Parkinson’s Disease-Cognitive Rating Scale: A New
Cognitive Scale Specific for Parkinson’s Disease
Javier Pagonabarraga, MD, Jaime Kulisevsky, MD, PhD,* Gisela Llebaria, MD,
Carmen Garcı ´a-Sa ´nchez, MD, PhD, Berta Pascual-Sedano, MD, PhD, and Alexandre Gironell, MD, PhD
Movement Disorders Unit, Neurology Department, Sant Pau Hospital, Autonomous University of Barcelona,
and Centro de Investigacio ´n Biome ´dica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Spain
Abstract: Cognitive defects associated with cortical pathology
may be a marker of dementia in Parkinson’s disease (PD).
There is a need to improve the diagnostic criteria of PD de-
mentia (PDD) and to clarify the cognitive impairment patterns
associated with PD. Current neuropsychological batteries
designed for PD are focused on fronto-subcortical deficits but
are not sensitive for cortical dysfunction. We developed a new
scale, the Parkinson’s Disease-Cognitive Rating Scale (PD-
CRS), that was designed to cover the full spectrum of cogni-
tive defects associated with PD. We prospectively studied
92 PD patients [30 cognitively intact (CogInt), 30 mild cogni-
tive impairment (MCI), 32 PDD] and 61 matched controls
who completed the PD-CRS and neuropsychological tests
assessing the cognitive domains included in the PD-CRS.
Acceptability, construct validity, reliability, and the discrimina-
tive properties of the PD-CRS were examined. The PD-CRS
included items assessing fronto-subcortical defects and items
assessing cortical dysfunction. Construct validity, test-retest
and inter-rater reliability of PD-CRS total scores showed an
intraclass correlation coefficient >0.70. The PD-CRS showed
an excellent test accuracy to diagnose PDD (sensitivity 94%,
specificity 94%). The PD-CRS total scores and confrontation
naming item scores-assessing ‘‘cortical’’ dysfunction—inde-
pendently differentiated PDD from non-demented PD. Alter-
nating verbal fluency and delayed verbal memory independ-
ently differentiated the MCI group from both controls and
CogInt. The PD-CRS appeared to be a reliable and valid PD-
specific battery that accurately diagnosed PDD and detected
subtle fronto-subcortical deficits. Performance on the PD-CRS
showed that PDD is characterized by the addition of cortical
dysfunction upon a predominant and progressive fronto-sub-
cortical impairment. ? 2008 Movement Disorder Society
Key words: Parkinson’s disease; cognition; dementia; rat-
ing scale; neuropsychological
Some degree of cognitive impairment is common in
nondemented Parkinson’s disease patients (PD-ND) and
eventually progress to dementia in 24 to 31% of
patients.1In population-based and cohort studies, from
23.5 to 55% of PD-ND patients show mild cognitive
defects since the early stages of the disease.2–4The cog-
nitive changes in PD are characterized by a frontal-sub-
cortical impairment with decreased attention and execu-
tive function leading to progressive impairment in pre-
frontal tasks, visuospatial skills, and memory.2,5,6Still,
20 to 25% of PD-ND patients may exhibit a pattern of
cortical impairment with memory tasks and confronta-
tion naming defects,2and cognitive findings associated
with cortical pathology, such as language errors, de-
velop in many patients with PD with dementia (PDD).7
Accordingly, neuroimaging and pathological studies
have variably correlated the development of PDD with
both frontal-subcortical and cortical alterations. De-
creased fronto-striatal dopaminergic activity,8,9wide-
spread decrease of cortical cholinergic activity,10,11and
higher cortical degeneration in the limbic/paralimbic
system have all been reported.12–14Thus, to capture the
whole spectrum of cognitive defects associated with PD,
neuropsychological evaluation should include items sen-
sitive to cortical and frontal-subcortical dysfunction.
Only two published neuropsychological batteries
have been specifically designed to target the most spe-
cific cognitive deficits in PD.15,16The Mini-Mental
This article contains supplementary material available via the Inter-
net at http://www.interscience.wiley.com/jpages/0885-3185/suppmat.
*Correspondence to: Jaime Kulisevsky, Movement Disorders Unit,
Neurology Department, Sant Pau Hospital, Sant Antoni M, Claret
167, 08025 Barcelona, Spain. E-mail: firstname.lastname@example.org
Received 2 October 2007; Accepted 6 February 2008
Published online 31 March 2008 in Wiley InterScience (www.
interscience.wiley.com). DOI: 10.1002/mds.22007
Vol. 23, No. 7, 2008, pp. 998–1005
? 2008 Movement Disorder Society
Parkinson15is a brief screening test aimed at identify-
ing PD patients who require a more comprehensive
cognitive assessment. Only the pilot study was pub-
lished, and the scale was not subjected to extensive
clinimetric evaluations.15The SCOPA-COG is a short,
reliable and valid instrument sensitive to measure cog-
nition in PD.16However, the scale was constructed
with items assessing frontal-subcortical functions, but
did not include items sensitive to cortical dysfunction.
To address the need for a more comprehensive but
still practical tool for cognitive assessment we devel-
oped the Parkinson’s Disease-Cognitive Rating Scale
(PD-CRS), a new PD-specific cognitive scale aiming to
capture the whole spectrum of cognitive functions
impaired over the course of PD. The information pro-
vided by the assessment of fronto-subcortical and corti-
cal cognitive functions may help to increase the sensi-
tivity and specificity to diagnose PDD, to separate sub-
groups of patients according to their pattern of
cognitive impairment since the early stages of the dis-
ease, and to detect those subjects with a higher risk to
eventually develop dementia.
PATIENTS AND METHODS
Ninety-two patients with idiopathic PD fulfilling
recruited from a sample of outpatients regularly attend-
ing the Movement Disorders Clinic at Sant Pau Hospi-
tal, Barcelona. Each patient was interviewed regarding
disease onset, education level, medication history and
current medications and dosage.18PD patients were at
stable doses of medication the 4 weeks before inclu-
sion and during the study.
Motor status and disease severity were assessed by
the Unified Parkinson’s Disease rating scale (UPDRS)
and Hoehn and Yahr scale (H&Y).19In accordance with
published research criteria, we used the 1-year rule to
distinguish between PDD and DLB.20,21All experimen-
tal participants scored <4 on the Ischemia Score of
Hachinski et al.22to rule out vascular dementia. To
screen for mood disorders, individuals with a cut-off
score ‡11 in the Hospital Anxiety and Depression Scale
(HADS)23were excluded, as were those with abnormal-
ities on brain CT or MRI in the past 12 months, abnor-
mal blood tests or non-compensated systemic diseases,
or inability to read or understand Spanish.
Patients were classified as cognitively intact (CgInt),
PD with mild cognitive impairment (MCI) or PD with
dementia (PDD). Intact cognition was diagnosed when
patients had a score of 0 on the Clinical Dementia Rat-
ing Scale (CDR),24MCI when the score was 0.5, and
PDD wshen the score was ‡one and when they met
294.1 criteria for PDD on the Diagnostic and Statistical
Manual of Mental Disorders, revised Fourth Edition
(DSM IV-TR).25Patients with motor fluctuations were
examined during the ‘‘on’’ state.
Sixty-one age-, sex- and education-matched healthy
subjects, most of whom were spouses or caregivers of
the patients, served as the control group. None of the
controls had cognitive complaints or prior history of
cardiovascular risk factors or neurological or psychiat-
Informed consent to participate in the study was
obtained from all patients or caregivers, as appropriate,
and from controls. The study was approved by the
Local Ethics Committee.
Procedure and Assessment of the Psychometric
Properties of the Scale
The PD-CRS items were selected to cover the full
spectrum of cognitive changes seen in PD. We divided
the items as either ‘‘subcortical-type’’ or ‘‘cortical-
type’’, depending on the neural correlates reported in
previous neuropsychological and neuroimaging studies.
Description and rationale for the selection of the PD-
CRS items included in the initial version of the scale
(content validity) is provided in E-Appendix 1.
The initial PD-CRS included 10 ‘‘subcortical-type’’
items (attention, working memory, Stroop test, phone-
mic, semantic, alternating, and action verbal fluencies,
immediate and delayed verbal memory, clock drawing),
and two ‘‘cortical-type’’ items (naming, copy of a clock).
Each patient was scheduled for four visits. At the first
visit, a neurologist (JP) administered the Mattis Demen-
tia Rating Scale (MDRS)26and classified patients into
cognitive groups according to the CDR. A neuropsy-
chologist (GL) blinded to the MDRS and CDR scores
administered a comprehensive neuropsychological bat-
tery with validated cognitive tasks that assessed the
same cognitive domains as those evaluated by the PD-
CRS. To assess concurrent validity and test-retest reli-
ability, the same neuropsychologist (GL) administered
the PD-CRS at the second and third visits. Inter-rater
reliability was assessed by another neuropsychologist
(CG) at the fourth visit. The interval between first and
second visits was 2 weeks and the interval between sec-
ond, third and fourth visits was 6 6 2 weeks.
Concurrent validity was assessed with the following
comparisons: total PD-CRS scores with total MDRS
scores; attention and working memory with the digit
span forward and backward subtests of the WAIS-III27;
verbal fluencies with the verbal fluencies in the WAIS-
999 COGNITIVE RATING SCALE FOR PD
Movement Disorders, Vol. 23, No. 7, 2008
III27; immediate and delayed verbal memory with the
Rey Auditory Verbal Learning Test28; naming with the
Boston Naming Test29; and drawing and copy of a clock
with the Judgement of Line Orientation Test (JLOT).30
Test-retest and inter-rater reliability were measured
for both total and individual item scores. Internal con-
sistency was assessed with Cronbach’s a. Intraclass
correlation coefficients31were used to calculate con-
current validity and reliability analysis.
Acceptability was considered appropriate for each
PD-CRS item if there was <5% of missing values and
<15% of the respondents with the lowest and highest
possible scores (floor and ceiling effect).32
An initial discriminative validity analysis was car-
ried out to determine the ability of the PD-CRS items
to differentiate between PD cognitive groups. One-way
ANOVA, the Kruskal-Wallis test, ANCOVA with age,
education, and the motor part of the UPDRS as covari-
ates, and logistic regression analysis were used.
To design the final version of the PD-CRS we
selected those items with an appropriate concurrent va-
lidity, reliability, and acceptability that showed the
best discriminative ability between cognitive groups.
Once the final version of the PD-CRS was constructed,
we calculated the discriminative properties of the total,
subcortical and cortical PD-CRS scores, and the valid-
ity and reliability of PD-CRS and individual item
scores. Finally, ROC curves were constructed to assess
PD-CRS screening test accuracy for dementia in PD.
Significance was set at P < 0.05 for all the analyses,
performed with the SPSS 13.0 statistical software.
Ninety-two PD patients and 61 control subjects par-
ticipated in the study. There were no significant differen-
ces in age, education or gender between PD patients and
controls (Table 1). The study population consisted of
three cognitive groups: 30 CgInt patients (age 64.1 6 9
years, education 11.1 6 5 years); 30 patients with MCI
(age 70 6 7 years, education 9.6 6 5 years); and 32
patients with PDD (age 77.7 6 5 years, education 6.6 6
4 years). One-way ANOVA showed significant differen-
ces between PD groups for both age [F(2, 89) 5 17.4;
P < 0.001] and education [F(2, 89) 5 6.4; P < 0.001].
Clinimetric Characteristics of the PD-CRS
The intraclass correlation coefficient (ICC) of total
scores on the initial version of the PD-CRS showed a
strong concurrent validity with the total score on the
MDRS (ICC 5 0.86). Strong concurrent validity was
also obtained for immediate (0.86) and delayed mem-
ory (0.85), alternating verbal fluency (VF) (0.80),
action VF (0.86), phonemic VF (0.87), semantic VF
(0.85), attention (0.80), naming (0.71), and both draw-
ing (0.71) and copy (0.73) of a clock. Scores on work-
ing memory showed a moderate concurrent validity
with digit span backward scores (0.64).
The ICC of individual items, and total, cortical and
subcortical scores on the initial version of the PD-CRS
showed both a high test-retest and a high inter-rater
reliability.33When evaluating test-retest reliability, the
ICC of individual items ranged from 0.84 to 0.91, and
the ICC of total, subcortical, and cortical scores were
?0.90. Evaluation of the inter-rater reliability showed
the ICC of individual items to range from 0.77 to 0.88,
and that of the total, subcortical and cortical scores
were ‡0.93. The PD-CRS showed a high internal con-
sistency (Cronbach’s a 5 0.85). The corrected item-
total correlations for the PD-CRS ranged from 0.73
(naming) to 0.87 (working memory). No item improved
Cronbach’s a if removed.
The percentage of missing values was <5% for all
items. No floor effect or outliers were observed for any
TABLE 1. Demographic and clinical characteristics
of matched PD and CG
(n 5 92)CG (n 5 61)
PD duration (yr)
Hoehn & Yahr (%)
Total LED (mg/day)
71.2 6 9.1
8.9 6 5.3
123 6 17
8.2 6 5
69 6 8.0
10.3 6 4
138.3 6 2.9
25.6 6 12
609.7 6 408
163.1 6 168
774.7 6 460
Values are expressed as mean 6 SD, or percentage of subjects (%).
PD: Parkinson’s disease; CG: Control Group; MDRS: Mattis de-
mentia rating scale; UPDRS: Unified Parkinson’s disease rating
scale; DA: Dopamine agonists; LED: levodopa equivalent dose.
aT-tests for independent samples.
1000J. PAGONABARRAGA ET AL.
Movement Disorders, Vol. 23, No. 7, 2008
cognitive item. When evaluating the PD group as a
whole, a ceiling effect (>15% of the respondents with
the highest possible score) was observed in naming
and the copy of a clock. However, this effect was
eliminated when we analyzed separately the PDD
group, in which no floor or ceiling effect was shown.
Therefore, none of the items was initially deleted.
Univariate Analysis. Both the one-way ANOVA
and Kruskal-Wallis test analysis showed significant dif-
ferences between controls, CgInt, MCI, and PDD
groups for all the PD-CRS items (all P-values <
0.001). Since age, education, and the motor part of the
UPDRS [F(2, 89) 5 28.7; P < 0.001] were signifi-
cantly different between cognitive groups in the uni-
variate analysis, we conducted an ANCOVA analysis
with age, education, and motor function as covariates.
Tests of between-subjects effects showed all the PD-
CRS items (P < 0.001) to be significantly different
between cognitive groups (Table 2). In the post hoc
comparisons, working memory was the only item to
differentiate controls from CgInt, all the ‘‘subcortical-
type’’ items -but neither naming nor the copy of the
clock- differed MCI from controls, MCI differed from
the CgInt group by the alternating VF and working
memory, and both the ‘‘cortical-type’’ and ‘‘subcortical-
type’’ items differentiated PDD from MCI, CgInt and
controls (see Fig. 1). Thus, ‘‘cortical-type’’ item scores
were not significantly different between the control,
CgInt and MCI groups, but selectively helped to differ-
entiate PDD from each cognitive group. A post hoc
analysis considering only PD patients with mild demen-
tia (CDR 5 1; n 5 10) also showed this group to score
significantly lower than the MCI group in the two
‘‘cortical-type’’ items [naming (P < 0.001), copy of the
clock (P 5 0.004)], and in action VF (P 5 0.01).
No evidence of heteroscedasticity was found when
examining the residuals for each cognitive item in the
Multivariate Analysis. Stepwise logistic regression
analysis (forward: conditional) showed that naming
(P 5 0.046; OR 5 0.18, CI95% 0.32–0.96), action VF
(P 5 0.034; OR 5 0.21, CI95% 0.05–0.89), and imme-
diate memory (P 5 0.02; OR 5 0.06, CI 95% 0.01–
0.36) independently differentiated PDD from the PD-
ND group. The MCI group was independently differen-
tiated from CgInt patients by the alternating VF (P 5
0.008; OR 5 0.79, CI 95% 0.66–0.94) and delayed
memory (P 5 0.04; OR 5 0.75, CI 95% 0.56–0.99).
Selection and Clinimetric Assessment of the Final
Version of the PD-CRS. For their greater ability to
discriminate between cognitive groups, alternating and
action verbal fluencies were finally selected. The com-
puterized version of the Stroop test did not displayed
appropriate discriminative properties, so that this item
was excluded from the final version of the scale.
Total scores of the final version of the PD-CRS
showed a strong concurrent validity with the total
MDRS scores (ICC 5 0.87, CI 95% 0.82–0.90). The
individual items, total, cortical and subcortical scores
of the final version of the PD-CRS showed also a high
test-retest and a inter-rater reliability, with ICC ranging
from 0.75 to 0.94, as well as a high internal consis-
tency (Cronbach’s a 5 0.82).
TABLE 2. Analysis of covariance (ANCOVA) between controls and PD cognitive groups (CgInt, MCI, and PDD), with age,
education, and UPDRS-III as covariates
Post-hoc analysis (significant relationships)
Immediate verbal memory
Delayed verbal memory
Copy of a clock
MCI vs. controls
-Immediate memory (P 5 0.016) -Delayed memory (P < 0.001)
-Phonemic VF (P < 0.001) -Semantic VF (P 5 0.001)
-Alternating VF (P < 0.001) - Action VF (P < 0.001)
-Attention (P 5 0.001) -Working memory (P < 0.001)
Clock drawing (P 5 0.03)
MCI vs. CgInt
-Alternating VF (P 5 0.006)
-Working memory (P 5 0.04)
PDD vs. MCI
All ‘‘subcortical-type’’ items (P < 0.003)
-Naming (P 5 0.0007)
-Copy of a clock (P < 0.001)
PD: Parkinson’s disease; CgInt: Cognitively intact; MCI: mild cognitive impairment; PDD: Parkinson’s disease with dementia; PD-CRS: Par-
kinson’s Disease-Cognitive Rating Scale; CDT: Clock Drawing Task.
1001COGNITIVE RATING SCALE FOR PD
Movement Disorders, Vol. 23, No. 7, 2008
One-way ANOVA and Kruskal-Wallis test analysis
showed significant differences between controls, CgInt,
MCI, and PDD groups for total (P 5 0.0002;), cortical
(P 5 0.0001), and subcortical (P 5 0.0009) PD-CRS
scores (see Fig. 2). In the ANCOVA analysis, both
total and subcortical PD-CRS scores did not separate
controls from CgInt patients, but separated controls
and CgInt from MCI, and MCI from PDD patients.
PD-CRS cortical scores differentiated PDD from MCI
and CgInt, but did not differentiate MCI from controls
or CgInt patients. All these relationships had a signifi-
cance level of P < 0.01.
In the multivariate analysis, PDD were independently
differentiated from the PD-ND group by the PD-CRS
total score (P 5 0.0002; OR 5 0.79, CI95% 0.70–0.89).
Then, we used Receiver Operating Characteristic (ROC)
curve analysis to determine the optimal cutoff score for
the screening of dementia in our sample. ROC curve
showed that a cut-off score of ?64 on the PD-CRS total
score yielded high sensitivity (94%) and specificity
(94%), and positive and negative predictive values (PPV
91%, NPV 96%). The area under the ROC curve was
0.98 (CI 95% 5 0.96–0.99). ROC curve analysis to dis-
criminate MCI from CgInt patients yielded moderate
sensitivity and specificity for total PD-CRS scores (sensi-
tivity 73%, specificity 84%) or subcortical PD-CRS
scores (sensitivity 77%, specificity 71%).
The overall duration of the final version of the PD-
CRS was 16 6 3.6 min in the PD-ND group and 24 6
7.8 min in the PDD group. The content, instructions
FIG. 1. Comparative progression
of impairment of ‘‘subcortical-
type" (A, B) and ‘‘cortical-type’’
(C,D) items in controls and PD
cognitive groups, showing an ab-
rupt decrease in ‘‘cortical-type’’
items scores in PDD. CG: control
group; MCI: mild cognitive im-
pairment; PDD: Parkinson’s dis-
ease with dementia.
FIG. 2. Comparison of PD-CRS total scores between cognitive
groups and controls (ANOVA, F[3, 149] 5 128.2; P 5 0.0002). *Post
hoc significant differences from controls (P < 0.01). y Post hoc signif-
icant differences from CgInt (P < 0.01). § Post hoc significant differ-
ences from MCI (P < 0.01). The box plots show the median values
(center line of box), the 25th (lower line of box), 75th (upper line of
box), 10th (lower T bar), and 90th centiles (upper T bar) in each
group. Open circles (*) indicate mild outliers (1.5–3 interquartile
range). CG: Control Group; CogInt: Cognitively intact; MCI: mild
cognitive impairment; PDD: Parkinson’s disease with dementia.
1002 J. PAGONABARRAGA ET AL.
Movement Disorders, Vol. 23, No. 7, 2008
and scoring of the final version of the PD-CRS are
provided in the E-Appendix 2.
The main results of our study show that the PD-
CRS: (1) is a valid, reliable and useful neuropsycho-
logical battery that accurately diagnoses PDD; (2)
detects mild fronto-subcortical deficits in PD-ND
patients; and (3) shows that the transition from MCI to
PDD is characterized by the addition of ‘‘cortical-
type’’ cognitive defects upon a progressive and pre-
dominant fronto-subcortical impairment.
The PD-CRS showed a strong concurrent validity
with the MDRS, a test of global cognitive function
that is specially useful in PD,34and with cognitive
tasks widely accepted to assess each cognitive domain
included in our scale. This reflects the ability of the
PD-CRS to measure the cognitive functions impaired
in PD. No floor effect was observed in any item of the
scale, which indicates that patients do not obtain mini-
mum scores before severe cognitive impairment is
reached. When analyzing the PD group as a whole, a
ceiling effect was observed only in the ‘‘cortical-type’’
items. This effect disappeared when the PDD group
was analyzed separately. This indicates the heterogene-
ous distribution of ‘‘cortical-type’’ item scores in our
sample, almost normal in the PD-ND group and
abruptly decreasing in the PDD group.
The discriminative analysis showed the ability of the
PD-CRS to detect the progressive decline in cognitive
function that is characteristic of PD.35Total, cortical,
and subcortical PD-CRS scores, and items assessing
executive function presented a progressive impairment
within PD cognitive groups (see Fig. 2).
Particularly, MCI subjects differed from controls
only in the ‘‘subcortical-type’’ items, whereas the two
‘‘cortical-type’’ items selectively appeared impaired in
the transition from MCI to PDD. Logistic regression
analysis showed the total PD-CRS score and naming to
although cortical PD-CRS score did not differentiate
PDD from PD-ND. We acknowledge that PDD is a
predominantly attentional-executive dementia. Never-
theless, a cortical item such as naming independently
predicted the presence of dementia in our sample. This
reinforces the hypothesis that cognitive decline leading
to PDD is associated with the development of cogni-
tive defects associated with cortical pathology,7and
replicates results showing that PDD subjects have a
higher frequency of aphasic features than PD-ND indi-
viduals.36On the other hand, in a recent community-
based longitudinal study of newly diagnosed PD
patients, picture copying and semantic rather than pho-
nemic verbal fluency appeared as the most significant
neuropsychological predictors of cognitive decline in
early PD-ND.37The higher value of semantic verbal
fluency in predicting cognitive decline suggests that a
breakdown in the semantic system, whose neural sub-
strate is thought to lie within the temporal neocortex,
accounts for a higher risk of developing PDD.38Nam-
ing is very sensitive to the integrity of the semantic
system and imposes fewer demands on effortful self-
initiated retrieval than semantic fluency.38Thus, if con-
firmed that PD-ND with cortical abnormalities repre-
sents a group at risk for PDD,37using a scale with
cortical items could provide us with a useful predictive
instrument for dementia in PD.
Further, detection of cortical cognitive alterations
might help elucidate whether the coexistence of Alz-
heimer-type pathology in PD has a relevant impact on
the pattern and evolution of cognitive impairment in
The finding that alternating VF independently differ-
entiated MCI from CgInt patients replicates previous
results pointing towards this verbal fluency as the most
sensitive task to detect mild cognitive defects in PD-
ND.40Consistent also with previous studies, action VF
appeared as an early indicator of the conversion from
PD-ND to PDD.41Although clock-drawing tasks might
be confounded by PD motor symptoms, ANCOVA
analysis with motor function as a covariate still
showed clock-drawing to differentiate MCI from con-
trols and the copy of a clock to differentiate PDD from
We acknowledge some limitations of our study.
First, as we have not compared the PD-CRS with exist-
ing rating scales for cognitive dysfunction in PD, we
cannot demonstrate that a scale with subcortical and
cortical items performs better than a scale with sub-
cortical items only. Second, the absence of a consensus
on the definition of ‘‘mild cognitive impairment" in
PD42led us to adopt MCI criteria used to classify sub-
jects at risk for Alzheimer’s disease to categorize our
subjects as CgInt or MCI subjects. A more appropriate
definition of MCI specific for PD could improve the
discriminant ability of PD-CRS to screen for MCI in
Overall, our results show that the PD-CRS appears
to be a valid neuropsychological battery specific for
PD. It may prove useful in clinical research since it is
sensitive to mild fronto-subcortical deficits, follows the
progressive impairment of executive function through-
out the course of the disease, and includes ‘‘cortical-
1003COGNITIVE RATING SCALE FOR PD
Movement Disorders, Vol. 23, No. 7, 2008
type" cognitive tasks that may improve detection of
the transition from MCI to dementia. The discriminant
ability to diagnose dementia in PD shown by the PD-
CRS total score in the ROC analysis suggests that this
scale may also be a good instrument for screening pur-
Future prospective studies could assess the ability of
the PD-CRS to detect patterns of cognitive impairment
with a distinct risk to develop dementia from the early
stages of the disease. Further investigation on the
transcultural validation of the PD-CRS, is also war-
Acknowledgments: (1) Work partially supported by public
research Grants from ‘Fondo de Investigaciones Sanitarias’
PI051916 and Centro de Investigaciones Biome ´dicas en Red-
Enfermedades Neurodegenerativas (CIBERNED), and from
La Fundacio ´ La Marato ´ de TV3, Expedient Number 060310.
(2) We thank the assistance and helpful comments given by
Dr. Christopher Goetz and Dr. Glenn Stebbins in writing this
manuscript. We also thank Ignasi Gich for his expert statisti-
1. Aarsland D, Zaccai J, Brayne C. A systematic review of preva-
lence studies of dementia in Parkinson’s disease. Mov Disord
2. Muslimovic D, Post B, Speelman JD, Schmand B. Cognitive pro-
file of patients with newly diagnosed Parkinson disease. Neurol-
3. Janvin C, Aarsland D, Larsen JP, Hugdahl K. Neuropsychologi-
cal profile of patients with Parkinson’s disease without dementia.
Dement Geriatr Cogn Disord 2003;15:126–131.
4. Foltynie T, Brayne CE, Robbins TW, Barker RA. The cognitive
ability of an incident cohort of Parkinson’s patients in the UK.
The CamPaIGN study. Brain 2004;127:550–560.
5. Jacobs DM, Marder K, Cote LJ, Sano M, Stern Y, Mayeux R.
Neuropsychological characteristics of preclinical dementia in Par-
kinson’s disease. Neurology 1995;45:1691–1696.
6. Ballard CG, Aarsland D, McKeith I, et al. Fluctuations in atten-
tion: PD dementia vs DLB with Parkinsonism. Neurology 2002;
7. Galvin JE, Pollack J, Morris JC. Clinical phenotype of Parkinson
disease dementia. Neurology 2006;67:1605–1611.
8. Ito K, Nagano-Saito A, Kato T, et al. Striatal and extrastriatal
dysfunction in Parkinson’s disease with dementia: a 6-[18F]fluo-
ro-L-dopa PET study. Brain 2002;125:1358–1365.
9. Lewis SJ, Dove A, Robbins TW, Barker RA, Owen AM. Cogni-
tive impairments in early Parkinson’s disease are accompanied
by reductions in activity in frontostriatal neural circuitry. J Neu-
10. Hilker R, Thomas AV, Klein JC, et al. Dementia in Parkinson
disease: functional imaging of cholinergic and dopaminergic
pathways. Neurology 2005;65:1716–1722.
11. Dubois B, Pilon B, Lhermitte F, Agid Y. Cholinergic deficiency
and frontal dysfunction in Parkinson’s disease. Ann Neurol
12. Burton EJ, McKeith IG, Burn DJ, Williams ED, O’Brien JT.
Cerebral atrophy in Parkinson’s disease with and without demen-
tia: a comparison with Alzheimer’s disease, dementia with Lewy
bodies and controls. Brain 2004;127:791–800.
13. Nagano-Saito A, Washimi Y, Arahata Y, et al. Cerebral atrophy
and its relation to cognitive impairment in Parkinson disease.
14. Beyer MK, Janvin CC, Larsen JP, Aarsland D. A magnetic res-
onance imaging study of patients with Parkinson’s disease with
mild cognitive impairment and dementia using voxel-based
morphometry. J Neurol Neurosurg Psychiatry 2007;78:254–
15. Mahieux FMM, Boller F, Fermanian J, Guillard G. Mini-mental
Parkinson: first validation study of a new bedside test constructed
for Parkinson’s disease. Behav Neurol 1995;8:15–22.
16. Marinus J, Visser M, Verwey NA, et al. Assessment of cognition
in Parkinson’s disease. Neurology 2003;61:1222–1228.
17. Hughes AJ, Daniel SE, Kilford L, Lees AJ. Accuracy of clinical
diagnosis of idiopathic Parkinson’s disease: a clinico-pathological
study of 100 cases. J Neurol Neurosurg Psychiatry 1992;55:181–
18. Grosset K, Needleman F, Macphee G, Grosset D. Switching
from ergot to nonergot dopamine agonists in Parkinson’s disease:
a clinical series and five-drug dose conversion table. Mov Disord
19. Hoehn MM, Yahr MD. Parkinsonism: onset, progression and
mortality. Neurology 1967;17:427–442.
20. McKeith IG, Dickson DW, Lowe J, et al. Diagnosis and manage-
ment of dementia with Lewy bodies: third report of the DLB
consortium. Neurology 2005;65:1863–1872.
21. Emre M, Aarsland D, Brown R, et al. Clinical diagnostic criteria
for dementia associated with Parkinson’s disease. Mov Disord
22. Hachinski VC, Iliff LD, Zilhka E, et al. Cerebral blood flow in
dementia. Arch Neurol 1975;32:632–637.
23. Marinus J, Leentjens AF, Visser M, Stiggelbout AM, van Hilten
JJ. Evaluation of the hospital anxiety and depression scale in
patients with Parkinson’s disease. Clin Neuropharmacol 2002;25:
24. Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A
new clinical scale for the staging of dementia. Br J Psychiatry
25. American Psychiatric Association. Diagnostic and statistical man-
ual of mental disorders, 4th ed: DSM-IV. Washington, DC: APA
26. Schmidt R, Freidl W, Fazekas F, et al. The Mattis Dementia rat-
ing scale: normative data from 1,001 healthy volunteers. Neurol-
27. Wechsler D. Wechsler intelligence scale. San Antonio, TX: The
Psychological Corporation; 1997.
28. Schmidt M. Rey auditory and verbal learning test: a handbook.
Los Angeles, CA: Western Psychological Services; 1996.
29. Kaplan EFGH, Weintraub S. The boston naming test. Boston:
Kaplan and Goodglass; 1978.
30. Benton AL HK, deS Hamsher K, Varney NR, Spreen O. Contri-
butions to neuropsychological assessment. New York: Oxford
University Press; 1983.
31. Landis JR, Koch GG. The measurement of observer agreement
for categorical data. Biometrics 1977;33:159–174.
32. McHorney CA, Tarlov AR. Individual-patient monitoring in clin-
ical practice: are available health status surveys adequate? Qual
Life Res 1995;4:293–307.
33. Nunally. Psychometric theory, 2nd ed. New York, NY: McGraw-
34. Aarsland D, Litvan I, Salmon D, Galasko D, Wentzel-Larsen T,
Larsen JP. Performance on the dementia rating scale in Parkin-
son’s disease with dementia and dementia with Lewy bodies:
comparison with progressive supranuclear palsy and Alzhei-
mer’s disease. J Neurol Neurosurg Psychiatry 2003;74:1215–
35. Aarsland D, Andersen K, Larsen JP, Lolk A, Nielsen H, Kragh-
Sorensen P. Risk of dementia in Parkinson’s disease: a commu-
nity-based, prospective study. Neurology 2001;56:730–736.
1004 J. PAGONABARRAGA ET AL.
Movement Disorders, Vol. 23, No. 7, 2008
36. Stern Y, Tang MX, Jacobs DM, et al. Prospective comparative
study of the evolution of probable Alzheimer’s disease and Parkin-
son’s disease dementia. J Int Neuropsychol Soc 1998;4:279–284.
37. Williams-Gray CH, Foltynie T, Brayne CE, Robbins TW, Barker
RA. Evolution of cognitive dysfunction in an incident Parkin-
son’s disease cohort. Brain 2007;130:1787–1798.
38. Hodges JR, Patterson K. Is semantic memory consistently
impaired early in the course of Alzheimer’s disease? Neuroana-
tomical and diagnostic implications. Neuropsychologia 1995;33:
39. Jellinger KA. The morphological basis of mental dysfunction in
Parkinson’s disease. J Neurol Sci 2006;248:167–172.
40. Zec RF, Landreth ES, Fritz S, et al. A comparison of phonemic,
semantic, and alternating word fluency in Parkinson’s disease.
Arch Clin Neuropsychol 1999;14:255–264.
41. Piatt AL, Fields JA, Paolo AM, Koller WC, Troster AI. Lexical,
semantic, and action verbal fluency in Parkinson’s disease with
and without dementia. J Clin Exp Neuropsychol 1999;21:435–443.
42. Dubois B. Is PD-MCI a useful concept? Mov Disord 2007;22:
1005COGNITIVE RATING SCALE FOR PD
Movement Disorders, Vol. 23, No. 7, 2008
Content, instructions, and scoring of the final version of the Parkinson’s
Disease Cognitive Rating Scale (PD-CRS)
Items are administered to the subjects in the same order as presented below.
1. Immediate free recall verbal memory.
Instruction: The subject is asked to read aloud the written words shown on 12
consecutive cards. Three trials are performed, and the subject is asked to recall as many
words as possible after each trial.
Score: 1 point for each word recalled. The highest number of words recalled in any one
trial is the score. (0-12)
2. Confrontation naming.
The subject is asked to name the line drawings shown on 20 consecutive cards. There is
no time limit for response, and only one trial is given. No semantic or phonemic cues are
provided. When objects are included in their context (bib, buckle, mane, hook, jingle bell,
and hoof), the examiner is allowed to indicate the part of the line drawing to be named.
Images (see line drawings at the end of
Score: 1 point for each line drawing correctly named. (0-20)
3. Sustained attention.
Instruction: An ascending series of letters and numbers are read to the subjects. The
subject is asked to report the number of letters in the sequence. Ten series of letters and
numbers are presented, divided into five levels of ascending complexity. Two training
series are provided at the beginning of the test.
2 L T
8 A 9
2 P 6 5 4
3 A 6 K L
B 9 0 4 L T
3 C P 5 7 3
3 9 5 L 4 Z A
I 1 A S Q 4 1
7 5 D A 4 T B 2
9 6 8 4 3 7 L C
Z 4 9 A T D 3 8 4
9 5 M D 4 S C 3 E
Score: 1 point for each correct series. (0-10)
4. Working memory.
Instructions: The examiner reads aloud a randomized list of numbers and letters ranging
in length from 2 to 6 letters and numbers. After each series the subject is asked to repeat
the numbers first, and then the letters. This test ends when the subject is unable to give
the correct answer in two consecutive series. Two training series are provided at the
beginning of the test.
L 2 T
8 A 9
2 L T
8 9 A
2 G 8 M 8 G M
9 I 6 9 6 I
T 0 4 A
7 V 6 J
0 4 T A
7 6 V J
M 6 4 N I
3 5 S G C
6 4 M N I
3 5 S C G
1 R 9 V B 3
M 2 7 4 Z 9
1 9 3 R V B
2 7 4 9 M Z
Score: 1 point for each correct series. (0-10)
Clock drawing task.
5. Umprompted drawing of a clock: The subject is asked to draw a clock face on a
blank sheet of paper, and to set the hands at “twenty-five minutes past ten”. (0-10)
6. Copy drawing of a clock: The patient is asked to copy the presented clock. (0-10)
9 • 3
Yes No Yes No
The figure looks like a clock.
The clock is not divided by lines or sectors.
There is a symmetric disposition of numbers.
Only 1 to 12 numbers are drawn.
Hour numbers are correctly sequenced.
Only two hands are drawn.
Clock hands are represented as arrows.
Hour hand is shorter than minute hand.
No words have been written.
The number ‘25’ has not been drawn.
Score: 1 point for each correct item. (0-10 for each task)
7. Delayed free recall verbal memory.
Instructions: The subject is asked to recall as many words as possible from the list of
words presented at the beginning of the scale.
TIE Download full-text
Score: 1 point for each word recalled. (0-12)
8. Alternating verbal fluency.
Instructions: The subject is asked to alternately generate as many different words as
possible beginning with the letter ‘S’ and words describing articles of clothing during 60
seconds. Participants are instructed not to use proper nouns or to repeat the same word
with a different ending (e.g., swim, swimming, swimsuit).
Score: 1 point for each correct answer maintaining the alternation between words
beginning with ‘s’ and articles of clothing. (0-20)
9. Action verbal fluency.
Instructions: We used the instructions listed in Piatt et al. (reference number 45 in the
manuscript) for the action verbal fluency task. The instructions are as follows: ‘‘During
60 seconds, I’d like you to tell me as many different things as you can think of that
people do. I don’t want you to use the same word with different endings, like eat, eating,
eaten. Also, just give me single words such as eat, or smell, rather than a sentence”.
Score: 1 point for each correct answer. (0-30)