Full text of DOI:10.1093/gerona/56.4.M212

Journal of Gerontology: MEDICAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2001, Vol. 56A, No. 4, M212–M216
Nutritional Parameters, Body Composition, and
Progression of Disability in Older Disabled Residents
Living in Nursing Homes
Giovanni Zuliani,1 Franco Romagnoni,2 Stefano Volpato,1 Lucia Soattin,2 Vincenzo Leoci,2
Maria Cristina Bollini,2 Mauro Buttarello,3 Daniela Lotto,4 and Renato Fellin1
1Second Department of Internal Medicine, University of Ferrara, Italy.
2Istituto di Riposo per Anziani, Padua, Italy.
3Geriatric Hospital, Padua, Italy.
4Institute of Statistics, University of Padua, Italy.
Background. The evaluation of nutritional status is one of the primary components of multidimensional geriatric as-
sessment. We investigated the relationship between some markers of malnutrition and the modifications in functional
status in a sample of older disabled residents living in nursing homes.
Methods. Ninety-eight subjects who were independent in at least two activities of daily living (ADLs) were enrolled
in a 2-year longitudinal study. Anthropometric, nutritional, and metabolic parameters, as well as body composition,
were measured at baseline and after 2 years.
Results. Deteriorating functional status (Ն2 additional lost ADLs) was associated with baseline albumin levels (Ter-
tile 3 vs Tertile 1; odds ratio [OR] 0.16, 95% confidence interval [CI] 0.04–0.67) and subscapular skinfold thickness
(Tertile 3 vs Tertile 1; OR 0.06, 95% CI 0.006–0.50). After multivariate adjustment, the OR for increasing disability was
Ͼ4 in subjects with decreasing body cell mass (BCM), compared with subjects with a stable BCM. The degree of BCM
reduction was strongly related to the number of additional ADLs lost at follow-up (test for trend, p ϭ .003).
Conclusions. In a sample of older disabled nursing home residents, signs of malnutrition seem to predict further
worsening in functional status. Furthermore, BCM declines proportionally to the loss in ADLs, suggesting the existence
of a strong relationship between BCM loss and the progressive deterioration of functional status.
HE evaluation of nutritional status is one of the primary
components of multidimensional geriatric assessment
between malnutrition and disability, owing to the cross-sec-
tional design of the survey.
T
(1,2). Indeed, it has been reported that 30% to 60% of older
subjects living in nursing homes are malnourished (3,4); for
example, Abbasi and Rudman (5) found that 30% to 50% of
elderly nursing home residents had substandard body weight
and midarm circumference and low albumin concentrations.
Involuntary weight loss and geriatric cachexia have been as-
sociated with poor health outcomes, such as morbidity and
mortality (6,7), and malnutrition has been associated with
considerable dysfunction and disability (7,8).
Disability in activities of daily living (ADLs) is mainly a
dynamic and progressive process that is the consequence of
the interaction of chronic diseases and age-related physio-
logical changes (9). It is an important risk factor for institu-
tionalization, recurrent hospitalization, falls, and acute ill-
nesses in older subjects (9).
We here report the results of the 2-year follow-up of the
IRA study. The aim of this longitudinal research was to
examine the relationship between some markers of malnu-
trition, including body cell mass (BCM), measured by tet-
rapolar bioelectric impedance analysis (BIA), and the modi-
fications in functional status over a period of 2 years. In this
study, we specifically focused our interest on the associa-
tion between nutritional status and the risk of further decline
in physical function in a sample of functionally impaired
older subjects. Although many factors have been associated
with disability (9), much less is known about the trajectory
of disability in institutionalized older subjects.
METHODS
In a large sample of older nursing home residents (the Is-
tituto di Riposo per Anziani [IRA] study), we recently
found that severe disability was associated with some indi-
cators of malnutrition. By multivariate analysis, we demon-
strated that this association was strong and independent
from the effects of age, gender, and comorbidity (10). Nev-
ertheless, we could not establish any temporal relationship
The IRA is a nursing home located in Padua, Northern It-
aly. The IRA study is a longitudinal study with the aim of
evaluating the relationship between a large number of bio-
logical parameters, functional status, and mortality. Func-
tional status was assessed at baseline and after 2 years.
Findings from the baseline cross-sectional survey have been
reported elsewhere (10).
M212

MALNUTRITION AND DISABILITY IN ELDERLY PERSONS
Health and Functional Status
M213
Subjects
From a total population of 410 individuals living in the
Age, gender, number of chronic medical conditions, and
drugs currently used were recorded at baseline. The ADLs
were evaluated at baseline by the geriatricians working in
the nursing home. The subjects were classified by the Katz
index (14) in reference to their progressive dependence in
feeding, continence, moving about, going to the toilet,
dressing, and bathing. The same physicians evaluated the
ADLs again after 24 months. Chronic medical conditions
(heart disease [including coronary heart disease and heart
failure], hypertension, stroke, diabetes mellitus, claudica-
tion, chronic obstructive pulmonary disease, parkinsonism,
rheumatic disease, depression, and dementia) were ascer-
tained by both medical examination and chart review.
nursing home at the time of the study’s inception, 344 estab-
lished residents (272 women and 72 men) were enrolled in
the IRA study. Inclusion criteria were as follows: subjects
must (i) have been Ն65 years old; (ii) have resided at the In-
stitute for at least 2 months; and (iii) have had no clinical
evidence of acute illness at the time of observation, nor in
the previous 30 days. Terminal patients with cancer or se-
vere liver and kidney disease were excluded.
At baseline, 172 subjects dependent in all or all but one
ADL (Katz classes F and G) were considered fully disabled
and were excluded. The remaining 172 subjects were inde-
pendent in at least two ADLs (Katz classes A to E) and were
considered at risk for worsening disability during the 2-year
follow-up period. Among these subjects, 52 died within the
2-year follow-up period. We defined a priori the presence of
worsening disability as a loss of two or more ADLs during
this 2-year follow-up period. According to this definition,
survivors were classified into two groups: group 1, 40 indi-
viduals with stable or improved functional status as defined
by their Katz score; group 2, 58 individuals with a signifi-
cant functional decline, defined as the loss of two or more
ADLs at follow-up.
Statistical Analysis
All results are reported as mean Ϯ standard deviation.
Baseline characteristics of participants were compared ac-
cording to functional status at the 2-years follow-up by us-
ing unpaired (paired when reported) Student’s t test for con-
2
tinuous variables and the ␹ test or Fischer’s Exact Test
(when necessary) for categorical variables. The association
of single variables with the 2-year modifications in ADLs
(worsening vs constant/improved) was evaluated by univari-
ate logistic regression analysis; afterwards, multivariate lo-
gistic regression analysis was used to identify the variables
independently associated with worsening functional status.
Variables were selected by the stepwise forward selection
method. The probability of the variable entering or being re-
moved from the model was .05 and .10, respectively.
At baseline, there were two individuals without disability
(no lost ADLs) in group 1 and six in group 2. Twenty-two
subjects who lost only one ADL at follow-up were not con-
sidered for this specific analysis.
Anthropometric Parameters
The following baseline parameters were forced into the
model: albumin, transferrin, total cholesterol, subscapular
skinfold thickness, and WHR (tertiles), age, gender, and
number of drugs (Tertile 1: Յ2 drugs; Tertile 2: Ͼ2 and Յ5
drugs; Tertile 3: Ͼ5 drugs). The baseline Katz status
(classes 0 to 6) was also included into the model to adjust
for the initial level of disability.
To investigate the association between the longitudinal
change in selected characteristics and the likelihood of be-
coming disabled in two or more additional functions, we
compared the average absolute change in these variables be-
tween the two groups of subjects by using the unpaired t test.
The 2-year absolute modification in the variables of interest
(⌬) was calculated by subtracting the baseline value from
the 2-year value (⌬ ϭ 2-y value Ϫ baseline value). The as-
sociation of the dichotomized 2-year modification in BCM
(decreased vs stable/increased BCM) with increasing dis-
ability (Ն2 ADLs lost) was evaluated by multivariate logis-
tic regression analysis. Systat (Systat, Inc., Evanston, IL)
for Windows, version 5.0, and SPSS for Windows, version
7.0 (SPSS, Inc, Chicago, IL) statistical packages were used.
According to standardized methods, weight, height, tricipi-
tal and subscapular skinfold thickness, and waist and hip cir-
cumference were measured by the same physician, who was
unaware of the subjects’ Katz score (11). The subjects’ body
mass index (BMI, kg/m2) and waist/hip ratio (WHR) were
calculated. A Holtain caliper (Holtain, Ltd., Crymych, UK)
was used to quantify skinfold thickness. Body composition
was determined by tetrapolar BIA (12,13) using a BIA 109
instrument (RJL Systems, Detroit, MI). The BIA measure-
ments were carried out after a 20-minute rest period. The
electrodes were placed just proximal to the right metacarpal
and third metacarpal joints, and between the right medial and
lateral right malleoli. A current at 50 Hz was then introduced.
Clinical Chemistry Parameters
All analyses were performed in the central laboratory of
the Geriatric Hospital of Padua, which complies with a qual-
ity-control program. Blood samples were obtained in the
morning after a 12-hour overnight fast, kept at 4ЊC for 1 hour,
and then centrifuged at 3000 rpm for 10 minutes at 4ЊC to
obtain serum or plasma. Serum total protein and iron were
measured by spectrophotometry, while albumin and transfer-
rin were assayed by nephelometry. Cell blood counts were
evaluated using a Bayer-H600 (Bayer, Tarrytown, NY) in-
strument. Total cholesterol and triglycerides were assayed
by the Trinder method. HDL cholesterol (HDL-C) was de-
termined after precipitation of apo B–containing lipoproteins
with MgCl2-phosphotungstic acid (in 84 of 98 subjects).
RESULTS
The baseline characteristics of the sample divided by the
2-year modifications in ADLs are reported in Table 1. Sub-
jects with deteriorating functional status at the 2-year fol-
low-up were older and had been treated with a higher num-
ber of drugs. On average, the number of lost ADLs was 1.6
and 2.0 in groups 1 and 2, respectively (p ϭ .34); after 2
years, lost ADLs were 1.4 in group 1 and 5.0 in group 2 (p ϭ

M214
ZULIANI ET AL.
Table 1. Principle Characteristics of Disabled Nursing
Table 3. Results of Logistic Regression Analysis (Stepwise
Forward) of 2-year Modifications in Functional Status (Stable/
Improved ADLs vs Ն2 Worsened ADLs)
Home Residents
Group 1
(2-y ADLs
constant or improved)
Group 2
(Ն2-y ADLs
worsened)
Variable
␤
SE
Wald Test
Odds Ratio
95% CI
Baseline Variables
p Value†
Albumin
Gender, % female
Age (yr)
No. Chronic Diseases
No. Drugs
74%
79.8 Ϯ 7.8
3.0 Ϯ 1.3
85%
82.7 Ϯ 6.6
3.4 Ϯ 1.3
.18
.05
.35
Tertile 1
Tertile 2
Tertile 3
Reference
0.97
Ϫ .011
Ϫ 1.79
0.82
0.71
0.0002
6.27
0.19–4.91
0.04–0.67
0.16
SST
Baseline
After 2 y
No. Lost ADLs
Baseline
2.9 Ϯ 1.7
3.4 Ϯ 1.8
4.2 Ϯ 2.3
4.4 Ϯ 2.3
.01
.03
Tertile 1
Tertile 2
Tertile 3
No. Drugs
Յ 2
2–5
Ͼ 5
Gender, male
Reference
0.17
Ϫ 1.73
Ϫ 2.86
1.10
1.08
2.45
6.99
0.02–1.5
0.006–0.5
0.06
1.6 Ϯ 0.9
1.4 Ϯ 0.9
4.48 Ϯ 0.4
260 Ϯ 59
226 Ϯ 42
51.2 Ϯ 17
13.9 Ϯ 1.3
15.8 Ϯ 4.9
0.97 Ϯ 0.1
26.9 Ϯ 3.9
17.9 Ϯ 5.2
2.0 Ϯ 1.1
5.0 Ϯ 1.0
4.27 Ϯ 0.35
228 Ϯ 38
.34
.001‡
.01
.003
.03
.04
.009
.01
After 2 y
Reference
7.5
10.5
Albumin (g/dl)
Transferrin (mg/dl)
Total Cholesterol (mg/dl)
HDL-C (mg/dl)
Hemoglobin (g/dl)
SST (mm)
WHR
BMI (kg/m2)
BCM (kg)
2.02
2.36
Ϫ 1.47
0.77
0.89
0.74
6.82
7.03
3.97
1.65–34.2
1.85–61.4
0.054–0.97
205 Ϯ 43
0.22
44.8 Ϯ 11.5
13.1 Ϯ 1.3
12.9 Ϯ 5.5
0.91 Ϯ 0.1
25.1 Ϯ 4.6
15.9 Ϯ 3.5
Notes: Logistic model: albumin, transferrin, total cholesterol, subscapular
skinfold thickness, waist/hip ratio, age, gender, number of drugs, baseline Katz
status. ADLs ϭ activities of daily living; CI ϭ confidence interval; SST ϭ sub-
scapular skinfold thickness.
.01
.07
.05
Note: Only subjects independent in Ն2 activities of daily living (ADLs) at
baseline were included in this study. HDL-C ϭ high density lipoprotein choles-
terol; SST ϭ subscapular skinfold thickness, WHR ϭ waist/hip ratio, BMI ϭ
body mass index, BCM ϭ body cell mass.
serum albumin levels, subscapular skinfold thickness, num-
ber of drugs, and female gender. The results did not change
by forcing BCM, BCM to fat-free mass ratio, or BCM to
weight ratio into the model (data not shown).
†Group 1 vs group 2, unpaired t test.
‡Group 1 vs group 2 (unpaired t test) and group 2 baseline vs after 2 years
(paired t test).
The 2-year modifications in the principal clinical-chemis-
try and anthropometric parameters are reported in Table 4.
A slight reduction in all parameters was observed at the 2-year
follow-up. The BCM decreased, on average, by 0.1 kg
(from 17.8 to 17.7 kg, or Ϫ0.5%) in group 1, and by 2.4 kg
(from 15.9 to 13.1 kg, or Ϫ15%) in group 2 (p Ͻ .01). The
odds ratio for increasing disability in subjects with decreas-
ing BCM was 4.94 (95% confidence interval [CI] 1.17–
20.21) compared with subjects with stable and/or increased
BCM, after adjustment for the principal baseline variables
(i.e., Katz score, BCM, albumin, total cholesterol, serum
iron, number of chronic medical conditions, and number of
drugs), age, and gender. In a different model, the odds ratio
was 5.60 (95% CI 1.03–30.11), after adjustment for base-
.001 for group 1 vs group 2, and group 2 baseline vs 2-y fol-
low-up). Group 2 patients were also characterized by lower
albumin, transferrin, total cholesterol, HDL-C, hemoglobin,
WHR, subscapular skinfold thickness, BMI, and BCM.
The principal diseases in the two groups of subjects at
baseline are reported in Table 2. No significant differences
emerged, but the prevalence of cognitive impairment was
much higher in group 2 (32%) compared with group 1 (13%).
Table 3 reports the results of multivariate logistic regres-
sion analysis for the 2-year modification in ADLs. Deterio-
rating functional status (Ն2 ADLs lost) was associated with
Table 4. Two-year Average Modifications (⌬ ϭ 2-y value Ϫ
baseline value) in the Principal Clinical-Chemistry and
Anthropometric Parameters
Table 2. Principal Baseline Diseases in Disabled Nursing Home
Residents Divided by 2-year Modification in Functional Status
Group 1
(2-y ADLs constant
or improved, %)
Group 1
(Ն2-y ADLs
worsened, %)
Group 1
(2-y ADLs
constant or improved)
Group 2
(Ն2-y ADLs
worsened)
Disease
p Value
Variables (⌬)
p Value†
Dementia
20
13
6
19
32
15
13
24
15
11
19
9.5
4
.80
.11
.45
.36
.70
.89
.45
.35
.80
.65
.50
Albumin (g/dl)
Transferrin (mg/dl)
Total cholesterol (mg/dl)
HDL-C (mg/dl)
Hemaglobin (g/dl)
SST (mm)
WHR
BMI (kg/m2)
BCM (kg)
Ϫ0.61
Ϫ22
Ϫ19
ϩ0.76
Ϫ0.53
Ϫ1.5
Ϫ0.04
Ϫ1.1
Ϫ0.67
Ϫ29
Ϫ16
Ϫ1.12
Ϫ0.90
Ϫ4.0
Ϫ0.08
Ϫ2.1
.35
.57
.36
.24
.46
.43
.09
.08
.01
Cognitive impairment
Parkinson’s disease
Depression
Cerebrovascular disease
Heart failure
Coronary heart disease
Hypertension
Diabetes
23
22
13
23
30
13
10
10
Ϫ0.1
Ϫ2.4
Neoplasm
COPD
Note: ADLs ϭ activities of daily living; HDL-C ϭ high density lipoprotein
cholesterol; SST ϭ subscapular skinfold thickness; WHR ϭ waist/hip ratio;
BMI ϭ body mass index; BCM ϭ body cell mass.
18
Note: COPD ϭ chronic obstructive pulmonary disease; ADLs ϭ activities
of daily living.
†Unpaired t test.

MALNUTRITION AND DISABILITY IN ELDERLY PERSONS
M215
Taken together, our results suggest the following: (i) at
baseline, several indicators of malnutrition were lower in
subjects who experienced a further decline in functional sta-
tus compared with those who remained stable over the 2-year
follow-up period; (ii) low albumin and subscapular skinfold
thickness values seem to predict future worsening in func-
tional status; and (iii) the loss in BCM and the loss of physi-
cal function are significantly associated over time.
It is not known why markers of malnutrition and disabil-
ity may be associated in older nursing home residents, nor is
the possible mechanism linking the loss in BCM and physi-
cal function known. Low albumin levels are considered a
marker of malnutrition and have been associated with poor
outcomes, such as morbidity and mortality (16,17). Albu-
min is also a negative protein of the acute phase (18), and it
has been proposed that low albumin might be a marker for
disease in elderly individuals (19). Of interest, Baumgartner
and colleagues (20) found that serum albumin was associ-
ated with muscle mass and suggested that the increased risk
of disability with low albumin levels in elderly persons
might reflect an association with sarcopenia.
The loss of BCM is a good indicator of malnutrition (15).
A reduction in BCM has been associated with chronic sys-
temic inflammation in patients with rheumatoid arthritis and
cancer (21–23), and a role of inflammatory cytokines has
been proposed in the pathogenesis of geriatric cachexia
(24). Cederholm and colleagues (25) found increased inter-
leukin 6 (IL-6) and IL-1␤ levels in older subjects with
cachexia but without any evidence of malignant or infec-
tious diseases. On the other hand, it has been shown that
disability is associated with increased levels of plasma IL-6
in elderly persons (26), and Ferrucci and colleagues (27)
found that higher IL-6 levels predict disability onset in com-
munity-dwelling older persons.
Figure 1. Two-year body cell mass (BCM) modification (⌬ BCM:
2-y value Ϫ baseline value). Subjects were divided into six groups
based on the number of additional activities of daily living (ADLs)
lost at follow-up (0–5 ADLs lost). Test for linear trend was ascer-
tained from the linear regression model with the number of ADLs
lost entered as an ordinal variable.
line Katz score, age, gender, and the 2-year changes (⌬)
in albumin, cholesterol, serum iron, hemoglobin, HDL-C,
transferrin, and number of drugs.
Figure 1 shows a stepwise dose-response relationship be-
tween the 2-year BCM reduction and the 2-year decline in
functional status; the degree of BCM reduction was strongly
related to the number of additional ADLs lost at the follow-
up (test for trend, p ϭ .003).
It is difficult to explain the relationship between markers
of malnutrition, in particular the loss in BCM, and disabil-
ity. One possible explanation is that the loss in BCM, possi-
bly driven by chronic inflammation, might lead to a loss in
physical function, perhaps through a loss in muscular
strength. This would be in accord with the cross-sectional
results of Zamboni and colleagues (28) who found, in a
sample of elderly women, an association between BCM and
leg strength, suggesting a possible role of muscle mass loss
as a risk factor for disability. Nevertheless, basal BCM was
not an independent predictor of increasing disability in our
sample, although it is well known that low muscular
strength is a predictor of future disability. A second possi-
bility is that deteriorating functional status might lead to a
decrease in BCM through undernutrition or inactivity. This
is not supported by the observation that most of the subjects
in group 2 were able to feed themselves independently. Fur-
thermore, BCM was lower at baseline in group 2 compared
with group 1, despite a very similar functional status. An-
other possibility is that a third factor might cause the de-
crease in both physical function and BMC. It is possible that
older residents with deteriorating functional status, unlike
those with stable ADLs, would be frail or “more frail.” If
this is the case, chronic diseases and the respective chronic
inflammation state would lead to weight loss, undernutri-
tion, and disability in these subjects; this would be consis-
DISCUSSION
We found that older disabled nursing home residents with
2-year deteriorating functional status were characterized, at
baseline, by a “worse” nutritional status when compared
with disabled subjects with stable and/or improved ADLs.
Even if the two groups were very similar in gender distribu-
tion, number of chronic diseases, and functional status,
group 2 subjects had lower values in several laboratory
(e.g., albumin, transferrin, and cholesterol) and anthropo-
metric (e.g., BCM, skinfold thickness, and WHR) parame-
ters compared with group 1. Among the variables consid-
ered, low albumin and subscapular skinfold thickness
values (Tertile 1) were associated with the risk of deteriorat-
ing functional status after multivariate adjustment.
We also measured the BCM, which is considered the
“working” portion of fat-free mass, by means of BIA. This is
important, as the term “malnutrition” refers to the loss of
structural body components, which is most accurately re-
flected by a reduction in the BCM (15). Not only was the
BCM lower at baseline in group 2 compared with group 1, but
it was also the only variable that decreased significantly at the
same time as functional status. After multivariate adjustment,
the risk of increasing disability was about fivefold in subjects
who lost BCM compared with individuals with stable and/or
increased BCM. Furthermore, the reduction in BCM was pro-
portional to the number of ADLs lost at follow-up.

M216
ZULIANI ET AL.
6. Marton KI, Sox HC Jr, Krupp JR. Involuntary weight loss: diagnosis
and prognosis significance. Ann Intern Med. 1981;95:568–574.
7. Buzby GP, Mullen JL, Matthews DC, et al. Prognostic nutritional in-
dex in gastrointestinal surgery. Am J Surg. 1980;139:160–167.
8. Rudman D, Feller AG. Protein-calorie undernutrition in the nursing
home. J Am Geriatr Soc. 1989;37:173–193.
9. Fried LP, Guralnik JM. Disability in older adults: evidence regarding
significance, etiology, and risk. J Am Geriatr Soc. 1997;45:92–100.
10. Romagnoni F, Zuliani G, Bollini C, et al. Disability is associated with
malnutrition in institutionalized elderly people. The I.R.A study. Ag-
ing Clin Exp Res. 1999;11:194–199.
11. Lohman TG, Roche AF, Martorell R. Anthropometric Standardization
Reference Manual. Champaign, IL: Human Kinetics Books; 1988.
12. McDougall D, Shizgal HM. Body composition measurements from the
whole body resistance and reactance. Surg Forum. 1986;37:42–45.
13. Shizgal HM. Validation of the measurements of body composition
from whole body bioelectric impedance analysis. Infusionstherapie.
1990;17(suppl):64–74.
14. Katz S, Downs TD, Cash HR, Grotz RC. Progress in the development
of the index of ADL. Gerontologist. 1970;1:20–30.
15. Moore FD, Olesen KH, McMurray JD. The Body Cell Mass and Its
Supporting Enviroment: Body Composition in Health and Disease.
Philadelphia: WB Saunders; 1963:23–27.
16. Corti MC, Guralnik JM, Salive ME, Sorkin JD. Serum albumin level
and physical disability as predictors of mortality in older persons.
JAMA. 1994;272:1036–1042.
17. Fried LP, Kronmal RA, Newman AB, et al. Risk factors for 5-year
mortality in older adults. The Cardiovascular Health Study. JAMA.
1998;279:585–592.
18. Ballmer PE, Ballmer-Hofer K, Repond F, et al. Acute suppression of
albumin synthesis in systemic inflammatory disease: an individually
graded response of rat hepatocytes. J Histochem Cytochem. 1992;40:
201–206.
19. Friedman PJ, Campbell AJ, Caradoc-Davies TH. Hypoalbuminemia in
the elderly is due to to disease not malnutrition. J Clin Exp Gerontol.
1985;7:191–203.
20. Baumgartner RN, Koehler KM, Romero L, Garry PJ. Serum albumin
is associated with skeletal muscle in elderly men and women. Am J
Clin Nutr. 1996;64:552–558.
21. Roubenoff R, Roubenoff RA, Cannon JG, et al. Rheumatoid cachexia:
cytokine-driven hypermetabolism accompanying reduced body cell
mass in chronic inflammation. J Clin Invest. 1994;93:2379–2386.
22. Simons JP, Schols AM, Buurman WA, Wouters EF. Weight loss and low
body cell mass in males with lung cancer. Clin Sci. 1999;97:215–223.
23. McMillan DC, Scott HR, Watson WS, Preston T, Milroy R, McArdle
CS. Longitudinal study of body cell mass depletion and the inflamma-
tory response in cancer patients. Nutr Cancer. 1998;31:101–105.
24. Yeh SS, Schuster MW. Geriatric cachexia: the role of cytokines. Am J
Clin Nutr. 1999;70:183–197.
25. Cederholm T, Wretlind B, Hellstrom K, et al. Enhanced generation of
interleukins 1␤ and 6 may contribute to cachexia of chronic disease.
Am J Clin Nutr. 1997;65:876–882.
26. Cohen HJ, Pieper CF, Harris T, et al. Plasma IL-6: an indicator of
functional disability in community dwelling elderly. J Gerontol Med
Sci. 1997;52A:M201–M208.
tent with both their baseline characteristics and their pro-
gressive loss in BCM and ADLs. The dose-response
relationship between the number of medications and the risk
of functional decline seems to reinforce the main role of
symptomatic diseases in the disablement process.
Finally, three limitations of the present study must be un-
derlined. First, our sample was made by a relatively small
number of older disabled nursing home residents. Not only
do the results need to be confirmed in larger samples, but
they also cannot be generalized to community-dwelling
older individuals. Second, we measured indirectly BCM by
BIA, and it is possible that age, sex, and other factors, in-
cluding hydration, could have led to over- or underestima-
tion of body compartments. Nevertheless, all the individuals
were clinically stable and had no evidence of acute illness at
the time of measurements. Furthermore, terminal patients
with cancer or severe liver and kidney disease were ex-
cluded. On the other hand, BIA is currently used to estimate
malnutrition in elderly individuals (28) and has been corre-
lated with other traditional methods, both in healthy sub-
jects and in patients with different diseases and nutritional
status (13,29,30). Third, we did not consider the new inci-
dental pathologies, which possibly appeared during the fol-
low-up. It is possible that, at least in some cases, a new dis-
ease might have caused a sudden worsening in functional
status with a consequent reduction in BCM.
In conclusion, the results of this longitudinal study sug-
gest that (i) signs of malnutrition seem to predict the wors-
ening in functional status in older disabled nursing home
residents and (ii) the BCM declines proportionally to the
loss in ADLs over time, suggesting the existence of a strong
relationship between BCM loss and the deterioration of
functional status in institutionalized older individuals. Nev-
ertheless, the underlying mechanism for such an association
has not been determined. Understanding whether the de-
cline in BCM is causally related to the decline of physical
function might provide important insight into the patho-
physiology of disability in older persons.
Acknowledgments
The first two authors (GZ and FR) contributed equally to this study.
Address correspondence to Prof. Renato Fellin, Istituto di Medicina In-
terna II, Universita’ degli Studi, via Savonarola no. 9, 44100 Ferrara, Italy.
E-mail: flr@ifeuniv.unife.it
27. Ferrucci L, Harris TB, Guralnik JM, et al. Serum IL-6 levels and the
development of disability in older persons. J Am Geriatr Soc. 1999;47:
639–646.
References
28. Zamboni M, Turcato E, Santana H, et al. The relationship between
body composition and physical performance in older women. J Am
Geriatr Soc. 1999;47:1403–1408.
29. Shizgal HM. The effect of malnutrition on body composition. Surg
Gynecol Obstet. 1981;152:22–26.
30. Segal KR, Van Loan M, Fiztgerald P, Hodgdon JA, Van Itallie TB.
Lean body mass estimation by bioelectrical impedance analysis: four
site cross-validation study. Am J Clin Nutr. 1988;47:7–14.
1. Frisoni GB, Franzoni S, Rozzini R, Ferrucci L, Boffelli S, Trabucchi
M. A nutritional index predicting mortality in the nursing home. J Am
Geriatr Soc. 1994;42:1167–1172.
2. Antonelli Incalzi R, Landi F, Cipriani L, et al. Nutritional assessment:
a primary component of multidimensional geriatric assessment in the
acute care setting. J Am Geriatr Soc. 1996;44:166–174.
3. Lewis EJ, Stacey JB. Nutritional assessment in the elderly. In: Morley
JE, Glick Z, Rubenstein LZ, eds. Geriatric Nutrition. New York:
Raven Press; 1990:73–87.
4. Silver AJ, Morley JE, Strome LS, et al. Nutritional status in academic
nursing homes. J Am Geriatr Soc. 1988;36:487–491.
Received September 14, 2000
5. Abbasi AA, Rudman D. Undernutrition in the nursing home: prevalence,
consequences, causes and prevention. Nutr Rev. 1994;52:113–122.
Accepted October 10, 2000
Decision Editor: John E. Morley, MB, BCh