This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Dietary Intake, Dietary Patterns, and Changes With Age

An Epidemiological Perspective

^a School of Public Health, University of California, Berkeley

Gladys Block, Earl Warren Hall, School of Public Health, University of California, Berkeley, CA 94720-7360 E-mail: gblock{at}uclink4.berkeley.edu.

	Abstract

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
Cohort and cross-sectional data were reviewed to describe the changes in dietary intake with age. Total energy intake decreases varied substantially with age, by 1000 to 1200 kcal in men and by 600 to 800 kcal in women. This resulted in concomitant declines in most nutrient intakes. For some nutrients, substantial numbers of older Americans consumed only one fifth to one third of the recommended dietary allowance. For most nutrients, research is lacking with which to judge the health impact of reduced nutrient consumption with age, although there is some evidence of an age-related decline in absorptive and metabolic function. With the aging of the population, more research is needed on nutrient requirements and health outcomes, and public health efforts are needed to increase physical activity and food intake among older people.

ALTHOUGH there is a growing understanding of the importance of diet in health promotion and disease prevention, our present knowledge of nutrient requirements of elderly individuals and changes in dietary intake with age is limited. This study presents data on how nutrient intakes and dietary habits change from adulthood through old age.

	Knowledge About Requirements of Older Persons

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
Although our science base is growing, knowledge of the nutrient requirements of elderly people remains inadequately documented. At present, the most widely accepted standards used for interpretation of dietary data are the National Academy of Sciences Recommended Dietary Allowances (RDAs) ⁽¹⁾. The RDAs are defined as the levels of intake of essential nutrients that, on the basis of scientific knowledge, are judged by the Food and Nutrition Board to be adequate to meet the known nutrient needs of practically all healthy persons ⁽¹⁾. There are several important considerations in assessing the adequacy of the RDA standards for older people. The RDAs provide for a single category of older people, applying a single set of standards to all persons aged 51 years and older. However, with the advancement of the aging process, changes in physical abilities, and the presence of chronic conditions, nutritional needs may be quite different as people progress from their 50s through the second half of their life span.

Although the RDAs have been the accepted reference available to health professionals and researchers for assessing diets of individuals and groups, they were not ideally suited for some of these purposes ⁽²⁾. In the near future, dietary reference intakes (DRIs) will be replacing the last issue of the RDA report. The decision to establish revised reference intakes was based on the following needs: (i) to critically review the underlying estimates and the criteria for adequacy; (ii) to assess new scientific evidence from a range of studies, including epidemiological studies of chronic disease; (iii) to include the concept of reduction in the risk of chronic disease; and (iv) to move away from deficiency diseases to more functional outcomes related to health and diseases. DRI is used as a collective term to accommodate various nutrient-based reference values, as shown in Table 1 ⁽³⁾.

	Issues in Estimating Change in Intake With Age

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
Data collection methods and databases
Assessing dietary change over time is limited by a number of methodological issues. Data collection methods differ (e.g., a single 24-hour recall vs multiple-day diet diaries). Even if the same dietary assessment method is used, there may be important differences from one time to another. Nutrient composition databases change with time. For instance, fiber values previously appeared as crude fiber (a subset of total fiber) and fiber data were available for only a limited number of food items. At present, U.S. Department of Agriculture (USDA) food tables contain values for total dietary fiber, not crude fiber, and for all food items. There have also been technological advances made in methods of measuring components of fiber. Valid comparisons of fiber intake over time are at best difficult to make. Similar problems exist in assessing changes in fat intake. For example, changes in the Third National Health and Nutrition Examination Survey (NHANES III) methods that may have influenced nutrient estimates included automated data collection, improved protocol for probing for information about food sources of dietary fat, and a different nutrient database ⁽¹⁰⁾.

Longitudinal versus cross-sectional data
What we know about dietary intake and patterns in elderly persons compared with younger adults comes from longitudinal and cross-sectional studies.

Most dietary data are cross-sectional. That is, data on intake in different age groups are obtained from different individuals, rather than by observing the same individuals as they age.

In a cross-sectional study, it is virtually impossible to separate changes attributable to cohort differences in food preferences from actual physiologic changes due to aging. As people age, they not only undergo physiological changes, but they also bring with them food behaviors that have evolved from the social, cultural, economic, and environmental history of their lifetime experiences. With the aging population, other issues also interfere with interpretation of cross-sectional studies. Nonresponse rate increases in the older age groups, potentially biasing the estimates. Fading memory may make it more difficult to report food intake accurately. Another issue is selective mortality, and the possibility that persons with certain dietary patterns may have better survival; in that case, we cannot be sure that those who have reached older ages—the survivors—demonstrate a pattern that represents changes with aging or one that is related to survival. In addition, food habits in older people are determined not only by lifetime preferences and physiological changes, but also by living arrangements, aloneness, finances, transportation access, and disability, among other issues. Thus, differences seen in cross-sectional studies must be interpreted with these issues in mind.

Longitudinal studies, which follow the same individuals through the life cycle, may seem to offer the best approach to examine changes over time. However, these studies, too, have shortcomings. As noted previously, dietary collection methods also evolve over time, making observed differences difficult to interpret. Nutrient intake is influenced not only by physiological factors associated with aging, but also by changes in the food supply (such as the availability of low-fat food items) and in public perceptions about what constitutes a desirable diet (such as recent emphasis on fiber intake). In addition, as noted previously, age-related changes in cognitive ability and differences between survivors and nonsurvivors can also make interpretation of longitudinal studies difficult.

In any longitudinal study, sample selection and representativeness of the cohort are issues that limit our confidence in the generalizability of these studies' results to the population at large. Most members of such cohorts are better-educated and more health-conscious. Finally, there are few longitudinal studies with large samples and with comprehensive dietary data collected over several decades, and even fewer with published results in the scientific literature.

Despite the problems with both types of data, both provide useful information and afford useful comparisons. Where findings are similar, our confidence in the interpretation of the cross-sectional studies is enhanced. Dissimilarities may also be useful, in aiding our understanding of factors such as cohort and survival effects. Despite the difficulty in attributing dietary changes to physiological, cohort, or cultural effects, the large national cross-sectional representative surveys provide the major source of accurate information on dietary intake of people at different stages of the life cycle.

	Description of the Data Sets

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
Longitudinal studies
There are several longitudinal studies that examine intakes of specific nutrients in a sample and health outcomes at a later date, including the Honolulu Heart Program, Nurses Health Study, Health Professionals Follow-Up Study, Adventists Health Study, Iowa Women's Study, and the Cardiovascular Heart Study ^{(11)(12)(13)(14)(15)(16)}. Emphasis in these studies is on health outcomes, and no published information was found on changes in nutrient intake in these populations over time. Cross-sectional results from the Framingham Study ⁽¹⁷⁾ have been reported on trends in diet over time, but no results have been published on changes with age. The sample included the original cohort, spouses, and offspring. For these reasons, data from the previously mentioned studies are not reported here.

Only a few longitudinal studies were identified that include dietary data on the whole diet, collected at more than one point in time, and on which results on dietary change over time have been published ^{(18)(19)(20)(21)(22)}. The studies are summarized in Table 2 .

In the 9-year longitudinal follow-up, analyses showed significant decreases in total fat and cholesterol consumption in the women ⁽¹⁹⁾. In the men, in addition to decreases in fat intake, there were decreases in energy and protein, whether expressed as absolute consumption or per kg of body weight.

The Baltimore Longitudinal Study of Aging (BLSA) ⁽²⁰⁾ gathered data between 1960 and 1987. The sample consisted of 180 men, who were predominately white, highly educated, upper-middle class, and living in the community. Their ages ranged from 35 to 74 years at entry into the study. Seven-day diet records were collected from 1961 to 1975. Subsequently, Hallfrisch and colleagues ⁽²¹⁾ conducted a longitudinal analysis on the sample of 105 men who completed at least 3 to 8 days of food records, over 3 decades. The strength of this study lies in the long duration of follow-up and in the fact that the same diet method and nutrient database was used for all analyses. Limitations arise from the nature of the sample, which represented primarily white, highly educated men.

In a longitudinal study of 20 years, Flynn and colleagues ⁽²²⁾ assessed diet change over time. The convenience sample included 144 University of Missouri male faculty members and staff. At entry, their ages ranged from the late 30s through 60s. The dietary assessment method consisted of 4-day diet records, every 4 to 6 years. Again, the follow-up period was long, but the sample size was small and unrepresentative.

National follow-up studies
Persons aged 55 years and older at the time of the NHANES I (1971–1975) were interviewed again in 1984 in the NHANES I Epidemiologic Follow-Up Study (NIEFS) ⁽²³⁾. A total of 2653 individuals with appropriate data were reexamined (1103 men, 1550 women; 2264 whites, 373 African Americans and 16 of other races). Of these, 922 were 55 to 64 years old and 1731 were aged 65 years and older at the NHANES I examination. At the time of the NIEFS, the subjects were 10 years older. The assessment instrument used in the NHANES I baseline was a 19-item food group questionnaire. The dietary assessment instrument used in the NIEFS analyses was an extensive 93-item food frequency questionnaire. Consequently, it is difficult to draw conclusions about dietary changes with any confidence.

National cross-sectional surveys
The National Center for Health Statistics and the USDA conduct periodic surveys of the health and nutritional status of representative samples of Americans. National-level cross-sectional surveys with comprehensive dietary data on the older American population include the Nationwide Food Consumption Surveys (NFCS), the National Health Interview Surveys (NHIS), the Continuing Survey of Food Intakes by Individuals (CSFII), and the NHANES ^{(24)(25)(26)(27)}. The surveys are summarized in Table 3 .

• The NHIS (1987 and 1992). The NHIS is conducted annually. In 1987 and 1992, the National Cancer Institute sponsored a special supplement that included dietary data. Data were obtained from 16,065 white, 2748 African American, and 1330 Hispanic respondents, 18 to 99 years old. The sample included 7412 persons (2873 men and 4539 women) 50 years and older, comprising approximately 63% of the total sample. There was more information available on dietary intake in elderly persons than was previously available in any national data set, and there was no upper age limit for inclusion. The 1987 and 1992 NHIS differ from other national surveys in that the dietary assessment consisted of a 60-item food frequency questionnaire rather than diet recalls or records. This was unique in that it provided estimates of usual intake of individuals, rather than 1-day's or few-days' intake, and estimates of the distribution of nutrient intake in the population. Nutrient intake data were reported by age, sex, and race based on data from the 1987 NHIS by Block and Subar ⁽²⁸⁾. Trend data were reported by Norris and colleagues ⁽²⁹⁾.
  
• The NHANES. These surveys include NHANES I (1971–1975), NIEFS (1976–1980), NHANES II (1976–1980), Hispanic HANES (HHANES; 1982–1984), and NHANES III (1988–1994). They are conducted by the National Center for Health Statistics. The first and second NHANES included persons only up to 74 years old. In NHANES III, there was no upper age limit for inclusion in the sample. The NHANES surveys used a 24-hour recall, conducted by a nutritionist or dietitian, and abstract three-dimensional volume models to improve assessment of portion size. NHANES III also included a 61-item food frequency questionnaire.
  

NHANES II (1976–1980). The study population over the age of 50 totaled approximately 9200; 40% of those were aged 66 to 74 years.

NHANES III (1989–1994). In NHANES III, there were over 6830 participants in the age categories of 50 years and older.

HHANES (1982–1984). The HHANES was a survey of three major Hispanic subgroups in the United States: Mexican Americans, Cubans, and Puerto Ricans. The sample consisted of civilian, noninstitutionalized Hispanics up to the age of 74. Respondents in the 50- to 74-year-old age bracket included 1413 Mexican Americans, 679 Cubans, and 575 Puerto Ricans. The Cuban population resided in Florida, the Puerto Rican population lived in the New York City metropolitan area, and the Mexican population resided in the five southwestern states. Samples of Cuban and Puerto Rican older adults were relatively small. There are no publications of findings on diet and the elderly population from this data set.

• The NFCS (1977–1978 and 1987–1988). These USDA surveys were conducted every 10 years, until replaced by the CSFII. Data from the 1977 to 1978 and the 1987 to 1988 NFCS are described later. The dietary data consisted of a 24-hour recall combined with 2 days of a diet record (3 consecutive days in all). The 1977 to 1978 total sample included approximately 36,000 participants, with 1012 men and 1655 women aged 65 years and older. The total number interviewed in the 1987 to 1988 NFCS was approximately 10,000, with approximately 1048 persons who were aged 65 years and older.
  
• CSFII (1994–1996). The CSFII has been conducted by the USDA since 1985. The 1985 and 1986 surveys included adults aged up to 50 years. Older persons were not included until 1989. Both the 1989 to 1991 and 1994 to 1996 surveys provide information on at least 2 days of food intakes among individuals of all ages. The 1994 to 1996 survey was completed by more than 15,000 individuals and included a larger sample of elderly individuals compared with the 1989 to 1991 survey. The 1994 to 1996 CSFII included more than 4700 men and approximately 4370 women aged 60 years and older. Only data from the later survey are reported here.
  

	Results

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
Energy and Macronutrients
Fig. 1, from NHANES II ⁽³⁰⁾ illustrates the energy intake pattern seen in both the cross-sectional and longitudinal studies: energy intake declines substantially across the age spectrum. The oldest age category in NHANES II, 70 to 74 years, had the lowest energy intakes. At all ages, men consumed more than women, but this difference decreased with age, as energy intake decreases faster in men than in women.

View larger version (13K): [in this window] [in a new window]   Figure 1. Mean daily caloric intake (95% confidence intervals), by age and sex. Data from NHANES II reproduced with permission from Block and colleagues (30).

Mean intakes of energy and macronutrients (fat, protein, and carbohydrates) from the first phase of NHANES III (1988–1991) are reported in detail in Table 5 ⁽³²⁾, for the entire age spectrum. Consistent across races, and for women and men, total food energy intake and macronutrients decreased with age. The percentage of calories from fat tended to be lower in those older than 60 than in younger age groups, usually reaching its lowest value in the oldest age groups. Although the percentage of calories from fat declined with age, the proportion of calories from protein tended to be higher than that of younger adults. Dietary cholesterol declined with age in both men and women and in all ethnic groups.

Estimates of nutrient intake are consistently lower in the CSFII (1994–1996) than NHANES III, but patterns of intake with age are similar to NHANES III. Total energy intake and macronutrients decreased with age.

The NHIS ⁽²⁸⁾, used an abbreviated food frequency questionnaire, and thus the absolute values are underestimates. The relative intakes, however, are consistent with the findings of the longitudinal and national survey data. Energy estimates decreased as age group increased with concurrent decreases in mean intake of macronutrients. This, in turn, has a considerable impact on intake of micronutrients.

Nutrient Density
As total energy intake decreases with age, the absolute amount of most nutrients also decreases. However, it is sometimes useful to consider the nutrient density of diets, that is, the nutrient intake as a proportion of total calories. It is often the case that energy intake declines more substantially than does intake of micronutrients. Fiber is a good illustration of this fact. In the Fig. 2 data from NHANES II ⁽³³⁾, it can be seen that absolute grams of fiber decrease with age in men. However, Fig. 2 shows fiber intake expressed as grams of fiber per 1000 kcal. It can be seen that, in terms of nutrient density, the proportion of dietary intake that is composed of fiber increases with age in men.

View larger version (15K): [in this window] [in a new window]   Figure 2. Dietary fiber intakes, by age, sex, and race in g/1000 kcal, A, and in g, B. Values are mean ± SEM. Adapted with permission from Lanza and colleagues (33).

Micronutrients
Table 6 presents the mean intake of several minerals, and Table 7 presents vitamins, by age group, from NHANES III ⁽³²⁾. With few exceptions, mineral intake declines with age after age 50, usually reaching its lowest point in the oldest age group. Decreases in calcium, iron, and zinc intake with age were seen in the following surveys: NHANES III, NHANES II, and CSFII (1994–1996), all based on 24-hour recalls; and in the NHIS, based on the food frequency questionnaire. In NHANES III data, white men aged 70 to 79 years obtained 832 mg of calcium and 12.2 mg of zinc, while white women obtained only 651 mg of calcium and 8.8 mg of zinc. Levels in other ethnic groups were lower. The current recommendations call for 1200 mg for calcium and 12 and 15 mg for zinc for women and men, respectively.

In contrast to the general fall in micronutrient intake with age, vitamin C and vitamin A or carotene intake did not decrease with age and, in fact, tended to increase with age. This pattern was consistent in NHANES II, NHANES III, and CSFII, all by 24-hour recall, and in NHIS by food frequency questionnaire. Vitamin A and carotene intake increased with age in both genders until the oldest age group, 80 years and older. Among whites and African Americans, vitamin C intake tended to decrease with age in men, but increased with age in women.

Distributions of Nutrient Intake
Distributions are of interest since they present a more accurate and complete picture of the population's nutrient intake. Distributions for most nutrients are not normally distributed, but are skewed to the right—that is, there is a long "tail" to the right, with some people consuming very high levels. This skewness drags up the mean, producing a higher mean than would be the case in a normal distribution. Consequently, means give an inflated estimate of the central tendency of the population. Medians are more appropriate, representing that value for which half the population consumed more and half the population consumed less of a nutrient.

The single 24-hour recall data collected in the NHANES surveys can provide precise estimates of mean intakes of nutrients, but are less appropriate for describing the distribution of intake. With 24-hour recall data, the distribution is wider (a wider, flatter bell-shaped curve), resulting in an exaggerated impression of the proportion of the population with very high or very low intakes. The 2-day nonconsecutive 24-hour recalls used in CSFII can provide more appropriate data on distributions, because intraindividual variability is somewhat smoothed over 2 days and there are fewer of the extreme values often seen in 24-hour recall data. (Several nonconsecutive 24-hour recalls would be ideal for describing distributions. With 2 days, the distribution of intakes obtained will still be wider for some nutrients than the distribution of true intakes.)

Fig. 3 illustrates the distribution of intakes of protein, vitamin A, and energy from the 2-day CSFII data. For protein, both the median and the entire distribution shift downward with age, for both men and women. Among young men, the median protein intake is slightly less than 100 g, although 10% of young men consume only about 50 g. In contrast to the pattern for protein intake, the distribution of vitamin A intake is shifted upward with age. Although the 75th percentile for young women is about 900 retinol equivalent (RE), the 75th percentile for older women is almost 1300 RE.

View larger version (40K): [in this window] [in a new window]   Figure 3. Distributions of nutrient intake in persons aged 20–29 and 80 years: A, protein; B, vitamin A; and C, energy distribution. — Men 20–29 years; –x– women 20–29 years; men 80+ years; and hourglass shape = women 80+ years.

Only about 5% of older women consume the DRI recommendation for calcium (1200 mg), and only 10% of older men consume the recommended 1200 mg of calcium. In addition, 75% of women do not regularly consume the 12 mg RDA for zinc. Seventy-five percent of men aged 50 to 69 years and 90% of men older than 70 years do not consume the recommendation of 15 mg of zinc. Twenty-five percent of older women and 10% of older men consume only 5 mg or less.

As many as 75% of older women and men consume less than the RDA for vitamin E—8 and 10 -tocopherol equivalents, respectively. Approximately 95% of older men and 90% of older women do not consume the DRI of folate, which is 400 µg. For the B vitamins, thiamin and riboflavin, more than 25% of both men and women have marginal levels compared with their respective DRIs.

Over half of men and women meet the current RDA for Vitamin C of 60 mg. However, a substantial minority of the population consumes far less, based on an average of 2 nonconsecutive days. Ten percent of older women consume only two thirds of the current RDA, and 10% of older men consume only about one fourth of the RDA.

Eating Patterns
One of the earliest publications characterizing the eating patterns of the elderly population in the United States, by Fanelli and Stevenhagen ⁽³⁵⁾, was based on data from the 1977 to 1978 NFCS. There were 159 participants older than 85 years. Core foods, those routinely consumed by a population group, were identified, and there were no marked differences between men and women or among three different age groups (55–64, 65–74, and 75 years and older). Whole milk, white bread, coffee, and sugar were the most frequently mentioned foods. Orange juice and bananas were the most frequently mentioned fruits. The most commonly used vegetables were tomatoes, potatoes, and lettuce. Eggs and milk were the more frequent protein-rich foods reported.

Murphy and colleagues ⁽³⁶⁾ examined diets of those aged over 65 years in this same data set and reported a higher percentage of the older age group, 85 years and older, had poorer quality diets. Diet quality was defined as a diet that provided two thirds or more of the RDA for nine selected nutrients.

Food group consumption and dietary patterns in elderly individuals were examined using data from NHANES I and the NIEFS ⁽³⁷⁾. Subjects were 55 years and older at baseline, and 65 and older at follow-up. NHANES I used a 19-item food frequency questionnaire representing broad food groups, while NIEFS used a 93-food-item questionnaire. The daily number of servings of food groups was calculated from these instruments. The later survey produced higher estimates of intake of milk and cheese, butter and/or margarine, salty snacks, fish and/or shellfish, servings of fruits and vegetables, cereals, and legumes and/or nuts, and lower estimates of servings of sweets and alcoholic beverages. Because the number and specificity of food items on a questionnaire influence responses, these results are difficult to interpret.

Popkin and colleagues ⁽³⁸⁾ reported on dietary changes in older Americans based on cross-sectional data from the 1977 to 1978 and the 1987 to 1988 NFCS. They found that the top ten sources of energy, fat, and fiber in 1977 remained the top ten in 1987. The investigators reported shifts from consumption of high-fat to low-fat milk and milk products, and from high-fat beef and pork to low-fat chicken and fish.

Similar data were derived from NHANES III by Alaimo and colleagues ⁽³²⁾. Table 10 lists the top 50 sources of energy intake for persons aged 20 to 30 years and those aged 70 years and older.

The most recent data on food patterns from a national survey come from the CSFII (1994–1996) ⁽⁴⁰⁾. Information collected included sources of food, eating-occasion variables, patterns, and other health-related variables. A higher proportion of elderly persons eat breakfast than do younger adults. Over 90% of both men and women aged 60 years and older eat breakfast, with breakfast contributing over 20% of daily calories, 17.2% to 19.0% of daily total fat for men, and 16.7% to 17.8% of daily total fat for women. Among adults, calories from foods obtained and eaten away from home are highest among those aged 20 to 29 years (approximately 37% of energy for men and 34% for women) and lowest among those aged 70 years and older (approximately 16% for men and 11% for women).

Patterson and colleagues ⁽⁴¹⁾ reported on the U.S. population's consumption of fruits and vegetables. The proportion of those meeting the recommended servings was higher among older than among younger persons, and this was true for both whites and African Americans. Nevertheless, fewer than one third of older persons ate the recommended servings of vegetables, and fewer than half ate the recommended servings of fruit. Among older persons, more men had adequate numbers of servings of vegetables (including potatoes) than did women. Women, on the other hand, were more likely than men to have adequate servings of fruit. Even fewer African Americans than whites met the USDA guidelines for recommended servings of fruits and vegetables. Among African Americans, less than one fourth of those in the older age group ate the recommended servings of vegetables, and fewer than one third ate the recommended servings of fruit.

In general, the cross-sectional studies on food patterns described previously have found that older people are less likely to consume red meat, whole milk, and other fatty foods than younger people, and are more likely to consume fruits and vegetables than are younger people. That this represents actual age differences and not just a cohort effect is supported by the longitudinal data cited above ^(21)(22)(31) and Sobell and colleagues ⁽³⁹⁾. This obviously reflects temporal trends to some extent (as in reference 38) that cannot be clearly separated from the changes due to the aging process. However, it appears that these changes in desirable directions may be larger among older than younger persons, and are larger and more consistent among older women than among older men.

Supplement Use
A significantly larger proportion of older persons than younger persons take vitamin supplements ⁽⁴²⁾. In data from NHANES I (1971–1975), persons aged 65 to 74 years consumed more supplements than persons in each of the younger age categories. Exceptions were for vitamin E, which was consumed more commonly in the middle years, and iron supplements, which were consumed most among young women. The proportion consuming single-entity supplements regularly was approximately twice as high in the oldest as compared with the youngest age groups.

More recent data on supplement use were obtained from the 1987 NHIS. As in earlier surveys, older people were significantly more likely to consume vitamin supplements ⁽⁴³⁾. Women were more likely to consume supplements regularly than were men of the same age and race. Daily usage among white women was highest within the age range of 55 to 65 years (39.9%) and 65 to 74 years (38.4%) and declined slightly for those aged 75 years and older (34.9%). Among age-sex categories, whites tended to consume the most supplements compared with all other races. Hispanics consumed supplements at a frequency intermediate between those of whites and African Americans.

Slesinski and colleagues ⁽⁴⁴⁾ reported on the trends in use of vitamin and mineral supplements in the U.S. population based on the 1987 and 1992 NHIS. There were few meaningful changes in the older age groups of 55 years and older, with the exception of calcium use, which declined between 1987 and 1992.

Total intake of nutrients, intake from supplements added to nutrient intake from food, is not available from most surveys because supplement data were not obtained in detail.

	Discussion

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
The most important observation from the information reviewed previously is also one that comes as no surprise: The data from both cohort and cross-sectional surveys are consistent in indicating a very substantial decline with age in energy intake and quantity of food consumed. Mean energy intake declines by 1000 to 1200 kcal in men and by 600 to 800 kcal in women between those aged in their 20s and those in their 80s (Table 5 ). By age 80, 10% of men consume 890 kcal or fewer per day, and 10% of women consume 750 kcal or fewer (Table 8 and Table 9 ), based on data from 2 nonconsecutive days of intake.

This decline presumably follows in part from a decrease in physical activity and in part from the decline in muscle mass with age. This results in a lower requirement for energy. (Substantial decrements in energy intake may, in turn, result in lower physical activity, a declining cycle.) Potential problems arise because, as total food intake declines, for most nutrients there is a concomitant decline in nutrient intake. Table 8 and Table 9 suggest potentially important declines with age in median protein, zinc (both down by approximately one third in men), calcium, vitamin E, and other nutrients, especially in men. Further examination of the lower end of the distribution (e.g., the 10th percentile) suggests that a substantial number of minority elders consume grossly less than the RDA. In the 2-day data in Table 9 , Table 10 % of men obtained only one fifth to one third of the recommendations for protein, zinc, calcium, vitamin E, thiamin, riboflavin, vitamin B6, and vitamin B12.

Unfortunately, there is little evidence on which to base a judgment about the adequacy of these nutrient intakes that are concomitant with lower energy intakes in elderly individuals. For protein, the RDA committee notes that "there is surprisingly little information on which to base recommendations for protein intake in the elderly." Similar statements are made for other nutrients.

Furthermore, it is noted repeatedly by the RDA committee that there is no current evidence for most nutrients that absolute requirements decrease with age. Indeed, there is some evidence that absorption and utilization efficiency declines with age for some nutrients, potentially resulting in increased rather than decreased requirements for intake.

Occasionally, nutrient intake is expressed as a proportion of energy intake (i.e., nutrient intake per 1000 kcal). When expressed this way, it is clear that for at least some nutrients, the nutrient density of the diet does not decrease with age. That is, the energy intake declines with age at a faster rate than the specific nutrient intake. However, there is no evidence that this is an appropriate reflection of nutrient needs, for most nutrients. For example, women in their 20s consume approximately 0.36 mg of calcium per 1000 kcal, while for women in their 80s, this figure is 0.40 mg of calcium per 1000 kcal. Expressed this way, calcium intake in older women appears no less than that in younger women. However, it seems more important from a health standpoint to note that 524 mg of calcium (the median among women in their 80s) is both severely insufficient and is lower than that of younger women.

Another observation of interest in these data is the apparent health consciousness of older women. Although energy intake declines with age, the absolute value of certain nutrients actually increases with age. Vitamin A, vitamin C, and potassium intakes are all substantially higher in women in their 80s compared with those in their 20s (Table 8 ). These nutrients are found especially in fruits and vegetables, and this observation is consistent with the food pattern data cited previously. Older people, particularly older women, are more likely to consume fruits and vegetables. This general health consciousness is also seen in the much higher proportion of older women who consume vitamin supplements regularly.

Again, however, it is always important to observe the distribution of intake. Not all older women eat healthfully. For vitamin C, for example, 10% of women in their 80s obtained only 19 mg of vitamin C or less (about one third of the RDA) in a 2-day average.

	Recommendations

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 
Nutrition research is urgently needed to determine metabolic changes and nutrient needs of older people.

Outcomes research is needed to determine prospectively whether persons who maintain a higher nutrient intake with age have better health outcomes.

Intervention research is needed to determine whether providing increased macro- and micronutrients to elderly individuals can prevent or mitigate some of their health problems.

Public health efforts are needed to promote the maintenance of physical activity levels and food intake in elderly persons.

	References

Top Abstract Knowledge About Requirements of... Issues in Estimating Change... Description of the Data... Results Discussion Recommendations References

 

1. National Research Council1989. Recommended Dietary Allowances 10th ed. National Academy Press;, Washington, DC.
2. National Academy of Sciences1997. Dietary reference intakes. Nutr Rev. 55:319-326. [Medline]
3. National Academy of Sciences1997. Uses of dietary reference intakes. Nutr Rev. 55:327-331. [Medline]
4. Campbell W, Crin M, Dallal G, Young V, Evans W, 1994. Increased protein requirement in elderly people: new data and retrospective reassessments. Am J Clin Nutr. 60:501-509. [Abstract/Free Full Text]
5. Christensen H, 1993. Riboflavin can protect tissues from oxidative injury. Nutr Rev. 51:149-150. [Medline]
6. Young R, 1992. Energy requirements in the elderly. Nutr Rev. 50:95-101. [Medline]
7. Russell RM, 1997. New views on the RDAs for older adults. J Am Diet Assoc. 97:515-518. [Medline]
8. Wood RJ, Suter PM, Russell RM, 1995. Mineral requirements of elderly people. Am J Clin Nutr. 62:493-505. [Abstract/Free Full Text]
9. Russell RM, Suter PM, 1993. Vitamin requirements of elderly people: an update. Am J Clin Nutr. 58:4-14. [Abstract/Free Full Text]
10. Centers for Disease Control and Prevention1994. Daily dietary fat and total food-energy intakes—Third National Health and Nutrition Examination Survey, Phase 1, 1988–91. MMWR Morb Mortal Wkly Rep CDC Surveill Summ. 43:116-120.
11. Yano K, Reed DM, McGee DL, 1984. Ten-year incidence of coronary heart disease in the Honolulu Heart Program: relationship to biologic and lifestyle characteristics. Am J Epidemiol. 119:653-666. [Abstract/Free Full Text]
12. Colditz GA, Manson JE, Hankinson SE, 1997. The Nurses' Health Study: 20-year contribution to the understanding of health among women. J Womens Health. 6:49-62. [Medline]
13. Rimm EB, Katan MB, Ascherio A, Stampfer MJ, Willett WC, 1996. Relation between intake of flavonoids and risk for coronary heart disease in male health professionals. Ann Intern Med. 125:384-389. [Abstract/Free Full Text]
14. French SA, Folsom AR, Jefery RW, Williamson DF, 1999. Prospective study of intentionality of weight loss and mortality in older women: the Iowa Women's Health Study. Am J Epidemiol. 149:504-514. [Abstract/Free Full Text]
15. Singh PN, Fraser GE, 1998. Dietary risk factors for colon cancer in a low-risk population. Am J Epidemiol. 148:761-774. [Abstract/Free Full Text]
16. Kumanyika S, Tell GS, Shemanski L, Polak J, Savage PJ, 1994. Eating patterns of community-dwelling older adults: the Cardiovascular Health Study. Ann Epidemiol. 4:404-415. [Medline]
17. Posner BM, Franz MM, Quatromoni PA, et al. 1995. Secular trends in diet and risk factors for cardiovascular disease: the Framingham Study. J Am Diet Assoc. 95:171-179. [Medline]
18. Garry PJ, Rhyne RL, Halioua L, Nicholson C, 1989. Changes in dietary patterns over a 6-year period in an elderly population. Ann N Y Acad Sci. 561:104-112. [Medline]
19. Garry PJ, Hunt WC, Koehler KM, VanderJagt D, Vellas BJ, 1992. Longitudinal study of dietary intakes and plasma lipids in healthy elderly men and women. Am J Clin Nutr. 55:682-689. [Abstract/Free Full Text]
20. Hallfrisch J, Tobin JD, Muller DC, Andres R, 1988. Fiber intake, age, and other coronary risk factors in men of the Baltimore Longitudinal Study (1959–1975). J Gerontol Med Sci. 43:M64-M68.
21. Hallfrisch J, Muller D, Drinkwater D, Tobin J, Andres R, 1990. Continuing diet trends in men: the Baltimore Longitudinal Study of Aging (1961–1987). J Gerontol Med Sci. 45:M186-M191.
22. Flynn MA, Nolph GB, Baker AS, Krause G, 1992. Aging in humans: a continuous 20-year study of physiologic and dietary parameters. J Am Coll Nutr. 11:660-672. [Abstract]
23. Cornoni-Huntley JC, Huntley RR, Feldman JJ. Health Status and Well-Being of the Elderly National Health and Nutrition Examination Survey-I Epidemiologic Follow-Up Study. New York: Oxford University Press; 1990.
24. U.S. Department of Agriculture. USDA Methodological Research for Large Scale Dietary Intake Surveys, 1975–1988. Washington, DC: U.S. Government Printing Office; 1989. Home Economics Research Report Number 49.
25. National Center for Health Statistics. The National Health Interview Survey Design 1973–1984 and Procedures 1975–1983. Hyattsville, MD: U.S. Department of Health and Human Services; 1985. DHHS publication PHS 85–1320, Series 1, No. 18.
26. Tippett KS, Cypel YS, eds. Design and Operation: The Continuing Survey of Food Intake by Individuals and 1996 Diet and Health Knowledge Survey, 1994–1996. Washington, DC: U.S. Department of Agriculture; 1996. Nationwide Food Surveys Report No. 96–1.
27. National Center for Health Statistics. Plan and Operation of the Second National Health and Nutrition Survey, 1976–1980. Rockville, MD: National Center for Health Statistics; 1981. Vital and Health Statistics Series 11, No. 231. DHHS publication PHS 83–1681.
28. Block G, Subar AF, 1992. Estimates of nutrient intake from a food frequency questionnaire: the 1987 National Health Interview Survey. J Am Diet Assoc. 92:969-977. [Medline]
29. Norris J, Harnack L, Carmichael S, Pouane T, Wakimoto P, Block G, 1997. US trends in nutrient intake: the 1987 and 1992 National Health Interview Surveys. Am J Public Health. 87:740-746. [Abstract/Free Full Text]
30. Block G, Rosenberger WF, Patterson BH, 1988. Calories, fat and cholesterol: intake patterns in the US population by race, sex and age. Am J Public Health. 78:1150-1154. [Abstract/Free Full Text]
31. Elahi VK, Elahi D, Andres R, Tobin JD, Butler MG, Norris AH, 1983. A longitudinal study of nutritional intake in men. J Gerontol. 38:162-180.
32. Alaimo K, McDowell MA, Briefel RR, et al. Dietary Intake of Vitamins, Minerals, and Fiber of Persons Ages 2 Months and Over in the United States: Third National Health and Nutrition Examination Survey, Phase 1, 1988–1991. Hyattsville, MD: National Center for Health Statistics; 1994. Advance Data From Vital and Health Statistics, No 258.
33. Lanza E, Jones DY, Block G, Kessler L, 1987. Dietary fiber intake in the US population. Am J Clin Nutr. 46:790-797. [Abstract/Free Full Text]
34. Murphy SP, Subar AF, Block G, 1990. Vitamin E intakes and sources in the United States. Am J Clin Nutr. 52:361-367. [Abstract/Free Full Text]
35. Fanelli MT, Stevenhagen KJ, 1985. Characterizing consumption patterns by food frequency methods: core foods and variety of foods in diets of older Americans. J Am Diet Assoc. 85:1570-1576. [Medline]
36. Murphy SP, Rose D, Hudes M, Viteri FE, 1992. Demographic and economic factors associated with dietary quality for adults in the 1987–1988 Nationwide Food Consumption Survey. J Am Diet Assoc. 92:1352-1357. [Medline]
37. Murphy SP, Everett DF, Dresser C, 1990. Dietary patterns. Cornoni-Huntley JC, Huntley RR, Feldman JJ, , ed.Health Status and Well-Being of the Elderly National Health and Nutrition Examination Survey-I Epidemiologic Follow-Up Study Oxford University Press, New York.
38. Popkin BM, Haines PS, Patterson RE, 1992. Dietary changes in older Americans, 1977–1987. Am J Clin Nutr. 55:823-830. [Abstract/Free Full Text]
39. Sobell J, Block G, Koslowe P, Tobin J, Andres R, 1989. Validation of a retrospective questionnaire assessing diet 10–15 years ago. Am J Epidemiol. 130:173-187. [Abstract/Free Full Text]
40. U.S. Department of Agriculture. Data tables: results from USDA's 1996 Continuing Survey of Food Intake by Individuals and 1996 Diet and Health Knowledge Survey [database online]. Available at: www.barc.usda.gov/bhnrc/pdf/trend.pdf
41. Patterson BH, Block G, Rosenberger WF, Pee D, Kahle LL, 1990. Fruit and vegetables in the American diet: data from the NHANES II survey. Am J Public Health. 80:1443-1449. [Abstract/Free Full Text]
42. Block G, Cox C, Madans J, Schreiber GB, Licitra L, Melia N, 1988. Vitamin supplement use, by demographic characteristics. Am J Epidemiol. 127:297-309. [Abstract/Free Full Text]
43. Subar AF, Block G, 1990. Use of vitamin and mineral supplements: demographics and amounts of nutrients consumed. The 1987 Health Interview Survey. Am J Epidemiol. 132:1091-1101. [Abstract/Free Full Text]
44. Slesinski MJ, Subar AF, Kahle LL, 1995. Trends in use of vitamin and mineral supplements in the United States: the 1987 and 1992 National Health Interview Surveys. J Am Diet Assoc. 95:921-923. [Medline]

This article has been cited by other articles:

				H. Schroder, J. Vila, J. Marrugat, and M.-I. Covas Low Energy Density Diets Are Associated with Favorable Nutrient Intake Profile and Adequacy in Free-Living Elderly Men and Women J. Nutr., August 1, 2008; 138(8): 1476 - 1481. [Abstract] [Full Text] [PDF]

				B. N. Ames Low micronutrient intake may accelerate the degenerative diseases of aging through allocation of scarce micronutrients by triage PNAS, November 21, 2006; 103(47): 17589 - 17594. [Abstract] [Full Text] [PDF]

				R. D. Semba, B. Bartali, J. Zhou, C. Blaum, C.-W. Ko, and L. P. Fried Low serum micronutrient concentrations predict frailty among older women living in the community. J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2006; 61(6): 594 - 599. [Abstract] [Full Text] [PDF]

				M Hickson Malnutrition and ageing Postgrad. Med. J., January 1, 2006; 82(963): 2 - 8. [Abstract] [Full Text] [PDF]

				D. M. Gordon, S. E. Stern, and P. J. Collignon Influence of the age and sex of human hosts on the distribution of Escherichia coli ECOR groups and virulence traits Microbiology, January 1, 2005; 151(1): 15 - 23. [Abstract] [Full Text] [PDF]

				B. N. Ames Supplements and Tuning Up Metabolism J. Nutr., November 1, 2004; 134(11): 3164S - 3168S. [Full Text] [PDF]

				M.-M. G. Wilson and J. E. Morley Invited Review: Aging and energy balance J Appl Physiol, October 1, 2003; 95(4): 1728 - 1736. [Abstract] [Full Text] [PDF]

				E. Ho, C. Courtemanche, and B. N. Ames Zinc Deficiency Induces Oxidative DNA Damage and Increases P53 Expression in Human Lung Fibroblasts J. Nutr., August 1, 2003; 133(8): 2543 - 2548. [Abstract] [Full Text] [PDF]

				C. Murphy, C. R. Schubert, K. J. Cruickshanks, B. E. K. Klein, R. Klein, and D. M. Nondahl Prevalence of Olfactory Impairment in Older Adults JAMA, November 13, 2002; 288(18): 2307 - 2312. [Abstract] [Full Text] [PDF]

				H. Atamna, D. W. Killilea, A. N. Killilea, and B. N. Ames Heme deficiency may be a factor in the mitochondrial and neuronal decay of aging PNAS, November 12, 2002; 99(23): 14807 - 14812. [Abstract] [Full Text] [PDF]

				J. A. Luchsinger, M.-X. Tang, S. Shea, and R. Mayeux Caloric Intake and the Risk of Alzheimer Disease Arch Neurol, August 1, 2002; 59(8): 1258 - 1263. [Abstract] [Full Text] [PDF]

				A. Drewnowski and W. J. Evans Nutrition, Physical Activity, and Quality of Life in Older Adults: Summary J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2001; 56(90002): 89 - 94. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation