. 4511 201 16 -- nutrition code: chicken. 4511 201 16 10 -- processing code: fresh. 4511 201 16 10 11 -- variation 1: with bone; not smoked, cured, or pickled. 4511 201 16 10 11 11 -- variation 2: ready to cook or cooked parts. For each food code, the units that were reported were then converted to pound weights with standard conversion factors. Use of equivalent weights permitted further calculation of milk equivalents (in terms of calcium) in dairy products, flour equivalents for cereals and bakery products, and the single-strength equivalent of fruit juices. The nutrient contribution of each food was computed by multiplying the reported quantity (in pounds) by the nutritive value in the edible part of a pound of that food. Nutrient valuesS were derived from standard reference tables in the USDA's Handbook No. 8, information from food manufacturers, or typical recipes. For each food and for the household diet, calculations were made of the contribution of food energy and 14 nutrients: protein, fat, carbohydrate, vitamins (A, B6, B12, C, thiamin, riboflavin, and niacin), and minerals (calcium, iron, magnesium, and phosphorus). The nutritive value of the house- hold diet was then compared with the calculated total of the 1974 NRC Recommended Dietary Allowances6 appropriate for the persons in the house- hold. Calculations were adjusted to account for meals eaten away from the home and for guest meals in the home. It should be noted that such calcu- lations tell only whether nutrients available in the household (prorated on the basis of the proportion of meals eaten from the family food supply) were sufficient to meet the calculated RDA for that household. The calculations cannot be used to determine whether any one person's food intake was adequate in nutrient content. Estimates of the adequacy of nutrient intake by indi- viduals were based on data on individual dietary intake. INDIVIDUAL INTAKE COMPONENT . . The individual intake component of the survey consisted of a 24-hour dietary recall, a 2-day food diary, questions on eating patterns and on the intake of specific foods and dietary supplements, and statements as to height, weight, and health status. - The format of the form used to record food intake was the same for the first day's recall and for the second and third days' diary records. The data collected were specific and detailed and consisted of:

OCR for page 13
19 ~ Description of all foods and beverages consumed, including foot obtained away from home. Quant ities eaten . Form in which inges ted . Sources of food. Eat ing occas ions . General information on diet and health. The food and beverage description included all food eaten at each eating occasion, beginning with midnight of the previous day. Teen-agers and adults responded directly to the interviewer, and the household food manager provided responses for children under 12 and other persons in the household unable to respond for themselves. For the 24-hour dietary recall, the interviewer helped the respondent start the recall procedure with the first time something was eaten. Then the respondent continued, providing information about what was eaten during the day. The interviewer probed as necessary to ensure complete descriptions of food and to check for omissions. For the 2-day diary record, the respondent was provided with a reference booklet for describing food and the amounts eaten. The quantity of each food and beverage ingested was reported in common household measures, dimensions, numbers of units, or weights. To help in visualizing quantities, each house- hold was provided with common measuring cups, spoons, and a ruler. The source of each food was obtained through a structured question. Sources included food from home supplies that was eaten at home, food from home supplies that was eaten away from home, food eaten at someone elders home, and other food obtained and eaten away from home. For food eaten away from home, information was requested about the place where food was obtained (e.g., restaurant, school, or fast-food establishment), the type of service, and the cost of the food, if it was purchased. Each eating occasion was identified by the time of its beginning. The respondent was asked what each eating occasion was usually called and with whom it was shared. Questions were also asked about the amount of water ingested each day and whether a given day's intake of water was typical. Each respondent was asked for some information only once, including: self- reported height and weight; whether the person was on a special diet and, if so, what kind; whether vitamin or vitamin-mineral supplements were used and, if so, what kinds (but not quantities); and health status and physical handicaps .

OCR for page 13
20 Four steed were involved in preparing data taper from the information collected on Individual dietary intake. Each food was assigned a food code. Then the reported quantity of each food was converted to the weight, in grams, of the edible portion of that food. The nutrient contribution of each ingested food was then calculated. Finally, the respondent's nutrient intake for each of 3 days and average daily nutrient intake for the 3 days were determined. The food coding manual for the individual survey contained over 4,500 food codes. The first digit of each food code identified one of nine major food groups. The remaining six digits contained information identifying food sub- groups, information relates to cooking method, and information about the com- bination of the given food with other foods. For example, chicken a la king would be coded as 271 4002, with the coding system indicating: 2 -- meat, poultry, or fish mixture. 27 -- meat, poultry, or fish combined with other nonmeat items. 271 -- meat, poultry, or fish in a gravy or sauce. 4002 -- chicken a la king. The energy and nutrient contents of food were calculated by using a second computer tape containing values for the amounts of food energy and 14 nutrients in 100 grams of each food, as eaten. The 14 nutrients were the same as those previously listed for the household data. The nutrient values7 were based on values from USDA's Agriculture Handbook No. 8, updated in 1977 and on food industry data. GENERAL QUALITY ASSURANCE_TECHNIQUES Several quality assurance procedures were followed during the 1977-1978 survey. Interviewers took 5-day training sessions. Separate training sessions were provided for data coders. Interviewers and respondents were recontacted by telephone (where possible) or in person (if necessary) when responses to key questions were missing or confusing. All entries were double-coded and checked by a supervisor. Extreme values for amounts consumed were reviewed and, if possible, verified with respondents. All computer entries were verified. Extreme values for amounts of food energy and 6 nutrients were reviewed. Traditional validation procedures (e.g., replication, observation, and weighing of foods) were not deemed to be feasible in a survey of this size and scope. Therefore, a number of partial validation procedures were undertaken. These included relatively small-scale validation studies, self-validation from repeat surveys, and comparisons with other national surveys. In preparation for the 1977-1978 NFCS, several data collection methods were tested: use of observers in the home, photographing of foods, preparation of duplicate meals, interviewer tagging of foods in inventory or as purchased, and CG1 lection of food-related trash and garbage.

OCR for page 13
- 21 - In addition, before the 1977-1978 survey, Burk ant Pao3 conducted a comprehensive review and evaluation of the literature on methods for large- scale surveys of household and individual diets. They appraised the reli- ability and validity of two types of records of current food intake, three types of recalls of past food intake, and a combination of record and recall. They concluded that, for the random samples in large-scale surveys, the 24-hour recall method hat high response rates and seemed to have the highest reliability. For the aspects of validity reviewed, no data collection method was seen to be clearly better than the others. The records (diaries) tended to be upwardly biased (a likelihood of higher estimated than actual nutrient intake) and the recall method downwardly biased. Thus, the 1977-1978 NFCS used a combination of the recall method with a record method to offset the opposing weaknesses of each.8 SURVEY REPORTING , The survey data collected and processed for the 1977-1978 survey are contained on 21 data tapes. These data have been made available to the public through the National Technical Information Service. Separate tapes contain household and individual data. Others contain data from the four quarters of the year for the 48 states; for Alaska, Hawaii, and Puerto Rico; and for the special low-income and elderly-subjects surveys. Descriptive data on the quantities and money value of food used by house- holds and on the nutrient content of household food supplies are contained in 13 preliminary reports. Additional preliminary reports contain information on the food and nutrient intake of individual household members, classified by sex, age, and whether pregnant or nursing. Final reports are more compre- hensive, but are also descriptive. Analyses of survey data that address a wide variety of issues have been conducted by the Human Nutrition Information Service (both within the agency and under agreements with universities and others), by other government agencies, and by private groups. Results of these analyses have been presented to the public in several ways, e.g., in administrative reports, monographs, articles in professional journals, public presentations, and pub l ished proceedings . Some survey resul ts have been used in food advertis ing, e.g;, to promote foods or supplements. RESEARCH INITIATI~S Since 1978, several methodologic studies have been undertaken to resolve questions arising from past surveys and to test methods to be used in a longi- tudinal survey of individuals. For example, the ability of men to report their food intake accurately is being assessed. Foods eaten by men in a

OCR for page 13
- 22 - company cafeteria were observed, and the discarded food was measured. Recall of that dietary intake was later obtained both from the men and from their spouses. Preliminary results indicate that, to obtain useful information about the dietary intake of working men, they themselves must be interviewed. Another question being addresses is whether the quantities of food reported to have been ingested by men are affected by the measuring devices used in a survey -- measuring cups and spoons, rather than food models. Results indicate that some sets of models yielded better results than when cups and spoons were used; in other tests, use of models had the reverse effect. A new study on measuring devices is being conducted. Respondent-reported data on height and weight collected in the 1977-1978 NFCS are being compared with measurements of height and weight collected in the National Heal th and Nutrition Examination Survey (NHANES) . Preliminary results indicate that respondent-reported heights and weights probably are adequate for the NFCS, although young people might tend to underreport height slightly and older people might tend to overreport height. The effect of a list of fewer than 500 food codes on the analysis of the nutrient content of diets is being studied. A shortened food list could lead to cost and time savings in data process ing and tabulation. It remains to be determined whether such savings are outweighed by disadvantages to data users. The food intake data from the recall and 2-day records of the NFCS are being compared with those from the 1-day recall of the NHANES, to determine whether estimates of dietary nutrients are generally consistent and whether inconsistencies can be explained. Some inconsistencies associated with procedures can be and are being standardized. Additional studies are in progress. Alternative procedures for repeated collection of food and beverage intake (up to 12 days) over a 1-year period are being tested. Nine groups of women in lowand middle-income households are involved in a year-long study. The results will be used in determining design and data collection methods for future USDA dietary surveys. The applicability of longitudinal or repeat surveys of food and beverage intake among segments of the low-income population is being studied in nine localities. The study is testing various methods of measuring, on a continuing basis, the dietary status of population groups at nutritional risk. Food- and nutrition-related support information is being re-examined and updated for use in the next survey. This support information includes the food codes, weights, equivalents, conversion factors, and nutrient composition tables required in survey data processing. Delineation of needs of data users is a part of the evaluation process.

OCR for page 13
Under an HNIS grant, the Food and Nutrition Board is evaluating standards of nutritional adequacy for survey diets that are estimated to contain lees than 1003 of the Recommended Dietary Allowances. From this evaluation, stan- dards of nutritional adequacy may be developed at three or four different levels of risk. Results from the 2-year study should be available for use in the planned continuing survey of food intake. FUTURE PLANS A longitudinal survey of individual dietary intake is scheduled to begin in 1985. To obtain information on usual food intake, individual subjects will be studied for 6 days per year. At a minimum, a monitoring group consisting of one sex-age category (e.g., women 19-50 years old and all children I-5 years of age of those women) would be studied. Each year, a new panel of subjects in the same sex-age group would be selected. As funds allow, sup- plemental surveys of additional sex-age categories of the population believed to be at nutritional risk would be conducted. The next Nationwide Food Consumption Survey is planned for 1987. The design and methods used in that survey will be influenced by the availability of funds, the emergence of new interests or concerns on the part of data users, and results from collateral research programs and will be subject to approval by the Office of Management and Budget. USES OF DIETARY DATA FROM NFCS SURVEYS . Five categories of uses of NFCS data are listed below with specific examples of each . The type of data -- household (H.) or individual ~ I) -- is indicated in each case; if both types of data are used, the primary or most important type is listed first. The examples listed here have been provided by HNIS staff and were selected to be representative. ASSESSMENT OF DIETARY INTAKE General Use To provide detailed benchmark data on the food and nutr lent intake of the population. .-( ~ H) Exampl e NFCS results are compared with food and nutrient intake derived from small surveys of special groups and with projected intakes under assumed conditions, such as increased enrichment of selected foods or supplementation of resources.

OCR for page 13
- 24 - To monitor the nutritional quality of diets of the population. (I, H.) To determine the ~ ize and nature of populations that might be at risk because of inadequate or excessive intake of nutrients. (I) To identify intervention (food assistance, fortification, or education) most appropriate for populations at risk. (I, H) To obtain a better understanding of the variation in intake of nutrients among individuals and of intake of individuals over time. (I) To explain the relationships of food and nutrient intake to nutritional and health status of the population. (I, H.) To show the dietary change required to meet guidel ines proposed f or improved heal th and we 11-be ing . ~ I, H.) To identify socioeconomic factors associated with diets that meet and fail to meet nutritional criteria. (I, H.) Current surveys show the dietary situation about every 10 years. (Starting in 1985, nutrient intakes of a control population segment, such as women 19-50 years old, will be compared over time in a continuous survey, and segments at high nutritional risk will be studied for shorter periods.) Problem nutrients and sex-age categories most frequently experiencing shortfalls were identified. Intakes were distributed by fat-to- energy ratios. Diets of most teen-aged girls and women of child-bearing age were found to have low intakes of many nutrients. Educational efforts have been initiated. One-day intakes were compared with 3-day intakes in 1977-1978 NFCS. Methods studies under way will compare data for up to 12 days collected over a year, i.e., four 3-day intakes. ~ Epidemiologis ts have studied the types and amounts of fats in diets from surveys relative to statistics on the incidence of heart disease. The dietary change required to meet the RDA or the "Dietary Goals" have been determined for use in exploring the practicality of the guidelines and in determining the direction of educational efforts. Differences in food and nutrient intake by sex and age of individual, household, income, region, ant urbanization of household, season of year, and participation in food stamp and school feeding programs have been explored.

OCR for page 13
- 25 - To identify food intake patterns and eating pattern~_(how often, when, where, and with whom food is eaten) associated with diets that meet and fail to meet nutritional criteria. (I) To identify foods that are primary contributors of nutrients in various populations. (I, H.) To provide a basis for the identification of foods on which composition data are needed. (I, H.) To predict the need and market for new products and characteristics of likely consumers. (H. I) ECONOMICS OF FOOD CONSUMPTION To provide detailed benchmark data on domestic consumption-- quantities of foods, costs of foods, and nutrients available from foods used by U.S. households. (H.) To assess demands for agricul- tural production, marketing facilities, and services. (H.) Snacks were found to be important contributors of nutrients. Diets that tended to conform to the Daily Food Guide (basic four) frequently exceeded moderate amounts of fat. These facts have implications for educators. That dark greens are rich sources of iron can be determined from tables on the composition of foods. That they provide 1% of iron intake, compared with 34X from meat, poultry, ant fish, according to the NFCS, is also of importance to educators. HNIS develops data on the composition of foods through extramural agreements as necessary for commonly used items. A corporation used survey data to predict demand for a beverage, foods that were eaten with that type of beverage, and eating occasion at which it was selected. _ Example Agricultural economists use data to explain how food production and marketing practices affect household food uses at a given time. Researchers use data to show usual consumption relative to consumption of small population groups surveyed. Demand analysis is used on survey data to show the demand s tructure for specific product groups, such as meat and dairy, and for special products, such as imitation milk and meat extenders .

OCR for page 13
- 26 - To study the effects of socio- economic factors (income , s ize, and compos Otis of households on total food consumption and expenditures and on food consumption-expenditure patterns for various food commodities. (H.) To analyze the relationships between socioeconomic character- istics of and nutrient availability for selected population groups. (H.) To determine the importance of home food production and of foods received as gifts or pay in food consumption of households. (H.) To determine the importance of food bought and eaten away from home on food consumption of households and individual household members. (H. I) To determine the nutritional significance of enrichment and fortification. (H. I) To determine the importance of the use of "convenience" foods on food consumption of households. (H.) Food stamp benefits for different- size households reflect the greater cost of equivalent diets in small that in large households, as determined from the survey data. Analyses us ing econometric models are under way at Washington State University and the Un iver s i ty of Missouri. Findings that household size and food expenditures were major determinants of nutrient availability reinforce the food stamp program benefit s tructure. USDA's Economic Research Service estimates of U. S. per capita food consumpt ion do no t inc. lude food produced at home. NFCS is the primary source of data on this important component of U.S. food consumption. As eating out becomes more prevalent, unders Landing 0 f the economic and nutritional effects of eating out is of increas ing importance to health professionals and the food industry. Survey data are being used to study this issue at the University of Illinois . Survey data are being explored as a source of information on the quantities of nutrients added to the food supply, for potential use in estimating the nutrient content of the food supply, and as a basis for policy on the enrichment and fortification of food. Many convenience foods are available in markets, and an understanding of their nutritional and economic effects is of increasing importance to educators and the incus try. These issues are being studied by the Virginia Polytechnic Institute and State University.

OCR for page 13
27 To determine the differences in k inds and amoun ts o f f cod the t households report as us ed and individual household members report as eaten (H. I) To determine the extent of seasonal variability in food nutrient consumption among households. (H.) FOOD PROGRAMS AND FOOD GUIDANCE General Use To determine factors affecting participation in food programs and the effect of participation on food expenditures and diet quality; to predict change in participation and effects that would result from modification in the program. (H.) To assess the dietary change associated with food program participation. (H. I) To analyze the effect of food programs on the market for commodities. (H)- To identify populations at nutritional risk for possible intervention with food assistance, forti fications, and education programs. (I, H) Adequate measures of household discard of edible food are not available. Survey data for hous ehold food use and food intake o f individual household members were adjusted to comparable bases and compared as background for a methodologic survey of household discard conducted by the University of Arizona. Seasonal availability of foods affects and consumption. The economic and nutritional e ffects of seasonal variation in consumption of selected foods and food groups are being s tudied by Washington State University. Example The marginal propens ity to consume food from food s tamp bonus income was found to be twice that of cost income. This finding indicates to administrators that a change to cash instead of food stamp benefits might be expected to reduce food-consumption effects of the program. NFCS data showed administrators and others that the school lunch program significantly improves diets of children. School breakfast participation was associated with improvement in nutrients provided by milk. Administrators could estimate the effect on the cheese market of providing cheese free to a given proportion of the low-income population. - The WIC program was initiated and expanded to improve resources and nutrition education of low-income women with infants and small children.

OCR for page 13
- 28 - To identify changes in food and nutrient_consumption that might be expected to reduce risk. (H. I) To develop for administrative and educational purposes food guides and food plans that reflect food consumption practices and meet nutritional and cost criteria. H. I) To determine reasonable amounts of foods and surplus to distribute to families. (H. I) HISTORICAL AND SECULAR TRENDS . General Use . . To appraise the history of food consumption and dietary s tatus relative to economic, technologic, and other factors. (H. I) To predict changes in food consumpt ion and die tary s ta tus as they may be inf luenced by economic, technologic, and other developments. (H. I) To attempt to correlate food consumpt ion and die tary s ta tus with incidence of diseases over- time. (I, H.) A national educational campaign "Making Food Dollars Count, " emphasizes the dietary changes that low-income survey households might make to achieve nutritious, low-cost diets. The USDA family food plans at four cos t levels have been revised. Household and individual data were used to es tab 1 ish the food consumption basis for these food plans. Base food prices are also from the survey. Guidelines for distribution of cheese, butter, nonfat dry milk, and cornmeal were based on survey results. Example_ l Shifts in the consumpt ion and nutrient contribution of foods introduced into the marke tplace -- lowfat milk, ethnic foods, fortified breads and cereals, and commercially frozen mixtures -- can be shown. The effect of changing the iron enrichment levels of flour and bread on diets of teen-aged girls and women and on diets at the 90th percentile of bread and flour consumption among men can be estimated. Thus, the potential advantages and d isadvantages could be predicted. The types and amounts of fat in diets, as reported in surveys, have been studied relative to changes in the incidence of heart disease.

OCR for page 13
- 29 - To assess the reasonableness of dietary standards relative to food intake of healthy populations over time. (I, H) To attempt to follow food consumption through the life cycle. ~ I) FOOD-SAFETY CONSIDERATIONS * . . General Use - To determine intake of incidental contaminants, food additives, and naturally occurring toxic substances. (I, H.) To identify the patterns of use of foods and food components in the diets of a population. (I, H.) To identify extreme and unusual patterns of intake of foods or food ingredients, including food additives. (I) To identify size and nature of populations at risk from use of par t icular foods and food products. (H. I) Foods available in households provide zinc and folacin at well below the RDA. Although mean amounts of calcium and iron in households are high, some households' and many household members' diets (especially those of teen-aged girls and women) do not provide the RDA. Teen-agers with high soft-drink consumption in 1965 continued their preference for soft drinks into adulthood as indicated by the 1977-1978 NFCS intakes. Adult onset of coffee consumption in 1977 was not high. Example Food patterns for FDA's Total Diet Study were partially derived from NFCS data. This study reports levels of some potentially toxic substances in the diet. Quantities of foods consumed in excess are identified. Characteristics of liquid protein users were identified and made known to FDA with other dietary information when sudden death of liquid protein dieters was reported. Manufacturers of food containers used information on quantities of foods in containers to estimate amounts of substances migrating into food from containers. *These include determining human exposure to food additives, pesticides, animal drugs, feed additives, GRAS substances, environmental contaminants, and naturally occurring toxic substances.

OCR for page 13
- 30 - To determine number of food items EPA estimated amounts of herbicides in which ~ food additive may be and pesticides that people were likely permitted and in what amounts. to ingest with leafy vegetables, on the (H. I) basis of data on quantities and frequency of use from NFCS. To determine the need to modify Canned-milk trade associations used regulations in response to data on users and quantities consumed changes in food consumption. (I) by infants and other sex-age groups to help to determine whether lead in canned milk constituted a problem. SCARY Data from the 1977-1978 Nationwide Food Consumption Survey provided information on U. S. households, the food used by households at home, and the food intake of individual household members both at home and away from home. Data collected include information on household characteristics, foods used from home food supplies, household income, and participation in food assistance programs. Data were also collected on the intake of specific foods by individuals over a 3-day period. The basic 1977-1978 NFCS sample was a representative sample of approximatly 15 ,000 U. S. households in the 48 conterminous states and involved approximately 34,000 persons. Data were collected primarily by interview in the household. Data processing and calculations provided information both on the food and money value of food used by U.S. households and on food intake at home and away from home by individuals, as well as estimates of the nutrient content of the food consumed . Repeat surveys prov ide a data base for es timates of changes over time .