See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/7236634

The Erasmus MC modifications to the (revised) Nottingham Sensory

Assessment: A reliable somatosensory assessment measure for
patients with intracranial disorders

Article  in  Clinical Rehabilitation · March 2006

DOI: 10.1191/0269215506cr932oa · Source: PubMed

CITATIONS READS

103 5,041

5 authors, including:

Cisca Stolk-Hornsveld J. Lesley Crow

Erasmus MC Erasmus University Rotterdam
6 PUBLICATIONS   114 CITATIONS    9 PUBLICATIONS   512 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by J. Lesley Crow on 10 February 2016.

The user has requested enhancement of the downloaded file.

 Clinical Rehabilitation
http://cre.sagepub.com

The Erasmus MC modifications to the (revised) Nottingham Sensory Assessment: a

reliable somatosensory assessment measure for patients with intracranial disorders
F Stolk-Hornsveld, J L Crow, E P Hendriks, R van der Baan and B C Harmeling-van der Wel
Clin Rehabil 2006; 20; 160
DOI: 10.1191/0269215506cr932oa

The online version of this article can be found at:

http://cre.sagepub.com/cgi/content/abstract/20/2/160

Published by:

http://www.sagepublications.com

Additional services and information for Clinical Rehabilitation can be found at:

Email Alerts: http://cre.sagepub.com/cgi/alerts

Subscriptions: http://cre.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.co.uk/journalsPermissions.nav

Citations (this article cites 16 articles hosted on the

SAGE Journals Online and HighWire Press platforms):
http://cre.sagepub.com/cgi/content/refs/20/2/160

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

Clinical Rehabilitation 2006; 20: 160-172

The Erasmus MC modifications to the (revised)

Nottingham Sensory Assessment: a reliable
somatosensory assessment measure for
patients with intracranial disorders
F Stolk-Hornsveld, JL Crow, EP Hendriks, R van der Baan and BC Harmeling-van der Wel Physiotherapy Department,
Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands
Received 23rd July 2004; returned for revisions 6th December 2004; revised manuscript accepted 3rd October 2005.

Objective: To investigate the intra-rater and inter-rater reliability of the Erasmus MC

modifications to the Nottingham Sensory Assessment (EmNSA).
Subjects: A consecutive sample of 18 inpatients, with a mean age of 57.7 years,
diagnosed with an intracranial disorder and referred for physiotherapy.
Setting: The inpatient neurology and neurosurgery wards of a university hospital.
Design: Through discussions between four experienced neurophysiotherapists, the
testing procedures of the revised Nottingham Sensory Assessment were further
standardized. Subsequently, the intra-rater and inter-rater reliabilities of the EmNSA
were investigated.
Results: The intra-rater reliability of the tactile sensations, sharp-blunt discrimination
and the proprioception items of the EmNSA were generally good to excellent for both
raters with a range of weighted kappa coefficients between 0.58 and 1.00. Likewise
the inter-rater reliabilities of these items were predominantly good to excellent with a
range of weighted kappa coefficients between 0.46 and 1.00. An exception was the
two-point discrimination that had a poor to good reliability, with the range for intra-
rater reliability of 0.11 -0.63 and for inter-rater reliability -0.10-0.66.
Conclusion: The EmNSA is a reliable screening tool to evaluate primary
somatosensory impairments in neurological and neurosurgical inpatients with
intracranial disorders. Further research is necessary to consolidate these results and
establish the validity and responsiveness of the Erasmus MC modifications to the
NJSA.

Introduction oped. This is remarkable considering the impor-

A reliable clinically useful screening tool to mea-
sure somatosensory impairments for patients with
intracranial disorders has to date not been devel-
Address for correspondence: Lesley Crow, Physiotherapy
Department, Erasmus MC, University Medical Center Rotter-
dam, Postbus 2040, 3000 CA Rotterdam, The Netherlands.
e-mail: berg.crow(hetnet.nl
Co 2006 Edward Arnold (Publishers) Ltd
tant role of the somatosensory system in motor
control13 and the high prevalence (35-60%) of
somatosensory deficits after stroke,4 especially
since somatosensory deficits have been associated
with less favourable functional outcomes.5-9 Such
results are not totally unexpected as the somato-
sensory system plays an important part in motor
control. Cutaneous reflexes, for example, assist in
motor co-ordination between arms and legs during
10.1 191/026921 5506cr932oa

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

 Erasmus MC modifications to Nottingham Sensory Assessment 161
walking and, along with the proprioceptive re-
flexes, contribute to the prevention of falls during
human locomotion." 2 In the absence of vision,
somatosensory inputs to the brain provide knowl-
edge of where our limbs and body are in space.3
A reliable somatosensory assessment measure
also provides a baseline at the beginning of
the rehabilitation process, so enabling any altera-
tions in sensory state to be monitored over
time. With the recent indications that sensory
retraining programmes with stroke patients are
effective0 -12 knowledge of somatosensory impair-
ment can provide useful information to predict
functional outcome and may be helpful in selecting
appropriate treatment goals.
Literature searches on the subject found three
screening tools: the Nottingham Sensory Assess-
ment (NSA),'3 the revised Nottingham study
(rNSA)'4 and the Rivermead Assessment of
Somatosensory Performance (RASP).'5 The relia-
bility of the RASP was not conclusive (see
Discussion) and the reliability of the rNSA was
not consistently good for all items. However, in
contrast to the RASP, no special expensive equip-
ment is required for the administration of the
rNSA. Therefore, the aim of this study was to
improve the reliability of the rNSA by providing
modifications and further standardization of the
testing procedures.
Methods
All the modifications to the chosen test items of
the rNSA were decided through regular discussions
between four experienced neurophysiotherapists
working in the acute care setting at the Erasmus
MC, University Medical Center in Rotterdam.
From the original items of the rNSA (light touch,
pressure, pinprick, temperature, kinaesthetic sense
and two-point discrimination) the following
changes were made. The temperature test was
excluded according to the recommendations in
the revised Nottingham study. A sharp-blunt
discrimination was added to assess pain sensa-
tion.16 The pinprick test remained as an item to
assess tactile sensation. The scoring of the two-
point discrimination test item was modified in
accordance with the assessment recommendations
Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008
of the American Society for Surgery of the Hand'7
and the kinaesthetic sensation item (propriocep-
tion) was further standardized. Details of these
modifications are incorporated in the guidelines in
Appendix 1, with Appendix 2 providing additional
standardizations for the defined points of contact
used with the tactile sensation items. Appendix 3
contains the specific standardizations for the
proprioception item. The general guidelines and
score sheet for the EmNSA (Erasmus MC mod-
ifications to the (revised) Nottingham Sensory
Assessment) were updated to include additional
standardizations and a uniform scoring system,
(see Appendices 1 and 4).
The assessment
This study was approved by the hospital medical
ethics committee and took place within a limited
two-month period (April-May 2003). All current
inpatients, along with all new admissions to the
neurology and neurosurgery wards of the Erasmus
MC and referred for physiotherapy, were screened
for eligibility for inclusion in the study. Inclusion
criteria were: an intracranial disorder, diagnosed
by a neurologist or neurosurgeon and by neuro-
imaging studies, at least 18 years of age. Excluded
from the study were patients with a peripheral
neurological disorder, a disorder at the level of the
spinal cord, Parkinson's disease or other hypoki-
netic rigid syndromes, multiple sclerosis, or a
severally impaired cognitive state. To exclude
patients with considerable cognitive impairments,
the observers administered the Mini-Mental State
Examination (MMSE).'8 Some cognitive impair-
ment was allowed, therefore, a cut-off point was
introduced. A score of at least 15 points was
required to take part in the study. Prior to enrolling
in the study a full explanation of the procedure was
given to the patient by an examiner. Informed
consent was then obtained.
Two physiotherapists acted as examiners for the
study. One examiner had over 20 years' experience
working with patients with intracranial disorders,
the other physiotherapist one year. Both phy-
siotherapists continued to have a clinical caseload
throughout the period of this study so helping to
minimize the recall of the results for the intra-rater
reliability.
To assess the intra-rater reliability each examiner
tested all the patients on two occasions. To
162 F Stolk-Hornsveld et al.
minimize any recall bias there was an interval of at
least 24 h between the initial and repeat test
occasions. To assess the inter-rater reliability, the
two examiners assessed each patient on the same
day, with an interval of at least 1 h, but not more
than 2 h. Throughout the study period the
examiners were blind to each other's results.
Descriptive statistics were carried out on SPSS
version 10.1 for Windows and Intercooled Stata
5.0. SPSS was used to calculate the weighted kappa
coefficients.
Results
Twenty-one patients were eligible for the study,
however three patients were excluded due to a
MMSE score of less than 15 points. Therefore, the
final study sample consisted of 18 inpatients (nine
men, nine women) with neurological (n = 14) or
neurosurgical (n = 4) disorders. Two patients were
assessed on both sides. The mean age was 57.7
years (range 20-84) and the mean number of days
since admission was 14.9 days (range 4-92).
Diagnostic details are described in Table 1.
Five patients (28%) achieved a normal score, for
all the test items. Some degree of sensory impair-
ment was seen in six patients (33%). Seven patients
(39%) had severe sensory loss; they scored 0 or 1
on five or more test items (see Table 2).
Intra-rater reliability
According to the Fleiss categories for Cohen's
kappa coefficient,'9 the intra-rater reliability of the
Erasmus MC modifications to the Nottingham
Table 1 Demographic details of the study sample (n = 18)
Left- Right- Bilateral Total
sided sided
Stroke 4 6 28 12
SAH 1 1
Cerebral tumour 2 2
Hydrocephalus 1 1 2 2
Traumatic brain 1 1
injury
n=18
aBoth sides assessed.
SAH, subarachnoid haemorrhage.
Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008
Sensory Assessment (EmNSA) for both raters is
generally good to excellent (Table 3).
Only two of the weighted kappa coefficients, of
the 84 calculated, represented a poor agreement
(both for two-point discrimination). Of the re-
maining 82 results, only two represented a fair
agreement and 11 a good agreement. Thus 81% of
the weighted kappa coefficients (68 values) repre-
sented an excellent agreement. Rater A had slightly
less favourable results than rater B.
Inter-rater reliability
The inter-rater reliability of the Erasmus MC
modifications to the Nottingham Sensory Assess-
ment (EmNSA) is predominantly good to excellent
(Table 4). The weighted kappa coefficient of the
two-point discrimination at the fingertip had a
negative score. At the thenar eminance the relia-
bility of the two-point discrimination was good.
Only one of the weighted kappa coefficients, of
the 52 calculated, represented a poor agreement
(two-point discrimination). Of the remaining 51
results only two represented a fair agreement and
nine a good agreement. Thus 77% of the weighted
kappa coefficients (40 values) represented an
excellent agreement.
Discussion
This study has developed standardized modifica-
tions to the revised Nottingham Sensory Assess-
ment for use with patients diagnosed with an
intracranial disorder. The reproducibility of the
chosen items included in the EmNSA is now
predominantly good to excellent. The original
studies (including only stroke patients) were unable
Sets to reach such high levels of reliability. The
of data improved intra-rater and inter-rater reliability is
14
most likely due to the additional standardized
1 modifications made to the guidelines of the
2 assessment measure. Feedback from the examiners
and observers confirmed that the EmNSA is a
1 clinically feasible assessment measure and easy to
administer. However, it still requires approximately
n=20 10- 15 min to administer. This amount of time
could possibly be reduced if a hierarchy were
established for the test items and the points of
contact.
a)Q)
E
7U

C
U)

0
0)

U0)
o

-o

0
0)
F
Q)

0
U)
0
co

0)
4C

a
n

-c

CU

Q
F
E0-
0)

0)
o
Un

0
U)+-

O C

4O0
0)
a
0
._
0
a

4-0

.Y0)
U)
CL

D
U)
a-
F=
QzQL-
.:?
E

C:5+co
._

.)
LL

LU
-D

LJ

-D

LU

LLJ

LU

LUL
I9
-r -E
0) m
U)Un) C

N CN

"N CN N

CN CN N

N C

N N N

N N N N

N N N
E
==0 0) 0)

N N
Erasmus MC modifications to Nottingham Sensory Assessment 163

0)0)

-
ao

-
a)E
O a O
> > CD > > >x > o
0) 0) 0) ) 0

CN

N N N 00

-
0 a)

CN
4
-)i

n
a)E E" s Z V

ZN
_

0Z N N N N N N

N N oZ N N N N N N '- N N N CN

N N N- 0 -
=

N
=

-
DCD5
DC

-
U-)-}CD CD
=0 0

-
c-

N N N N N N N N N N

ON (N N N N N
-
00

-
'

Z Z
__

C
E E

ZZ

<

NJ N N NC4

N N
°

CN

1OClJNro
aQ
co
Q-

co
._
Q

o
O

.
X

0)
o
o
CU
4_
0

E
>-

0
O

>Gn
co

>-
.t
E
a)
.

a~~~~~~)
43

00)
0

et)

0
o

(1)~~~~~~~~~~~~~~~~~U
w
4-
a)
c-

_D
~0

0-
01)
4-C
0
a)
-o
_-

5: LU
D N NNN N -0Z N N N
N NN- N
N CN
Q)
0)
co
0
o)
LU
0) N N N N r- N- 0 Z N NN N N N '- N N N N
Un
Q)
0 c-
0)

c-9 LU
D ( N CN1 -
RJ
Z CN N CN CN (N CN N CEJC>lC'
CD

co
co
-o
U)0a) Cj t Lo C LO (CD
FH01)U)
r- 00 h-CD O)0
I ,- v r- - , N

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

164 F Stolk-Hornsveld et al.

It LO 00 0 C
C: 00
00 0) C 0o0 OCc a)
9 co 0' 0 00 '- 0 E
a>
a1)
0

0
0 C) 0
0
't
CC
LO0C) C
00 0 0 0) a)
0Q < o a)
C)
0
0
F Q)
a)
C C
a)
a) a)
a)
-o
X X 0
co
C LL5U)JU
o.iTO 0
0
c
a)
C aa)
E C0
0
C:
U) O 00 (N c?o c00) 0
10 00 0)010 a)
co 0 00 0 0

00 N C 0 .-
:3I
_C
co 00' 0)O
<) a-
a) 0 100 0 C) 0(
CN 0)
O C
z 0 OC 0
U )
F
LL
a) (0 '-100 (Nb
o cu
C+
0) N 0 0 a)a.._
0
.0 0100 0 0) 0)1000 0)
CC) 0 00 00
coa
c
a) 0 ._
a) OC
- a)
-o a
Q- 000 0 00N N
r-o o C 0) o0 010
CD 10
Q- < 0o 0' 00 0 00 -CaI
. E
a)
a)
6 t0 - 1
a) C O)
C
0 '10
0(0co0
oo
-
0 CD 0) 100)0)0) 0) 10
co 00'- 0 a)
a) ° 0n
a 0)
a) o 000' 0 co
O)
L-a) '-CD 0
'- (O000 7 LO ) I)
0 N
0 010
0. N- mD oo
6 6;l,C
6666;
C
C O)
0 Qa)
CO
0- CY) CD N .0 L
co co
a) a)o
0Y)
0) 1.0 C'- 0
(C(0'-0 C)
N0 E
0 < CO r- - - 10 1000 N 0
CY) CD a)E
(C CN
U) C 00'- 0 0 000'- 0 < 66 U) Q)
a) 0
a1) -D 0
:3 :3
_C
a) co ._
L- 0 000'- 0 a
co CN o
C)114 0
a) Oco a
Q a)
F-

a )-C ) a)
a1) L:
Qa)
.0) C1) L CS) 0 M01-
0
C
co
Q0U)
U-
a)C
C:
CV)
a) a) E oU)
0) C ao a0 a)
4) C0 a .C a)
L ." co
:5 +1 LI tDO Q
LLI1
a
D O O )C
, L IJ-
0
-J 10 ILl-)

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

 Erasmus MC modifications to Nottingham Sensory Assessment 165
Table 4 Weighted kappa values for the inter-rater reliability on the EmNSA
Light Pressure Pinprick Sharp-blunt Proprioception
touch discrimination
Fingers 0.89 1.00 0.87 1.00 Fingers 0.71
Hand 0.87 1.00 0.76 0.84 Wrist 0.63
Forearm 0.87 1.00 1.00 0.84 Elbow 1.00
Upper arm 0.71 0.84 1.00 1.00 Shoulder 0.46
Upper extremity 0.90 1.00 0.88 0.86 Upper extremity 0.74
Toes 0.86 0.83 0.90 0.53 Toes 0.69
Foot 0.85 0.92 0.90 0.81 Ankle a
Leg 0.83 0.83 0.79 0.90 Knee 1.00
Thigh 0.89 0.90 0.76 0.61 Hip 1.00
Lower extremity 0.81 0.83 0.88 0.70 Lower extremity 0.66
Two-point discrimination
Fingertips - 0.10
Thenar eminance 0.66

aKappa values could not be calculated because the contingency table had a row with zero elements.
Fleiss categories for Cohen's kappa: 0.00-0.39 poor agreement; 0.40-0.59 fair agreement; 0.60-0.74 good agreement; 0.75-
1.00 excellent agreement.19

Although only 18 patients were involved in this
study (resulting in 20 sets of data) this number
does meet the minimal size requirements for a
sample, as suggested by Cicchetti.20 When using
the kappa statistic the sample size is calculated in
relation to the number of categories in the scoring
system. In reliability studies it has to be taken into
account that the (weighted) kappa statistic is
influenced by the selection of subjects over which
it is defined (i.e. the kappa statistic is dependent on
the distribution across all the categories). If the
subjects are highly variable then the kappa statistic
tends to be high, whereas for a more homogeneous
group of subjects it will be lower.2' This implies
that the disorder not the disease (i.e. the level of
somatosensory impairments in the research popu-
lation) has consequences for the kappa statistic.
Another problem of the kappa statistic is that
when one of the possible scoring categories is not
chosen, in other words the contingency table has a
row or column with all zero elements, the kappa
statistic cannot be calculated. This occurred in
three out of the 136 results in our study. Despite
some drawbacks, the weighted kappa remains an
appropriate statistical method of analysis of ordi-
nal data for agreement within and between raters.
The use of parametric statistical tests for non-
parametric data should only be used with caution.
In the RASP study the use of the Pearson corre-
lation coefficient could lead to false-positive results
(exaggerated high coefficient values) as association
is examined, not agreement.21'22 Likewise, the use
of Bland and Altman plots may also be inapp-
ropriate, as although they consider rater agree-
ment, they are based on parametric principles.23
In our study the research population included all
eligible patients with intracranial disorders. Over
half of these were stroke patients and of the total
sample, 72% were found to have some degree of
somatosensory impairments (compared with 60%
expected in a stroke population).2 Despite the high
percentage of patients with one or more sensory
impairments, the overall level of sensory impair-
ments in our subjects was low. This is a disadvan-
tage of our study population. Our research
population probably differs at this point from a
population of stroke patients, the level of somato-
sensory impairments in stroke patients should be
overall higher. Therefore, to make reliability stu-
dies more comparable the distribution across the
categories should be published, especially with
small sample sizes, which require more variability
in data. We included all eligible patients with

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

166 F Stolk-Hornsveld et al.

Clinical messages
0 The modifications made in the Erasmus
MC modifications to the Nottingham
Sensory Assessment (EmNSA) have im-
proved the reproducibility of the majority
of the test items such that they now have
predominantly good to excellent inter-
rater and intra-rater reliability.
- Two-point discrimination remains an un-
reliable test item.
The EmNSA is an inexpensive and easy
assessment measure to administer as a
screening tool in routine clinical practice.
intracranial disorders because this is an exact copy
of our clinical situation. We wanted the assessment
of somatosensory impairments to be reliable and
clinically useful not only in stroke patients but for
most patients with intracranial lesions admitted to
the neurology and neurosurgery wards. Hence the
inclusion of patients with mild cognitive disorders.
Although the kappa coefficients for propriocep-
tion have improved in this study, they have not
reached such high levels of reliability as the tactile
sensations. Two-point discrimination remains a
test item that is difficult to reproduce and is
therefore not included on the definitive version of References
the EmNSA score sheet.
From the intra-rater reliability results it is 1 Zehr EP, Stein RB, Komiyama T. Function of sural
evident that whilst both physiotherapists achieved
good to excellent levels of agreement those
achieved by the more experienced physiotherapist 2 Haridas C, Zehr EP. Coordinated interlimb
(rater B) were generally higher. This may suggest
that clinical experience positively influenced the
intra-rater reliability. Both the intra-rater reliabil- 3 BerlucchiJG,Neurophysiol
ity and the inter-rater reliability of the upper
extremity have higher levels of agreement than
those of the lower extremity. This is consistent with 4 Carey LM. Somatosensory loss after stroke. Crit
the results of the revised NSA study and may relate
to the larger representation of the arm in the brain
(Penfield's sensory homunculus).
Conclusion
The EmNSA is a reliable, standardized assessment
measure, which is easy, inexpensive and relatively
quick to administer. Therefore, it is a useful
screening tool in the clinical setting for the
assessment of primary somatosensory impairments
in acute patients with an intracranial disorder.
With such promising results from this preliminary
study, further work is now planned with a larger,
diverse population. Such research should support
these conclusions and aim to establish the validity
and responsiveness of the Erasmus MC modifica-
tions to the revised Nottingham Sensory Assess-
ment in the acute care setting.
Acknowledgements
This project was undertaken in conjunction with
the Hogeschool Rotterdam, with Diane Breedijk
(lecturer in physiotherapy) providing valuable
advice throughout the study.
We would like to thank Sylvia Gorter and
Marieke Sizoo for acting as examiners and along
with four of the authors participating in the initial
discussions. We would also like to thank Elizabeth
Carr and Denise Coleman for their valuable
comments on this article prior to its submission.
Copies of the EmNSA guidelines can be ob-
tained from the authors in English and Dutch.
nerve reflexes during human walking. J Physiol
1998; 507: 305-14.
compensatory responses to electrical stimulation of
cutaneous nerves in the hand and foot during
walking.
Aglioti
2003; 90: 2850-61.
S. The body in the brain:
neural bases of corporeal awareness. Trends
Neurosci 1997; 20: 560-64.
Rev Phys Rehabil Med 1995; 7: 51-91.
5 Patel AT, Duncan PW, Lai S, Studenski S. The
relation between impairments and functional
outcomes poststroke. Arch Phys Med Rehabil 2000;
81: 1357-63.
6 Han L, Law-Gibson D, Reding M. Key
neurological impairments influence function-related
group outcomes after stroke. Stroke 2002; 33:
1920-24.
7 Ken C, Leo MA, Soderberg GL. Relationship
between perception of joint position sense and limb

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

 Erasmus MC modifications to Nottingham Sensory Assessment 167
synergies in patients with hemiplegia. Phys Ther
1981; 61: 1433-37.
8 Anderson EK. Sensory impairments in hemiplegia.
Arch Phys Med Rehabil 1971; 293-97.
9 Sommerfeld DK, von Arbin MH. The impact of
somatosensory function on activitiy performance
and length of hospital stay geriatric patients with
stroke. Clin Rehabil 2004; 18: 149-55.
10 Carey LM, Matyas TA, Oke LE. Sensory loss in
stroke patients: effective training of tactile and
proprioceptive discrimination. Arch Phys Med
Rehabil 1993; 74: 602-10.
11 Smania N, Montagnana B, Faccioli S, Fiaschi A,
Aglioti SM. Rehabilitation of somatic sensation and
related deficit of motor control in patients with pure
sensory stroke. Arch Phys Med Rehabil 2003; 84:
1692-702.
12 Peurala SH, Pitkanen K, Sivenius J, Tarkka IM.
Cutaneous electrical stimulation may enhance
sensorimotor recovery in chronic stroke. Clin
Rehabil 2002; 16: 709-16.
13 Lincoln NB, Crow JL, Jackson JM, Waters GR,
Adams SA, Hodgson P. The unreliability of sensory
assessments. Clin Rehabil 1991; 5: 273-82.
14 Lincoln NB, Jackson JM, Adams SA. Reliability
and revision of the Nottingham Sensory
Assessment for stroke patients. Physiotherapy 1998;
84: 358-65.
15 Winward CE, Halligan PW The Rivermead
Assessment of Somatosensory Performance
(RASP): standardization and reliability data. Clin
Rehabil 2002; 16: 523-33.
16 Oosterhuis HJGH. Klinische neurologie. Bohn
Stafieu Van Loghum, 1992.
17 Aurora CO. The hand examination and diagnosis.
American Society for Surgery of the Hand, 1978.
18 Wade DT. Measures of cognitive impairment and
disability. In Measurement in neurological
rehabilitation. Oxford University Press, 1994:
133-34.
19 Fleiss JL. Statisitical methods for rates and
proportions. John Wiley and Sons, 1981.
20 Chichetti DV. On a model for assessing the security
of infantile attachement: issues of observer
reliability and validity. Behav Brain Sci 1984; 7:
149-50.
21 Armitage P, Berry G, Matthews. Kappa measure of
agreement. In Brown A, Pattison F eds. Statistical
methods in medical research, fourth edition.
Blackwell Science, 2002: 698-707.
22 Maher C. Pitfalls in reliability studies: some
suggestions for change. Aust J Physiother 1993; 39:
5-7.
23 Streiner DL, Norman GR. Health measurements
scales, second edition. Oxford Medical
Publications, 1995.

Appendix 1 Guidelines for the Erasmus MC Modifications to the (revised)

-

Nottingham Sensory Assessment (EmNSA)

General instructions
The patient, suitably undressed, lies on a bed in the supine position, with the forearms in supination.
The patient is asked to close their eyes whilst the actual testing is in progress. Each test item is explained
in turn to the patient and, if necessary, demonstrated on the unaffected side. Testing should begin distally
(i.e. fingers, toes). As a screening tool the affected side only is tested. However for more detailed
information both sides can be tested. Each body section is tested once, in any order, at the three defined
points of contact as described in Appendix 2. No more than 2-5 s should elapse between the stimuli of a
specific test item.
A reminder of the defined points of contact can be found on the scoresheet.

Tactile sensation (light touch, pressure and pinprick)

For each test item, the skin is stimulated three times, at each location, in a random order, using the
defined points of contact. The patient indicates verbally or non-verballya whenever he or she feels the test
stimulus.

'Patient and physiotherapist agree, in advance, the most suitable manner in which the test sensation is indicated, e.g. by specific
hand movements. whenever he feels the test stimulus.

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

168 F Stolk-Hornsveld et al.
Scoring criteria for light touch, pressure and pinprick.
0 Absent Patient fails to identify the test sensation on all three occasions.
1 Impaired Patient identifies the test sensation on only one or two occasions.
2 Normal Patient identifies the test sensation on all three occasions.
With light touch, if a score of 2 is assigned for all of a limb, then automatically assign a score of 2 for all
the pressure and pinprick test items.
* Light touch Touch the skin, at the defined points of contact, lightly with a cotton wool ball.
* Pressure Apply pressure to the skin, using the index finger, at the defined points of contact,
sufficient enough to just deform the skin contour.
* Pinprick Prick the skin using a cocktail stick, at the defined points of contact, sufficient
enough to just deform the skin contour.
Sharp-blunt discrimination
Not tested if score 0 or 1 on tactile sensations.
Stimulate the skin six times at each location, in a random order, three times with a cocktail stick and
i;hree times with the index finger, using the defined points of contact.
The patient must verbally describe or non-verbally indicate the test sensation as sharp or blunt.
Scoring criteria for sharp-blunt discrimination
0 Absent Patient fails to correctly describe/ indicate the test sensation on all six occasions.
1 Impaired Patient correctly describes/ indicates the test sensation, but on less than six occasions.
2 Normal Patient correctly describes/ indicates the test sensation on all six occasions.

Two-point discrimination
Only test two-point discrimination if the patient has been assigned normal scores for all light touch,
pressure and pinprick test items of the upper limb.
Set dividers at decreasing intervals. Apply the two points simultaneously to the skin of the index finger
and then to the thenar eminence for approximately 0.5 s. Ask the patient to indicate if one or two points
are felt. Record the last interval at which two points are felt.
Recommended starting intervals are for the thumb 10 mm and index finger 20 mm.
Scoring
0 Absent Patient is not able to detect two points.
1 Impaired Patient detects two points with an interval of 10 mm on the fingertip and 20 mm
on the thenar eminance.
2 Normal Patient detects two points with an interval of 5 mm on the finger tip, and 12 mm
on the thenar eminance.

Proprioception
Specified passive movements are tested in only one joint at a time. The starting positions, specific hand
grips for the physiotherapist to use, along with the directions of the movement to be tested are described
in Appendix 3. The large joints (hip and knee, shoulder and elbow) are moved through approximately a
quarter of their total range of motion (ROM). The other joints (wrist and fingers, ankle and toes) are

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

 Erasmus MC modifications to Nottingham Sensory Assessment 169
moved throughout the full available range of movement. To demonstrate the procedure, three practice
movements are allowed (with the patient's eyes open.) Each joint is then moved three times. The patient is
asked, using specific questions, to indicate verbally or non-verbally the direction of the movement taking
place. If the patient is incapable of doing this, he is then asked to identify (verbally or non-verbally) when
movement is taking place.
0 Absent Patient does not detect the movement taking place.
1 Impaired Patient detects the movement taking place but the direction is not correct on all
three occasions.
2 Normal Patient correctly detects the direction of the movement taking place on all three
occasions.

Appendix 2 - The Erasmus MC Modifications to the (revised) Nottingham Sensory

Assessment
Defined points of contact for stimulating the light touch, pressure, pinprick and sharp-blunt sensations
A Fingers:
1) Distal phalanx 5th digit, palmar aspect
2) Distal phalanx 3rd digit, palmar aspect
3) Distal phalanx 1st digit, palmar aspect
B Hand
1) Second metacarpal, distal palmar aspect
2) Fifth metacarpal, distal palmar aspect
3) Centre of the thenar eminence
C Forearm
1) Ulnar styloid, anterior aspect
2) Centre of the forearm, anterior aspect
3) 2 cm distal to the elbow joint line, anterolateral aspect

D Arm:
1) 2 cm proximal to the elbow joint, anteromedial aspect
2) Centre of anterior aspect of the humerus
3) 2 cm distal to the acromion, lateral aspect
E Toes
1) Distal phalanx, 5th digit, plantar aspect
2) Distal phalanx, 3rd digit, plantar aspect
3) Distal phalanx, 1st digit, plantar aspect
F Foot
1) Base of 5th metatarsal bone, dorsal aspect
2) Second metatarsal, dorsal aspect
3) Centre of midtarsal line, dorsal aspect

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

170 F Stolk-Hornsveld et al.
G Leg
1) Medial malleolus, medial aspect
2) Centre of the anterior border of the tibia
3) Fibular head, lateral aspect
H Thigh
1) Medial femoral epicondyle, medial aspect
2) Centre of line of femur, anterior aspect
3) Greater trochanter
A reminder of these defined points of contact are represented diagramatically on the score sheet.

Appendix 3 - Specific starting positions, movements and hand grips for testing
proprioception
Starting position
Unless otherwise stated, the patient remains in supine lying with the forarm in supination. The large
joints (hip and knee, shoulder and elbow) are moved through approximately a quarter of their total range
of motion (ROM). The other joints (wrist and fingers, ankle and toes) are moved throughout the full
available range of movement.
Fingers
Movement: flexion and extension of the distal phalanx of the thumb.
Ask the patient: 'Is your thumb being bent or straightened?'
Hand grips
Distal (moving) hand: place the thumb laterally and the index finger medially on the distal phalanx of the
thumb.
Proximal (fixing) hand: fix the proximal phalanx between thumb and index finger.
Wrist
Movement: flexion and extension of the wrist. Place the elbow in a starting position 150- 1600
extension.
Ask the patient: 'Is your hand moving upwards or downwards?'
Hand grips
Distal (moving) hand: place the thumb on the lateral aspect and the index finger on the medial aspect of
the hand.
Proximal (fixing) hand: fix the distal end of the forearm.
Elbow
Movement: flexion and extension of the elbow. Place the elbow in a starting position of 900 flexion.
Ask the patient: 'Is your elbow being bent or straightened?'
Hand grips
Distal (moving) hand: Grasp the distal end of the forearm, placing the thumb anteriorly and the fingers
posteriorly.
Proximal (fixing) hand: Fix the distal end of the humerus

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

 Erasmus MC modifications to Nottingham Sensory Assessment 171
Shoulder
Movement: abduction and adduction of the shoulder. Place the elbow in 90° flexion, and lift the arm
sufficient to allow the movements to occur.
Ask the patient 'is your arm moving towards you or away from you?'
Hand grips
Distal (guiding) hand: grasp the distal end of the forearm with the thumb anteriorly and the fingers
posteriorly.
Proximal (guiding) hand: grasp the flexed elbow in a cupped manner.
Toes
Movement: flexion and extension of the first metatarsophalangeal joint.
Ask the patient: ' Is your toe moving upwards or moving downwards?'
Hand grips
Distal (moving) hand: place the thumb lateral and the index finger medial on the distal phalanx of the
great toe.
Proximal (fixing) hand: fix the first metatarsal bone, just proximal to the metatarsophalangeal joint with
the thumb lateral and the index medial.
Ankle
Movement: flexion and extension of the ankle joint.
Ask the patient: 'Is your foot moving upwards or moving downwards?'
Hand grips
Distal (moving) hand: grasp the foot with thumb placed on the lateral margin of the foot and fingers on
the medial margin of the foot.
Proximal (fixing) hand: fix the distal end of the tibia and fibula.
Knee
Movement: flexion and extension of the knee, with the hip joint in 900 flexion.
Ask the patient: 'Is your knee being bent or straightened?'
Hand grips
Distal (moving) hand: grasp the calcaneus with the thumb medially and the fingers cupped inferiorly. The
foot should be supported by the lower forearm.
Proximal (fixing) hand: grasp the distal end of the femur, with the thumb laterally and the fingers
medially.
Hip
Movement: flexion and extension of the hip joint, starting with the hip in 900 flexion.
Ask the patient: Is your thigh moving towards you or away from you?'
Hand grips
Distal (guiding) hand: grasp the calcaneus with the thumb medially and the fingers cupped inferiorly. The
foot should be supported by the lower forearm.
Proximal (moving) hand: grasp the distal end of the femur and with the thumb laterally and the fingers
medially. Maintain knee position accurately as you flex the hip.

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

 172 F Stolk-Hornsveld et al.

Appendix 4 - EmNSA score sheet

ERASMUS MC MODIFICATIONS TO THE (REVISED) PATIENT NAME ;
NOTTINGHAM SENSORY ASSESSMENT (EMNSA) DATE OF BIRTH ;

DATE: RIGHT LE E
____ ___ ____ ___ ____ ___ ____ ___ LT PR PP SB. LT P p S
UPPER ARM
TACTILE
FOREARM
SENSATION HAND
FINGERS
TOTAL SCORE UE
SHARP-BLUNT T:IGH
THIGH
DISCRIMINATION LEG
FOOT
TOES
TOTAL SCORE LE
SHOULDER
PROPRIOCEPTION ELBOW
WRIST
FINGERS
TOTAL SCORE UE ..
HIP
KNEE
ANKLE
TOES

TOTAL SCORE LE
- light touch PR - Pressure PpiUE pinprick s-e sharp-blurt discimination
upper extremity LE lower extremity
=
0 = absent I impaired 2 = normal
=

RIGHT UPPER EXTREMITY RIGHT L-OWER EXTREMITY

Exact points of contact for

tactile sensation &
sharp-blunt discrimination
I
*ij

Full description of points of contact available in appendix 2

Downloaded from http://cre.sagepub.com at Erasmus Univ Rotterdam on October 3, 2008

View publication stats