Filetype PDF | Posted on 07 Jan 2023 | 3 years ago
Authentication
digital resources
374x Filetype PDF File size 0.25 MB Source: www.irspen.ie
Clinical Nutrition 32 (2013) 1007e1011
Contents lists available at SciVerse ScienceDirect
Clinical Nutrition
journal homepage: http://www.elsevier.com/locate/clnu
Original article
Prevalence and determinants for malnutrition in geriatric outpatients
a,* b,d
Marian A.E. van Bokhorst-de van der Schueren , Sabine Lonterman-Monasch ,
c,e c,e c,e c,e
Oscar J. de Vries , Sven A. Danner , Mark H.H. Kramer , Majon Muller
aDepartment of Nutrition and Dietetics, Internal Medicine, VU University Medical Center, PO Box 7057, 1007 MB Amsterdam, The Netherlands
bDepartment of Internal Medicine, Haga Hospital, Leyweg 275, 2454 CH The Hague, The Netherlands
cDepartment of Internal Medicine, VU University Medical Center, PO Box 7057, 1007 MB Amsterdam, The Netherlands
articleinfo summary
Article history: Background &aims:Fewdataisavailableonthenutritionalstatusof geriatric outpatients. The aim of this
Received 13 February 2013 study is to describe the nutritional status and its clinical correlates of independently living geriatric older
Accepted 13 May 2013 individuals visiting a geriatric outpatient department.
Methods: From 2005 to 2010, all consecutive patients visiting a geriatric outpatient department in the
Keywords: Netherlands were screened for malnutrition. Nutritional status was assessed by the Mini Nutritional
Geriatric outpatients Assessment (MNA). Determinants of malnutrition were categorized into somatic factors (medicine use,
Nutritional status comorbidity, walking aid, falls, urinary incontinence), psychological factors (GDS-15 depression scale,
Mini nutritional assessment MMSEcognition scale), functional status (Activities of Daily Life (ADL), Instrumental ADL (IADL)), social
Malnutrition factors (children, marital status), and life style factors (smoking, alcohol use). Univariate and multivariate
logistic regression analyses, adjusted for age and sex and all other risk factors were performed to identify
correlates of malnutrition (MNA < 17).
Results: Included were 448 outpatients, mean (SD) age was 80 (7) years and 38% was men. Prevalence of
malnutrition and risk for malnutrition were 17% and 58%. Depression, being IADL dependent, and
smoking were independently associated with an increased risk of malnutrition with OR’s (95%CI) of 2.6
(1.3e5.3), 2.8 (1.3e6.4), 5.5 (1.9e16.4) respectively. Alcohol use was associated with a decreased risk (OR
0.4 (0.2e0.9)).
Conclusion: Malnutrition is highly prevalent among geriatric outpatients and is independently associated
with depressive symptoms, poor functional status, and life style factors. Our results emphasize the
importance of integrating nutritional assessment within a comprehensive geriatric assessment. Future
longitudinal studies should be performed to examine the effects of causal relationships and multifac-
torial interventions.
2013Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
1,2
1. Introduction conditions put older individuals at a higher risk of malnutrition.
Malnutrition is a prognostic factor associated with morbidity,
3,4
Aging maycome with an accumulation of diseases and impair- mortality and costs of care. It is therefore important to detect
ments, including cognitive and physical decline, depressive symp- those older individuals who are at risk for malnutrition.
toms and emotional changes, all of which may directly influence The reported prevalence rates of malnutrition in the
1
the balance between nutritional needs and intake. Dietary Netherlands are relatively low in community dwelling older per-
behavior of older individuals may change because of health or so- sons (2%e12%), but rise considerably in older individuals receiving
cial reasons, decrease in taste and smell, or a reduced ability to home care (18%e35%) or in the hospitalized or institutionalized
purchase and prepare food. This combination of symptoms or older patients (30%e60%).5e9
Dataontheprevalenceofmalnutritionandclinicalcorrelatesof
nutritional status of geriatric patients whovisit geriatric outpatient
* Corresponding author. Tel.: þ31 20 4443410. departments is not available. These patients are referred to an
E-mail addresses: m.vanbokhorst@vumc.nl (M.A.E. van Bokhorst-de van der outpatient clinic with multiple problems in somatic functioning,
Schueren), s.lonterman-monasch@hagaziekenhuis.nl (S. Lonterman-Monasch), psychological functioning, and/or with functional or social prob-
oj.devries@vumc.nl (O.J. de Vries), s.danner@vumc.nl (S.A. Danner), m.kramer@ 10
vumc.nl (M.H.H. Kramer), m.muller@vumc.nl (M. Muller). lems. Multimorbidity is thought to have a direct influence on the
d Tel.: þ31 70 2100000. balance between nutritional needs and nutritional intake and to
e 11
Tel.: þ31 20 44443009. contribute to a high prevalence of malnutrition.
0261-5614/$ e see front matter 2013 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
http://dx.doi.org/10.1016/j.clnu.2013.05.007
1008 M.A.E. van Bokhorst-de van der Schueren et al. / Clinical Nutrition 32 (2013) 1007e1011
In this study we aimed to investigate the malnutrition preva- wheelchair. Falls were classified as never vs. ever. Urinary inconti-
lence rates among older patients visiting a geriatric outpatient nence was classified as absent vs. present.
department of a large teaching hospital in the Netherlands. Psychological characteristics included depressive symptoms
Furthermore, we investigated which somatic, psychological, func- andcognitive functioning. Depressive symptoms were assessed by
tional, social or life style characteristics were associated with the Geriatric Depression Scale with 15 items (GDS-15). A higher
14
malnutrition. score indicates more depressive symptoms. Acut-off value of 5
was used to indicate clinically important depressive symptoms.
2. Methods Global cognitive functioning was assessed with the Mini Mental
State Examination (MMSE). Cognitive dysfunction was defined as
15
2.1. Study design and population an MMSEscore <24.
Functional characteristics included activities of daily life (ADL)
For this cross-sectional study, aiming to investigate the clinical and instrumental ADL (IADL). ADL was assessed by asking if the
correlates of nutritional status of geriatric patients, we included patient was able to dress or wash himself independently, partly
448consecutive patients at their first visit to a geriatric outpatient independent, or with help only. IADL was assessed by asking the
clinic of a large teaching hospital in the Netherlands between patientifhe/shewasabletodotheshopping,financesandcleaning
October 2005 and March 2010. All patients were living indepen- the household independently, partly independent, or with help
dently (in their own home or in an assisted care facility). Patients only. Both ADL and IADL were classified as independent or partly
living in a nursing home were excluded. Patients were referred for independent vs. dependent.
multiple problems in the somatic, psychological, social or func- Social characteristics included education, marital status, and
tional domain. Data collectionwas performed prospectivelyas part whether the patient had children. Education was classified as low
of the routine measurements during the outpatient visits. All pa- (no education/primary school), middle (lower vocational educa-
tients underwent a comprehensive geriatric assessment including tion/intermediate vocational education), or higher education (pre-
physical examination, laboratory tests and functional screening. university education/higher vocational education/university).
Nutritional status, cognitive functioning and depressive symptoms Marital status was classified as married/living together or unmar-
were assessed with questionnaires. Furthermore, patients were ried/divorcedvs.widow(-er).Presenceofchildrenwasclassifiedas
asked about demographics, medical history, medication use, and zero vs. 1 child(ren).
life style. Finally, it was inquired whether a patient was a current smoker
(vs. former smoker or never smoker) or a current alcohol user (vs.
2.2. Nutritional status former or never alcohol user).
Nutritional status was assessed with the Mini-Nutritional 2.4. Other variables
Assessment (MNA), a validated questionnaire for older in-
12
dividuals, recommended by the European Society for Clinical Height was measured with a stadiometer to the nearest cen-
13 timetre (cm) and weight was assessed by a non-electronically
Nutrition and Metabolism (ESPEN). The questionnaire consists of
18questionsclusteredinfoursections:anthropometricassessment scale (Seca, model 761) to the nearest kilogram (kg). Patients
(weight, height, weight loss); general assessment (living situation, were weighed with their clothes on and the measured body
medicine use, mobility); dietary assessment (number of meals, weight was corrected for clothing (2 kg). BMI was calculated as
2
food and fluid intake, and autonomy of feeding), and subjective weight in kg divided by the square of height in meters (kg/m ).
assessment (self-perception of health and nutritional status). A Waist circumference was measured to the nearest cm with a
maximumscoreof 30 can be obtained. A score below 17 indicates flexible tape measured while the patient was in standing position.
malnutrition,ascoreof17e23.5indicatesariskofmalnutritionand Thetapewasplacedapproximately3cmbelowthebellybuttonof
a score of 24 or higher indicates a satisfactory nutritional status. If the patient.
the patient was suspected not to be able to give reliable answers,
the MNA questionnaire was confirmed by proxy. 2.5. Statistical analyses
2.3. Conditions associated with malnutrition Patient characteristics were calculated for the nutritional status
categories (MNA <17.0, 17.0e23.5, and >23.5). Differences across
Possible clinical determinants of malnutrition were classified as categories were tested with ANOVA for normally distributed vari-
somatic, psychological, functional, social, and life style factors. ables, KruskalleWallis tests for not normally distributed variables,
Somatic characteristics included medication use, co-morbidity, and with c2 tests for categorical variables.
fall-events, use of a walking aid, and urinary incontinence. The Logistic regression analyses were performed to assess the in-
number of drugs was derived from the patients’ medical records dependent association of the clinical covariates with presence of
and was checked by asking the patient or the caregiver. Both pre- malnutrition (MNA < 17). Somatic, psychological, functional, social
scription drugs and over-the counter-drugs were included. Poly- andlife style characteristics were separately included as covariates
pharmacywasclassifiedasusing<6drugsvs.6drugs(6beingthe in the model. Regression analyses wereadjustedforageandsex.To
median number of drugs taken). Comorbidity was assessed by assess the independent association of the clinical characteristics
summingthenumbersofunderlyingchronic diseases of a patient. with presence of malnutrition, all covariates for malnutrition (so-
Multimorbidity was classified as having <4vs.4 chronic diseases matic,psychological,functional,social,andlifestyle)wereincluded
(divided by median number of comorbidities). Information about in one logistic regression model using backward elimination.
underlying diseases was obtained from the patients’ medical re- Finally, all somatic, psychological, functional, and social corre-
cords.Thefollowingchronicdiseaseswereclassified:hypertension, lates were summed and mean adjusted MNA scores were calcu-
diabetes mellitus, cardiovascular disease, cerebrovascular disease, lated for categories of number of clinical problems (2, 3e4, 5e6,
renal impairment, osteoporosis, chronic obstructive pulmonary 7) using analysis of covariance (ANCOVA).
disease (COPD), and malignancy. The use of a walking aid was Statistical analyses were performed with Statistical Package for
classified as none vs. use of a walking stick/trolley walker/ theSocialSciences(SPSSInc,Chicago,IL)version20.0forWindows.
M.A.E. van Bokhorst-de van der Schueren et al. / Clinical Nutrition 32 (2013) 1007e1011 1009
3. Results Table 2
Prevalence of comorbidity across categories of nutritional status.
Mean(SD)ageofthetotalpopulation(n¼448)was80(7)years *
MNA<17 MNA MNA>23.5 p-Value
and 38% was male. In this population of geriatric outpatients, 17% 17e23.5
was malnourished (MNA < 17.0), and 58% were at risk for malnu- N¼76 N¼261 N¼111
trition (MNA 17.0e23.5). Hypertension 55% 55% 56% 0.98
Table 1 presents the patient characteristics across categories of Diabetes 24% 31% 25% 0.27
MNA. Patients with malnutrition had a lower weight, BMI and Cardiovascular disease 51% 48% 37% 0.09
smallerwaistcircumference,andlessfrequentlydrankalcoholthan Cerebrovascular disease 17% 15% 7% 0.09
patients with better nutritional status. Also, patients with malnu- Renal impairment 15% 17% 15% 0.78
tritionusedmoremedication,weredependentonwalkingaidmore COPD 15% 11% 9% 0.52
frequently, and more often had urinary incontinence than patients Osteoporosis 20% 9% 11% 0.03
Malignancy 28% 23% 27% 0.59
with better nutritional status. Finally, patients with malnutrition COPD: chronic obstructive pulmonary disease; MNA: mini nutritional assessment.
had more depressive symptoms, had a higher prevalence of poor *p-Value derived from c2 test.
functional status, and were lower educated.
Table 2 shows the prevalence of comorbidities across categories
of nutritional status. Patients with malnutrition more often had 5.5 (1.9e16.4) respectively. Alcohol use was associated with a
osteoporosis; a trend was observed for cardiovascular and cere- decreased risk (OR 0.4 (0.2e0.9)).
brovascular diseases. Furthermore, increasing numbers of correlates were associated
Univariate logistic regression analyses, adjusted for age and sex with lower mean MNA scores; the p-value for trend was <0.001
showedthatpatientswhosmoked,patientwhousedawalkingaid, (Fig. 1).
patientswithdepressivesymptoms,andpatientsbeingADLorIADL
dependentwereatincreasedriskformalnutrition(Table3).Patient 4. Discussion
currently using alcohol were at decreased risk of being malnour-
ished. Polypharmacy, multimorbidity, falls, urine incontinence, The present study among 448 independently living geriatric
level of education, cognitive functioning, level of education, marital outpatients indicates a high prevalence of malnutrition and risk of
status,or‘havingchildren’werenotsignificantlyassociatedwithan malnutrition(17%and58%).Multimorbidity,poorfunctionalstatus,
increased risk of malnutrition (Table 3). depressive symptomsandsmokingwereindependentlyassociated
In the multivariate model depression, being IADL dependent, with an increased risk of malnutrition. Also, the more somatic,
andsmokingremainedindependentlyassociatedwithanincreased psychological, social, or functional problems a patient experienced,
riskofmalnutritionwithOR’s(95%CI)of2.6(1.3e5.3),2.8(1.3e6.4), the higher the risk of being malnourished. Alcohol use was asso-
ciated with a decreased risk of malnutrition.
This is one of the first studies describing malnutrition preva-
Table 1 lence rates among older persons visiting an outpatient clinic. We
Characteristics of the study population (N ¼ 448) according to categories of nutri- are aware of two other European studies including older patients,
tional status. both showing lower prevalence rates than the present study.
**
MNA<17 MNA MNA>23.5 p-Value Howeverinthesestudiesthepatientsincludedweremuchyounger
17e23.5 11,16
thanours. In the ‘middle old’ (75e84 y) and ‘oldest old’ (85 y)
N¼76 N¼261 N¼111 11
subpopulations in Saka’s study he found data very much in
General characteristics accordance with ours. Our study population consisted of relatively
Sex, % male 33% 38% 40% 0.63 unhealthy older patients (with many patients having functional
Age (yr)a 82 7807807 0.11 limitations, being incontinent, having depressive symptoms or
Weight (kg)a 62 13 72 14 76 13 <0.01
2 a cognitive decline). The sample is thought to be representative for
BMI (kg/m ) a 22 4254273 <0.01 older patients attending Dutch geriatric outpatient clinics, but may
Waist circumference (cm) 94 12 101 12 102 9 <0.01
Life-style characteristics not be representable across Europe. The poor medical condition of
Smoking, % current 20% 12% 8% 0.12
Alcohol use, % current 30% 43% 57% 0.01
Somatic characteristics Table 3
Medication use, % 6 40% 42% 25% 0.02 Somatic,psychological,functional,andsocialcharacteristicsandriskofmalnutrition
Comorbidities, % 4 53% 52% 46% 0.29 (MNA<17).
Using walking aid, % 65% 46% 35% <0.01
Falls, % ever 63% 61% 55% 0.53 Characteristics MNA<17
Urinary incontinence, % 59% 66% 52% 0.04 OR(95%CI)
Psychological characteristics
GDS, % 5 46% 32% 8% <0.01 Lifestyle Smoking (current) 4.3 (1.9e 9.9)
Cognition, % MMSE <24 61% 53% 45% 0.09 Alcohol (current) 0.4 (0.2e 0.8)
Functional characteristics Somatic Medication use (6) 1.0 (0.6e 1.7)
ADL, % dependent 33% 11% 5% <0.01 Comorbidities (4) 1.1 (0.6e 1.8)
IADL, % dependent 50% 28% 17% <0.01 Use of walking aid (yes) 2.1 (1.2e 3.6)
Social characteristics Fall incident (ever) 1.0 (0.6e 1.8)
Education, % low 41% 33% 21% 0.01 Urinary incontinence (yes) 0.8 (0.5e 1.4)
Marital status, % 41% 36% 32% 0.52 Psychological GDS-15 (5) 2.8 (1.6e 4.9)
widow(-er) MMSE(<24) 1.5 (0.8e 2.5)
Children, % no children 17% 13% 12% 0.57 Functional ADL(dependent) 4.9 (2.6e 9.3)
ADL: activities of daily living; BMI: body mass index; GDS: geriatric depression IADL (dependent) 3.1 (1.8e 5.3)
scale; IADL: instrumental activities of daily living; MMSE: mini mental state ex- Social Education (low) 1.6 (0.9e 2.8)
amination; MNA: mini nutritional assessment. Marital status (widow) 1.0 (0.6e 1.8)
**P-value derived from either ANOVA, KruskalleWallis, or c2 test. Children (no) 1.3 (0.7e 2.6)
a Mean SD. Adjusted for age, sex.
1010 M.A.E. van Bokhorst-de van der Schueren et al. / Clinical Nutrition 32 (2013) 1007e1011
16
24 explain the association between malnutrition and depression.
Malnutrition has been associated with progressive loss of muscle
23 massandmusclestrength,whichcouldbeexplainedbydecreased
e 22
r activity pattern or inadequate nutritional intake.
o
c Life style factors were found to be associated with malnutrition
s 22
t aswell,wherebysmokingincreasedtheriskofbeingmalnourished
n
e and alcohol use decreased this risk. Being a current smoker has
m 21
s beenassociatedwithapoorernutritionalstatusinCOPDpatientsin
e
s earlier studies, after adjustment for age, social deprivation and
s 20 23,24
A disease severity. The authors of these manuscripts hypothe-
l
a sized that the association between smoking and malnutrition may
n
o 19
i be linked to how taste and appetite are affected in smokers, or to
t
i
r
t the pro-inflammatory effect of smoking. Our finding that alcohol
u
N 18 use was associated with a decreased risk of malnutrition is in line
i 6
n
i witharecentcross-sectionalstudyincommunitydwellingelderly.
M 17 The beneficial effect of alcohol consumption may be found in the
high energy content of alcohol consumptions, thus preventing
16 involuntary weight loss, or in the context of the alcohol drinking
<=2 3 to 4 5 to 6 >=7 (often in companionship).
Malnutritionisassociatedwithadverseclinicaloutcomes,ashas
Number of somatic, psychological, functional, or 25e27
beenshowninalargenumberofstudies. Sincemalnutritionis
social determinants mostly thought to be modifiable, it is important to develop and
Fig. 1. Mean (SE) mini nutritional assessment scores for categories of number of implement adequate interventions to prevent, diagnose and treat
clinical determinants. malnutrition. Early identification of malnutrition is a first step. The
MNAfulfils many criteria for both screening and diagnostic mea-
sures. However, critics may argue that it is a too time-consuming
our patient cohort may in part explain the higher prevalence rates method to use in daily clinical practice. In this study, we have
of malnutrition. shownthatitsimplementationisfeasibleintheoutpatientsetting.
The two earlier studies describe a correlation between malnu- Moreover, most questions of the MNA are already covered by the
trition and other comorbidities, among others: depression, fecal Comprehensive Geriatric Assessment, which has been adapted as
incontinence, bone mineral density, several biochemical parame- the basis for diagnosis and treatment in Dutch geriatric medicine.
ters, decreased cognitive functioning and functional de- TheMNAhastheadvantageovereasierscreeninginstrumentsthat
11,16
pendency. Theseresultsaswellasoursunderlinetheimportance it identifies not only (the degree of) malnutrition, but also the
of multimorbidityas a risk factor for malnutrition and vice versa. possible underlying causes. For these reasons, a Dutch geriatric
More studies have described the nutritional status of commu- consensus group has defined the MNA as the preferred instrument
5e9,17 21
nity dwelling older individuals. These studies observed prev- for diagnosis and screening in the Netherlands.
alence rates of malnutrition ranging from 0 to 35%. However, these As a follow-up to diagnosis of malnutrition, adequate in-
studies used different definitions for malnutrition and different terventions are required. Recent studies indicate that protein and
settings such as private households, general practice, communities, energy supplementation to malnourished older subjects not only
28
and institutions. Because of these differences, the prevalences leads to increase body weight, but also to improved function and
29,30
found are difficult to compare. decreasednumberoffalls. Followingtheideathatmalnutrition
Our finding that the MNA score was lower in patients with is a multifactorial problem, the intervention should preferably
multiple burden of somatic, psychological, functional, or social target not only the nutritional status, but also the underlying
characteristics provides further evidence that malnutrition could problems in the somatic, social, functional, or psychological
be regarded as a geriatric syndrome, next to already established domain. These studies are, so far, lacking for malnourished older
geriatric syndromes such as falls, incontinence, pressure sores, and patients.
delirium.Ageriatricsyndromereferstoonesymptomoracomplex One of the strengths of this study was the availability of a
of symptoms with high prevalence in geriatric populations, complete and extensive dataset, including data on nutritional sta-
resulting from multiple diseases and multiple risk factors and tus. This data was prospectively collected by one clinical geriatri-
18
leading to decreased functioning. From literature study it has cian, so data collection was performed consistently. Another
been shown that the chance of having a geriatric syndrome is strength of this study was the use of validated questionnaires like
higher when more risk factors e especially: older age, cognitive MNA, GDS-15, and MMSE. ADL and IADL were, at the time of the
impairment, functional impairment, and impaired mobility e are study, assessed by interview, not using a formal instrument. This
19,20 mustbeconsideredasaweakness.Presently,alsoADLandIADLare
present. This etiological principle has recently also been shown
for malnutrition. In Saka’s study, involving 413 geriatric inpatients assessed using validated questionnaires.
and outpatients, nutritional status of patients was found to corre- Ourinterpretationoftheresultsmaybelimitedbyafewfactors.
late with the number of established geriatric syndromes. The Thefirst limitation of our study is its cross-sectional design, which
higherthenumberofgeriatricsyndromes,thehigherthechanceof limits conclusions regarding within-person change or direction of
11
being malnourished. Only recently, a consensus group of Dutch causality. Second,ourdatawasderivedfromaclinicaldatabase,not
geriatricians has defined that they would like to approach malnu- specifically designed with the purpose to investigate the preva-
trition as a geriatric syndrome as well, having multiple underlying lence and risk factors of malnutrition. Data collection was depen-
causes and needing a multifactorial approach.21 dentonthereportingofindividualpatientsoftheirmedicalhistory
Depressive symptoms and poor functional status were identi- and medication use. This could have led to both under and over
fied as independent determinants for malnutrition. Lack of appe- reporting of comorbidities and drug use. If we assume that
tite, loss of interest in self-care, apathy and physical weakness can misclassification was non-differential, this might have led to an
no reviews yet
Please Login to review.
