REVIEW PAPER

Systematic review and meta-analysis of the diagnostic accuracy of the

water swallow test for screening aspiration in stroke patients
Po-Cheng Chen, Ching-Hui Chuang, Chau-Peng Leong, Su-Er Guo & Yi-Jung Hsin

Accepted for publication 5 April 2016

Correspondence to C.-H. Chuang:
e-mail: chinhui@cgmh.org.tw
Po-Cheng Chen MD
Doctor
Department of Physical Medicine and
Rehabilitation, Chang Gung Memorial
Hospital-Kaohsiung Medical Center, Chang
Gung University College of Medicine,
Kaohsiung, Taiwan
Ching-Hui Chuang MSN RN
Nursing Lecturer
College of Nursing, Chang Gung University
of Science and Technology, Chiayi Campus,
Taiwan and Department of Nursing, Chang
Gung Memorial Hospital-Kaohsiung
Medical Center, Taiwan
Chau-Peng Leong MD
Doctor
Department of Physical Medicine and
Rehabilitation, Chang Gung Memorial
Hospital-Kaohsiung Medical Center, Chang
Gung University College of Medicine,
Kaohsiung, Taiwan
Su-Er Guo PhD RN
Director/Professor
Chronic Diseases and Health Promotion
Research Center, and College of Nursing
and Graduate Institute of Nursing, Chang
Gung University of Science and Technology,
Chiayi, Taiwan
CHEN P.-C., CHUANG C.-H., LEONG C.-P., GUO S.-E. & HSIN Y.-J. (2016)
Systematic review and meta-analysis of the diagnostic accuracy of the water swal-
low test for screening aspiration in stroke patients. Journal of Advanced Nursing
00(0), 000–000. doi: 10.1111/jan.13013
Abstract
Aim. The aim of this study was to determine the diagnostic accuracy of the water
swallow test for screening aspirations in stroke patients.
Background. The water swallow test is a simple bedside screening tool for
aspiration among stroke patients in nursing practice, but results from different
studies have not been combined before.
Design. A systematic review and meta-analysis was conducted to provide a
synthetic and critical appraisal of the included studies.
Data sources. Electronic literature in MEDLINE, EMBASE, CINAHL and other
sources were searched systemically in this study. Databases and registers were
searched from inception up to 30 April 2015.
Review methods. This systematic review was conducted using the
recommendations from Cochrane Collaboration for Systematic Reviews of
Diagnostic Test Accuracy. Bivariate random-effects models were used to estimate
the diagnostic accuracy across those studies. The tool named Quality Assessment
of Diagnostic Accuracy Studies (QUADAS-2) was used to evaluate the quality of
the studies.
Results. There were 770 stroke patients in the 11 studies for the meta-analysis.
The water swallow test had sensitivities between 64-79% and specificities
between 61-81%. Meta-regression analysis indicated that increasing water volume
resulted in higher sensitivity but lower specificity of the water swallow test.
Conclusions. This systematic review showed that the water swallow test was a
useful screening tool for aspiration among stroke patients. The test accuracy was
related to the water volume and a 3-oz water swallow test was recommended for
aspiration screening in stroke patients.

Yi-Jung Hsin MD Keywords: aspiration, nursing, stroke, systematic review and meta-analysis, water
Doctor
swallow test
Department of Physical Medicine and
Rehabilitation, Chang Gung Memorial
Hospital-Kaohsiung Medical Center,
Kaohsiung, Taiwan

© 2016 John Wiley & Sons Ltd 1

P.-C. Chen et al.
Why is this review needed?
● Dysphagia is very common in patients with stroke.
● Aspiration is one of the critical consequences of dysphagia.
● Aspiration screening is important to minimize the risk of
aspiration pneumonia in stroke patients.
What are the key findings?
● This systematic review and meta-analysis investigates the
sensitivity and specificity of the water swallow test in
screening aspiration among stroke patients.
● Increasing water volume results in higher sensitivity but
lower specificity of the water swallow test.
● A 3-oz water swallow test is recommended for aspiration
screening in stroke patients.
How should the finding be used to influence policy/
practice/research/education?
● The water swallow test plays an important role in screen-
ing aspiration among stroke patients. Standardized testing
protocols should be developed and used in the future.
Introduction
Stroke is considered to be the primary cause of dysphagia
(Martino et al. 2005). Approximately 22-65% of the
patients with stroke have dysphagia (Jauch et al. 2013).
There are two types of dysphagia: oropharyngeal and oeso-
phageal dysphagia (Singh & Hamdy 2006). Oropharyngeal
dysphagia is defined as disturbance of food or fluid through
mouth and pharynx. Oesophageal dysphagia is defined as
difficulty in moving food or fluid from the oesophagus to
the stomach. Oropharyngeal dysphagia is the most common
type of dysphagia among patients with stroke (Falsetti et al.
2009). When the liquid is given to stroke patients, they can-
not hold the liquid in the oral cavity adequately. Therefore,
the liquid enters the pharynx accidentally and subsequently
induces aspiration. Aspiration (subglottic penetration of
food or liquid) is one of the critical consequences of dys-
phagia, giving rise to an increased risk of aspiration pneu-
monia (Martino et al. 2005). It has been estimated that
40-50% of stroke patients with dysphagia encountered
aspiration problems (Marik & Kaplan 2003).
One third of stroke patients demonstrating signs and
symptoms of aspiration will suffer from pneumonia (Dono-
van et al. 2013). Aspiration pneumonia leads to prolonged
hospital stay and high mortality (Brogan et al. 2014).
2 © 2016 John Wiley & Sons Ltd
Because food straying into the respiratory tracts can result
in aspiration pneumonia, proper evaluation of aspiration is
important in stroke patients. Silent aspiration occurs when
a patient aspirates particles without having a protective
cough reflex and about 40-70% of stroke patients suffer
from silent aspiration (Matsuo & Palmer 2008). Therefore,
we cannot merely judge whether a stroke patient has aspi-
ration by the existence of cough reflex. In fact, the videoflu-
oroscopy (VFS) and the fiberoptic endoscopic evaluations of
swallowing (FEES) are the gold standards to identify aspira-
tion in stroke patients, but inconvenience (special equip-
ment is required) and danger to some stroke patients
(allergy to contrast media in the VFS or mucosal perfora-
tion in the FEES) are primary concerns of these studies
(Rugiu 2007). There are many bedside screening tools for
aspiration in stroke patients and the water swallow test
(WST) is easy and safe in nursing practice. However, the
diagnostic accuracy of the WST for screening aspiration has
not been summarized in previous studies. From this point
of view, a systematic review and meta-analysis of the diag-
nostic accuracy of the WST for screening aspiration in
stroke patients is indicated.
Background
Aspiration screening, a quick and minimally invasive proce-
dure, is statistically significant in stroke patients and
enables timely nutritional nursing intervention while refer-
ring to speech-language pathologists for evaluation and
treatment process (Raoufi & Shade 2014). Early awareness
of aspiration allows immediate management, which may
reduce morbidity, length of hospitalization and medical
costs (Odderson & McKenna 1993, Hinchey et al. 2005).
According to the American Heart Association/American
Stroke Association guidelines for early stroke management,
it is the primary step to screen aspiration before the admin-
istration of food, liquid or medication in stroke patients
(Jauch et al. 2013). Although instrumental screenings such
as videofluoroscopy (VFS) or fiberoptic endoscopic evalua-
tion of swallowing (FEES) are the gold standards to identify
aspiration in stroke patients, radiation exposure in VFS or
the necessity of skilled operators and specialized equipment
in FEES still limit their widespread use (Rao et al. 2003,
Bax et al. 2014).
Many screening tools for aspiration in stroke patients,
including water swallow test (WST), Toronto Bedside Swal-
lowing Screening test (TORBSST), Barnes Jewish Hospital
Stroke Dysphagia Screen (BJH-SDS), Modified Mann Assess-
ment of Swallowing Ability (Modified MASA) and Emer-
gency Physician Swallowing Screening (Antonios et al. 2010,
JAN: REVIEW PAPER
Schepp et al. 2012, Edmiaston et al. 2014), are with good
reliability. Among these assessment tools, the WST may be
the most convenient screening tool to evaluate aspiration
among stroke patients in nursing practice (Daniels et al.
2012). In this systematic review and meta-analysis, we sum-
marized the published data by assessing the diagnostic accu-
racy of the WST for screening aspiration in stroke patients.
The review
Aim
The aim of this review was to determine the diagnostic
accuracy of the WST for screening aspiration in stroke
patients.
Definitions
The WST is used as an index test in our review. Although it is
a test commonly used all over the world, the water volume
given in the test varies depending on the examiners (Osawa
et al. 2013). Many WSTs start with teaspoons of water and
proceed to 30 ml, 50 ml or even 3 oz (approximately 90 ml)
of water. In these cases, we define the water volume in one
WST as the largest volume used in the WST (for example, if
30 ml of water is given as the largest volume in a step-by-step
WST, it is defined as a 30-ml WST). The endpoint of the
WST is defined as inability to complete the test, cough, chok-
ing or voice change after water swallowing.
The VFS and FEES are widely recognized as the gold
standards for evaluation of aspiration in stroke patients
(Rao et al. 2003, Bax et al. 2014). Aspiration is defined as
the entry of food or fluid below the level of the vocal cords,
while penetration is defined as the entry of food or fluid in
the laryngeal vestibule but remaining above the levels of the
vocal cords (Murray 1999). As penetration usually leads to
high risk of aspiration, we define the endpoint of the VFS
or FEES as aspiration or penetration. The VFS and FEES
are the reference standards with which the WST is com-
pared.
Design
This was a systematic review and meta-analysis including
diagnostic accuracy tests of the WST for screening aspira-
tion in stroke patients. We conformed to the standard
reporting guidelines and recommendations from Cochrane
Collaboration for Systematic Reviews of Diagnostic Test
Accuracy throughout the duration of the investigation
(Deeks et al. 2013). The structure of the search strategies
was based on the key concepts: the screening test(s) of
© 2016 John Wiley & Sons Ltd
Water swallow test for screening aspiration in stroke patients
interest (index test), the clinical condition of interest (target
condition) and a defined study population (patient descrip-
tion) (de Vet et al. 2008).
Search methods
The goal of this systematic review and meta-analysis was to
gather evidence describing the diagnostic accuracy of the
WST for screening aspiration in stroke patients, so any
quantitative studies with reported sensitivities and specifici-
ties were sought. These diagnostic accuracy studies were
principally cross-sectional studies. There were no limita-
tions on study designs (e.g. prospective or retrospective
study), publication dates or languages. The inclusion crite-
ria for this systematic review were as follows:
1 Stroke patients were diagnosed clinically or by the medi-
cal images without limitations on ages, countries, stroke
phases or latencies from stroke onset to study day.
2 Aspiration was evaluated using index tests (the WST) in
comparison with using reference standards (the VFS or
the FEES).
3 Sufficient information for diagnostic accuracy measures
were provided in eligible studies.
Those studies with non-stroke patients, no reporting of
reference standards, insufficient information for diagnostic
accuracy measures or non-diagnostic test accuracy studies
(such as prognostic or predictive accuracy studies) were
excluded.
Electronic databases of MEDLINE, EMBASE and
CINAHL were searched. The MEDLINE searching strategy
was created by two reviewers and it was adapted to syntax
and subject headings for other databases. In addition, the
Current Controlled Trials (ISRCTN Register) and Clini-
calTrials.gov were searched for ongoing or recently com-
pleted trials and PROSPERO was searched for ongoing or
recently completed systematic reviews. As relevant studies
were identified, we also checked cited articles from these
relevant studies by hand searching. These search strategies
were presented in Appendix S1. Databases and registers
were searched from inception up to 30 April 2015.
Search outcome
Two reviewers independently reviewed the titles or
abstracts of the searching results matching the inclusion cri-
teria. We included studies evaluating the diagnostic accu-
racy of the WST for screening aspiration in stroke patients
and provided enough information for calculation of sensi-
tivity and specificity.
3
P.-C. Chen et al.
Quality appraisal
Two reviewers independently appraised the methodological
quality of the included studies using the Quality Assessment
of Diagnostic Accuracy Studies-2 (QUADAS-2) tool (Whit-
ing et al. 2011). QUADAS-2 consisted of four domains:
patient selection, index test, reference standard and flow
and timing. Each domain was assessed in terms of risk of
bias and the first 3 domains were also assessed in terms of
concerns regarding applicability (Mowatt et al. 2013). The
answer to each item or domain in QUADAS-2 was ‘low
risk’, ‘high risk’ or ‘unclear’. Any differences between items
were solved through discussion.
Data extraction
Data extraction was performed independently by two
reviewers. Discrepancies were resolved through discussion.
If disagreements were not resolved by consensus, a third
reviewer was consulted. We extracted data using a form
including the author, publication year, characteristics of
patients, characteristics of studies, true positive (TP) results
(diseased cases correctly diagnosed as diseased), true nega-
tive (TN) results (non-diseased cases correctly identified as
non-diseased), false-positive (FP) results (non-diseased cases
incorrectly identified as diseased), false negative (FN) results
(diseased cases incorrectly identified as non-diseased), sensi-
tivity (TP/(TP+FN)), specificity (TN/(TN+FP)), positive
likelihood ratio (LR+, sensitivity/(1-specificity)), negative
likelihood ratio (LR-, (1-sensitivity)/specificity) and diagnos-
tic odds ratio (DOR, LR+/LR-).
To ensure that the results of this study were not affected
by publication biases, we plotted a funnel plot. A more
appropriate method for detecting funnel plot asymmetry in
reviews of diagnostic studies has been developed (Deeks
et al. 2005). It tested for an association between the lnDOR
and the ‘effective sample size’, a simple function of the
number of diseased and non-diseased individuals (Macaskill
et al. 2010).
Synthesis
We used the bivariate random-effects model for analysing
and pooling the diagnostic accuracy measures across stud-
ies (Reitsma et al. 2005, Hamza et al. 2008). The bivari-
ate model estimated paired logic transformed sensitivity
and specificity of the studies, incorporating the correlation
that might exist between sensitivity and specificity.
Investigating the possible causes of heterogeneity between
studies was an important part of conducting a meta-
4 © 2016 John Wiley & Sons Ltd
analysis (Petitti 2001). We explored heterogeneity using uni-
variate meta-regression (Baker et al. 2009). The following
factors were specified a priori as potential sources of hetero-
geneity: study location, publication year, delay between
tests, reference test, methodological quality, age, male per-
centage, water volume and aspiration prevalence. The statis-
tical significance level was set at 5%. Statistical analyses
were performed with a commercial software program
(STATA, version 13.1; StataCorp LP:Taipei, Taiwan).
Results
Characteristics of the included studies
The study flow diagram in Figure 1 showed the literature
search process. We identified 1374 articles from electronic
database and other sources, of which 11 met the inclu-
sion criteria (DePippo et al. 1992, Kidd et al. 1993,
Daniels et al. 1997, Smithard et al. 1998, Lim et al.
2001, Mann & Hankey 2001, Chong et al. 2003, Nishi-
waki et al. 2005, Zhou et al. 2011, Osawa et al. 2013,
Somasundaram et al. 2014). There were 770 stroke
patients in the 11 studies that were eventually available
for this meta-analysis.
Table 1 (DePippo et al. 1992, Kidd et al. 1993, Daniels
et al. 1997, Smithard et al. 1998, Lim et al. 2001, Mann
& Hankey 2001, Chong et al. 2003, Nishiwaki et al. 2005,
Zhou et al. 2011, Osawa et al. 2013, Somasundaram et al.
2014) summed up the clinical characteristics of the included
studies. The studies originated from three continents (Asia,
Europe and America) and six countries (Japan, Singapore,
France, Germany, UK and USA). The included patients
were mostly from medical centres and over half of these
patients were old males. The stroke phase was not
restricted to acute, sub-acute or chronic stage. The WST
was performed using various water volumes, ranging from
3-90 ml). The FEES was used as reference tests in three
studies (Lim et al. 2001, Chong et al. 2003, Somasundaram
et al. 2014) and the VFS was used as reference tests in eight
studies (DePippo et al. 1992, Kidd et al. 1993, Daniels
et al. 1997, Smithard et al. 1998, Mann & Hankey 2001,
Nishiwaki et al. 2005, Zhou et al. 2011, Osawa et al.
2013).
Diagnostic accuracy, methodological quality and
publication bias
Our meta-analysis contained 11 studies and each measure
of diagnostic accuracy was listed in Table 2. These diagnos-
tic accuracy measures were pooled by the random-effects
JAN: REVIEW PAPER Water swallow test for screening aspiration in stroke patients

Records identified through Additional records identified

Identification
database searching through other sources
(n = 1326) (n = 48)

Records after duplicates removed

Screening (n = 1374)

Records screened Records excluded by titles

(n = 1374) or abstracts
(n = 1313)

Full-text articles excluded,

with reasons
Full-text articles (n = 50)
Eligibility

assessed for (1) 29 not using WST as

eligibility (n = 61) index tests
(2) 17 not using VFS or FEES
as reference standards
(3) 3 review articles
(4) 1 economic study
Included

Studies included in
quantitative synthesis
(meta-analysis)
(n = 11)

Figure 1 Study flow diagram. WST, water swallow test; FEES, fiberoptic endoscopic examination of swallowing; VFS, videofluoroscopy.

model. The pooled sensitivity was 0
72 (95% CI 0
64-
0
79), the pooled specificity was 0
72 (95% CI 0
61-0
81)
and the pooled DOR was 6
78 (95% CI 4
63-9
92) respec-
tively.
Figure 2 summarized the QUADAS-2 assessment of these
11 eligible studies. Patient selection augmented the risk of
bias and the applicability was concerned in five studies
because of non-consecutive samples or patients with specific
types of stroke (DePippo et al. 1992, Daniels et al. 1997,
Lim et al. 2001, Osawa et al. 2013, Somasundaram et al.
2014). The index test domain in one study (DePippo et al.
1992) was labelled as unclear, because we did not know
whether the index test results interpreted without the
knowledge of the results from the reference standard. The
reference standard domain in four studies (DePippo et al.
1992, Nishiwaki et al. 2005, Osawa et al. 2013, Somasun-
daram et al. 2014) was labelled as unclear since the results
of the reference standard interpreted without the knowledge
of the results from the index test was unknown. The flow
and timing domain in two studies (DePippo et al. 1992,
Chong et al. 2003) was labelled as unclear because of
unknown interval between the index test and the reference
standard and it was labelled as high in two studies (Mann
& Hankey 2001, Osawa et al. 2013) because inappropriate
interval between the index test and the reference standard,
or not all patients were included in the analysis.
The funnel plot (Figure 3) and regression test indicated
no statistically significant publication bias (P = 0
30).
Exploration of heterogeneity
Between-study heterogeneity of sensitivity and specificity
were presented among studies. We used univariate meta-
regression to explore the heterogeneity. Table 3 listed the
results of univariate meta-regression analysis for identifying
potential sources of heterogeneity. Increasing water volume
resulted in higher sensitivity (P = 0
002) but lower speci-
ficity (P = 0
025) of the WST for screening aspiration in
stroke patients. The study location, publication year, delay
between tests, reference test, methodological quality, age,
male percentage or aspiration prevalence did not cause any
heterogeneity.

© 2016 John Wiley & Sons Ltd 5

 6
Table 1 Characteristics of the studies included in the systematic review.
P.-C. Chen et al.

Mean Sample Reference test,

age, size, Index test/practice by; endpoint of Delay between
Reference Nation Setting Population years Men, % no. endpoint of index test reference test tests

Somasundaram Germany Medical centre Acute stroke 68 67 67 WST/SLP, 50 ml; cough or voice FEES, penetration Average 4
4 hours
et al. (2014) patients change or aspiration
Osawa et al. (2013) Japan Medical centre Acute stroke 70
2 64 50 MWST/physician, 3 ml; inability to VFS, aspiration Approximately the
patients complete the test, dyspnoea, cough same time
or dysphonia
Zhou et al. (2011) France Medical centre Acute stroke 67
8 69
2 107 WST/physician, 3 oz; inability to VFS, aspiration Within 48 h
patients complete the test, cough, choking
or voice change
Nishiwaki et al. (2005) Japan Medical centre Stroke 70
4 65
6 61 WST/physician or SLP, 30 ml; cough VFS, aspiration Within 7 days
and non- patients or voice change
medical centre
Chong et al. (2003) Singapore Medical centre Recent and 75
5 52 50 WST/physician, 50 ml; cough, choking FEES, aspiration Unclear
previous or voice change or penetration
stroke
patients
Lim et al. (2001) Singapore Medical centre Acute stroke 67
5 62 50 WST/physician, 50 ml; cough, choking FEES, aspiration Within 48 h
patients or voice change or penetration
Mann et al. (2001) USA Medical centre First stroke 71 64
1 128 WST/SLP, 10 ml; cough or voice VFS, penetration About 7 days
patients change or aspiration
Smithard et al. (1998) UK Medical centre Acute stroke 79 50 94 WST/physician, 60 ml; cough, choking VFS, aspiration Within 3 days
and non- patients or voice change
medical centre
Daniels et al. (1997) USA Medical centre Non- 66 100 59 WST/unclear, 70 ml; cough or voice VFS, aspiration Within 48 h
haemorrhagic change or penetration
stroke patients
Kidd et al. (1993) UK Medical centre Acute stroke 72 41
7 60 WST/physician, 50 ml; Cough, choking VFS, aspiration Within 72 h
patients or voice change
DePippo et al. (1992) USA Medical centre Stroke patients 71 Unclear* 44 WST/unclear, 3 oz, cough or voice VFS, aspiration Unclear
change

*Suppose that male percentage was 50% for further meta-regression analysis.
h, hours; y, year; no., number; 3 oz, 3 ounce; WST, water swallow test; MWST, modified water swallowing test; VFS, videofluoroscopy; FEES, fibreoptic endoscopic evaluation of
swallowing; SLP, speech-language pathologist.

© 2016 John Wiley & Sons Ltd

JAN: REVIEW PAPER Water swallow test for screening aspiration in stroke patients

[4
63, 9
92]

No., number; CI, confidence interval; TP, true positive; FP, false positive; FN, false negative; TN, true negative; LR+, positive likelihood ratio; LR-, negative likelihood ratio; DOR,
Discussion

6
56
2
54
5
03
5
39
6
43
16
50
18
09
4
57
7
02
24
00
4
73
6
78
DOR Poststroke screening for aspiration is an important part in
the investigation of swallowing function (Jauch et al. 2013)
and the WST is one of the most convenient screening tools
in nursing practice. Although there are some systematic

[0
31, 0
48]
reviews about bedside screening tools for aspiration in
stroke patients (Bours et al. 2009, Daniels et al. 2012,
0
48
0
62
0
30
0
41
0
33
0
21
0
57
0
45
0
44
0
23
0
37
0
39
LR

O’Horo et al. 2015), none of these articles used quantita-

tive data analysis to synthesize the diagnostic accuracy of
the WST. To the best of our knowledge, this study was the

[1
93, 3
53]
first systematic review and meta-analysis to summarize the
diagnostic accuracy measures of the WST for screening
3
17
1
57
1
51
2
22
2
12
3
38
10
38
2
07
3
08
5
60
1
75
2
61
LR+

aspiration in stroke patients and discuss the factors result-

ing in between-study heterogeneity of sensitivity and speci-
ficity. This systematic review and meta-analysis included 11
[0
61, 0
81]

studies that addressed the WST in different clinical settings.

Specificity

Only three studies (Kidd et al. 1993, Smithard et al. 1998,

0
81
0
60
0
42
0
67
0
63
0
75
0
96
0
66
0
79
0
86
0
54
0
72

Zhou et al. 2011) satisfactorily met the quality criteria

adapted from the QUADAS-2 tool and most studies had
unclear or high risk of bias and applicability concerns in at
[0
64, 0
79]

least one item of the QUADAS-2 tool. These findings sug-

Sensitivity

gested that future studies of the WST for screening aspira-

0
61
0
63
0
87
0
72
0
79
0
85
0
45
0
70
0
65
0
80
0
80
0
72

tion in stroke patients should be designed according to the

guideline of the QUADAS-2 tool to minimize the impact of
bias. We found that the overall sensitivity and specificity of
Prevalence

the WST for screening aspiration in stroke patients were

0
72 (95% CI 0
64-0
79) and 0
72 (95% CI 0
61-0
81)
0
69
0
62
0
65
0
69
0
74
0
80
0
63
0
67
0
73
0
83
0
66

respectively. The pooled diagnostic odds ratio (DOR) was

6
78 (95% CI 4
63-9
92). These results indicated that the
Table 2 Diagnostic accuracy for screening aspiration in stroke patients.

WST was a useful screening tool for aspiration in stroke

TN

21

22
29
10
18
44
49
26
30
13
9

patients, especially for its convenience and cost-effectiveness

in clinical practice.
Test result, No. of patients

The strength of our study was that the proposed sys-

FN

7
5
7
4

6
9
5
4
16
13

45

tematic review was based on recommended methodologi-

cal approaches (Deeks 2001, Liberati et al. 2009, Whiting
et al. 2011). To explain the heterogeneity, we performed
FP

30
14

25

11
5
6

6
6
2

7
5

an univariate meta-regression. After these analyses, there

was only one source of heterogeneity in our study: water
volume. The WST was frequently used in clinical practice
TP

25
22
48
13
27
22
37
14
17
20
16

to screen aspiration and was initially described as a

30 ml-WST in Kubota’s study (Kubota et al. 1982). How-
Somasundaram et al. (2014)

ever, the amount of water given in the test varied

Pooled estimate [95% CI]

depending on the examiners. Considering the risk of aspi-

Nishiwaki et al. (2005)

Smithard et al. (1998)

DePippo et al. (1992)

diagnostic odds ratio.

ration in stroke patients and the elders while swallowing

Daniels et al. (1997)
Osawa et al. (2013)

Chong et al. (2003)

Mann et al. (2001)

Zhou et al. (2011)

Kidd et al. (1993)

Lim et al. (2001)

large amounts of water, a modified 3 ml-water swallow test

(MWST) was therefore developed (Tohara et al. 2003, Mae-
shima et al. 2011). Nevertheless, if a mouthful volume was
Source

not large enough, the swallow reflex was hard to be induced

(Lazarus et al. 1993). In our meta-analysis, we found that

© 2016 John Wiley & Sons Ltd 7

P.-C. Chen et al.

(a)
Patient Selection
Index Test
Reference Standard
Flow and Timing

0% 25% 50% 75% 100% 0% 25% 50% 75% 100%

Risk of Bias Applicability Concerns
High Unclear Low

(b) Risk of Bias Applicability Concerns

Reference Standard

Reference Standard
Patient Selection

Patient Selection
Flow and Timing
Index Test

Index Test
Chong et al. 2003 + + + ? + + +

Daniels et al. 1997 + + + + – + +

DePippo et al. 1992 ? ? ? ? + + +

Kidd et al. 1993 + + + + + + +

Lim et al. 2001 ? + + + + + +

Mann et al. 2001 + + + – + + +

Nishiwaki et al. 2005 + + ? + + + +

Osawa et al. 2013 – + ? – + + +

Smithard et al. 1998 + + + + + + +

Somasundaram et al. 2014 + + ? + – + +

Zhou et al. 2011 + + + + + + +

– High ? Unclear + Low

Figure 2 Risk of bias and applicability concerns graph (a) and summary (b): review author’s judgements about each domain presented as
percentages across included studies.

increasing water volume resulted in higher sensitivity but
lower specificity of the WST for screening aspiration in
stroke patients. As the stroke population had higher risk of
aspiration than that of normal population, screening tests
with higher sensitivity was preferred to a better diagnostic
accuracy. Therefore, we recommended a 3-oz WST for
screening aspiration in stroke patients since the largest water
volume used in a WST was 3 oz among these studies.
Limitations
There were still some limitations in our study. First, we nei-
ther identified unpublished studies nor related conference
reports. However, we performed a comprehensive system-
atic review from three electronic databases and two clinical
trial registries and no publication bias was identified among
the retrieved studies. Second, patients in these studies were
mostly from medical centres and only two studies (Smithard
et al. 1998, Nishiwaki et al. 2005) included patients from
non-medical centres. The diagnostic accuracy of WST might
not be the same when the test was applied to general hospi-
tals. Third, the WST was performed by physicians and
speech-language pathologists among these studies. The
results should be cautiously applied to nursing practice.
Finally, we discussed only the diagnostic accuracy of the
WST for screening aspiration in stroke patients in this
review. However, bedside screening tests for aspiration in
stroke patients usually included many items, including eval-
uation of tongue function, gag reflex or swallow reflex.
However, we could not identify enough papers allowing the
comparison of the diagnostic accuracy between the WST
and other bedside screening tools in this meta-analysis.

8 © 2016 John Wiley & Sons Ltd

JAN: REVIEW PAPER Water swallow test for screening aspiration in stroke patients

Deeks' funnel plot asymmetry test Future studies regarding diagnostic accuracy between differ-
·08 P-value = 0·30 ent bedside screening tools are thereby needed. The WST
will add depth to the analysis presented in this review and
7 provide better choices of bedside screening tools for aspira-
3 tion in stroke patients.
·1

Conclusions
1/root(ESS)

·12 This systematic review and meta-analysis provided evidence

8 1 that the WST was a useful bedside screening tool for aspi-
9 10 ration in stroke patients, especially for its convenience and
cost-effectiveness in clinical practice. The test accuracy was
·14 4
6 related to the water volume and a 3-oz WST for screening
Study aspiration in stroke patients was recommended.
11 5 Regression It is important to identify aspiration in stroke patients
2
Line
and thus provide early and timely nutritional nursing
·16
1 10 100 1000 interventions or refer to speech-language pathologists for
Diagnostic odds ratio evaluation and treatment. The findings of this review can
help decrease complications of aspiration pneumonia and
Figure 3 Funnel plot with superimposed regression line. ESS,
effective sample size.
malnutrition in stroke patients.

Table 3 Univariate meta-regression analysis for identifying potential sources of heterogeneity in the diagnostic accuracy of screening tests.
Sensitivity Specificity
No. of
Categorical variable studies Adjusted [95% CI] P value Adjusted [95% CI] P value

Study location
Western countries 7 0
71 [0
52, 0
84] 0
620 0
75 [0
51, 0
89] 0
512
Eastern countries 4 0
75 [0
61, 0
85] 0
68 [0
48, 0
83]
Publication year
After 2001 5 0
74 [0
56, 0
86] 0
786 0
64 [0
42, 0
81] 0
133
Before or during 2001 6 0
71 [0
59, 0
81] 0
78 [0
66, 0
87]
Delay between tests
>3 days or unclear 4 0
70 [0
51, 0
83] 0
644 0
74 [0
50, 0
89] 0
813
≤3 days 7 0
73 [0
63, 0
82] 0
71 [0
58, 0
82]
Reference test
FEES 3 0
75 [0
56, 0
87] 0
683 0
74 [0
48, 0
90] 0
836
VFS 8 0
71 [0
61, 0
79] 0
72 [0
59, 0
82]
Methodological quality†
Low risk for all items 3 0
81 [0
65, 0
90] 0
118 0
65 [0
40, 0
84] 0
383
Not 8 0
68 [0
59, 0
76] 0
75 [0
62, 0
84]

Logit (sensitivity)‡ Logit (specificity)‡

Continuous variable Coefficient [95% CI] P Value Coefficient [95% CI] P value

Age 0
00 [ 0
11, 0
11] 0
970 0
02 [ 0
15, 0
11] 0
761

Male percentage 1
20 [ 3
84, 1
44] 0
373 0
48 [ 2
86, 3
82] 0
778
Water volume 0
02 [0
01, 0
03] 0
002** 0
02 [ 0
03, 0
00] 0
025*
Aspiration prevalence 1
34 [ 3
93, 1
25] 0
311 1
11 [ 2
13, 4
34] 0
502

P value is for testing the difference in sensitivity or specificity between the two groups of a covariate. *P < 0
05; **P < 0
01.
†
Assessed by QUADAS-2 tool.
‡
Logit (sensitivity) = log (sensitivity/(1 sensitivity)); Logit (specificity) = log (specificity/(1 specificity)).
CI, confidence interval; no., numbers; FEES, fiberoptic endoscopic examination of swallowing; VFS, videofluoroscopy.

© 2016 John Wiley & Sons Ltd 9

P.-C. Chen et al.
Funding
This research received no specific grant from any funding
agency in the public, commercial or not-for-profit sectors.
Conflict of interest
No conflict of interest has been declared by the authors.
Author contributions
All authors have agreed on the final version and meet at
least one of the following criteria [recommended by the
ICMJE (http://www.icmje.org/recommendations/)]:
• substantial contributions to conception and design,
acquisition of data or analysis and interpretation of
data;
• drafting the article or revising it critically for important
intellectual content.
Supporting Information
Additional Supporting Information may be found in the
online version of this article at the publisher’s web-site.
References
Antonios N., Carnaby-Mann G., Crary M., Miller L., Hubbard H.,
Hood K., Sambandam R., Xavier A. & Silliman S. (2010)
Analysis of a physician tool for evaluating dysphagia on an
inpatient stroke unit: the modified Mann Assessment of
Swallowing Ability. Journal of Stroke and Cerebrovascular
Diseases 19(1), 49–57. doi:10.1016/j.jstrokecerebrovasdis.2009.
03.007
Baker W.L., White C.M., Cappelleri J.C., Kluger J. & Coleman
C.I. (2009) Understanding heterogeneity in meta-analysis: the
role of meta-regression. International Journal of Clinical Practice
63(10), 1426–1434. doi:10.1111/j.1742-1241.2009.02168.x
Bax L., McFarlane M., Green E. & Miles A. (2014) Speech-
language pathologist-led fiberoptic endoscopic evaluation of
swallowing: functional outcomes for patients after stroke.
Journal of Stroke and Cerebrovascular Diseases 23(3), e195–200.
doi:10.1016/j.jstrokecerebrovasdis.2013.09.031
Bours G.J., Speyer R., Lemmens J., Limburg M. & de Wit R.
(2009) Bedside screening tests vs. videofluoroscopy or fibreoptic
endoscopic evaluation of swallowing to detect dysphagia in
patients with neurological disorders: systematic review. Journal
of Advanced Nursing 65(3), 477–493. doi:10.1111/j.1365-
2648.2008.04915.x
Brogan E., Langdon C., Brookes K., Budgeon C. & Blacker D.
(2014) Dysphagia and factors associated with respiratory
infections in the first week post stroke. Neuroepidemiology 43
(2), 140–144. doi:10.1159/000366423
10
Chong M.S., Lieu P.K., Sitoh Y.Y., Meng Y.Y. & Leow L.P.
(2003) Bedside clinical methods useful as screening test for
aspiration in elderly patients with recent and previous strokes.
Annals of the Academy of Medicine, Singapore 32(6), 790–794.
Daniels S.K., McAdam C.P., Brailey K. & Foundas A.L. (1997)
Clinical Assessment of Swallowing and Prediction of Dysphagia
Severity. American Journal of Speech-Language Pathology 6(4),
17–24. doi:10.1044/1058-0360.0604.17
Daniels S.K., Anderson J.A. & Willson P.C. (2012) Valid items for
screening dysphagia risk in patients with stroke: a systematic
review. Stroke 43(3), 892–897. doi:10.1161/STROKEAHA.111.
640946
Deeks J.J. (2001) Systematic reviews in health care: Systematic
reviews of evaluations of diagnostic and screening tests. BMJ
323(7305), 157–162. doi:10.1136/bmj.323.7305.157
Deeks J.J., Macaskill P. & Irwig L. (2005) The performance of
tests of publication bias and other sample size effects in
systematic reviews of diagnostic test accuracy was assessed.
Journal of Clinical Epidemiology 58(9), 882–893. doi:10.1016/
j.jclinepi.2005.01.016
Deeks J.J., Bossuyt P.M. & Gatsonis C. (2013) Cochrane
Handbook for Systematic Reviews of Diagnostic Test Accuracy
Version 1.0.0. The Cochrane Collaboration. Retrieved from
http://srdta.cochrane.org/ on 01 June 2015.
DePippo K.L., Holas M.A. & Reding M.J. (1992) Validation of the
3-oz water swallow test for aspiration following stroke. Archives
of neurology 49(12), 1259–1261. doi:10.1001/archneur.1992.005
30360057018
Donovan N.J., Daniels S.K., Edmiaston J., Weinhardt J., Summers
D. & Mitchell P.H. (2013) Dysphagia screening: state of the art:
invitational conference proceeding from the State-of-the-Art
Nursing Symposium, International Stroke Conference 2012.
Stroke 44(4), e24–e31. doi:10.1161/STR.0b013e3182877f57
Edmiaston J., Connor L.T., Steger-May K. & Ford A.L. (2014) A
simple bedside stroke dysphagia screen, validated against
videofluoroscopy, detects dysphagia and aspiration with high
sensitivity. Journal of Stroke and Cerebrovascular Diseases 23(4),
712–716. doi:10.1016/j.jstrokecerebrovasdis.2013.06.030
Falsetti P., Acciai C., Palilla R., Bosi M., Carpinteri F., Zingarelli
A., Pedace C. & Lenzi L. (2009) Oropharyngeal dysphagia after
stroke: incidence, diagnosis and clinical predictors in patients
admitted to a neurorehabilitation unit. Journal of Stroke and
Cerebrovascular Diseases 18(5), 329–335. doi:10.1016/
j.jstrokecerebrovasdis.2009.01.009
Hamza T.H., van Houwelingen H.C. & Stijnen T. (2008) The
binomial distribution of meta-analysis was preferred to model
within-study variability. Journal of Clinical Epidemiology 61(1),
41–51. doi:10.1016/j.jclinepi.2007.03.016
Hinchey J.A., Shephard T., Furie K., Smith D., Wang D. & Tonn
S. (2005) Formal dysphagia screening protocols prevent
pneumonia. Stroke 36(9), 1972–1976. doi:10.1161/01.STR.
0000177529.86868.8d
Jauch E.C., Saver J.L., Adams H.P. Jr, Bruno A., Connors J.J.,
Demaerschalk B.M., Khatri P., McMullan P.W. Jr, Qureshi A.I.,
Rosenfield K., Scott P.A., Summers D.R., Wang D.Z.,
Wintermark M. & Yonas H. (2013) Guidelines for the early
management of patients with acute ischemic stroke: a guideline
for healthcare professionals from the American Heart
© 2016 John Wiley & Sons Ltd
JAN: REVIEW PAPER
Association/American Stroke Association. Stroke 44(3), 870–947.
doi:10.1161/STR.0b013e318284056a
Kidd D., Lawson J., Nesbitt R. & MacMahon J. (1993) Aspiration
in acute stroke: a clinical study with videofluoroscopy. QJM 86
(12), 825–829.
Kubota T., Mishima H., Hanada M., Minami I. & Kojima Y. (1982)
Paralytic dysphagia in cerebrovascular disorder – screening tests
and their clinical application. Sogo Rehabilitation 10, 271–276.
Lazarus C.L., Logemann J.A., Rademaker A.W., Kahrilas P.J.,
Pajak T., Lazar R. & Halper A. (1993) Effects of bolus volume,
viscosity and repeated swallows in nonstroke subjects and stroke
patients. Archives of Physical Medicine and Rehabilitation 74
(10), 1066–1070. doi:10.1016/0003-9993(93)90063-G
Liberati A., Altman D.G., Tetzlaff J., Mulrow C., Gøtzsche P.C.,
Ioannidis J.P., Clarke M., Devereaux P.J., Kleijnen J. & Moher
D. (2009) The PRISMA statement for reporting systematic
reviews and meta-analyses of studies that evaluate healthcare
interventions: explanation and elaboration. BMJ 339, b2700.
doi:10.1136/bmj.b2700
Lim S.H., Lieu P.K., Phua S.Y., Seshadri R., Venketasubramanian
N., Lee S.H. & Choo P.W. (2001) Accuracy of bedside clinical
methods compared with fiberoptic endoscopic examination of
swallowing (FEES) in determining the risk of aspiration in acute
stroke patients. Dysphagia 16(1), 1–6. doi:10.1007/
s004550000038
Macaskill P., Gatsonis C., Deeks J.J., Harbord R.M. & Takwoingi
Y. (2010) Chapter 10: analysing and Presenting Results. In
Cochrane Handbook for Systematic Reviews of Diagnostic Test
Accuracy (Deeks J.J., Bossuyt P.M. & Gatsonis C., eds), Version
1.0. The Cochrane Collaboration. Retrieved from http://
srdta.cochrane.org/ on 1 June 2015.
Maeshima S., Osawa A., Miyazaki Y., Seki Y., Miura C., Tazawa
Y. & Tanahashi N. (2011) Influence of dysphagia on short-term
outcome in patients with acute stroke. American Journal of
Physical Medicine and Rehabilitation 90(4), 316–320.
doi:10.1097/PHM.0b013e31820b13b2
Mann G. & Hankey G.J. (2001) Initial clinical and demographic
predictors of swallowing impairment following acute stroke.
Dysphagia 16(3), 208–215. doi:10.1007/s00455-001-0069-5
Marik P.E. & Kaplan D. (2003) Aspiration pneumonia and
dysphagia in the elderly. Chest 124(1), 328–336. doi:10.1378/
chest.124.1.328
Martino R., Foley N., Bhogal S., Diamant N., Speechley M. &
Teasell R. (2005) Dysphagia after stroke: incidence, diagnosis
and pulmonary complications. Stroke 36(12), 2756–2763.
doi:10.1161/01.STR.0000190056.76543.eb
Matsuo K. & Palmer J.B. (2008) Anatomy and physiology of
feeding and swallowing: normal and abnormal. Physical
Medicine and Rehabilitation Clinics of North America 19(4),
691–707. doi:10.1016/j.pmr.2008.06.001
Mowatt G., Scotland G., Boachie C., Cruickshank M., Ford J.A.,
Fraser C., Kurban L., Lam T.B., Padhani A.R., Royle J.,
Scheenen T.W. & Tassie E. (2013) The diagnostic accuracy and
cost-effectiveness of magnetic resonance spectroscopy and
enhanced magnetic resonance imaging techniques in aiding the
localisation of prostate abnormalities for biopsy: a systematic
review and economic evaluation. Health Technology Assessment
17(20), 1–281. doi:10.3310/hta17200
© 2016 John Wiley & Sons Ltd
Water swallow test for screening aspiration in stroke patients
Murray J. (1999) Manual of Dysphagia Assessment in Adults.
EUA: Singular Publishing Group, San Diego (CA).
Nishiwaki K., Tsuji T., Liu M., Hase K., Tanaka N. & Fujiwara
T. (2005) Identification of a simple screening tool for dysphagia
in patients with stroke using factor analysis of multiple
dysphagia variables. Journal of Rehabilitation Medicine 37(4),
247–251. doi:10.1080/16501970510026999
Odderson I.R. & McKenna B.S. (1993) A model for management
of patients with stroke during the acute phase. Outcome and
economic implications. Stroke 24(12), 1823–1827. doi:10.1161/
01.STR.24.12.1823
O’Horo J.C., Rogus-Pulia N., Garcia-Arguello L., Robbins J. & Safdar
N. (2015) Bedside diagnosis of dysphagia: a systematic review.
Journal of Hospital Medicine 10(4), 256–265. doi:10.1002/jhm.2313
Osawa A., Maeshima S., Matsuda H. & Tanahashi N. (2013)
Functional lesions in dysphagia due to acute stroke: discordance
between abnormal findings of bedside swallowing assessment and
aspiration on videofluorography. Neuroradiology 55(4), 413–
421. doi:10.1007/s00234-012-1117-6
Petitti D.B. (2001) Approaches to heterogeneity in meta-analysis.
Statistics in Medicine 20(23), 3625–3633. doi:10.1002/sim.1091
Rao N., Brady S.L., Chaudhuri G., Donzelli J.J. & Wesling M.W.
(2003) Gold-standard?: analysis of the videofluoroscopic and
fiberoptic endoscopic swallow examinations. The Journal of
Applied Research 3(1), 89–96.
Raoufi F. & Shade K. (2014) Evaluation of nursing dysphagia
screening tools among patients with stroke: a systematic review
protocol. The JBI Database of Systematic Reviews and
Implementation Reports 12(10), 61–72. doi: 10.11124/jbisrir-
2014-1560. Retrieved from http://www.joannabriggslibrary.org/
index.php/jbisrir/article/view/1560 on 01 June 2015.
Reitsma J.B., Glas A.S., Rutjes A.W., Scholten R.J., Bossuyt P.M.
& Zwinderman A.H. (2005) Bivariate analysis of sensitivity and
specificity produces informative summary measures in diagnostic
reviews. Journal of Clinical Epidemiology 58(10), 982–990.
doi:10.1016/j.jclinepi.2005.02.022
Rugiu M.G. (2007) Role of videofluoroscopy in evaluation of
neurologic dysphagia. Acta Otorhinolaryngologica Italica 27(6),
306–316.
Schepp S.K., Tirschwell D.L., Miller R.M. & Longstreth W.T. Jr
(2012) Swallowing screens after acute stroke: a systematic
review. Stroke 43(3), 869–871. doi:10.1161/
STROKEAHA.111.638254
Singh S. & Hamdy S. (2006) Dysphagia in stroke patients.
Postgraduate Medical Journal 82(968), 383–391. doi:10.1136/
pgmj.2005.043281
Smithard D.G., O’Neill P.A., Park C., England R., Renwick D.S.,
Wyatt R., Morris J. & Martin D.F. (1998) Can bedside
assessment reliably exclude aspiration following acute stroke?
Age and Ageing 27(2), 99–106. doi:10.1093/ageing/27.2.99
Somasundaram S., Henke C., Neumann-Haefelin T., Isenmann S.,
Hattingen E., Lorenz M.W. & Singer O.C. (2014) Dysphagia risk
assessment in acute left-hemispheric middle cerebral artery stroke.
Cerebrovasc Diseases 37(3), 217–222. doi:10.1159/000358118
Tohara H., Saitoh E., Mays K.A., Kuhlemeier K. & Palmer J.B.
(2003) Three tests for predicting aspiration without
videofluorography. Dysphagia 18(2), 126–134. doi:10.1007/
s00455-002-0095-y
11
P.-C. Chen et al.

de Vet H.C.W., Eisinga A., Riphagen I.I., Aertgeerts B. & Pewsner
D. (2008) Chapter 7: Searching for Studies. In Cochrane
Handbook for Systematic Reviews of Diagnostic Test Accuracy
(Eisinga A., ed.), Version 0.4 [updated September 2008]. The
Cochrane Collaboration. Retrieved from http://
srdta.cochrane.org/ on 1 June 2015.
Whiting P.F., Rutjes A.W., Westwood M.E., Mallett S., Deeks J.J.,
Reitsma J.B., Leeflang M.M., Sterne J.A. & Bossuyt P.M. (2011)
QUADAS-2: a revised tool for the quality assessment of
diagnostic accuracy studies. Annals of Internal Medicine 155(8),
529–536. doi:10.7326/0003-4819-155-8-201110180-00009
Zhou Z., Salle J., Daviet J., Stuit A. & Nguyen C. (2011)
Combined approach in bedside assessment of aspiration risk post
stroke: PASS. European Journal of Physical and Rehabilitation
Medicine 47(3), 441–446.

The Journal of Advanced Nursing (JAN) is an international, peer-reviewed, scientific journal. JAN contributes to the advancement of
evidence-based nursing, midwifery and health care by disseminating high quality research and scholarship of contemporary relevance
and with potential to advance knowledge for practice, education, management or policy. JAN publishes research reviews, original
research reports and methodological and theoretical papers.

For further information, please visit JAN on the Wiley Online Library website: www.wileyonlinelibrary.com/journal/jan

Reasons to publish your work in JAN:

• High-impact forum: the world’s most cited nursing journal, with an Impact Factor of 1·527 – ranked 14/101 in the 2012 ISI Jour-
nal Citation Reports © (Nursing (Social Science)).
• Most read nursing journal in the world: over 3 million articles downloaded online per year and accessible in over 10,000 libraries
worldwide (including over 3,500 in developing countries with free or low cost access).
• Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jan.
• Positive publishing experience: rapid double-blind peer review with constructive feedback.
• Rapid online publication in five weeks: average time from final manuscript arriving in production to online publication.
• Online Open: the option to pay to make your article freely and openly accessible to non-subscribers upon publication on Wiley
Online Library, as well as the option to deposit the article in your own or your funding agency’s preferred archive (e.g. PubMed).

12 © 2016 John Wiley & Sons Ltd