![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
The Department of Nutrition, University of North Carolina at Chapel Hill, Chapel Hill, NC 27516
3To whom correspondence should be addressed. E-mail: jstookey{at}email.unc.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
< 1.3 cm2], or gain
of MAMA with concurrent loss, maintenance [ < 2
cm2] or gain of body fat), were observed for this sample.
After controlling for sex, baseline age, urban residence, height,
weight, income, MAMA, waist circumference, smoking status, activity
level, mean daily energy and protein intakes (from three 24-h recalls),
and change in height, it was determined that subjects who lost both arm
muscle and body fat were distinguished from subjects who lost arm
muscle but gained body fat by lower income and energy intake at
baseline. Although protein intakes at baseline did not differ between
the groups that lost arm muscle, protein intakes were significantly
higher for subjects who gained both muscle and fat. Patterns of change
involving gains in arm muscle were associated with increased protein
intake, urban residence, as well as moderate or heavy levels of
physical activity at baseline. Variation in protein intake, physical
activity, and urban residence also differentiated between the groups
that gained fat. Patterns of age-related change in body composition
appear associated with modifiable variables, including income, urban
residence, activity and protein and energy intake.
KEY WORDS: • body composition • sarcopenia • diet • activity • aging
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
2
3)
. These age-related
changes are associated with increased risk of morbidity, functional
impairment and mortality, and they constitute important public health
problems.
The role of diet in age-related changes in body composition remains
unclear. Although research indicates that age-related changes in
body composition may be modifiable by environmental variables,
particularly physical activity, a potential role for diet remains
unclear. Despite experimental evidence in support of a direct
relationship between energy and/or protein intakes and muscle outcomes
(4
5
6
7
8)
, many observational studies of diet and sarcopenia
report null effects (9
10
11
12
13)
. The role of nutrition in
age-related increases in body fat has so far been limited to
decreases in energy expenditure. Age-related increases in body fat
are primarily attributed to declines in physical activity and basal
metabolic rate, rather than to dietary intake. Because body fat
increases while food intake is decreasing, the role of diet has even
been described as a "conundrum" (3)
.
Relationships between diet and age-related changes in lean and fat
mass may become clearer if concurrent changes are studied together.
Considerable evidence indicates that patterns of change in body
composition over the short-term are explained by particular
profiles of energy and protein intake. Clinical studies of obesity
(overfeeding studies), starvation and the treatment of obesity
(underfeeding studies), as well as the treatment of malnourished
individuals (refeeding studies), demonstrate that energy and protein
intakes, relative to dietary requirements, predict simultaneous change
in both the lean and fat compartments (see 14
15
16
17
18
).
Longer-term, age-related changes in body composition may be
associated with long-term dietary intake profiles.
To motivate work on the dietary determinants of sarcopenia and
concurrent changes in body fat, information is needed regarding
prevalent patterns of change in both lean and fat compartments.
Cross-sectional data suggest that different patterns exist, such
that sarcopenia may be accompanied by maintenance or loss of fat mass
(19
20
21)
. While young elders tend to have less muscle and
greater percent body fat than younger adults, older more frail elderly
may appear wasted with little subcutaneous fat. Few longitudinal
studies report changes in both compartments, with the exceptions only
presenting mean changes for the sample at large, without consideration
of particular subcategories or combinations of change. Studies
describing how modifiable sociodemographic or environmental variables
covary with different patterns of change are lacking. The epidemiology
of patterns of age-related changes in body composition—their
prevalence, environmental correlates and associated risks— has still
yet to be fully explained by existing knowledge.
As a first step toward understanding existing patterns of age-related change in body composition, this article describes longitudinal, simultaneous changes in the arm muscle and body fat of older adults in China. We report the prevalence of different patterns of change in body composition and examine to what extent the observed patterns correlate with key, modifiable, sociodemographic variables, such as income and urban residence, as well as with potential risk factors for sarcopenia and change in body fat, namely physical activity and protein and energy intakes.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Longitudinal data on 608 healthy adult Chinese between the ages of 50
and 70 y who participated in both the 1993 and 1997 China Health
and Nutrition Surveys (CHNS)4 were used for this analysis.
The CHNS provide unique, detailed data on longitudinal changes in body
composition, dietary intake, physical activity, as well as
sociodemographic variables. Due to the extremely rapid ongoing
economic, demographic and epidemiologic transitions in China
(22
23
24
25)
, these data capture unusually large variation in
the lifestyle and anthropometry of free-living individuals. The
sampling frame and survey methodology have been described in detail
previously (24
,26)
. The CHNS are in accord with human
subjects procedures that have been approved by both the University of
North Carolina School of Public Health and the Chinese Academy of
Preventive Medicine human subjects protection committees.
Selection criteria were chosen to define a sample of healthy older
adults who would be relatively homogeneous with respect to measurement
error in the available body composition indices. All subjects included
in the sample answered "no" to the question "Over the past three
months have you had any difficulty in carrying out your daily
activities and work due to illness?" in both the 1993 and 1997
surveys. To avoid confusing acute or illness-related changes in
body composition with chronic changes, subjects who experienced a
weight change of >20% from baseline or reported energy intakes below
levels shown to be adequate for short-term weight maintenance
(125–138 or 33 kcal/kg body wt) (27
,28)
were also
excluded. Subjects who were obese in either survey year according to
the World Health Organization (WHO) cutoffs (BMI 
30) were also
excluded to minimize bias related to extreme differences in the
reliability of triceps skinfold measurements.
Of the original 2,164 persons aged 50–70 y who were interviewed in 1993 and living in provinces that were resurveyed in 1997, 583 were missing both health status and anthropometric information, 116 reported poor health in the 3 mo preceding the survey and 1,465 subjects reported no health-related difficulties. Of these 1,465 subjects, 1,410 had nonmissing anthropometric data and were nonobese. Subjects who were missing dietary intake information (n = 10) or reported baseline energy intakes < 125–138or 33 kcal/kg body wt (n = 216) were next excluded from the sample.
Between 1993 and 1997, 261 (22% of the sample) were lost to follow-up. Of the remaining 923, 858 reported no health-related difficulties before the survey in 1997. Although the subjects who were lost to follow-up were significantly older and richer and had larger waist circumferences than those who participated in both surveys, it was not possible to test for selection bias related to this loss. Tests for selection bias would require information on the health status of missing subjects in 1997, which is unavailable. The target sample only included subjects who were healthy in both survey years.
Of the 858 healthy subjects followed from 1993 to 1997, follow-up
anthropometric measures were available for 653. Further exclusion of
subjects who experienced a weight change of >20% from baseline
(n = 40) or had a BMI > 30 in 1997 reduced
the sample size to 608. Sample characteristics are shown in Table 1
.
|
Specially trained interviewers obtained detailed individual-level
diet data via 24-h recall for three consecutive days as well as
household-level information on changes in food inventory. The
household-level data were used to estimate individual consumption
of edible oil (29)
and to cross-check the diet recall
data. The 1991 Food Composition Table
for China (30)
was
used to calculate 3-d mean daily energy and protein intakes from the
food consumption data for each individual. Energy and protein intakes
were expressed in both absolute and relative units (kilojoules,
kilojoules per kilogram body weight, and percent of the Chinese age-,
sex- and activity-specific RDA for energy intake, and grams, grams
per kilogram body weight, and percent of energy for protein intake).
The protein RDA was not used in this study, because protein
requirements among older adults remain controversial
(31
32
33)
. According to Willett and Stampfer
(34)
, 3-d mean energy and protein intakes adequately
represent usual intakes for these nutrients. Although the energy
intakes observed in the CHNS have not been validated using a gold
standard technique, they have been shown to predict BMI in CHNS adults
(26)
. A validation study of CHNS dietary intake,
undertaken by Sue Roberts and others at the Tufts Human Nutrition
Research Center in Beijing, has not been analyzed yet.
Physical activity was recorded as a four-level variable. Very light
activity was characterized by working in a sitting position or as an
office worker; light activity as work in a standing position; moderate
activity as work carried out by drivers or electricians; and heavy
activity as the work of farmers. The activity variable was designed by
the Chinese Nutrition Society to reflect total energy expenditure and
was intended for use in calculating the Chinese RDAs. The variable
significantly predicts energy intake and weight status among Chinese
adults (26)
. Missing values for the activity variable
(n = 13) were imputed using age, sex, urban/rural
residence, occupation and self-reported information on the present
condition of the upper and lower extremities (functioning normally,
having some problems but not affecting daily activities and work,
slightly affecting daily activities, affecting daily activities [help
is required]). The four-level activity variable was collapsed into
a dichotomous variable (very light or light versus moderate or heavy
activity).
In addition to age (years) and sex, the CHNS collect information on
type of residence and income. Given the rapid urbanization-related
changes occurring in China, these are key indicators for the present
sample. Questions on income probed for any income-producing
activity each person might have engaged in during the previous year.
Income from market and nonmarket activities and nonmonetary government
subsidies (food subsidies in the form of ration coupons) were included
in the income estimate. Per capita household income was deflated by the
relative Retail Price Index, which was 100 at the base year of 1985
(35)
. Urban/rural residence was defined according to the
Chinese census definition, which considers small towns and city
neighborhoods as urban and villages and suburbs as rural.
Body composition indices.
Midarm muscle area (MAMA) was chosen as the indicator of arm muscle
mass for the present analyses. Estimates of MAMA were calculated from
the midarm circumference and triceps skinfold measures for each
individual using the equation recommended by WHO (27)
. We
arbitrarily chose a cut point of 1.3 cm2 to distinguish
between real change in MAMA over time and apparent change resulting
from measurement error. This decision was based on calculations for a
hypothetical individual with an arm circumference of 24 cm and triceps
skinfold of 11 mm, assuming a worst case scenario whereby both measures
were underestimated at baseline then overestimated at follow-up,
and by using the technical errors of measurement reported by Frisancho
(36)
(intra- plus interexaminer measurement errors in
midarm circumference: 0.7 cm; intra- plus interexaminer errors in
triceps skinfold: 2.7 mm). Only changes in MAMA > 1.3
cm2 were considered possible gains or losses.
Waist circumference was chosen as the indicator of body fat for this
analysis. In validation studies in adults over age 18 y and in
healthy elderly, waist circumference has consistently appeared to be a
good measure (0.70 < r < 0.95) of total body
fat as well as abdominal fat for both sexes (37
38
39
40)
.
Because repeated waist circumference measurements for an individual
typically fall within 1 cm (41)
, gains and losses in waist
circumference were defined as changes > 2 cm between the 1993 and
1997 surveys.
To explore patterns of simultaneous change in arm muscle and body fat, nine possible combinations of change were defined: loss (L), maintenance (M), or gain (G) of arm muscle (M) concurrent with loss, maintenance or gain in body fat (F). Throughout the article, the nine patterns are labeled using the following abbreviations: M for arm muscle, F for body fat, L for lost, M for maintenance, and G for gain. For example, LMLF represents subjects who lost arm muscle and lost body fat.
Analysis.
All procedures were carried out using the STATA statistical package
(42)
. The proportion of the sample experiencing each of
the nine possible combinations of change in body composition was
determined. To compare the patterns of change with respect to their
anthropometric, sociodemographic and/or lifestyle characteristics,
adjusted means or prevalences were estimated for the following
variables: baseline age, height, weight, arm muscle, waist
circumference, income, urban/rural residence, smoking status
(current/nonsmoker), activity level and energy and protein intakes. For
each of the continuous descriptor variables, a sex-specific
ordinary least squares (OLS) regression model was created with the
descriptor variable as outcome. Similarly, for each of the dichotomous
descriptor variables, sex-specific logistic regression models were
fit. The independent variables included in these models were the dummy
variables representing the patterns of longitudinal change and all of
the remaining covariates. Change in height was controlled in all of the
models to account for height-related error in change in waist
circumference (43)
. To account for the potential
nonindependence of data for subjects from the same household, the
standard errors for the estimated coefficients were adjusted using the
cluster command available with the STATA statistical package
(42)
. Differences with a probability level <0.05 were
considered statistically significant. Finally, using the results from
each multivariable model, the adjusted mean or prevalence of each
descriptor variable for each pattern of change was predicted.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
presents descriptive statistics for this sample of Chinese
adults. Over the survey period, arm muscle decreased by 1.9
(SD = 7.4) cm2, while waist
circumference increased by 1.1 (7.1) cm for the whole sample. These
changes were not statistically different from zero. Prevalence of particular patterns of longitudinal change in body composition.
Figure 1
illustrates that between 1993 and 1997 all combinations of change in
both arm muscle and body fat occurred for this sample. Over a third of
the sample maintained their arm muscle within 1.3
cm2 and/or their waist circumference within 2 cm
over the 4-y period. Loss of arm muscle occurred with decreases in body
fat for 18% of the sample and with increases in body fat for another
17% of the sample. Gains in arm muscle were observed with gains and
losses in body fat for 18 and 9% of the sample, respectively.
Table 2
summarizes the prevalence and magnitudes of change in MAMA and waist
circumference observed for each pattern combination.
|
|
The patterns of change in body composition were associated with
different baseline anthropometric and sociodemographic characteristics
(Fig. 2
). For both sexes, subjects who lost arm muscle had significantly
greater baseline arm muscle than those who gained arm muscle. Decreases
in waist circumference were, similarly, significantly associated with
greater baseline waist circumference.
|
Older age at baseline was significantly associated with loss of muscle. Females who lost arm muscle, either with gains or losses of body fat, were significantly older than subjects who maintained arm muscle and body fat (MMMF). Males who lost muscle were significantly older than GMMF males. Older age also differentiated between the patterns of change in body fat for males. Among the males who gained arm muscle, those who lost body fat were significantly younger by an average of 4.2 y.
Urban residence characterized males and females who maintained body fat and gained muscle (GMMF). LMMF subjects were significantly more likely than GMMF subjects to live in rural areas. Among the male subjects who maintained or gained muscle over the survey period (GMMF or GMGF), those who also gained fat were significantly more urban than those who lost fat (GMLF).
Female subjects who lost both arm muscle and body fat had lower incomes at baseline than subjects who lost muscle but gained body fat (LMGF). LMLF females also had lower incomes than subjects who lost fat but gained or maintained arm muscle (GMLF). Although not statistically significant among men, lower income also appeared to differentiate between the LMLF and LMGF groups.
Lifestyle variables associated with the patterns of longitudinal change.
Adjusted, sex-specific, mean energy and protein intakes and
prevalences of smoking and activity levels at baseline are shown in
Figure 3
. For the females, energy and protein intakes and activity level varied
significantly between the pattern groups. Significantly lower energy
intakes were observed for the LMLF pattern of change compared with the
LMMF group. Significantly higher protein intake was observed for the
GMGF group compared with the LMGF group. A significantly greater
proportion of subjects who gained arm muscle but lost body fat (GMLF)
reported moderate or heavy activity levels than subjects who lost arm
muscle (LMLF, LMMF, LMGF or MMLF). Females who gained both (GMGF) were
also significantly less likely to report moderate or heavy activity
levels than GMLF subjects.
|
For both males and females, the patterns observed with energy and
protein intakes expressed in relative units (energy: kilojoules per
kilogram body weight; or percent RDA: grams protein per kilogram body
weight or percent of energy) paralleled those shown in Figure 3
(data
not shown).
The highest prevalence of smoking was observed for males who lost arm muscle and maintained or gained body fat. Current smoking was significantly more prevalent in these groups compared with males who maintained their baseline levels of arm muscle and body fat.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
All possible combinations of longitudinal change in both arm muscle and body fat were observed for this sample of older adults in China. Loss of arm muscle occurred with either gains or loss of body fat. Similarly, gains of arm muscle occurred with gains or loss of body fat. We remark only on the occurrence of all possible patterns, rather than the prevalence of each pattern type. Given the potential for selection bias related to study attrition and the probable dependence of prevalence estimates on the length of follow-up time, no attempt is made to generalize the prevalence estimates reported here to wider population groups.
Subjects who lost both arm muscle and body fat were distinguished from subjects who lost arm muscle but gained body fat by lower incomes and a lower mean daily energy intake at baseline. Although baseline protein intakes did not differ between the groups that lost arm muscle, protein intakes were significantly higher for subjects who gained both muscle and fat. Patterns of change involving gains in arm muscle were associated with increased protein intake, urban residence and moderate or heavy levels of physical activity at baseline. Variation in protein intake, physical activity and urban residence also differentiated between the groups that gained fat.
The findings with respect to physical activity and protein and energy
intakes are consistent with data from a wide variety of
shorter-term experimental studies. Resistance training has
repeatedly been shown to result in increases in muscle mass or area
among elderly populations (4
,5
,44)
.
Several types of clinical studies—starvation, semistarvation, nitrogen balance experiments and weight loss interventions—show that energy and protein intakes below requirements lead to loss of both lean and fat mass (LMLF). Experimental work on the "protein-sparing modified fast" shows that when protein intake is adequate relative to requirements, but energy intake is deficient, a GMLF pattern of change in body composition is observed. Consistent with this evidence, mean protein intake for the GMLF pattern was higher than that for the LMLF pattern.
Although few experiments in the literature report losses of muscle with
gains in fat mass, the LMGF pattern of change, two short-term
studies do indicate that among young adults, loss of lean mass occurs
with gains in body fat with consumption of diets low in protein but
adequate in energy. On feeding undernourished young adults a diet
adequate in energy (9372 kJ/d) but low in protein (26 g/d) for 45 d, Barac-Nieto et al. (45)
observed significant
increases in body fat and decreases in body cell mass. Miller and
Mumford (46)
overfed young adult subjects a diet providing
an excess 147 MJ containing 2–3% of energy from protein for 3–4 wk,
and observed increases in body weight and triceps skinfold but losses
of body potassium. Despite the older sample and longer-term
follow-up in the present study, the significantly higher energy intake
among LMGF subjects compared with LMLF subjects is consistent with
these results. The present results also parallel short-term
relationships between protein and energy and the spectrum (from
kwashiorkor to nutritional marasmus) of acute
protein-energy-malnutrition syndromes (47)
.
The results of the present study were consistent with results from shorter-term experiments, despite considerable differences in study design. The present study involved a much longer follow-up period and nonobese healthy adults who experienced no dramatic weight changes and whose diets were above cutoffs for short-term weight maintenance. Under these conditions, short-term experiments would likely produce null results. Also unlike the experiments, particular diet-activity combinations were not assigned to subjects in this study. Differences in the particular combination of diet profile and diet requirements experienced by subjects and differences in the choice of reference category might also be expected to produce discrepant results.
Baseline protein and energy intakes were studied instead of changes over time, because the exposure of interest is the level of dietary intake over the long term. This study assumes that 3-d mean protein and energy intakes adequately reflect the habitual or "usual" intake of these nutrients.
The findings with respect to income and urban residence suggest that deprivation may contribute to the etiology and presentation of sarcopenia. Low income, rural residence and low protein and energy intake over time among older adults may prove to be mechanisms amenable to intervention.
Subjects who lost both arm muscle and body fat had significantly greater baseline arm muscle and waist circumference than subjects who also lost arm muscle but gained body fat. While the greater loss associated with larger baseline arm muscle or waist circumference measures may represent a statistical regression-to-the-mean phenomenon, a biological explanation involving increased dietary requirements for tissue maintenance may also apply. Any given level of energy or protein intake is more likely to fall short of requirements among larger individuals than among smaller individuals. The finding that taller subjects appeared more likely to lose muscle supports this argument.
The changes in arm muscle and waist circumference reported here should
be interpreted with caution. Anthropometric measures are not precise
enough to detect subtle changes in muscle or body fat
(48)
, and they may not reflect changes in whole body
compartments. While limited information suggests that waist
circumference and MAMA are meaningful among Chinese groups
(49
,50)
, validation studies of change in either measure
are not available in the literature. Arm lean area is a poor index of
total lean body mass (51)
. Given the limitations of the
MAMA measure, we interpret MAMA as an index of arm muscle mass rather
than total body muscle and recognize that changes in this variable may
not reflect sarcopenia in the whole body.
The present sample of nonobese individuals aged 50 to 70 y was
chosen to minimize bias related to differentially distributed
differences in measurement errors in the MAMA measure. Errors in this
measure are known to increase with age and obesity
(52
,53)
. Although MAMA has been shown to correlate
reasonably well with criterion measures in healthy younger adults
(54
55
56)
, in older adults, MAMA correlates less well with
arm muscle and correlates poorly with total body muscle
(52)
. In addition to errors in the measurement of triceps
skinfold and midarm circumference, invalid geometric assumptions may
contribute to overestimation of the true arm muscle area. Among elderly
subjects in the U.S., bone-free MAMA has been shown to overestimate
true muscle and bone area by as much as 30% in men and 50% in women
(53)
. If geometric deviations from circularity increase
over time, the overestimation in MAMA may be worse at follow-up
than at baseline, leading to apparent increases in muscle mass over
time, particularly in older subjects.
In conclusion, this study constitutes a first step toward filling gaps
in knowledge about patterns of concurrent, age-related change in
body muscle and fat. Previous studies (e.g., 1
,2
) have
reported mean changes for each compartment, but not for the prevalence
of different patterns of change and their correlates. Such information
is necessary for a deeper understanding of links between dietary
intake, activity levels, sarcopenia and age-related gains in body
fat. Research on concurrent changes in lean and fat mass may help
highlight the role of diet in age-related changes in body
composition.
|
FOOTNOTES |
|---|
2 Funding for parts of the project design, data
collection and computerization has been provided by the Chinese Academy
of Preventive Medicine (CAPM), the Carolina Population Center (CPC) of
the University of North Carolina at Chapel Hill (UNC-CH), and the
National Institutes of Health (NIH) (R01-HD38700 and R01-HD30880).
Funds for the research reported in this article were provided by a
National Institute on Aging training grant, the UNC Institute on Aging,
and the UNC Institute of Nutrition. This article is part of a
collaborative research project between CAPM, directed by Ge Keyou with
coprincipal investigators Zhai Fengying and Jin Shuigao, and a group
from UNC-CH and CPC, directed by Barry M. Popkin, with coprincipal
investigators Barbara Entwisle and Gail E. Henderson.
4 Abbreviations: CHNS, China Health and Nutrition
Surveys; WHO, World Health Organization; MAMA, midarm muscle area; OLS,
ordinary least squares.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Muller D C., Elahi D., Sorkin J. D. & Andres R. (1995) Muscle mass: it’s measurement and influence on aging: the Baltimore Longitudinal Study of Aging. Rosenberg I.H. eds. Nutritional Assessment of Elderly Populations Measure and Function 1995:50-62 Raven Press New York .
2. Borkan G. A. & Norris A H. (1977) Fat redistribution and the changing body dimensions of the adult male. Human Biol 49:495-513.[Medline]
3.
Morley J. E. (1997) Anorexia of aging: physiolic and pathologic. Am. J. Clin. Nutr. 66:760-773.
4. Meredith C. N., Frontera W. R., O’Reilly K. P. & Evans W. J. (1992) Body composition in elderly men: effect of dietary modification during strength training. J. Am. Geriatr. Soc. 40:155-162.[Medline]
5.
Fiatarone M. A., O’Neil E. F., Ryan N. D., Clements K M., Solares G. R., Nelson M. E., Roberts S. B., Kehayias J. J., Lipsitz L. A. & Evans W. J. (1994) Exercise training and nutritional supplementation for physical frailty in very elderly people. N. Eng. J. Med. 330:1769-1775.
6. Gray-Donald K., Payette H. & Boutier V. (1995) Randomized clinical trial of nutritional supplementation shows little effect on functional status among free-living frail elderly. J. Nutr. 125:2965-2971.
7.
Castaneda C., Charnley J M., Evans W. J. & Crim M. C. (1995) Elderly women accommodate to low-protein diet with losses of body cell mass, muscle function, and immune response. Am. J. Clin. Nutr. 62:30-39.
8.
Campbell W. W., Crim M. C., Young V. R., Joseph L. J. & Evans W. J. (1995) Effects of resistance training and dietary protein intake on protein metabolism in older adults. Am. J. Physiol. 268:E1143-E1153.
9.
Munro H. N., McGandy R. B., Hartz S. C., Rusell R. M., Jacob R. A. & Otradovec C. L. (1987) Protein nutriture of a group of free-living elderly. Am. J. Clin. Nutr. 46:586-592.
10. Neiman D. C., Sherman K. M., Arabatzisk K., Underwood B. C., Barbosa J. C., Johnson M., Shultz T. D. & Lee J. (1989) Hematological, anthropometric, and metabolic comparisons between vegetarian and non-vegetarian elderly women. Int. J. Sports Med. 10:243-250.[Medline]
11.
Baumgartner R. N., Koehler K. M., Gallagher D., Romero L. & Heymsfield S. B. (1998) Epidemiology of sarcopenia among the elderly in New Mexico. Am. J. Epidemiol. 147:755-763.
12. Baumgartner R. N., Waters D. L., Gallagher D., Morley J. E. & Garry P. J. (1999) Predictors of skeletal muscle mass in elderly men and women. Mech. Ageing Dev. 107:123-136.[Medline]
13.
Starling R. D., Ades P. A. & Poehlman E. T. (1999) Physical activity, protein intake, and appendicular skeletal muscle mass in older men. Am. J. Clin. Nutr. 70:91-96.
14. Keys A., Brozek J., Henschel A., Michelson O. & Taylor H. L. (1950) The Biology of Human Starvation 1950 University of Minnesota Press Minneapolis, MN. .
15. Apfelbaum F., Bostsarron J. & Lacatis D. (1971) Effect of caloric restriction and excessive caloric intake on energy expenditure. Am. J. Clin. Nutr. 24:1405-1409.[Abstract]
16.
Van Gaal L. F., Snyders D., De Leeuw I. H. & Bekaert J. L. (1985) Anthropometric and calorimetric evidence for the protein sparing effects of a new protein supplemented low calorie preparation. Am. J. Clin. Nutr. 41:540-544.
17. Vansant G., Van Gaal L., Van acker K. & DeLeeuw I. (1989) Short and long term effects of a very low calorie diet on RMR and body composition. Int. J. Obesity 13(Suppl.):87-89.
18.
Orphanidou C. I., McCargar L. J., Birmingham C. L. & Belzberg A. S. (1997) Changes in body composition and fat distribution after short-term weight gain in patients with anorexia nervosa. Am. J. Clin. Nutr. 65:1034-1041.
19. Patrick J. M., Bassey E. J. & Fentem P. H. (1982) Changes in body fat and muscle in manual workers at and after retirement. Eur. J. Appl. Physiol. 49:187-196.
20.
Cohn S. H., Vartsky D., Yasumura S., Savitsky A., Zanzi I., Vaswani A. & Ellis K. J. (1980) Compartmental body composition based on total-body potassium and calcium. Am. J. Physiol. 239:E524-E530.
21. Evans W. & Campbell W. (1993) Sarcopenia and age-related changes in body composition and functional capacity. J. Nutr. 123:465-468.
22. Popkin B. M., Keyou G., Zhai F., Guo X., Ma H. & Zohoori N. (1993) The nutrition transition in China: a cross-sectional analysis. Eur. J. Clin. Nutr. 47:333-246.[Medline]
23. Popkin B. M. (1994) The nutrition transition in low-income countries: an emerging crisis. Nutr. Rev. 52:285-298.[Medline]
24. Popkin B. M., Paeratakul S., Zhai F. & Ge K. (1995) Dietary and environmental correlates of obesity in a population study in China. Obes. Res. 3(Suppl.):135s-143s.
25. Popkin B. M. (1998) The nutrition transition and its health implications in lower income countries Pub. Health Nutr 1:5-21.
26. Paeratakul S., Popkin B. M., Ge K., Adair L. S. & Stevens J. (1998) Changes in diet and physical activity affect the body mass index of Chinese adults. Intl. J. Obesity 22:424-432.
27. World Health Organization (1985) Energy and protein requirements: report of joint FAO-WHO-UNU expert consultation 1985 World Health Organization Geneva, Switzerland. .
28. Roberts S. B., Young V. R. & Fuss P. (1992) What are the dietary energy needs of elderly adults?. Int. J. Obes. Relat. Metab. Disord. 16:969-976.[Medline]
29.
Guo X., Popkin B. M., Mroz T. A. & Zhai F. (1999) Food price policy can significantly reduce the negative effects of the nutrition transition in China. J. Nutr. 129:994-1001.
30. Institute of Nutrition and Food Hygiene, Chinese Academy of Preventive Medicine (INFH-CAPM) (1991) The Food Consumption Tables 1991 People’s Medical Publication House Bejing China. .
31. Campbel W. W. & Evans W. J. (1996) Protein requirements of elderly people. Eur. J. Clin. Nutr. 50(Suppl. 1):S180-S183.
32. Bidlack W. R. & Wang W. (1995) Nutrition requirements of the elderly. Morley J. E. Glick Z. Rubenstein L. Z. eds. Geriatric Nutrition. 2nd ed. 1995:25-28 Raven Press New York .
33.
Millward D. J., Fereday A., Gibson N. & Pacy P. J. (1997) Aging, protein requirements, and protein turnover. Am. J. Clin. Nutr. 66:774-786.
34. Willett W. & Stampfer M. (1998) Nature of variation in diet. Willett W. eds. Nutritional Epidemiology 2nd ed. 1998:33-49 Oxford University Press New York. .
35. China State Statistical Bureau (SSB) (1992) China’s Statistical Yearbook 1992 China Statistical Press Beijing, China (in Chinese). .
36. Frisancho A. R. (1990) Anthropometric Standardization for the Assessment of Growth and Nutritional Status 1990:6 The University of Michigan Press Ann Arbor, Michigan. .
37. Seidell J. C., Bjorntorp P., Sjostrom L., Sannerstedt R., Krotkiewski M. & Kvist H. (1989) Regional distribution of muscle and fat mass in men: new insight into the risk of abdominal obesity using computed tomography. Int. J. Obesity 13:289-303.[Medline]
38. Seidell J. C., Oosterlee A., Deurenberg P., Hautvast J.G.A.J. & Ruijs J.H.J. (1988) Abdominal fat depots measured with computed tomography: effects of degree of obesity, sex, and age. Eur. J. Clin. Nutr. 42:805-815.[Medline]
39.
Ross R., Leger L., Morris D., De Guise J. & Guardo R. (1992) Quantification of adipose tissue by MRI: relationship with anthropometric variables. J. Appl. Physiol. 72:787-795.
40.
Ross R., Shaw K. D., Martel Y., de Guise J. & Avruch L. (1993) Adipose tissue distribution measured by magnetic resonance imaging in obese women. Am. J. Clin. Nutr. 57:470-475.
41. Lohman T. G., Roche A. F. & Martorell R. (1988) Anthropometric Standardization Reference Manual 1988 Human Kinetics Books Champaign, IL. .
42. StataCorp (1999) Stata Statistical Software: Release 6.0 1999 Stata Corporation College Station TX .
43. Chumlea W. C., Roche A. F. & Webb P. (1984) Body size, subcutaneous fatness and total body fat in older adults. Int. J. Obesity 8:311-317.[Medline]
44.
Fiatarone M. A., Marks E. C. & Ryan N. C. (1990) High-intensity strength training in nonagenarians: effects on skeletal muscle. JAMA 263:3029-3034.
45.
Barac-Nieto M., Spurr G., Lotero H., Maksud M. & Dahners H. (1979) Body composition during nutritional repletion of severely undernourished men. Am. J. Clin. Nutr. 32:981-999.
46. Miller D. & Mumford P. (1967) Gluttony, an experimental study of overeating low- or high-protein diets. Am. J. Clin. Nutr. 20:1212-1222.[Abstract]
47. Waterlow J. C. (1993) Protein energy malnutrition 1993 Edward Arnold London, UK. .
48. Wright R. A. & Heymsfield S. (1984) Nutritional Assessment 1984 Blackwell Scientific Publications Boston, MA. .
49.
Eston R. G., Fu F. & Fung L. (1995) Validity of conventional anthropometric techniques for predicting body composition in healthy Chinese adults. Br. J. Sp. Med. 29:52-56.
50. Kwok T., Woo J., Chan H.H.L. & Lau E. (1997) The reliability of upper limb anthropometry in older Chinese people. Int. J. Obesity 21:542-547.
51. Borkan G. A., Hults D. E., Gerzof S. G., Burrows B. A. & Robbins A. H. (1983) Relationships between computed tomography tissue areas, thicknesses and total body composition. Ann. Hum. Biol. 10:537-546.[Medline]
52.
Reid J. R., Evans M. C. & Ames R. (1992) Relationships between upper-arm anthropometry and soft-tissue composition in postmenopausal women. Am. J. Clin. Nutr. 56:463-466.
53. Baumgartner R. N., Rhyne R. L., Garry P. J. & Heymsfield S. B. (1993) Imaging techniques and anatomical body composition in aging. J. Nutr. 123:444-448.
54.
Heymsfield S. B., McManus C., Smith J., Stevens V. & Nixon D. W. (1982) Anthropometric measurement of muscle mass: revised equations for calculating bone-free arm muscle area. Am. J. Clin. Nutr. 36:680-690.
55.
Heymsfield S. B., Olafson R. P., Kutner M & H. & Nixon D. W. (1979) A radiographic method of quantifying protein-calorie undernutrition. Am. J. Clin. Nutr. 32:693-702.
56.
Heymsfield S. B., Smith R., Aulet M., Bensen B., Lichtman S., Wang J. & Pierson R. N. (1990) Appendicular skeletal muscle mass: measurement by dual-photon absorptiometry. Am. J. Clin. Nutr. 52:214-218.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
These adjusted means and prevalences
were predicted from OLS or logistic regression models controlling for
age, sex, urban residence, height, weight, income, MAMA, waist
circumference, current smoking, past smoking history, activity level
and energy and protein intakes.