Filetype PDF | Posted on 05 Jan 2023 | 2 years ago
Authentication
digital resources
206x Filetype PDF File size 0.09 MB Source: paulogentil.com
Nutritional Epidemiology
Mineral Intakes of Elderly Adult Supplement and Non-Supplement Users in
the Third National Health and Nutrition Examination Survey
R. Bethene Ervin1 and Jocelyn Kennedy-Stephenson
Centers for Disease Control/National Center for Health Statistics, Hyattsville, MD 20782
ABSTRACT Calcium, iron and zinc are important in many of the body’s functions. We report dietary and
combined(dietsupplements)intakesforthesemineralsforelderlysupplementandnon-supplementusersinthe
United States and the prevalence of inadequate intakes. We calculated usual dietary intakes for adults 60 y and
older from third National Health and Nutrition Examination Survey, 1988–94; mineral intakes from supplements and
calcium-containing antacids were added to usual dietary intakes. We evaluated iron and zinc intakes using the
dietary reference intakes, recommendeddietaryallowancesandestimatedaveragerequirementsforelderlyadults,
as well as calcium intakes using the Adequate Intake and the Healthy People 2010 objective. The highest
prevalences of inadequate dietary intakes was for calcium (males, 70–75%; females, 87%) and zinc (males,
35–41%; females, 36–45%). Dietary supplements improved intakes, but nearly two-thirds of elderly adults had
combined intakes below the calcium objective. Non-Hispanic blacks usually had lower intakes than non-Hispanic
whites and higher prevalences of intakes below the standards. Supplement users had significantly higher mean
dietary intakes than non-supplement users for all three minerals for total females and non-Hispanic white females
(P 0.05foreachmineral).Manyelderlyadultshadinadequatedietaryzincintakes,andcalciumintakesfellbelow
the Healthy People 2010 objective; dietary supplements improved intakes. Even with supplements most older
adults still had intakes below the calcium objective, partly because the supplements they took usually contained
lowdosesofcalcium.Totalfemaleandnon-Hispanicwhitefemalesupplementusersweretheonlygroupsthathad
higher dietary intakes than non-supplement users for all three minerals. J. Nutr. 132: 3422–3427, 2002.
KEY WORDS: ● dietary intake ● dietary supplements ● elderly ● minerals ● NHANES III
Diet and nutrition play important roles in maintaining strength and increased risk of fractures (11). Increased calcium
health and preventing disease (1,2). This is especially impor- intakes may improve calcium retention and reduce fracture
tant for elderly adults, for whom proper nutrition plays a rates (9).
crucial role in helping them maintain good health and func- Iron is part of many proteins, including ones involved in
tioning. Many older adults are at increased risk of inadequate oxygentransporttotissues, and acts as an electron carrier (12).
nutritional intakes due to lower intakes of energy and other Iron deficiency can lead to anemia and to impaired cellular
nutrients. Other risk factors for poor nutrition include disease, and humoral immunity (12,13). Iron overload may contribute
physical limitations and chewing difficulties, polypharmacy, to coronary artery disease, but the evidence is conflicting. Iron
living alone, lack of transportation and limited income (3,4). is essential for normal neurological functioning, and changes
Although many seniors use dietary supplements, the supple- in iron metabolism in the brain may be associated with certain
ment users often already have adequate dietary intakes (5–8). neurological diseases, such as Alzheimer’s disease (13).
This paper focuses on calcium, iron and zinc intakes among Zincplays a catalytic role for many enzymes and a structural
older adults in the United States. Calcium provides strength role for other proteins and enzymes. It also plays a role in
and hardness to bones and teeth and mediates vascular con- regulating gene expression (14). Zinc deficiency may result in
striction and vasodilation, muscle contraction, transmission of growth retardation and sexual immaturity, impaired immune
nerve impulses and blood clotting (9,10). Osteoporosis, a function, taste and smell dysfunction and eye and skin lesions
disease affecting 28 million Americans over the age of 65 y, (14,15).
is characterized by a decline in bone mass, reduced bone Weexamined the calcium, iron and zinc intakes of elderly
adults from the third National Health and Nutrition Exami-
nation Survey (NHANES III), 1988–1994, comparing non-
1 To whom correspondence should be addressed. E-mail: bervin@cdc.gov. supplement users’ dietary intakes to supplement users’ dietary
2 Abbreviations used: AI, adequate intake; BNSS, Boston Nutritional Status intakes and to their combined intakes (dietary intakes sup-
Survey; CSFII, Continuing Survey of Food Intakes by Individuals; DRI, dietary plements). Also, we examined the prevalence of dietary and
reference intake; EAR, estimated average requirement; MEC, mobile examination combined intakes below dietary reference intakes (DRI)
center; NHANES, National Health and Nutrition Examination Survey; RDA, rec-
ommended dietary allowance. (12,14) or Healthy People 2010 objectives (16).
0022-3166/02 $3.00 © 2002 American Society for Nutritional Sciences.
Manuscript received 16 May 2002. Initial review completed 14 June 2002. Revision accepted 6 August 2002.
3422
MINERAL INTAKES BY SUPPLEMENT USE PATTERN 3423
where SD is the between-person standard deviation and
METHODS SD Between
Sample population and definitions. The Centers for Disease Total observed is the total observed standard deviation (23). We
Control and Prevention’s National Center for Health Statistics con- transformed the adjusted values back to their original scale for sub-
ducted NHANES III, which was designed to collect information on sequent analysis.
the civilian, noninstitutionalized U.S. population 2 mo of age. A Estimating usual dietary supplement and antacid intakes. We
detailed description of the NHANES III plan, operation, and sample calculated the average daily supplement dose for supplement users by
design appears elsewhere (17). All procedures were approved by the dividing the product of the frequency of use and dosage by 30.4 d/mo.
NCHSInstitutionalReviewBoard,andwritteninformedconsentwas The average daily mineral intake was calculated by multiplying the
obtained from all subjects. Interviewers collected data during a house- respondent’s average daily supplement dose by the concentration of
hold interview and at a follow-up physical examination at a mobile the mineral in one dose of the supplement. Finally, we summed the
examination center (MEC). intakes from all the supplements containing these minerals. We
Age was the self-reported age at the time of the household identified 18 calcium-containing antacids reported in this survey and
interview. The self-reported race and ethnic group classifications were calculated the average daily calcium intake from these antacids using
combined to create the following categories: non-Hispanic white, the same procedures described for the dietary supplements.
non-Hispanic black, Mexican American and “other” races/ethnici- Evaluating nutrient adequacy. We compared respondents’ mean
ties. iron and zinc intakes to the recommended dietary allowances (RDA)
Data collection. Trained interviewers, bilingual in English and and the estimated average requirements (EAR) for elderly males and
Spanish, collected a 24-h dietary recall during the respondent’s visit females from the DRI. The RDA is the intake level sufficient to meet
to the MEC using an automated dietary interview and coding system the needs of nearly all healthy individuals in a group, whereas the
(17,18). Nutrient values were assigned to the dietary recalls using EAR is the amount used to assess the prevalence of inadequate
food composition data from the U.S. Department of Agriculture’s intakes within a group. The RDA for iron for elderly males and
Survey Nutrient Database (19). NHANES staff added nutrient values females is 8 mg/d; the RDA for zinc is 11 mg/d for males and 8 mg/d
for new products and product reformulations to the database (17,18). for females. The iron and zinc EARs for elderly males are 6 and 9.4
During the household interview, respondents were asked whether mg/d, respectively. The iron and zinc EARs for elderly females are 5
they took any dietary supplements or antacids during the past month, and 6.8 mg/d, respectively (12,14).
and the frequency and dose. We defined a dietary supplement as any The Food and Nutrition Board determined there was insufficient
vitamin and/or mineral supplement or other dietary supplement such scientific evidence to calculate an RDA and EAR for calcium, but
as herbs or botanical products. See Reference 20 for a detailed instead created an adequate intake (AI) (9). An AI is a nutrient
description of the dietary supplement and antacid questions and the intake goal for individuals but should not be used to calculate the
coding used. prevalence of inadequate intakes for groups (9,24). Mean intakes
Atotal of 8375 elderly adults 60 y of age and older participated in were compared with the calcium AI of 1200 mg/d for males and
NHANESIII.Ofthe5302elderlyadultsexaminedintheMEC,5039 females 51 y of age and older. We estimated the proportion of elderly
(95%) provided complete and reliable 24-h dietary recalls. We ex- adults whose calcium intakes met the Healthy People 2010 calcium
cluded five respondents due to missing or unknown responses for the objective (Objective 19–11), which is 77% of the AI for calcium, or
supplement use question and, for the calcium analyses, we excluded 924 mg (16).
anadditional 32 respondents due to missing or unknown responses for Data analyses. Dietary intakes consisted of calculated usual in-
the antacid use question. No apparent bias was introduced into the takes from foods and beverages. Combined intakes consisted of di-
results based on excluding these participants. The final sample sizes etary intakes plus mineral intakes from supplements plus calcium
were 5034 for iron and zinc and 5002 for calcium. from antacids. We defined a supplement user as anyone who took a
Estimating usual dietary intakes. We estimated usual dietary dietary supplement during the past month, regardless of whether it
mineral intakes from a single 24-h dietary recall per person and a containedcalcium,ironorzinc.Forthecalciumanalyses,respondents
second independent 24-h recall from a nonrandom subsample of who took a calcium-containing antacid were grouped with supple-
8% of the examined older adults. We did not impute values for ment users. Because mineral intakes were positively skewed, both
missing 24-h recall data. mean and median intakes are reported. Intakes for each sex are
Estimates from one dietary recall may contain substantial within- reported separately because women have lower energy and nutrient
person variation. The usual intake estimates presented in this paper intakes than men.
were adjusted to remove within-person variation using a modified TheSUDAANDescriptandCrosstabsprocedures (25) were used
version of the model developed by the National Research Council to calculate mean and median intakes and standard errors, and to
(21). Because the model is based on an assumption of normality, we calculate prevalences of intakes below the iron and zinc EARs or the
log-transformed the original mineral values to improve their distri- Healthy People 2010 calcium objective. We used the survey sampling
bution. Using the model developed by Feinleib et al. (22), we weights in all analyses to produce estimates that were representative
calculated adjusted values (x¢) for each respondent after removing of the civilian, noninstitutionalized U.S. population. The SUDAAN
the within-person variation from their original values (x). The model program incorporates the sample weights and adjusts for the survey’s
is: complex sample design in calculating the appropriate standard errors.
We used Statistical Analysis System (SAS) (26) to calculate the
SD percentage of supplement users whose supplements or antacids con-
x x Between tained each of these minerals.
SDTotalobserved Using SUDAAN, we performed tests for significant differences
We estimated the ratio of within-person variability to between- betweennon-supplementandsupplementusers’meandietaryintakes,
person variability using the formula: and between non-supplement users’ dietary intakes and supplement
users’ combined intakes, stratified by sex and race/ethnicity. Tests for
s2 1r significant differences among race-ethnic groups were performed on
w meandietary intakes for both supplement and non-supplement users,
2 r andonmeancombinedintakesforsupplementusers,stratifiedbysex.
s
b We used a critical value of 0.05 for significance tests and used the
where r is the correlation coefficient between the nutrient intakes Bonferroni method of adjusting for the family of pairwise comparisons
2
from the first and second dietary recalls; s is the within-person across race-ethnic groups (27).
variance and s2 w
b is the between-person variance. This formula can be
used to estimate the following ratio:
RESULTS
SD 1
Total observed Meandietary and combined calcium intakes for total males
SD 1r
Between 1 and total females were below the AI for calcium (Table 1).
r
3424 ERVIN AND KENNEDY-STEPHENSON
TABLE 1
Mean and median dietary and combined mineral intakes and prevalence of inadequate intakes among adults 60 y and older in NHANES III1,2
Non-supplement users Supplement users
Diet only Diet only Combined intakes2
Mineral, sex and Below Below Below
race/ethnicity n Mean Median standard Mean Median standard Mean Median standard
Calcium mg % mg % mg %
Male
3 4,5
Total 2432 735 11 690 13 75 789 15 716 28 70 909 23 819 36 60
Non-Hispanic
5,6,7 6,7 6
white 1396 762 14 716 17 73 803 16 732 30 68 924 24 843 38 58
Non-Hispanic
5,8 8 8
black 488 574 20 498 21 87 637 35 553 30 84 698 42 610 32 80
Mexican
American 489 679 305 626 35 77 736 19 674 42 79 946 84 802 50 64
Female
3 4,5
Total 2570 582 11 523 13 87 632 11 590 12 87 864 16 747 16 66
Non-Hispanic
4,5,6 6,7 6,7
white 1501 597 14 542 16 86 644 13 595 16 86 888 16 773 20 64
Non-Hispanic
5 8 8
black 514 494 13 446 16 91 501 19 445 22 94 610 27 563 28 86
Mexican
American 469 572 295 505 34 88 586 17 558 24 90 735 25 689 33 75
Iron
Male
3 4,5
Total 2447 15.4 0.22 13.9 0.18 2 16.9 0.39 14.7 0.32 1 26.7 1.11 21.7 0.83 1
Non-Hispanic
white 1401 15.7 0.274,5,6,7 14.1 0.21 1 17.3 0.436,7 14.9 0.39 1 27.2 1.25 22.0 0.94 1
Non-Hispanic
black 491 12.6 0.435 11.0 0.38 9 13.6 0.56 12.3 0.62 3 22.0 1.82 17.4 2.07 2
Mexican
American 496 13.7 0.515 12.6 0.44 4 14.7 0.44 14.3 0.45 1 28.4 6.04 17.9 1.40 1
Female
3 4,5
Total 2587 11.4 0.20 10.0 0.14 4 12.3 0.23 11.1 0.18 2 23.6 1.35 15.5 0.76 2
Non-Hispanic
4,5,7 6,7 6
white 1510 11.6 0.25 10.0 0.20 3 12.5 0.26 11.2 0.18 2 24.1 1.52 16.3 0.92 2
Non-Hispanic
black 518 10.6 0.335 9.7 0.38 7 10.4 0.30 9.9 0.30 7 17.1 1.45 10.6 0.42 6
Mexican
4,5
American 472 9.7 0.34 9.3 0.37 9 11.0 0.37 9.7 0.32 2 21.2 1.43 14.4 1.21 2
Zinc
Male
3 5
Total 2447 10.9 0.21 10.2 0.12 41 11.4 0.18 10.7 0.21 35 18.8 0.53 14.7 0.68 20
Non-Hispanic
5,6,7 6 6,7
white 1401 11.1 0.26 10.3 0.14 40 11.6 0.18 10.9 0.22 34 19.1 0.57 14.8 0.84 19
Non-Hispanic
5
black 491 9.5 0.28 8.8 0.16 58 9.7 0.30 9.5 0.27 49 14.9 0.80 11.9 0.84 31
Mexican
American 496 10.1 0.305 9.6 0.31 48 10.6 0.31 10.1 0.34 43 16.0 1.01 12.6 1.09 27
Female
3 4,5
Total 2587 7.8 0.14 7.2 0.16 45 8.3 0.18 7.6 0.14 36 15.1 0.54 10.6 0.79 25
Non-Hispanic
4,5 6 6
white 1510 7.9 0.16 7.2 0.18 44 8.5 0.21 7.8 0.18 34 15.5 0.60 11.0 0.97 23
Non-Hispanic
5
black 518 7.3 0.24 6.6 0.16 54 7.4 0.22 6.8 0.23 48 11.0 0.68 8.0 0.41 38
Mexican
5
American 472 7.5 0.26 6.9 0.24 49 8.0 0.22 7.4 0.20 41 14.6 1.39 9.2 0.91 31
1 Values are means or medians SE. Standards used: EAR for iron: males, 6 mg/d; females, 5 mg/d; EAR for zinc: males, 9.4 mg/d; females, 6.8
mg/d; Healthy People 2010 calcium objective: 924 mg/d for both sexes.
2 Combined calcium intakes include dietary intakes plus calcium from supplements plus calcium from antacids. Combined iron and zinc intakes
include dietary intakes plus mineral intakes from supplements.
3 Total includes race/ethnic groups not shown separately.
4 Significant difference between non-supplement users’ mean dietary intakes vs supplement users’ mean dietary intakes stratified by sex and
race/ethnicity, P 0.05.
5 Significant difference between non-supplement users’ mean dietary intakes vs supplement users’ mean combined intakes stratified by sex and
race/ethnicity, P 0.05.
6 Comparing within each supplement use classification, differences between non-Hispanic whites’ mean dietary or combined intakes vs
non-Hispanic blacks’ comparable intakes, P 0.05.
7 Comparingwithineachsupplementuseclassification,differencesbetweennon-Hispanicwhites’meandietaryorcombinedintakesandMexican
Americans’ comparable intakes, P 0.05.
8 Comparing within each supplement use classification, differences between Mexican Americans’ mean dietary or combined intakes and
non-Hispanic blacks’ comparable intakes, P 0.05.
MINERAL INTAKES BY SUPPLEMENT USE PATTERN 3425
Seventy to 75% of males and 87% of females had mean dietary Mean dietary calcium intakes for supplement (P 0.001)
calcium intakes below the Healthy People 2010 calcium ob- and non-supplement (P 0.001) users were significantly
jective, but the proportion with combined intakes below this different between non-Hispanic white and non-Hispanic
objective fell to 60% for males and 66% for females. In black females (Table 1). Non-Hispanic whites’ dietary calcium
contrast, mean dietary and combined iron intakes for total intakes were 103 mg higher than non-Hispanic blacks’ intakes
males and total females were above the RDA, and very few for non-supplement users and 143 mg higher for supplement
elderly adults had inadequate dietary or combined iron intakes users. There were also significant differences for supplement
(Table 1). Mean dietary and combined zinc intakes were users for iron (P 0.001) and zinc (P 0.001). Non-Hispanic
either near the RDA or above it for total males and total whites had a mean dietary iron intake 2.1 mg higher and a
females. Anywhere from 35 to 45% of elderly adults had mean dietary zinc intake 1.1 mg higher than non-Hispanic
inadequate dietary zinc intakes, but only 20–25% had inade- blacks. There were significant differences between these two
quate combined zinc intakes (Table 1). race-ethnic groups for combined intakes for all three minerals
There were significant differences in mean dietary mineral (P 0.005foriron; P 0.001 for both calcium and zinc), and
intakes between supplement and non-supplement users for the magnitude of the differences was larger than the differ-
total females for all three minerals (P 0.001 for calcium; P ences for the dietary intakes. There were significant differences
0.005foriron; and P 0.01 for zinc) and for total males for between non-Hispanic white and Mexican-American female
calcium (P 0.02) and iron (P 0.002) (Table 1). When the supplement users for dietary (P 0.01) and combined (P
data were stratified by race/ethnicity within sex this difference 0.001) calcium intakes (Table 1). Non-Hispanic whites’
remainedfornon-Hispanicwhitefemalesforallthreeminerals dietary calcium intakes were 58 mg higher than Mexican
(P 0.005 for calcium; P 0.01 for both iron and zinc), and Americans’ intakes and combined intakes were 153 mg higher.
for iron for non-Hispanic white males (P 0.005) and Mex- There were also significant differences for dietary iron intakes
ican-American females (P 0.05). In each case supplement for supplement (P 0.002) and non-supplement users (P
users had higher dietary intakes than non-supplement users. 0.001). Non-Hispanic white supplement users’ dietary in-
Supplement users’ combined intakes were significantly dif- takes were 1.5 mg higher and non-supplement users’ intakes
ferent from non-supplement users’ dietary intakes for all three were 1.9 mg higher than those of Mexican Americans.
minerals for total males (P 0.001 for each mineral) and total Non-Hispanic white females usually had the lowest preva-
females (P 0.001 for each mineral) (Table 1). When the lences of intakes below the calcium objective and inadequate
data were stratified by race/ethnicity within sex, the significant iron and zinc intakes, followed by Mexican-American females;
differences remained (P 0.02 for calcium for non-Hispanic non-Hispanic black females had the highest prevalences (Ta-
black males; P 0.05 for iron for Mexican-American males; P ble 1). The only exception to this pattern was non-supplement
0.001 for all other comparisons). As expected, supplement users’ dietary iron intakes, where Mexican-American females
users’ combined intakes were higher than non-supplement had the highest prevalence of inadequacy.
users’ dietary intakes. Forty-two percent of males and 54% of females 60yin
Non-Hispanic whites generally had higher intakes than the this sample reported taking a dietary supplement during the
other two race-ethnic groups. Mean dietary intakes were sig- previous month (data not shown). More importantly, between
nificantly different between non-Hispanic white and non- 41 and 44% of total male and total female supplement users
Hispanicblackmalesforallthreeminerals(P0.001foreach tooksupplementsthatcontainedironorzinc,and42%oftotal
comparison) (Table 1). These results applied to supplement male supplement users took supplements that contained cal-
and non-supplement users. Non-Hispanic whites’ dietary in- cium (Table 2). A larger proportion of total female supple-
takes were 166 and 188 mg higher for calcium, 3.7 and 3.1 mg ment users (54%) took supplements that contained calcium.
higher for iron and 1.9 and 1.6 mg higher for zinc for supple- There were differences by race/ethnicity in use of supplements
ment and non-supplement users, respectively. Combined in- containing these minerals. In general, non-Hispanic whites
takes for calcium (P 0.001) and zinc (P 0.001) were also and Mexican Americans were more likely to take supplements
significantly different between these groups, and the magni- that contained these minerals than non-Hispanic blacks.
tude of the differences was larger than those for dietary in- Among the elderly adults in this sample who reported
takes. taking an antacid during the previous month, a little less than
Mean dietary intakes were also significantly different be- one-third of the total males and total females took an antacid
tween non-Hispanic white and Mexican-American males for that contained calcium (Table 2). Stratified by race/ethnicity,
calcium (P 0.02) and iron (P 0.001) (Table 1). These a much smaller proportion of the non-Hispanic blacks took
results applied to supplement and non-supplement users. The calcium-containing antacids than the other two race-ethnic
only significant differences for zinc were for non-supplement groups, both for males and females.
users’ dietary intakes (P 0.02) and for combined intakes (P
0.01). Non-Hispanic whites had larger intakes, but the
magnitude of the differences were smaller than for non-His- DISCUSSION
panic blacks.
Non-Hispanic blacks and Mexican Americans generally Results from NHANES III indicated that most elderly
had larger prevalences of intakes below the Healthy People adults had dietary calcium intakes below the Healthy People
2010 calcium objective and inadequate iron and zinc intakes 2010 objective, and many also had inadequate dietary zinc
than non-Hispanic whites, with non-Hispanic blacks having intakes. Intakes improved when minerals from dietary supple-
the largest prevalences (Table 1). The differences in preva- ments and calcium-containing antacids were added to dietary
lences of inadequate iron intakes between non-Hispanic intakes. Even so, a substantial proportion of supplement users
whites and either non-Hispanic blacks or Mexican Americans still had combined intakes below the calcium objective and
were usually minimal. In fact, there were no differences in the EARforzinc.Non-Hispanicwhitesgenerallyhadhighermean
prevalences of inadequacy between non-Hispanic white and dietary intakes and lower prevalences of inadequate iron and
Mexican-American supplement users for either dietary or zinc intakes or intakes below the calcium objective than
combined intakes. non-Hispanic blacks and Mexican Americans. Many of these
no reviews yet
Please Login to review.
