/clinical/,/clinical/cckm-tools/,/clinical/cckm-tools/content/,/clinical/cckm-tools/content/questionnaires/,/clinical/cckm-tools/content/questionnaires/related/,

/clinical/cckm-tools/content/questionnaires/related/name-97247-en.cckm

201606168

page

100

UWHC,UWMF,

Clinical Hub,UW Health Clinical Tool Search,UW Health Clinical Tool Search,Questionnaires,Related

Development of a Valid and Reliable Malnutrition Screening Tool for Adult Acute Hospital Patients

Development of a Valid and Reliable Malnutrition Screening Tool for Adult Acute Hospital Patients - Clinical Hub, UW Health Clinical Tool Search, UW Health Clinical Tool Search, Questionnaires, Related


APPLIED NUTRITIONAL INVESTIGATION
Nutrition Vol. 15, No. 6, 1999
Development of a Valid and Reliable
Malnutrition Screening Tool for Adult Acute
Hospital Patients
MAREE FERGUSON, PHD,*† SANDRA CAPRA, PHD,* JUDY BAUER, M HLTH SC,† AND
MERRILYN BANKS, M HLTH SC‡
From the *Centre for Public Health Research, Queensland University of Technology, Brisbane, †Nutrition
Services Department, The Wesley Hospital, Brisbane, and ‡Nutrition and Dietetics Department,
The Redcliffe Hospital, Redcliffe, Australia
Date accepted: 30 August 1998
ABSTRACT
Nutrition screening identifies individuals who are malnourished or at risk of becoming malnourished and who may
benefit from nutrition support. The aim of this study was to develop a simple, reliable and valid malnutrition screening tool
that could be used at hospital admission to identify adult acute patients at risk of malnutrition. The sample population
included 408 patients admitted to an Australian hospital, excluding pediatric, maternity, and psychiatric patients. The
ability of various nutrition screening questions to predict subjective global assessment (SGA) were examined in
contingency tables. The combination of nutrition screening questions with the highest sensitivity and specificity at
predicting SGA was termed the malnutrition screening tool (MST), and consisted of two questions regarding appetite and
recent unintentional weight loss. Subjects who were at risk of malnutrition according to the MST had significantly lower
mean values for the objective nutrition parameters (except immunologic parameters) and longer length of stays than
subjects who were not at risk of malnutrition. Therefore convergent and predictive validity of the MST was established.
The interrater reliability of the malnutrition screening tool was high (93–97%). The MST is a simple, quick, valid, and
reliable tool which can be used to identify patients at risk of malnutrition. Nutrition 1999;15:458–464. Elsevier Science
Inc. 1999
Key words: malnutrition, nutrition screening, malnutrition screening tool, nutrition assessment, subjective global assessment,
nutritional status
INTRODUCTION
“Nutrition screening is the process of identifying parameters
known to be associated with nutritional problems.”
1
The pur-
pose of nutrition screening is to identify individuals who are
malnourished or at risk of becoming malnourished and who
may benefit from nutrition support.
1–4
In the absence of nutri-
tion screening, patients with malnutrition may remain
unrecognized.
Nutrition assessment methods such as subjective global assess-
ment are too detailed and time consuming to complete on all
patients admitted to hospital, therefore nutrition screening is a
feasible alternative for identifying patients at risk of malnutrition.
Numerous nutrition screening tools have been developed to iden-
tify hospital patients at risk of malnutrition.
2,4-16
Many of the
published nutrition screening tools have one or more of the fol-
lowing limitations: 1) their effectiveness, in terms of sensitivity,
specificity, validity, reliability, and cost effectiveness have not
been well established; 2) the screening parameters included have
been based on clinical judgment and intuition; 3) they were
developed in specific patient populations; 4) no practical informa-
tion on how to implement them has been provided; 5) they may be
complicated, time intensive, and invasive; 6) dietitians are often
required to collect the data, which are therefore too specialized for
implementation on a hospital-wide basis by nursing or adminis-
tration staff; and 7) they use nutrition parameters that are not
routinely or immediately available.
4,16
Correspondence to: Maree Ferguson, PhD, Nutrition Services Department, The Wesley Hospital, 499, Toowong QLD 4066, Australia. E-mail:
mferguso@wesley.com.au
Nutrition 15:458–464, 1999
Elsevier Science Inc. 1999 0899-9007/99/$20.00
Printed in the USA. All rights reserved. PII S0899-9007(99)00084-2

Few published nutrition screening tools have demonstrated
their reliability and validity. Only three studies could be identified
that have established the reliability and validity of their nutrition
screening tools.
13,15,16
However, these tools were complicated,
time consuming, and required a health professional to perform
calculations (for example to determine body mass index or per-
centage weight loss) and identify stress factors or diagnoses.
Recent studies have moved away from complex nutrition
screening tools towards those containing two or three simple
questions.
17–19
These studies demonstrated that the simpler tools
were as accurate at detecting nutritional risk, while reducing the
time and cost involved. However, these tools were not assessed for
reliability and validity.
As no quick, simple, reliable, and valid nutrition screening tool
could be located in the published literature, the objective of this
study was to develop a simple, reliable, and valid malnutrition
screening tool that could be used to identify acutely ill adult
patients at risk of malnutrition upon admission to hospital.
MATERIALS AND METHODS
Selection/Development of Nutrition Screening Questions
Criteria for development of the malnutrition screening tool
included that it 1) be applicable for use in a heterogeneous adult
patient population; 2) use routinely available data; 3) be conve-
nient to use, therefore simple, quick, and easily completed by
non-professional staff, patient, or family; 4) be non-invasive and
inexpensive; and 5) be valid and reproducible.
1,13
Therefore, an-
thropometric and biochemical data were not considered. Parame-
ters requiring calculations, such as body mass index and percent-
age weight loss, were also not considered.
Nutrition screening questions were selected and/or developed
from the literature
1–22
and clinical experience. Refer to Table I for
these nutrition screening questions.
Sample Population
All patients admitted to The Wesley Hospital in Brisbane,
Queensland, Australia, during a 3-mo study period were eligible
for inclusion in the study, with the exception of the following:
persons under the age of 18 y, psychiatric and maternity patients,
and patients with whom it was impossible to communicate. Ma-
ternity patients were excluded as these patients were found to be
well nourished in a pilot study. Psychiatric patients were excluded
at the hospital’s request. Pediatric patients were excluded because
the nutrition screening tool was being developed for adult patients.
A convenience sample of 408 subjects fulfilled the selection
criteria and participated in the study. This represented 10.3% of
the eligible hospital patient population during the 3-mo period. Of
these subjects, 201 (49.3%) were male and 207 (50.7%) were
female. The average age was 57.7 6 16.5 (19–94) y and the
average length of stay was 6.0 6 9.3 (0–77) d. The ward distri-
bution of subjects is shown in Table II.
TABLE I.
NUTRITION SCREENING QUESTIONS USED IN THE STUDY
Nutrition screening question Score
Have you lost weight recently without trying?
12
yes 5 1, no 5 0, unsure 5 1
If yes, how much weight have you lost (kilograms) and over what time period have you lost
the weight (weeks)?
n/a
n/a
Is your current appetite: poor 5 2, fair 5 1, good 5 0, unsure 5 1
What is your appetite/food intake like usually? poor 5 2, fair 5 1, good 5 0, unsure 5 1
Has your appetite/food intake been less than usual lately? yes 5 1, no 5 0, unsure 5 1
Have you been eating poorly because of a decreased appetite?
12
yes 5 1, no 5 0, unsure 5 1
Do you have an illness or condition that has made you change the kind and/or amount of food
you eat?
20–22
yes 5 1, no 5 0, unsure 5 1
Do you have tooth, mouth, or swallowing problems that make it hard for you to eat?
20–22
yes 5 1, no 5 0, unsure 5 1
Have you had nausea, vomiting, or diarrhea for the past 3 days or longer?
12
yes 5 1, no 5 0, unsure 5 1
Do you regularly skip meals? yes 5 1, no 5 0, unsure 5 1
Do you eat alone most of the time?
20,22
yes 5 1, no 5 0, unsure 5 1
Are you always physically able to shop, cook, and/or feed yourself?
20,22
yes 5 0, no 5 1, unsure 5 1
Do you wear dentures? yes 5 1, no 5 0, unsure 5 1
Do you have any allergies or intolerances for food? yes 5 1, no 5 0, unsure 5 1
Are you on any special diet/s? yes 5 1, no 5 0, unsure 5 1
How many medications prescribed by your doctor or bought over the counter are you taking?
20,22
0–2 5 0, $3 5 1
Have you been in a hospital overnight or longer in the past 12 months?
12
yes 5 1, no 5 0, unsure 5 1
If yes, how many different times did you stay in a hospital overnight or longer in the past 12
months?
21
0–2 times 5 0, $3 times 5 1
Have you had surgery in the past 6 months?
22
yes 5 1, no 5 0, unsure 5 1
Have you had an illness that kept you in bed during the past month?
21
yes 5 1, no 5 0, unsure 5 1
In general, would you say your health is:
12
poor 5 5, fair 5 4, good 5 3, very good
5 2, excellent 5 1, unsure 5 3
Compared to 1 year ago, how would you rate your health in general now?
21
much worse 5 5, somewhat worse 5 4,
about the same 5 3, somewhat better 5 2,
much better 5 1, unsure 5 3
n/a, not applicable.
DEVELOPMENT OF A MALNUTRITION SCREENING TOOL 459

Development of the Malnutrition Screening Tool
Subjective global assessment (SGA) was chosen as the “gold
standard” for defining malnutrition. The nutrition screening ques-
tions that had the highest sensitivity and specificity at predicting
nutritional status according to SGA were incorporated into the
final malnutrition screening tool.
SGA was performed as described by Detsky and colleagues.
23
SGA is a valid and reliable tool which assesses nutritional status
based on the features of a medical history (weight change, dietary
intake change, gastrointestinal symptoms that have persisted for
greater than 2 wk, changes in functional capacity) and physical
examination (loss of subcutaneous fat, muscle wasting, ankle/
sacral edema, and ascites). Features were combined subjectively
into an overall or global assessment, where subjects were rated as
being well nourished (SGA A); moderately, or suspected of being,
malnourished (SGA B); or severely malnourished (SGA C).
23
Subjects rated as SGA B or C were considered to be malnourished.
Responses to the nutrition screening questions listed in Table I
were collected and SGA was conducted (by M. F.) within2dof
admission. Responses to each of the nutrition screening questions
were given a score as indicated in Table I.
Statistical analysis was carried out using the Statgraphics Plus
version 2.1 package 1996 (Manugistics, Rockville, MD, US). A
level of significance of P5 0.05 was used. The nutrition screening
questions were tested individually against SGA for significance
using the chi-square test. Contingency tables were used to deter-
mine the sensitivity and specificity of both individual and combi-
nations of nutrition screening questions at predicting SGA. The
combination of questions and cutoff value that resulted in the
highest sensitivity and specificity at predicting SGA was termed
the malnutrition screening tool (MST). Subjects below the optimal
cutoff value for sensitivity and specificity were identified as being
not at risk of malnutrition and subjects above the cutoff value were
rated as being at risk of malnutrition.
Validity of the Malnutrition Screening Tool
Convergent validity was established by comparing the MST to
anthropometric (body mass index, midarm circumference, tricep
skinfold thickness, midarm muscle circumference, midarm muscle
area, hand grip strength) and biochemical (total protein, albumin,
prealbumin, hemoglobin, hematocrit, total lymphocyte count,
white cell count, C-reactive protein) parameters using analysis of
variance. Predictive validity was determined by comparing the
MST tool to length of stay using analysis of variance. The MST
scores were dichotomized as not at risk of malnutrition and at risk
of malnutrition for these statistical analyses.
Weight (kilograms) and height (meters) were self reported by
the subject. Body mass index was calculated from self-reported
weight (kilograms) divided by height (meters) squared. Midarm
circumference (MAC) and tricep skinfold thickness (TSF) were
measured according to standard techniques.
24
The average of three
measurements taken on the right side of the body, where possible,
was recorded. The left arm was used when it was impractical to
use the right arm, for instance in the case of injury. Midarm
muscle circumference (MAMC) was calculated using the formula
MAMC 5 MAC 2 p 3 TSF.
25
Midarm muscle area (MAMA)
was calculated using the formula MAMA 5 (MAC 2 p[TSF])
2
/
4p.
25
A Takei Kiki Kogyo Dynamometer (Japan) was used to
measure grip strength according to standard methods.
26
The mean
of three measurements taken using the dominant hand was
recorded.
Anthropometric parameters were obtained at the same time
SGA was performed. Results for biochemical parameters within
3 d of admission were noted if available from the medical record
in a retrospective medical record audit. Information on age, gen-
der, ward, and length of stay was also obtained retrospectively
from the medical record.
Reliability of the Malnutrition Screening Tool
Thirty-two additional subjects were selected using the selec-
tion criteria previously outlined. These subjects had the MST
completed by at least two dietitians and/or a nutrition assistant,
independently of one another, on the same day. The kappa statistic
was used to determine interrater reliability.
The study was approved by The Wesley Hospital Executive
and Ethics Committee, and the Queensland University of Tech-
nology Ethics Committee. Informed written consent was obtained
from each participant.
RESULTS
Development of the Malnutrition Screening Tool
The nutrition screening questions were tested individually
against SGA for possible associations using the chi-square test
(Table III). Screening questions that had a sensitivity or specificity
greater than 90% were identified (Table III). Several combinations
of these screening questions were compared with SGA in contin-
gency tables. The combination that resulted in the highest sensi-
tivity and specificity included the questions “Have you been eating
poorly because of a decreased appetite?” and “Have you lost
weight recently without trying?”
The addition of the questions “How much weight have you
lost?” and “Over what time period have you lost the weight?”
(Table I) was also examined. The question regarding the time
period of weight loss resulted in no improvement in the sensitivity
and specificity of the malnutrition screening tool. Different scor-
ing systems for the amount of weight loss were compared with
SGA in contingency tables. The best scoring system for the
amount of weight loss in terms of sensitivity and specificity was
that outlined in Table IV.
No subjects answered “unsure” to the appetite question but
some subjects answered “unsure” to the weight loss question.
Therefore different values, from 1 to 4, corresponding to the amount
of weight loss scoring system (Table IV), were assigned to the unsure
variable. A final value of 2 was assigned to “unsure,” as this had the
highest sensitivity and specificity at predicting SGA.
The particular combination and scoring system of nutrition
screening questions that resulted in the highest sensitivity and
specificity was termed the malnutrition screening tool (MST), and
is shown in Table IV.
TABLE II.
WARD DISTRIBUTION OF SUBJECTS IN THE SAMPLE
POPULATION
Ward
Number (percent)
of subjects
General medical/respiratory 38 (9.3)
Cardiac 65 (15.9)
Gynecology 57 (14.0)
Orthopedic 20 (4.9)
Oncology 39 (9.6)
Plastic surgery 84 (20.6)
Gastrointestinal 51 (12.5)
General surgical/urology 54 (13.2)
Total 408 (100.0)
DEVELOPMENT OF A MALNUTRITION SCREENING TOOL460

The ability of the MST to predict SGA is shown in Table V.
Subjects obtained a MST score between 0 and 5. Different cutoff
values were selected to determine if subjects were at risk of
malnutrition or not at risk of malnutrition. The effect of selecting
a particular MST score as a cutoff value on the sensitivity and
specificity of the MST was examined in Table VI. The cutoff
value with the highest sensitivity and specificity was 2. Subjects
with a score of 2 or more were subsequently classified as at risk of
malnutrition and subjects with a score of 0 or 1 were classified as
not at risk of malnutrition.
A total of 64 subjects were correctly classified as being mal-
nourished (true positives) and 315 subjects were correctly classi-
fied as being well nourished (true negatives). A total of 24 subjects
(5.9% of 408 subjects) were misclassified as being malnourished
(false positives) and 5 subjects (1.2%) were misclassified as being
well nourished (false negatives). Therefore the misclassification
rate was 7.1%. Both the sensitivity and the specificity were 93%.
The positive predictive value or the proportion of subjects who
were at risk of malnutrition and were malnourished was 98.4%,
and the negative predictive value or the proportion of subjects who
were not at risk of malnutrition and were well nourished was
72.7%.
Validity of the Malnutrition Screening Tool
The MST was compared with objective nutrition parameters
and outcome variables using analysis of variance (Table VII).
There was a significant difference in the mean values of the
objective nutrition parameters, except total lymphocyte count and
white cell count, between subjects who were at risk of malnutri-
tion and subjects who were not at risk of malnutrition. Subjects
who were at risk of malnutrition had significantly worse values for
the objective nutrition parameters compared with subjects who
were not at risk of malnutrition. The length of stay of subjects who
were at risk of malnutrition was significantly longer than subjects
who were not at risk of malnutrition.
Reliability of the Malnutrition Screening Tool
Agreement on the MST rating by the two dietitians occurred in
96% (22/23) of cases (kappa5 0.88, P, 0.01). Agreement on the
MST rating by the nutrition assistant with one of the dietitians
occurred in 93% (27/29) of cases (kappa 5 0.84, P , 0.01), and
TABLE III.
CHI-SQUARE, SENSITIVITY AND SPECIFICITY OF NUTRITION SCREENING QUESTIONS AT PREDICTING SUBJECTIVE GLOBAL
ASSESSMENT (SGA) OF NUTRITIONAL STATUS
Variable N Chi-square P value Sensitivity Specificity
Do you have any allergies or intolerances for food? 404 0.06 0.80 16.7 84.6
Do you have tooth, mouth, or swallowing problems that make it hard for
you to eat?
408 4.53 0.06 18.3 90.8
Do you wear dentures? 408 1.45 0.23 41.0 53.0
Are you on any special diets? 408 2.24 0.13 19.7 73.0
What is your appetite/food intake like usually? 408 53.08 ,0.001 48.3 82.2
Has your appetite/food intake been less than usual lately? 408 172.24 ,0.001 83.3 90.5
Have you been eating poorly because of a decreased appetite? 408 207.58 ,0.001 86.6 92.5
Is your current appetite: 408 151.10 ,0.001 59.0 95.2
Do you have an illness or condition that has made you change the kind
and/or amount of food you eat?
408 162.44 ,0.001 78.3 91.4
In general, would you say your health is: 405 92.10 ,0.001 78.0 72.5
Compared to 1 year ago, how would you rate your health in general now? 405 54.59 ,0.001 69.5 67.1
Have you had an illness that kept you in bed during the past month? 405 60.05 ,0.001 52.5 88.7
Have you had nausea, vomiting or diarrhea for the past 3 days or longer? 408 80.24 ,0.001 43.3 95.1
Have you been in a hospital overnight or longer in the past 12 months? 406 26.49 ,0.001 78.0 58.2
If yes, how many different times did you stay in a hospital overnight or
longer in the past 12 months?
405 32.3 ,0.001 46.7 86.9
How many medications prescribed by your doctor or bought over the
counter are you taking?
403 4.07 0.04 21.3 86.3
Are you always physically able to shop, cook, and/or feed yourself? 408 70.14 ,0.001 45.0 93.4
Do you eat alone most of the time? 408 0.84 0.36 30.0 75.6
Do you regularly skip meals? 408 5.69 0.02 30.0 83.0
Have you had surgery in the past 6 months? 405 13.34 ,0.001 35.6 84.4
Have you lost weight recently without trying? 408 195.85 ,0.001 98.4 83.1
TABLE IV.
MALNUTRITION SCREENING TOOL (MST)
Have you lost weight recently without trying?
No 0
Unsure 2
If yes, how much weight (kilograms) have you lost?
1–5 1
6–10 2
11–15 3
.15 4
Unsure 2
Have you been eating poorly because of a decreased appetite?
No 0
Yes 1
Total
Score of 2 or more 5 patient at risk of malnutrition.
DEVELOPMENT OF A MALNUTRITION SCREENING TOOL 461

with the other dietitian in 97% (31/32) of cases (kappa 5 0.93,
P , 0.01).
DISCUSSION
Development of the Malnutrition Screening Tool
The MST is a simple, quick, valid, and reliable tool which can
be used to identify patients at risk of malnutrition. The sensitivity
and specificity of the malnutrition screening tool was 93%. The
sensitivity of the MST is higher than that of previously published
nutrition screening tools,
13,15,16
although the ideal screening tool
would be 100% sensitive and specific. As this is generally not
achievable, the need to correctly classify all patients who are
malnourished (sensitivity) takes precedence over misclassifying
well-nourished patients (specificity).
To ensure high sensitivity, some well-nourished subjects were
identified as being at risk of malnutrition and would have been
unnecessarily referred for nutrition support. A nutrition assess-
ment would quickly determine that these subjects were well nour-
ished and therefore in no need of nutrition support. An improve-
ment in the specificity of the screening tool could have been
achieved by changing the cutoff value for being at risk of malnu-
trition from 2 or more to 3 or more. This would have reduced the
number of well-nourished subjects misclassified as at risk of
malnutrition, hence reducing the false positives. However, chang-
ing the cutoff value would also have increased the number of false
negatives and thereby have reduced the sensitivity.
Some subjects who were at risk of malnutrition were not
identified as such by the MST. An improvement in the sensitivity
of the screening tool could have been achieved by changing the
cutoff value for being at risk of malnutrition from 2 or more to 1
or more. This would have reduced the number of false negatives,
but could have increased the number of false positives. The cutoff
value chosen maximizes both the sensitivity and specificity of the
tool. Changing the scoring system for each question would also
have the same effect of changing the sensitivity and specificity of
the tool. The particular scoring system chosen maximized both the
sensitivity and specificity of the tool.
Validity of the Malnutrition Screening Tool
There was a significant difference in the mean values of the
objective parameters, except immunologic parameters, between
subjects who were at risk of malnutrition and subjects who were
not at risk of malnutrition. The length of stay of subjects who were
at risk of malnutrition was significantly longer than subjects who
were not at risk of malnutrition. Convergent and predictive valid-
ity of the MST has therefore been established.
Reliability of the Malnutrition Screening Tool
The interrater reliability of the malnutrition screening tool was
high, with the level of agreement between raters being between 93
and 97%. This is similar to or higher than the interrater reliability
reported of other published nutrition screening tools.
15,16
A high
interrater reliability would be expected with only two questions
comprising the malnutrition screening tool.
Implementation of the Malnutrition Screening Tool
It is recommended that nutrition screening be performed within
24 h of hospital admission. Patients who are not at risk of mal-
nutrition (MST score 0–1) should be rescreened weekly until they
are discharged to monitor changes in nutritional status. Patients
who are at risk of malnutrition (MST score 2–5) should undergo
a more detailed nutrition assessment (such as SGA) to identify if
they are malnourished and to determine the most appropriate form
of nutrition support. An advantage of using this scoring system is
that within the range of 2–5, prioritization of patients requiring
more urgent treatment can be performed. Hence the MST facili-
TABLE V.
COMPARISON OF THE NUMBER OF SUBJECTS CLASSIFIED AS BEING AT RISK OF MALNUTRITION USING THE MALNUTRITION
SCREENING TOOL (MST) AND THE NUMBER OF SUBJECTS RATED AS MALNOURISHED USING
SUBJECTIVE GLOBAL ASSESSMENT (SGA)
MST SGA
Total number
of subjectsMST score
At risk of
malnutrition
A
(well nourished)
B (moderately malnourished)
and
C (severely malnourished)
0 no 276 1 277
1no39 4 43
2 yes 18 22 40
3 yes 6 31 37
4 yes 0 6 6
5 yes 0 5 5
Total number of subjects 339 69 408
TABLE VI.
SENSITIVITY AND SPECIFICITY OF THE MALNUTRITION
SCREENING TOOL (MST) WHEN DIFFERENT CUTOFF VALUES
WERE USED TO DETERMINE IF SUBJECTS WERE AT RISK OF
MALNUTRITION
MST score
cutoff value
MST score rated
as not at risk of
malnutrition
MST score rated
as at risk of
malnutrition
Sensitivity
(%)
Specificity
(%)
1 0 1–5 99 81
2 0–1 2–5 93 93
3 0–2 3–5 61 98
4 0–3 4–5 16 100
5 0–4 5 7 100
DEVELOPMENT OF A MALNUTRITION SCREENING TOOL462

tates a more effective use of dietetic time and resources by
decreasing inappropriate referrals.
SUMMARY
The MST is a simple, quick, reliable, valid, tool which can be
completed by medical, nursing, dietetic, or administrative staff, as
well as by family, friends, or the patients themselves on admission
to hospital. Its high sensitivity and specificity indicates that it
strongly predicts nutritional status as defined by SGA. Hence, the
MST will consistently identify those patients at risk of malnutri-
tion so that nutrition care can be initiated promptly.
ACKNOWLEDGMENT
This research was supported by a scholarship from Queensland
University of Technology and Spotless Services Limited. The
authors would like to thank the staff and patients of The Wesley
Hospital.
REFERENCES
1. American Dietetic Association. Identifying patients at risk: ADA’s
definitions for nutrition screening and nutrition assessment. J Am Diet
Assoc 1994;94:838
2. Hunt DR, Maslovitz A, Rowlands BJ, Brooks B. A simple nutrition
screening procedure for hospital patients. J Am Diet Assoc 1985;85:
332
3. Pettigrew RA, Hill GL. Indicators of surgical risk and clinical judge-
ment. Br J Surg 1986;73:47
4. Nagel MR. Nutrition screening: identifying patients at risk for mal-
nutrition. Nutr Clin Prac 1993;8:171
5. Potosnak L, Chudnow LP, Simko MD. A simple tool for identifying
patients at nutritional risk. QRB 1983;9:81
6. Thompson JS, Burrough CA, Green JL, Brown GL. Nutritional
screening in surgical patients. J Am Diet Assoc 1984;3:337
7. Christensen KS, Gstundtner KM. Hospitalwide screening improves
basis for nutrition intervention. J Am Diet Assoc 1985;85:704
8. Hannaman KN, Penner SF. A nutrition assessment tool that includes
diagnosis. J Am Diet Assoc 1985;85:607
9. Pinchofsky GD, Kaminski MV. Increasing malnutrition during hos-
pitalisation: documentation by a nutritional screening program. J Am
Coll Nutr 1985;4:471
10. Hedberg AM, Garcia N, Trejus IJ, et al. Nutritional risk screening:
development of a standardised protocol using dietetic technicians.
J Am Diet Assoc 1988;88:1553
11. Grant AD, Dehoog S. Nutritional assessment and support. 4th ed.
Seattle: Grant/DeHoog, 1991:205
12. Department of Nutrition and Dietetics, Thomas Jefferson University
Hospital. Nutrition manual. Gaithersburg: ASPEN, 1993:1
13. Elmore MF, Wagner DR, Knoll DM, et al. Developing an effective
adult nutrition screening tool for a community hospital. J Am Diet
Assoc 1994;94:1113
14. Bauer J. Nutrition screening in an Australian private hospital:
development of a level 1 admission screen. Master of Health
Science thesis. Brisbane, Queensland: Queensland University of
Technology, 1995
15. Kovacevich DS, Boney AR, Braunschweig CL, Perez A, Stevens M.
Nutrition risk classification: a reproducible and valid tool for nurses.
Nutr Clin Prac 1997;12:20
16. Reilly HM, Martineau JK, Moran A, Kennedy H. Nutritional
screening—evaluation and implementation of a simple nutrition
risk score. Clin Nutr 1995;14:269
17. Fraker P, Christy-Given J. Successful nutrition screening protocol.
J Am Diet Assoc 1995;95(suppl):A79
18. Murtaugh M, Sowa D, Dalton C, Ratz N, Lipson S. Validation of a
rapid 2 factor screen for malnutrition against the current Rush-Pres-
byterian-St. Luke’s Medical Center 7 factor nutrition screen. J Am
Diet Assoc 1995;95(suppl):A33
19. Ohm-Pitock P, Kapica C. Improved detection of nutritional risk in a
hospital nutrition screening system. J Am Diet Assoc 1995;95(suppl):
A79
20. Lipschitz DA. Screening for nutritional status in the elderly. Primary
Care 1994;21:55
TABLE VII.
RELATIONSHIP BETWEEN THE MALNUTRITION SCREENING TOOL (MST), OBJECTIVE NUTRITION PARAMETERS, AND
OUTCOME VARIABLES
Nutrition parameter/outcome variable
Number of
subjects
MST score mean 6 SD
Not at risk of malnutrition
(0–1)
At risk of malnutrition
(2–5)
Analysis of
variance
(P value)
Body mass index (kg/m
2
) 408 26.16 4.6 23.66 4.7 ,0.001
Tricep skinfold thickness (mm) 374 18.06 8.4 15.76 8.0 0.03
Midarm circumference (cm) 374 29.86 3.9 27.26 4.0 ,0.001
Midarm muscle circumference (cm) 374 29.26 3.8 26.56 4.2 ,0.001
Midarm muscle area (mm
2
) 374 47876 1444 41046 1332 ,0.001
Grip strength (kg) 396 31.06 11.6 25.36 12.0 ,0.001
Albumin (g/L) 177 39.36 4.7 36.36 5.7 ,0.001
Total protein (g/L) 175 70.56 5.7 67.46 10.0 0.01
Prealbumin (mg/L) 134 265.86 74.1 233.36 101.2 0.046
White cell count (310
9
/L) 177 8.46 4.2 9.96 7.2 0.08
Lymphocyte count (310
9
/L) 177 2.16 2.7 1.76 1.8 0.28
C-reactive protein (mg/L) 136 20.96 39.4 59.86 87.5 ,0.001
Hemoglobin (g/L) 200 1356 22 1256 24 0.003
Hematocrit (g/dL) 195 0.406 0.06 0.376 0.07 ,0.001
length of stay (d) 408 4.96 8.2 9.56 11.6 ,0.001
DEVELOPMENT OF A MALNUTRITION SCREENING TOOL 463

21. Posner BM, Jette AM, Smith KW, Miller DR. Nutrition and health
risks in the elderly: the nutrition screening initiative. Am J Pub Health
1993;83:972
22. Wellman N. The nutrition screening initiative. Nutr Rev 1994;52:44
23. Detsky AS, McLaughlin JR, Baker JP, et al. What is subjective global
assessment of nutritional status? JPEN 1987;11:8
24. Norton K, Olds T. Anthropometrica. Sydney: University of New
South Wales Press, 1996:47
25. Gibson R. Principles of nutrition assessment. New York: Oxford
University Press, 1990:202
26. Tiddy JA, Magarey AM, Wilson PC, Hill PJ. Handgrip dynamometry:
percentiles for people aged 55–95 years. Aust J Nutr Diet 1990;47:113
DEVELOPMENT OF A MALNUTRITION SCREENING TOOL464