点击显示 收起
1 From the Division of Cancer Control and Population Sciences, National Cancer Institute, National Institutes of Health, Bethesda, MD, and the Division of Health Examination Statistics, National Center for Health Statistics, Centers for Disease Control and Prevention, Hyattsville, MD.
2 Presented at the symposium Fat Intake During Childhood, held in Houston, June 8–9, 1998. 3 Address reprint requests to RP Troiano, National Cancer Institute, DCCPS, ARP Executive Plaza North, Room 4005, 6130 Executive Boulevard, Bethesda, MD 20892-7344. E-mail: troiano{at}mail.nih.gov.
ABSTRACT
Background: Dietary factors related to body weight and chronic disease risk are of interest because of recent increases in the prevalence of overweight.
Objective: Secular trends in energy and fat intakes of youths aged 2–19 y were assessed. Current intakes were compared with recommendations.
Design: Dietary 24-h recall data from the third National Health and Nutrition Examination Survey (1988–1994) and earlier national surveys were examined.
Results: Mean energy intake changed little from the 1970s to 1988–1994 except for an increase among adolescent females. Over the same time period, the mean percentage of energy from total and saturated fat decreased, but remained above recommendations, with overall means of 33.5% of energy from fat and 12.2% of energy from saturated fat. In 1988–1994, 1 in 4 youths met the recommendations for intakes of fat and saturated fat and 3 in 4 met the recommendation for cholesterol intake. Beverages contributed 20–24% of energy across all ages and soft drinks provided 8% of energy in adolescents. Except for adolescent girls, beverage energy contributions were generally higher among overweight than nonoverweight youths; soft drink energy contribution was higher among overweight youths than among nonoverweight youths for all groups.
Conclusions: The lack of evidence of a general increase in energy intake among youths despite an increase in the prevalence of overweight suggests that physical inactivity is a major public health challenge in this age group. Efforts to increase physical activity and decrease nonnutritive sources of energy may be important approaches to counter the rise in overweight prevalence.Am J Clin Nutr 2000;72(suppl):1343S–53S.
Key Words: National Health and Nutrition Examination Survey • NHANES • energy intake • dietary fats • beverages • health surveys • nutrition surveys • trends • child • preschool • adolescence • United States
INTRODUCTION
The proportion of the preschool, school-age, and adolescent population classified as overweight has increased dramatically over approximately the past decade after remaining relative stabile from the 1960s through the late 1970s (1–3). The proportion of adults who are overweight has followed a similar secular trend (4, 5). These secular increases cross racial-ethnic and socioeconomic categories as well as age groups. Overweight or obesity in childhood is associated with increased blood lipids (6), blood pressure (7), and glucose intolerance (8), as well as psychosocial difficulties and increased risk of persistence of obesity and its related risks into adulthood (9). The health care costs of obesity, which are already high (10), can be expected to grow in an increasingly overweight population as it ages.
These observed rapid changes in the prevalence of overweight are unlikely to be due to genetic changes in the population (11); thus, there is much interest in the behavioral and environmental changes that have led to increased overweight prevalence. Increased fatness in an individual or increased overweight prevalence in a population can occur only as a result of a sustained positive energy balance, which occurs when energy intake exceeds energy expenditure. A shift to positive energy balance could be due to increased intake, decreased expenditure, or a combination of the 2 changes. The aspect of energy expenditure most subject to behavioral change is physical activity, but information on physical activity in the population, particularly in youths over time, is limited. However, dietary data have been collected with use of standardized 24-h recall methods in national surveys conducted by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC) since the early 1970s. The data from the National Health and Nutrition Examination Surveys (NHANESs) provide a unique opportunity to investigate concurrent US population trends in diet and measured weight status.
Regardless of any effect on body weight, dietary fat intake in children is of interest because of concerns about coronary artery disease risk. The atherosclerotic process begins in childhood (12, 13) and early experiences with food can shape long-term eating patterns (14). There is also evidence that dietary intake early in life is related to cardiovascular disease risk factors (15), although the relation is weaker for older children than for younger ones (16).
Nutrition and Your Health Dietary Guidelines for Americans recommends that individuals aged 2 y balance the food they eat with physical activity (17). Another guideline recommends that individuals "choose a diet low in fat, saturated fat, and cholesterol," and goes on to specify that a diet should be chosen that provides no more than 30% of energy from fat and <10% of energy from saturated fat. The dietary guidelines advise that children should gradually adopt a diet that meets these recommended intakes between the ages of 2 and 5 y.
This article presents data from NHANES III, 1988–1994, on current dietary intake of energy, percentage of energy from fat (total, saturated, monounsaturated, and polyunsaturated), and dietary intake of cholesterol, updating previous publications on the first 3 y of NHANES III (18–21) with data from the entire 6 y of the survey. We compare current intakes with dietary recommendations and examine secular trends in consumption of these nutrients by sex, race, and age. We also consider the contribution of several beverage sources of energy and fat for US children and adolescents and compare intakes of overweight and nonoverweight youths.
SUBJECTS AND METHODS
Cross-sectional information
NHANES III was the most recent in a series of health examination surveys carried out by the NCHS (22). Like previous NHANESs, NHANES III was a cross-sectional, multipurpose survey designed to provide nationally representative reference data and prevalence estimates for a variety of nutrition and health status measures and related risk factors. A sample representative of the US civilian, noninstitutionalized population was selected with use of a complex, stratified, multistage, probability cluster sampling design. Mexican Americans, black Americans, children aged 2–5 y, and persons aged 60 y were oversampled in NHANES III to produce more reliable estimates for these groups. Race-ethnicity was based on proxy or self-report and was categorized as non-Hispanic white, non-Hispanic black, Mexican American, or other. The "other" racial-ethnic group sample was too small for separate reporting, but was included in estimates for the total population. Detailed descriptions of the survey design and operation are available (22, 23).
The survey protocol included a household interview followed by a standardized physical examination of selected participants in a mobile examination center (MEC). Dietary interviews were administered to all examinees by a trained dietary interviewer fluent in English and Spanish in a private room in the MEC. Respondents or their proxies reported all foods and beverages consumed except plain drinking water for the previous 24-h time period (midnight to midnight). Proxy respondents provided information for 97% of children under 6 y of age (3% were by proxy and self-report). Children aged 6–11 y were permitted to report their own intake and 55% did so, but 22% of recalls in this age group were completed by proxy and 23% were completed with both the child and a proxy. About 99% of recalls for children aged 12–19 y were obtained by self-report. Daycare providers and school personnel were contacted as needed to obtain complete intakes for children.
An automated, microcomputer-based dietary interview and coding system known as the NHANES III Dietary Data Collection System was used to collect all dietary recall data (22, 24, 25). The NHANES III 24-h dietary recall protocol entailed a multiple-pass cognitive approach; detailed probes for food preparation methods, including fat and salt added to foods; and probes for frequently forgotten foods. Data collection covered all days of the week to improve overall population dietary estimates.
There were 12821 children and adolescents (aged 2–19 y) selected to participate in NHANES III. Of those selected, 11723 were interviewed (91.4%) and 11026 were interviewed and examined (86.0%). Of those interviewed and examined, 10371 children had satisfactory (ie, complete and reliable) 24-h recalls and were included in the analysis. Nutrient intakes were not calculated for children who were breast-feeding. This resulted in an overall response rate of 80.9%.
Prior national samples for trends
Two national surveys conducted before NHANES III [NHANES I, 1971–1974, ages 1–74 y (26), and NHANES II, 1976–1980, ages 6 mo to 74 y (27)] used a similar design and standardized methods to collect dietary data from nationally representative samples of children and adolescents. Before NHANES III, dietary interviews were recorded on paper forms by trained nutritionists using standardized interview techniques and probes. Because of examination schedules, limited dietary data were available for weekend intakes. Overall response rates for earlier NHANESs were reported previously and are comparable with those for NHANES III (20).
In addition, the NCHS conducted the Hispanic Health and Nutrition Examination Survey [HHANES, 1982–1984, ages 6 mo to 74 y (28)], a cross-sectional survey of the 3 major Hispanic subgroups in selected areas of the United States: Mexican Americans in the southwest; Cuban Americans in Dade County, Florida; and Puerto Ricans in the New York City metropolitan area. Dietary intake data from HHANES were reported previously (20) and allow comparison of intakes by Mexican Americans between HHANES and NHANES III.
The collection of data on ethnicity was limited in NHANES I and the sample in NHANES II did not provide for estimates based on ethnicity. Race was observed in NHANES I, NHANES II, and HHANES and was self-reported in NHANES III. In all the surveys, race was categorized as white, black, or other. Therefore, trend analyses were carried out by race (white and black) rather than by race-ethnicity. Sample sizes for the "other" race category were insufficient for separate reporting but were included in total population estimates.
Food-composition databases
The primary source of food-composition data for NHANES III was the US Department of Agriculture (USDA) survey nutrient database (22, 29, 30). All nutrient and dietary fiber intakes for individuals were calculated by using the gram amounts of food consumed and the nutrient values for the food as contained in the USDA survey nutrient database, expressed per 100 g food. The NCHS computed the percentage of total food energy intake from each macronutrient energy source. The energy conversion factor used for total fat and fatty acids was 9 kcal/g (38 kJ/g). The total energy from each macronutrient energy source was divided by the overall total energy intake and multiplied by 100 to produce the percentage contribution. The energy from total fat is greater than the sum of the contributions from saturated, monounsaturated, and polyunsaturated fatty acids because total fat includes lipid components that are not fatty acids.
The food-composition database used for NHANES I and II was maintained by the NCHS and was based primarily on the USDA's Handbook no. 8 and on manufacturers' data (31, 32). Dietary cholesterol intake values for NHANES I and NHANES II were reduced by 10% for trends comparison to HHANES and NHANES III. This was necessary to interpret dietary cholesterol trends properly because the data for eggs were changed in 1989 to incorporate data collected with use of improved analytic methods for measurement of cholesterol (33).
Determining beverage sources of energy and dietary fat
Preliminary analyses of the NHANES III foods data showed that beverages consistently appeared as one of the top contributors to energy for children and adolescents; thus, these sources were examined in further detail. Food groups were constructed that combined individual food items into the following 15 groups: fluid whole milk products, including condensed milk, evaporated milk, reconstituted dry milk, and chocolate milk; 2%-fat fluid milk products; 1%-fat and skim fluid milk products; milk shakes, malts, and other milk-based drinks; cream; meal replacement and cereal drinks; citrus juices, including citrus and noncitrus combinations; noncitrus juices and nectars; juice drinks (not 100% juice); vegetable juices; soft drinks; coffee; tea; alcoholic beverages; and all nondrink food items. The fluid groups primarily came from beverages, but could also include items such as milk added to cereal or coffee. A listing of food items included in each group is available on request.
The resulting food groups were analyzed for their mean proportional contribution to energy, total fat, and saturated fatty acids (34). The contributions of each group were also analyzed separately by overweight status. Overweight individuals were those whose body mass index (BMI) was greater than or equal to the sex- and age-specific 95th percentile of BMI from preliminary data for the revised NCHS-CDC growth charts (3). Results for children aged 2–5 y and 6–11 y are presented for boys and girls combined because of minimal variation by sex. Results for adolescents are presented by sex.
Statistical methods
Statistical analyses were carried out by using SAS version 6.09 (35) and SUDAAN release 7.00 (36). For each survey, sample weights were calculated that took into account the unequal probabilities of selection resulting from the cluster design, planned oversampling of selected subgroups, and unit nonresponse. All analyses incorporated the sample weights. Standard errors were calculated by using SUDAAN, a program that takes into account the sample weights and complex sample design for calculating variance estimates. To correct for changes in the age distribution over time when the different surveys were compared, estimates from each survey were age-adjusted by the direct method to the 1980 US Census population figures based on single years of age (37).
RESULTS
Current intakes
Mean and median intakes of energy, the percentage of energy from fat, and the percentage of energy from saturated fatty acids are presented in Table 1 by age, sex, and race-ethnicity. As expected, energy intake increased with age. Among males, mean energy intake increased by 1000 kJ with each successive age group and peaked at 12811 kJ for males aged 16–19 y. Among females, the increases were smaller and energy intake peaked at 8325 kJ for ages 12–15 y and then declined slightly at 16–19 y. Energy intake was greater for males than for females, particularly among adolescents. There was no clear ordering pattern of energy intake among racial-ethnic groups.
View this table:
TABLE 1. Intakes of energy, fat, and saturated fat of children and adolescents aged 2–19 y in the United States, 1988–19941
The percentage of energy from fat and saturated fatty acids varied little by age or sex, with overall means of 33.5% of energy from fat and 12.2% from saturated fatty acids for ages 2–19 y. Among the sex and age groupings examined by race-ethnicity, the mean percentage of energy from total fat was highest for non-Hispanic blacks. Among males aged 12–19 y, fat accounted for 35.7% of energy for non-Hispanic blacks compared with 33.2% and 34.1% for non-Hispanic whites and Mexican Americans, respectively. For females aged 12–19 y, corresponding values were 36.1% for non-Hispanic blacks compared with 33.4% for non-Hispanic whites and 34.1% for Mexican Americans.
There was less variation and no ordering pattern by sex and age for mean percentage of energy from saturated fatty acids. Saturated fat intake ranged from 11.6% of energy for females aged 12–15 y to 12.8% of energy for males aged 6–8 y. There also was no consistent pattern by race-ethnicity. Mean saturated fat as a percentage of energy ranged from 11.7% for non-Hispanic white females aged 12–19 y to 12.8% for Mexican American males aged 6–11 y.
Corresponding data for the percentage of energy from monounsaturated and polyunsaturated fatty acids and milligrams of cholesterol are shown in Table 2. Monounsaturated fatty acid intake ranged from 12.1% of energy for males aged 2–3 and 4–5 y to 12.9% of energy for males aged 16–19 y. Polyunsaturated fatty acid intake ranged from 5.9% of energy for males aged 2–3 and 4–5 y to 7.1% for females aged 12–15 y. Consistent with the differences in total fat intake by race-ethnicity, the mean percentage of energy from monounsaturated and polyunsaturated fatty acids was highest for non-Hispanic blacks. Among males aged 12–19 y, monounsaturated fatty acids accounted for 13.7% of energy for non-Hispanic blacks compared with 12.6% and 12.8% for non-Hispanic whites and Mexican Americans, respectively. The mean polyunsaturated fatty acid intake among males aged 12–19 y was 6.9% of energy for non-Hispanic blacks compared with 6.2% for non-Hispanic whites and 6.6% for Mexican Americans. For females aged 12–19 y, monounsaturated fatty acid intake was 13.5% for non-Hispanic blacks compared with 12.4% for non-Hispanic whites and 12.7% for Mexican Americans. Polyunsaturated fatty acid intake among females aged 12–19 y was 7.6% of energy for non-Hispanic blacks and 6.9% for non-Hispanic whites and Mexican Americans.
View this table:
TABLE 2. Intake of monounsaturated fat, polyunsaturated fat, and cholesterol of children and adolescents aged 2–19 y in the United States, 1988–19941
The pattern for cholesterol intake was similar to that for energy. There was a consistent increase with age for males; the highest mean intake was 375 mg for males aged 16–19 y. Among females, a less marked increase with age was observed and the peak intake (233 mg) occurred at ages 9–11 y, after which cholesterol intake declined in adolescence. Among males and females aged 2–5 and 6–11 y, Mexican Americans had the highest mean intake of cholesterol. The ordering by race-ethnicity did not hold for ages 12–19 y. For each racial-ethnic group among males aged 12–19 y, the mean, but not median, cholesterol intake exceeded 300 mg.
Secular trends
Energy intake
Mean energy intakes as reported in NHANES I, II, and III are shown in Table 3 by sex, age, and race. With the exception of adolescent females, there was little change in energy intake from NHANES I in the early 1970s through NHANES III. The NHANES III energy intake of 8264 kJ (1975 kcal) for females aged 12–19 y was 940 kJ (225 kcal) greater than energy intake in NHANES II and 669 kJ (160 kcal) greater than energy intake in NHANES I. The increase between surveys for black females was even larger, with the NHANES III value of 8647 kJ (2067 kcal) being 1040 kJ (249 kcal) greater than in NHANES II and 861 kJ (206 kcal) greater than in NHANES I.
View this table:
TABLE 3. Trends in age-adjusted mean energy intake of children and adolescents in the United States1
Total and saturated fat and cholesterol
Mean percentages of energy from total and saturated fat and cholesterol intake as reported in NHANES I, II, and III are shown in Table 4 by sex, age, and race. Over the period of the 3 national surveys, the contribution of both total fat and saturated fatty acids to energy intake declined slowly in all sex and age groups. The values for NHANES I and II, which were conducted about 5 y apart, tend to be similar, whereas NHANES III values are noticeably lower in all cases. Between the 1970s and NHANES III, the age-adjusted mean percentage of energy from total fat decreased from 36–37% to 33–34% and that for saturated fatty acids decreased from 14% to 12% across age and sex groups. Between NHANES I, II, and III, mean cholesterol intakes declined consistently across all age, race, and sex groups studied, with the exception of black males aged 12–19 y.
View this table:
TABLE 4. Trends in age-adjusted mean fat and saturated fat intake as a percentage of energy and in cholesterol intake of children and adolescents in the United States1
Beverage sources of energy and fat
The proportion of energy provided by beverages is shown in Table 5. Beverages contributed from 20% to 24% of energy across age and sex groups. In children aged <12 y, one-half or more of the beverage contribution to energy came from milk. However, in adolescents, the major beverage contributor to energy was soft drinks, providing 8% of total energy. For the groups aged 6 y, the contribution of juice drinks was greater than that from citrus juices and noncitrus juices combined. Among males aged 12–19 y, alcohol contributed 2% of total energy.
View this table:
TABLE 5. Percentage of energy intake contributed by beverages1
The proportion of total fat and saturated fatty acids provided by beverages is shown in Table 6. Only dairy beverages are shown because other beverage groups contributed 0.1% fat. The contribution of these beverages to total fat intake ranged from 16% in the youngest group to 7% among adolescent females. The contribution to saturated fatty acid intake was greater, ranging from 26% among ages 2–5 y to 12% among adolescent females.
View this table:
TABLE 6. Proportion of fat intake contributed by beverages1
When the contribution of beverages to energy and fat intakes was stratified by overweight status, with the exception of adolescent females, each group of overweight children and adolescents had 2% greater proportions of energy from beverages than did corresponding nonoverweight children and adolescents (data not shown). Overweight adolescent females obtained 2% less energy from beverages than did the nonoverweight group. For each age and sex group, soft drinks contributed a higher proportion of energy for overweight than for nonoverweight children and adolescents (2–5 y: 3.1% compared with 2.4%; 6–11 y: 5.4% compared with 4.0%; males, 12–19 y: 10.3% compared with 7.6%; females, 12–19 y: 8.6% compared with 7.9%).
DISCUSSION
US dietary recommendations and population-based guidelines to reduce cardiovascular disease risk and prevent coronary artery disease suggest that all healthy children aged 2 y eat a variety of foods to meet their nutritional needs and consume an adequate amount of energy to support normal growth and development and maintain a desirable body weight (12, 17). A change in the 1995 version of the Dietary Guidelines for Americans relative to earlier versions was an emphasis on energy balance through physical activity to promote weight maintenance (38). Secular trends in overweight prevalence clearly indicate that many children are in positive energy balance (3).
Mean energy intakes across the 3 NHANESs (from the early 1970s to 1988–1994) decreased slightly or remained stable for children aged 2–5 y and 6–11 y, but increased by 400 kJ (95 kcal) for adolescents aged 12–19 y. Mean energy intakes as reported in the USDA's national food consumption surveys changed little for young children but were slightly lower for adolescents in 1989–1991 than in 1977–1978 (20). These secular trends in energy intake suggest that increased intake over time is not the major contributor to the increased prevalence of overweight among youths in the United States (3).
Although energy intake for adolescent females in NHANES III was substantially greater than in earlier surveys, other sex and age groups that had comparable increases in overweight prevalence did not have higher reported energy intakes in the later surveys. Improved data collection methods, such as the use of automated dietary interviews, may account for part of the observed increase in energy intake among adolescent females, although an actual increase in intake or a combination of a real increase and improved assessment are also possible. However, it is not clear how such an improvement in assessment would be limited to adolescent girls and adults. This lack of association between energy intake and overweight prevalence at the population level does not rule out the possibility that individuals became overweight as a result of increased energy consumption. However, at least for children and younger adolescents, decreased physical activity was probably more important than increased energy intake as a contributor to the increase in overweight prevalence. It appears that energy intake did not decline sufficiently to balance the apparent secular decline in physical activity.
Previous studies documented a systematic bias of up to 25% in the underreporting of energy intake by adults in surveys using 24-h recalls (18, 21, 39, 40). The NHANES anthropometric and dietary data suggest that the degree of underreporting of energy intake in adults was lower in NHANES III than in the earlier NHANES (40). Improvements in the 24-h recall method, representative inclusion of weekend days, and quality-control procedures may have reduced underreporting in NHANES III compared with NHANES I and II and the USDA's Food Consumption Surveys. Even with improved methods, however, it is likely that underreporting of energy exists to some degree for a proportion of the survey population. Between NHANES II and NHANES III, mean energy intake increased 1–4% among most of the age groups under 20 y; mean intakes declined 3% for ages 6–11 y and increased 16% in females aged 16–19 y (21). Underreporting of energy has been studied less extensively in children and the effect of proxy reporting on underreporting of total energy intake is unclear.
Different nutrient databases were used in the different NHANESs. Although the effect of changes in food composition on energy estimates cannot be readily assessed, it is likely that changes in the dietary protocol to improve reporting of intake had a larger effect on energy estimates than did changes in the food-composition database. When improved methods to analyze the dietary cholesterol content of foods were introduced, adjustments were made to earlier survey data to allow comparison of cholesterol intakes over time.
The dietary recommendations suggest that total dietary fat not exceed 30% of energy, that saturated fatty acids be limited to <10% of energy, and that cholesterol intake be <300 mg/d (12, 17). Data from 3 NHANESs show that since the early 1970s US children and adolescents have continued to reduce their intakes of total fat and saturated fatty acids as a percentage of energy. On average, fat intake has declined from 36–37% of energy to 33–34% of energy, and the intake of saturated fatty acids has declined from 14% to 12% of energy. Declines were observed across all race, sex, and age groups. Similar declines in fat and cholesterol intakes were observed for Mexican Americans between HHANES and NHANES III; total fat decreased from 36% in 1982–1984 to 33–34% in 1988–1994 and saturated fatty acids decreased from 14% to 12–13% across age and sex groups (20).
Mean fat and saturated fat intakes from a single day, expressed as percentages of energy, still exceed dietary recommendations and targets set for the year 2000 health objectives for the nation (41–43). About 23% of children aged 2–5 y, 16% of children aged 6–11 y, and 15% of adolescents aged 12–19 y met the dietary recommendations for total fat intake in 1988–1991. Respective figures for meeting the saturated fat goal were 9% for ages 2–5 y and 6–11 y and 7% for ages 12–19 y, indicating that a smaller proportion met the saturated fat goals than the total fat goals (41–43). Data for 1 d from 1988–1994 showed further improvement, with 25% of individuals under the age of 20 y meeting the recommendations for both fat and saturated fat (44). Similarly, data from 3 d of intake in the 1989–1991 USDA national food consumption survey, the Continuing Survey of Food Intakes by Individuals (CSFII), found that <20% of males and females aged 6–11 and 12–19 y consumed 30% of energy from fat (20). More recent data from the 1994–1996 CSFII based on 2 d of intake indicated continued improvement and less disparity between the proportion of the population meeting goals for total fat and saturated fatty acid intakes. These data show that 30% of males and 34–35% of females aged 6–19 y had total fat intake at or below 30% of energy, and from 23.2% (females aged 6–11 y) to 33.5% (females aged 12–19 y) of youths aged 6–19 y had saturated fatty acid intakes <10% of energy (45).
With the exception of adolescent males, mean and median cholesterol intakes from NHANES III were below the recommended maximum intake of 300 mg/d. Analysis of CSFII 1989–1991 (46) also found that, except for white males, mean cholesterol intake for teens was close to, but not greater than, the recommended 300 mg/d. The higher cholesterol intake for adolescent males is consistent with their higher energy intake. On the basis of data from NHANES III, almost 75% of individuals under the age of 20 y had mean cholesterol intakes <300 mg (44).
In agreement with the declines in saturated fatty acid intakes, mean serum cholesterol values among adolescents aged 12–17 y declined by 7 mg/dL (0.181 mmol/L) between 1966–1970 and 1988–1994 (47). Adolescent trends in dietary fat and serum lipids are consistent with but less than those observed for adults. This positive trend in serum lipids in adolescents occurred despite an increase in the prevalence of overweight (3).
Intakes of energy, total fat, and saturated fatty acids did not differ significantly between overweight and nonoverweight children and adolescents (data not shown). However, overweight youths tended to consume a greater proportion of energy from beverages than did their nonoverweight counterparts. This tendency toward higher consumption was consistently evident in the contribution of soft drinks to total energy intake. Soft drinks provided 10.3% and 8.6% of energy for overweight males and females, respectively. Soft drinks and juice drinks are energy sources that provide proportionately fewer nutrients relative to milk or 100%-fruit juices.
Soft drinks contributed 8% of total energy to the intake of adolescents. The USDA national food consumption surveys documented an increase in the consumption of beverages by youths between the late 1970s and the mid-1990s, accompanied by a decline in milk consumption. The beverage increase was primarily observed in noncitrus juices and soft drinks. Soft drink consumption among teenage boys nearly tripled between the surveys in 1977–1978 and 1994 (48). These changes are cause for concern because soft drink consumption appears to contribute to low intake of calcium, magnesium, riboflavin, vitamin A, and vitamin C among teenagers (49). Low calcium intake and the resulting effect on bone health are a particularly critical issue for adolescent females (20, 46).
Although 2%-fat, 1%-fat, and skim milk are consumed more frequently than is whole milk by youths over the age of 5 y (48), the NHANES III data show that there is still room for further reduction in fat consumption from milk. Among youths aged 6 y, 10–18% of saturated fatty acid intake and 5–10% of total fat intake came from whole and 2%-fat milk products. For these age groups, most milk consumption was in the 2%-fat content group. Compared with the nonoverweight group, overweight adolescents tended to consume a greater proportion of the total milk products they reported as whole and 2%-fat products than as 1%-fat and skim products. Further shifts from whole or 2%-milk products to 1%-fat or skim products could reduce energy, fat, and saturated fat intakes while retaining the important nutrient contributions of dairy products.
The dramatic increase in overweight prevalence among youths and adults in the United States highlights the need to better understand both aspects of energy balance. Dietary intake methods and interpretation of data continue to improve and investigations of the errors in these methods are sophisticated (50). Measurements on the expenditure side of energy balance are comparatively immature. Measures of reported physical activity have not been used consistently in national surveys (51). Objective measures of total energy expenditure are available, but are still expensive and are not well suited to large-scale survey application (52, 53). Motion-detecting activity monitors hold promise for measuring the discretionary aspect of energy expenditure (51). Technological advances have led to reliable monitors that are simple for users and relatively inexpensive. The use of such devices in survey settings may be possible in the future.
Although more extensive than research on physical activity, most of the research on dietary intake methods has been conducted in adults. Underreporting has not been fully explored in children and adolescents. In younger children, the inclusion or sole use of proxy information further complicates the situation. Because precursors of chronic disease are found at younger ages, methods to improve monitoring and interpretation of dietary intake and physical activity in these groups are needed.
Future health objectives for the nation will continue to focus on behavioral changes in areas such as diet and physical activity to promote health and prevent disease. The absence of evidence for a general increase in energy intake among youths despite an increase in the prevalence of overweight supports the observation that inactivity is a public health challenge in our technological society (51). Efforts to increase physical activity and decrease nonnutritive sources of energy among youths may be important approaches to counter the rise in overweight prevalence. Beverage intake appears to be a potential target for improving the dietary intake of children and adolescents through a shift to lower-fat dairy products and a limiting of consumption of soft drinks. Population-based approaches and strategies will need to include health professionals, schools, families, government, the food industry, and the mass media.
ACKNOWLEDGMENTS
We thank Margaret McDowell for providing guidance on the structure and layout of the NHANES III foods files and general advice related to their use and Debra Reed-Gillette for invaluable programming assistance.
REFERENCES