Abstract
Background
The authors studied the measurement properties of the Parkinson's Disease‐Cognitive Rating Scale (PD‐CRS) compared with Movement Disorders Society Task Force (MDS‐TF) criteria for the diagnosis of dementia in patients with Parkinson's disease.
Methods
The sample consisted of 223 patients who were diagnosed in accordance with the United Kingdom Parkinson's Disease Society Brain Bank who were assessed with both the MDS‐TF and the PD‐CRS criteria (in addition to other instruments) without the assessors' knowledge of previous results. Internal consistency was studied (homogeneity of the items and Guttmann's λ values were obtained) in addition to convergent, divergent, and discriminative validity. The receiver operating characteristic curve was obtained, and the cutoff point at which the PD‐CRS had the greatest efficiency was analyzed.
Results
The internal consistency was shown to be adequate, with a λ value of 0.821. A floor effect was observed in 4 of the items (Sustained Attention, Working Memory, Immediate Verbal Memory, and Alternating Verbal Fluency), and 1 item showed a ceiling effect (Clock Copying). The scale adequately discriminated patients with and without dementia (Kruskal‐Wallis; P ≤ 0.000). The area under the curve was 0.899. With a cutoff score of 62 (from a possible score of 134), the scale achieved 94% sensitivity and 99% specificity.
Conclusions
The PD‐CRS has adequate measurement properties and is a valid tool for studying the presence of dementia in patients with Parkinson's disease.
Keywords: Parkinson's disease, cognitive deterioration, Parkinson's Disease Cognitive Rating Scale (PD‐CRS), validation
Parkinson's disease (PD) is a degenerative disorder of the central nervous system that causes motor and nonmotor symptomatology; the latter includes cognitive deterioration and dementia in up to 40% of cases. Both PD and associated dementia are independent predictors of an elevated risk of mortality in older adults.1, 2
The Parkinson's Disease Cognitive Rating Scale (PD‐CRS)3 is a scale used to evaluate cognitive deterioration in patients with PD. Its original validation study demonstrated a sensitivity and specificity of 94% and good internal consistency (Cronbach's α = 0.85). It has been externally validated by investigators other than its authors, demonstrating adequate measurement properties4, 5; and, recently,6 it was stated that the PD‐CRS is 1 of the tools designed for evaluating cognitive disorder in patients with PD.
The Movement Disorders Society Task Force (MDS‐TF), has established diagnostic criteria for dementia in PD7 and has developed a method for using this tool.8 The objective of this study was to evaluate the measurement properties of the PD‐CRS compared with the dementia criteria of the MDS‐TF.
Patients and Methods
Patients
A representative sample of 223 consecutive, ambulatory patients were chosen who were diagnosed with PD in accordance with the United Kingdom Parkinson's Disease Society Brain Bank criteria.9 All patients were regularly followed at the Abnormal Movement Unit of the Neurology Service at the Carlos Andrade Marin Hospital in Quito, Ecuador.
The number of patients who refused to participate was not registered, but it was a very small amount. In a previous study using the MDS‐TF criteria, the prevalence of dementia in Parkinson's patients10 was 36.5%. The present study was carried out over 18 months. Exclusion criteria included patients with any serious, acute, concomitant illness; amputated extremity; or sensory deficiencies (significant visual or auditory impairment).
Ethical Aspects
This study was approved by the Head of Teaching and Research at the Carlos Andrade Marin Hospital. All participants provided written informed consent.
Evaluations
Relevant demographic and clinical data were obtained for all participating patients and included age, sex, years of illness, years on levodopa (l‐dopa), and years of education.
All patients were evaluated using the following measures: (1) the Short Parkinson's Evaluation Scale (SPES)/Scales for Outcomes in Parkinson's Disease (SCOPA) (SPES‐Scopa),11 (2) categorization according to Hoehn & Yahr (H&Y) stage,12 (3) the Schwab & England (S&E) measure of daily life activities,13 and (4) the Clinical Impression of Severity Index for Parkinson's Disease (CISI‐PD).14 In addition, all patients were evaluated first using the MDS‐TF criteria,3, 4 which include the Mini‐Mental Status Examination (MMSE), by 1 of the researchers.15
Later, another researcher evaluated the patient in a different room without knowing the previous results, using the following measures: (1) the PD‐CRS5 (possibly with the assistance of the person accompanying the patient), (2) the Hospital Anxiety and Depression Scale (HADS),16 (3) the Parkinson's Impact Scale (PIMS),17 (4) the Scales for Outcomes in Parkinson's Disease (SCOPA) (SCOPA‐Psychosocial),18 and (5) the modified Parkinson's Psychosis Rating Scale (PPRS).19
All patients were evaluated at the time of best functionality (on state). The patients were asked to return 7 days later for the test‐retest (performed by the researcher who evaluated the patient at the first assessment) of the PD‐CRS.
The PD‐CRS5 is a specific neuropsychological battery for evaluating cognitive deterioration in patients with PD. It consists of 9 tests or items: (1) Immediate Verbal Memory (free recall; score range, 0–12), (2) Confrontation Naming (score range, 0–20), (3) Sustained Attention (score range, 0–10), (4) Working Memory (score range, 0–10), (5) Spontaneous Clock Drawing (score range, 0–10), (6) Clock Copying (score range, 0–10), (7) Delayed Verbal Memory (free recall; score range, 0–12), (8) Alternating Verbal Fluency (score range, 0–20), and (9) Action Verbal Fluency (score range, 0–30).
It has a total score of 134 (best score). The sum of the scores for test items 1, 3, 4, 5, 7, 8, and 9 is the subcortical score, and the sum of scores for items 2 and 6 is the cortical score. The maximum subcortical and cortical scores are 104 and 30 points, respectively.
Semantic Equivalence
The PD‐CRS was developed in Castilian Spanish (as spoken in Spain). Prior to this validation, a pilot study was performed with 50 patients to check semantic equivalence. Consequently, the word “taburete” was replaced with “banco” (different words used for “stool”), the word “bombillo” was replaced with “foco” (different words used for “light bulb”), the word “cereza” was replaced with “capuli” (different words used for a small red fruit), and the word “cerrojo” was replaced with “aldaba” (different words used for the “door lock”).
Patients were diagnosed as demented based on the MDS‐TF3, 4 criteria at levels I and II.
Statistical Analysis
Validation of the PD‐CRS
Data Quality and Acceptability
(1) Missing data must not exceed 5%; (2) computable data must be greater than 95%; (3) the mean‐median difference must be ≤ 10% of the maximum score; and (4) the floor and ceiling effects must not exceed 15%.20
Reliability
(1) Internal consistency: The homogeneity index of the items must be ≥ 0.3.21 (2) Reliability: An adequate reliability coefficient was obtained using the suggestions of Feldt and Charter22 and Helms et al.23
As an extension of reliability, we obtained the standard error of measurement (SEM) (SEM = SD × √1 − reliability coefficient), where SD indicates decision time. The SEM was compared with 0.5 of the standard deviation. It was accepted that it must be less, which yielded a precision ≥ 75%.24
Stability
The test‐retest correlation was analyzed using the intraclass correlation coefficient (ICC) for absolute correspondence (2‐way, for randomization purposes).
Validity
Validity was evaluated using: (1) convergent validity,25 (2) discriminant validity,26 and (3) known‐group validity.27 For this study, we used the H&Y stage as a segmentation variable. For statistical analysis using the Kruskal‐Wallis test, P values ≤ 0.05 were accepted as significant.
Other Analyses
The sample was also segmented according to age (into 2 categories: ≤ 65 years and ≥ 66 years), by education (into 2 categories: ≤ 9 years and ≥ 10 years), and by time with the illness (into 2 categories: ≤ 8 years and ≥ 9 years). Either the Mann‐Whitney U test or a t test was performed, depending on whether the variables had a normal distribution. P values ≤ 0.05 were accepted as significant.
A receiver operating characteristic curve was obtained to observe the area under the curve.28 This was performed using the MDS‐TF dementia criteria,3 and the sample was divided into those with and without dementia. The PD‐CRS cutoff score was the point that had the best properties of sensitivity, specificity, and maximum verisimilitude indicators.
Finally, the patients who had PD without dementia were compared with those who had PD with dementia (PDD). Either the Mann‐Whitney U test or the t test was used, depending on whether the variables had a normal distribution (Kolmogorov‐Smirnov test). In either case, P values ≤ 0.05 were accepted as significant.
Results
Seventy percent of the sample was made up of men (156 participants). The distribution of the total number of patients was as follows: 11 patients had in H&Y stage I disease; 49 had stage II disease; 139 (62.33%) had stage III disease; and 24 had stage IV disease.
The patients had the following average values: age, 69.4 years; disease duration, 7.7 years; and dosage of l‐dopa, 716.96 mg/daily. The average scores on the PD‐CRS were as follows: subcortical, 51.5; cortical, 24.4; and total, 76. Most variables did not show a normal distribution, with the following exceptions: age; section Results of the SPES‐Scopa and total SPES‐Scopa scores; the PIMS; and the subcortical, cortical, and total scores on the PD‐CRS (see Table 1).
Table 1.
Demographic and clinical characteristics of the sample
| Characteristic | Median | Mean ± SD | 95% CI (Mean) | Skewness | Kurtosis | CV |
|---|---|---|---|---|---|---|
| Age, y | 71 | 69.4 ± 9.9 | 68.0–70.7 | −0.5 | 0.2 | 0.1 |
| Education, y | 8.0 | 9.2 ± 4.7 | 8.6–9.8 | 0.6 | −0.5 | 0.5 |
| Disease duration, y | 7.0 | 7.7 ± 5.4 | 6.9–8.4 | 1.5 | 3.3 | 0.7 |
| Levodopa | ||||||
| Treatment duration, y | 5.0 | 6.0 ± 4.6 | 5.4–6.6 | 0.9 | 0.4 | 0.7 |
| Dose, mg/d | 750.0 | 716.9 ± 306.7 | 676.4–757.4 | −0.1 | 0.2 | 0.4 |
| SPES‐Scopa | ||||||
| Section 1 | 18.0 | 18.0 ± 5.5 | 17.3–18.7 | 0.2 | −0.5 | 0.3 |
| Section 2 | 8.0 | 9.2 ± 3.5 | 8.7–9.7 | 0.8 | 1.1 | 0.3 |
| Section 3 | 4.0 | 3.6 ± 2.9 | 3.2–4.0 | 0.5 | −0.4 | 0.8 |
| Total | 31 | 30.8 ± 10.9 | 29.4–32.2 | 0.5 | −0.1 | 0.3 |
| Schwab & England | 70.0 | 69.3 ± 15.0 | 67.3–71.3 | −1.1 | 1.3 | 0.2 |
| MMSE | 25.0 | 23.4 ± 4.9 | 22.7–24.0 | −1.9 | 5.2 | 0.2 |
| HADS | ||||||
| Anxiety | 6.0 | 7.3 ± 3.4 | 6.9–7.8 | 0.6 | −0.1 | 0.4 |
| Depression | 7.0 | 6.8 ± 3.7 | 6.3–7.3 | 0.4 | −0.5 | 0.5 |
| PIMS | 19.0 | 19.8 ± 9.4 | 18.5–21.0 | 0.1 | −0.6 | 0.4 |
| SCOPA‐Psychosocial | 9 | 10.2 ± 5.9 | 9.5–11.0 | 0.6 | −0.4 | 0.5 |
| PPRS | 1.0 | 1.5 ± 1.5 | 1.3–1.7 | 1.3 | 1.7 | 0.9 |
| CISI | 11.0 | 11.3 ± 4.4 | 10.8–11.9 | 0.6 | 0.5 | 0.3 |
| PD‐CRS | ||||||
| Subcortical | 51.6 | 51.5 ± 16.3 | 48.6–54.5 | −0.4 | 0.2 | 0.3 |
| Cortical | 24.4 | 24.4 ± 4.2 | 23.9–25.0 | −0.4 | −0.1 | 0.1 |
| Total | 76.0 | 76.0 ± 19.3 | 72.5–79.5 | −0.4 | 0.1 | 0.2 |
SD, standard deviation; CI, confidence interval; CV, coefficient of variation; SPES‐Scopa, Short Parkinson's Evaluation Scale‐Scales for Outcomes in Parkinson's Disease; MMSE, Mini‐Mental Status Examination; HADS, Hospital Anxiety and Depression Scale; PIMS, Parkinson's Impact Scale; SCOPA, Scales for Outcomes in Parkinson's Disease; PPRS, Parkinson's Psychosis Rating Scale; CISI, Clinical Impression of Severity Index; PD‐CRS, Parkinson's Disease Cognitive Rating Scale.
Fifty‐three patients (23.7%) were categorized with dementia in accordance with MDS‐TF criteria.3 The mean time for evaluation with the PD‐CRS was 19.72 minutes (range, 13.89–22.67 minutes). The test‐retest was performed with a mean difference of 6.71 days (range, 5.32–9.54 days).
Data Quality and Acceptability
There were no missing data, and 100% of the data were analyzed. The mean‐median difference surpassed the standard of ≤ 10% of the maximum score, except for the items Sustained Attention and Clock Copying. A floor effect was noted in the items Sustained Attention, Working Memory, Immediate Verbal Memory, and Alternating Verbal Fluency. A ceiling effect was noted in the item Clock Copying (see Table 2).
Table 2.
Characteristic metrics of the items, domains, and total scores on the Parkinson's Disease Cognitive Rating Scale
| Variable | Score | 10% Score Maximum | Floor Effect | Ceiling Effect | ||
|---|---|---|---|---|---|---|
| Mean | Median | Difference: Mean‐Median | ||||
| Immediate Verbal Memory (free recall) | 6.4 | 6.0 | 0.4 | 1.2 | 4.0 | 0.9 |
| Confrontation Naming | 15.4 | 17.0 | −1.6 | 2.0 | 2.2 | 0.9 |
| Sustained Attention | 4.1 | 2.0 | 2.1 | 1.0 | 39.4 | 1.7 |
| Working Memory | 5.2 | 5.0 | 0.2 | 1.0 | 15.7 | 1.7 |
| Spontaneous Clock Drawing | 8.6 | 8.0 | 0.6 | 1 | 6.2 | 6.2 |
| Clock Copying | 7.9 | 9.0 | −1.1 | 1.0 | 4.9 | 32.2 |
| Delayed Verbal Memory (free recall) | 2.7 | 2.0 | 0.7 | 1.2 | 39.4 | 1.7 |
| Alternating Verbal Fluency | 5.1 | 5.0 | 0.1 | 2.0 | 29.6 | 0.9 |
| Action Verbal Fluency | 15.5 | 13.0 | 2.5 | 3.0 | 7.6 | 0.9 |
| Score subcortical | 51.5 | 51.6 | −0.1 | 10.4 | 1.3 | 0.9 |
| Score cortical | 24.4 | 24.4 | 0.0 | 3.0 | 2.2 | 2.6 |
| Total score | 76.0 | 76.0 | 0.0 | 13.4 | 2.6 | 1.7 |
Reliability
Internal Consistency
The homogeneity index of the items was greater than the standard of ≥ 0.3. The reliability index obtained at the end was a Guttman's lambda (Guttman's λ) value of 0.821, which also surpassed the standard minimum of ≥ 0.7; the λ value rose to 0.831 with the elimination of the Action Verbal Fluency item (see Table 3). Regarding precision, we obtained a value for the SEM (SEM = 19.37 × √1 − 0.821 = 8.2) of less than 9.7 (a result of 0.5 × 19.37).
Table 3.
Reliability coefficient and items characteristics
| Guttmann λ = 0.821 | Homogeneity Index | Guttmann λ if Item Deleted |
|---|---|---|
| Immediate Verbal Memory (free recall) | 0.61 | 0.80 |
| Confrontation Naming | 0.70 | 0.78 |
| Sustained Attention | 0.63 | 0.79 |
| Working Memory | 0.70 | 0.79 |
| Spontaneous Clock Drawing | 0.38 | 0.81 |
| Clock Copying | 0.54 | 0.80 |
| Delayed Verbal Memory (free recall) | 0.43 | 0.81 |
| Alternating Verbal Fluency | 0.62 | 0.79 |
| Action Verbal Fluency | 0.71 | 0.83 |
Stability
The test‐retest correspondence was 0.81 (95% confidence interval [CI], 0.77–0.85; f = 5.01; P ≤ 0.000).
Validity
The maximum convergent validity of the PD‐CRS was 0.763 with the MMSE. Overall, the convergence values varied between 0.3 and 0.59 (see Table 4).
Table 4.
Convergent validitya
| Variable | PD‐CRS | Total PD‐CDR | |
|---|---|---|---|
| Subcortical | Cortical | ||
| Age, y | −0.54 | −0.41 | −0.54 |
| Education, y | 0.58 | 0.54 | 0.58 |
| Disease duration, y | −0.27 | −0.14 | −0.24 |
| Levodopa | |||
| Time on treatment, y | −0.21 | −0.74 | −0.18 |
| Dose, mg/d | −0.31 | −0.20 | −0.29 |
| SPES‐Scopa | |||
| Section 1 | −0.48 | −0.40 | −0.47 |
| Section 2 | −0.50 | −0.45 | −0.49 |
| Section 3 | −0.22 | −0.09 | −0.19 |
| Total | −0.47 | −0.38 | −0.45 |
| Schwab & England | 0.51 | 0.47 | 0.50 |
| MMSE | 0.73 | 0.71 | 0.76 |
| HADS | |||
| Anxiety | −0.32 | −0.28 | −0.34 |
| Depression | −0.36 | −0.28 | −0.35 |
| PIMS | −0.34 | −0.21 | −0.33 |
| SCOPA‐Psychosocial | −0.36 | −0.32 | −0.35 |
| PPRS | −0.30 | −0.35 | −0.34 |
| CISI‐PD | −0.59 | −0.51 | −0.59 |
Data are Spearman rho values.
PD‐CRS, Parkinson's Disease Cognitive Rating Scale; SPES‐Scopa, Short Parkinson's Evaluation Scale‐Scales for Outcomes in Parkinson's Disease; MMSE, Mini‐Mental Status Examination; HADS, Hospital Anxiety and Depression Scale; PIMS, Parkinson's Impact Scale; SCOPA, Scales for Outcomes in Parkinson's Disease; PPRS, Parkinson's Psychosis Rating Scale; CISI‐PD, Clinical Impression of Severity Index for Parkinson's Disease.
Known‐Groups Validity
Depending on the H&Y stage, the PD‐CRS scores varied significantly. We also found significant differences between the 2 age groups (P ≤ 0.05) and according to the level of education. We did not find differences according to years with the disease.
The receiver operating characteristic curve showed an area under curve value of 0.89. The PD‐CRS cutoff score that obtained the best efficiency (98%) for differentiating patients with PDD from those without dementia was 62. At that value, we obtained sensitivity of 94%, specificity of 99%, a positive predictive value of 96.7%, a negative predictive value of 98.1%, a positive likelihood ratio of 94, a negative likelihood ratio of 0.06, and a prevalence and bias‐adjusted κ value of 0.96.
When patients who had PPD were compared with those who had PD without dementia, significant differences were found (P ≤ 0.05) in the clinical and demographic variables, with the exception of years on l‐dopa (P ≤ 0.244) and section 3 of the SPES‐Scopa (P ≤ 0.98) (see Table 5).
Table 5.
Comparison of clinical and demographic characteristics of patients with and without dementia
| Characteristic | With Dementia | Without Dementia | P a | ||
|---|---|---|---|---|---|
| Mean | 95% CI | Mean | 95% CI | ||
| Age, y | 77.5 | 75.8–79.1 | 66.8 | 65.4–68.3 | 0.000 |
| Education, y | 7.0 | 6.0–7.9 | 10.0 | 9.2–10.7 | 0.000 |
| Disease duration, y | 10.4 | 8.3–12.4 | 6.9 | 6.2–7.5 | 0.007 |
| Levodopa | |||||
| Time on treatment, y | 7.3 | 5.7–8.9 | 5.7 | 5.0–6.3 | 0.244 |
| Dose, mg/d | 828.7 | 727.5–929.9 | 681.3 | 639.5–723.0 | 0.003 |
| SPES‐Scopa | |||||
| Section 1 | 20.9 | 19.3–22.6 | 17.1 | 16.4–17.9 | 0.000 |
| Section 2 | 11.7 | 10.5–12.9 | 8.4 | 8.0–8.9 | 0.000 |
| Section 3 | 3.7 | 2.8–4.6 | 3.6 | 3.1–4.0 | 0.980 |
| Total | 36.4 | 33.0–39.8 | 29.1 | 27.6–30.5 | 0.000 |
| Schwab & England | 58.9 | 53.4–64.3 | 72.7 | 71.0–74.4 | 0.000 |
| MMSE | 17.6 | 16.0–19.2 | 25.3 | 24.9–25.7 | 0.000 |
| HADS | |||||
| Anxiety | 8.9 | 7.9–9.9 | 6.9 | 6.4–7.4 | 0.001 |
| Depression | 8.9 | 7.8–9.9 | 6.2 | 5.7–6.7 | 0.000 |
| PIMS | 24.8 | 22.3–27.3 | 18.2 | 16.9–19.6 | 0.000 |
| SCOPA‐Psychosocial | 13.3 | 11.4–15.2 | 9.3 | 8.5–10.1 | 0.000 |
| PPRS | 2.6 | 2.1–3.1 | 1.2 | 1.0–1.4 | 0.000 |
| CISI‐PD | 15.0 | 13.6–16.4 | 10.2 | 9.7–10.8 | 0.000 |
| PD‐CRS | |||||
| Subcortical | 22.7 | 18.3–27.2 | 56.2 | 52.4–59.9 | 0.000 |
| Cortical | 19.9 | 17.2–22.5 | 25.0 | 24.1–29.9 | 0.000 |
| Total | 41.7 | 28.5–48.2 | 87.4 | 83.0–91.8 | 0.000 |
P values were determined using a Mann‐Whitney U test or a t test; in either case, P values ≤ 0.05 were considered significant.
CI, confidence interval; SPES‐Scopa, Short Parkinson's Evaluation Scale‐Scales for Outcomes in Parkinson's Disease; MMSE, Mini‐Mental Status Examination; HADS, Hospital Anxiety and Depression Scale; PIMS, Parkinson's Impact Scale; SCOPA, Scales for Outcomes in Parkinson's Disease; PPRS, Parkinson's Psychosis Rating Scale; CISI‐PD, Clinical Impression of Severity Index for Parkinson's Disease; PD‐CRS, Parkinson's Disease Cognitive Rating Scale.
Discussion
When analyzing our results and comparing them with results from previous publications about PD‐CRS (mean values),3, 4, 5, 29, 30 we observed that both age (69.4 years in our patients and 67.4 years in patients from the other publications) and disease duration (7.4 years in our patients and 7.2 years in those reported in previous publications) were very similar.
There was a difference in the H&Y stage of patients; 62.3% of our patients had stage III disease compared with 20.7% of the other patients. This is why the degree of affectation measured with the SPES‐Scopa was an average of 30.8, and that measured with the CISI‐PD was 11.3 versus values of 20.6 and 9.5, respectively, obtained by Martínez‐Martín et al.4
Our patients had a lower educational level. The ratio between patients with high and low educational levels was 2.4/1.0, similar to the ratio reported by Santangelo et al.5 It is known that the educational level has a significant influence on cognitive performance in patients with PD.31, 32, 33, 34
The authors of the original work3 found a ceiling effect in Naming as well as Clock Copying. Santangelo et al.5 reported a ceiling effect in their study on the items of Attention (25.2%), Clock Drawing (26.7%), and Clock Copying (47.5%). We observed this effect in Clock Copying. It did not occur with all of the cortical scores, which did present a ceiling effect in the study by Martínez‐Martín et al.4 It could be concluded that Clock Copying is an item with a tendency to present a ceiling effect; in other words, it is a “simple” item.
In our patients, we found that 4 items had a floor effect (Table 2). We are not aware of this having been reported before. We do not have an explanation for this, except that maybe, for this sample, those items were “complicated” items. Despite that, the difference between mean and median values, which theoretically should not be more than 10% of the maximum score, met that standard, except for Sustained Attention (Table 2).
Test stability, which was measured by concordance using the ICC, obtained an appropriate value of 0.81 (95% CI, 0.77–0.85). The mean application time was 19.7 minutes for patients with and without dementia, which is a moderate application time.
The reliability index—a Guttmann's λ value of 0.821—was slightly lower than that previously obtained3, 4, 5 but was still higher than the standard of 0.7, which is accepted as adequate. Regarding precision, the SEM value obtained (8.2 < 9.7) indicated adequate values, yielding a reliability of greater than 75%.24
Regarding convergent validity, we found an elevated correlation with the MMSE (0.76) in the work by Martínez‐Martín et al.,4 which was 0.53. In general, the values for correlation with tests like the PIMS, the SCOPA‐Psychosocial, the HADS (for both anxiety and depression), and the PPRS were between 0.3 and 0.4; whereas the correlation with the CISI–PD was 0.59. This last correlation value is important, because it could be attributed to the fact that the CISI–PD includes a clinical evaluation of cognitive state in its last item.
Discriminative validity (validity for known groups according to H&Y) demonstrated that the PD‐CRS adequately and significantly discriminated between participants who had less or more of the evaluated attribute, unlike the previous publication in which this was not found.4 As an extension, upon analyzing the participants according to age and education, we observed significant differences between them. It is noteworthy that we did not find a correspondence between these differences and the number of years with the illness that was reported in the Italian validation study.7 This could be because many patients do not remember exactly when their illness began.
The area under the PD‐CRS curve was 0.89 (95% CI, 0.84–0.94), which was adequate and highly discriminative. The cutoff point established as the most efficient for separating patients without dementia from those with dementia was 62. The original publication cited that value as 64, but we do not know what the mean was of the scores obtained. With that cutoff point, the scale reached optimum sensitivity (94%), specificity (99%), and likelihood ratio (96.7) values, which means that the post‐test probability is very high. With that cutoff point, the prevalence of dementia in our study was 23.7%, which is within the estimated prevalence range of 24.5% (95% CI, 17.4%–31.5%).35
It has been stated that the dementia criteria of the Movement Disorders Society are more sensitive than the traditional criteria from the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders.10 On the other hand, it has been said that the MDS criteria are not very sensitive, with 100% specificity but only 46.7% sensitivity.36 That finding has been observed on other occasions37 and by various investigators; Isella et al.38 demonstrated a specificity of 95% and a sensitivity of 65%.
It has been proposed that the MMSE should be replaced by tools specifically developed for evaluating cognitive deterioration in PD,38, 39 or, to gain sensitivity, to exclusively employ level 1 of the MDS criteria.37, 39, 40 Although the MMSE remains 1 of the fundamental elements for diagnosing dementia in the MDS‐TF criteria, from our perspective, it will continue to have low sensitivity if it does not more broadly include subcortical deterioration, which is very important in PD.
This scale requires an average of 19 minutes for its execution, and that time must be taken into account when applying it. This could be an element that complicates its use in general practice.
In conclusion, the PD‐CRS has demonstrated adequate measurement properties for evaluating cognitive deterioration and the presence of dementia compared with the criteria proposed by the MDS‐TF.
Author Roles
1. Research Project: A. Conception, B. Organization, C. Execution; 2. Statistical Analysis: A. Design, B. Execution, C. Review and Critique; 3. Manuscript Preparation: A. Writing the First Draft, B. Review and Critique.
M.S.‐D.: 1A, 1B, 1C, 2A, 2B, 2C, 3A, 3B
M.A.: 1C, 2B, 3B
D.V.: 1C, 3A, 3B
D.G.: 1C, 3B
Disclosures
Funding Sources and Conflicts of Interest: No specific funding was received for this work. The authors declare that there are no conflicts of interest relevant to this work.
Financial Disclosures for the previous 12 months: Marcos Serrano‐Dueñas is a member of the Medical Staff in the Neurological Service at the Carlos Andrade Marín Hospital (Ecuadorian Institute of Social Security) and Professor of the Medicine Faculty in the Pontifical Catholic University of Ecuador. Maite Serrano, Diana Villena, and David Granda receive support from their private practices.
Relevant disclosures and conflicts of interest are listed at the end of this article.
References
- 1. Posada IJ, Benito‐León J, Louis ED, et al. Mortality from Parkinson's disease: a population‐based prospective study (NEDICES). Mov Disord 2011;26:2522–2529. [DOI] [PubMed] [Google Scholar]
- 2. Aarsland D, Zaccai J, Brayne C. A systematic review of prevalence studies of dementia in Parkinson's disease. Mov Disord 2005;20:1255–1263. [DOI] [PubMed] [Google Scholar]
- 3. Pagonabarraga J, Kulisevsky J, Llebaria G, García‐Sánchez C, Pascual‐Sedano B, Gironell A. Parkinson's disease‐cognitive rating scale: a new cognitive scale specific for Parkinson's disease. Mov Disord 2008;23:998–1005. [DOI] [PubMed] [Google Scholar]
- 4. Martínez‐Martín P, Prieto‐Jurczynska C, Frades‐Payo B. Psychometric attributes of the Parkinson's disease‐cognitive rating scale. An independent validation study. Rev Neurol 2009;49:393–398. [PubMed] [Google Scholar]
- 5. Santangelo G, Barone P, Abbruzzese G, Ferini‐Strambi L, Antonini A; IRIS Study Group . Validation of the Italian version of Parkinson's disease‐cognitive rating scale (PD‐CRS). Neurol Sci 2014;35:537–544. [DOI] [PubMed] [Google Scholar]
- 6. Marras C, Troster AI, Kulisevsky J, Stebbins GT. The tools of the trade: a state of the art “How to Assess Cognition” in the patient with Parkinson's disease. Mov Disord 2014;29:584–596. [DOI] [PubMed] [Google Scholar]
- 7. Emre M, Aarsland D, Brown R, et al. Clinical diagnostic criteria for dementia associated with Parkinson's disease. Mov Disord 2007;22:1689–1707. [DOI] [PubMed] [Google Scholar]
- 8. Dubois B, Burn D, Goetz C, et al. Diagnostic procedures for Parkinson's disease dementia: recommendations from the movement disorder society task force. Mov Disord 2007;22:2314–2324. [DOI] [PubMed] [Google Scholar]
- 9. Gibb WR, Lees AJ. The relevance of the Lewy body to the pathogenesis of idiopathic Parkinson's disease. J Neurol Neurosurg Psychiatry 1988;51:745–752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Martinez‐Martin P, Falup‐Pecurariu C, Rodriguez‐Blazquez C, et al. Dementia associated with Parkinson's disease: applying the movement disorder society task force criteria. Parkinsonism Relat Disord 2011;17:621–624. [DOI] [PubMed] [Google Scholar]
- 11. Marinus J, Visser M, Stiggelbout AM, et al. A short scale for the assessment of motor impairments and disabilities in Parkinson's disease: the SPES/SCOPA. J Neurol Neurosurg Psychiatry 2004;75:388–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology 1967;17:427–442. [DOI] [PubMed] [Google Scholar]
- 13. Schwab RS, England AC. Projection technique for evaluating surgery in Parkinson's disease In: Gillingham FJ, Donaldson IML, eds. Third Symposium on Parkinson's Disease. Edinburgh, UK: E&S Livingstone; 1969:152–157. [Google Scholar]
- 14. Martínez‐Martín P, Forjaz J, Cubo E, Frades B; and the ELEP Project Members . Global versus factor related impression of severity in Parkinson's disease: a new clinimetric index (CISI‐PD). Mov Disord 2006;21:208–214. [DOI] [PubMed] [Google Scholar]
- 15. Folstein MF, Folstein SE, McHugh P. “Mini‐Mental State”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–198. [DOI] [PubMed] [Google Scholar]
- 16. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983;67:361–370. [DOI] [PubMed] [Google Scholar]
- 17. Calne S, Schulzer M, Mak E, et al. Validating a quality of life rating scale for idiopathic Parkinsonism: Parkinson's Impact Scale (PIMS). Parkinsonism Relat Disord 1986;2:55–61. [DOI] [PubMed] [Google Scholar]
- 18. Marinus J, Visser M, Martínez‐Martín P, van Hitten JJ, Stiggelbout AM. A short psychosocial questionnaire for patients with Parkinson's disease: the SCOPA‐PS. J Clin Epidemiol 2003;56:61–67. [DOI] [PubMed] [Google Scholar]
- 19. Visser M, Verbaan D, van Rooden SM, Stiggelbout AM, Marinus J, van Hilten JJ. Assessment of psychiatric complications in Parkinson's disease: the SCOPA‐PC. Mov Disord 2007;22:2221–2228. [DOI] [PubMed] [Google Scholar]
- 20. Hays RD, Anderson R, Revicki D. Psychometric considerations in evaluating health‐related quality of life measures. Qual Life Res 1993;2:441–449. [DOI] [PubMed] [Google Scholar]
- 21. Nunnally JC, Bernstein IH. Psychometric Theory. New York: McGraw‐Hill; 1994. [Google Scholar]
- 22. Feldt LS, Charter RA. Estimating the reliability of a test split into two parts of equal or unequal length. Psychol Methods 2003;8:102–109. [DOI] [PubMed] [Google Scholar]
- 23. Helms JE, Henze KT, Sass TL, Mifsud VA. Treating Cronbach's alpha reliability coefficients as data in counseling research. Couns Psychol 2006;34:630–660. [Google Scholar]
- 24. Norman G, Sloan J, Wyrwich K. Interpretation of changes in health‐related quality of life: the remarkable universality of half a standard deviation. Med Care 2003;41:582–592. [DOI] [PubMed] [Google Scholar]
- 25. Fisk JD, Brown MG, Sketris IS, Metz LM, Murray TJ, Stadnyk KJ. A comparison of health utility measures for the evaluation of multiple sclerosis treatments. J Neurol Neurosurg Psychiatry 2005;76:58–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Morgan GA, Gliner JA, Harmon RJ. Understanding and Evaluating Research in Applied and Clinical Settings. Mahwah, NJ: Lawrence Erlbaum Associates Publisher; 2006. [Google Scholar]
- 27. Armitage P, Berry G, Matthews JNS. Statistical Methods in Medical Research, 4th ed Malden, MA: Blackwell Publishing; 2002. [Google Scholar]
- 28. Pepe MS. The Statistical Evaluation on Medical Tests for Classification and Prediction. New York: Oxford University Press; 2003. [Google Scholar]
- 29. Fernández de Bobadilla R, Pagonabarraga J , Martínez‐Horta S , Pascual‐Sedano B , Campolongo A , Kulisevsky J . Parkinson's Disease‐Cognitive Rating Scale: psychometrics for mild cognitive impairment. Mov Disord 2013;28:1376–1383. [DOI] [PubMed] [Google Scholar]
- 30. Kulisevsky J, Fernández de Bobadilla R, Pagonabarraga J, et al. Measuring functional impact of cognitive impairment: validation of the Parkinson's disease cognitive functional rating scale. Parkinsonism Relat Disord 2013;19:812–817. [DOI] [PubMed] [Google Scholar]
- 31. Aarsland D, Janvin C. Clinical aspects of Parkinson dementia Chapter 28. In: Duyckaerts C, Litvan I, eds. Handbook of Clinical Neurology, Vol 89, 3rd series: Dementias. Philadelphia, PA: Elsevier; 2008:303–306. [DOI] [PubMed] [Google Scholar]
- 32. Wang Q, Zhang Z, Lih L, Wen H, Xu Q. Assessment of cognitive impairment in patients with Parkinson's disease: prevalence and risk factors. Clin Interv Aging 2014;9:275–281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. More SF, Barker RA. Predictors of Parkinson's disease dementia: towards targeted therapies for heterogeneous disease. Parkinsonism Relat Disord 2014;20(suppl 1):S104–S107.24262158 [Google Scholar]
- 34. Marder K. Cognitive impairment and dementia in Parkinson's disease. Mov Disord 2010;25(suppl 1):S110–S116. [DOI] [PubMed] [Google Scholar]
- 35. Aarsland D, Kurz MW. The epidemiology of dementia associated with Parkinson disease. J Neurol Sci 2010;289:18–22. [DOI] [PubMed] [Google Scholar]
- 36. Barton B, Grabli D, Bernard B, et al. Clinical validation of Movement Disorder Society‐recommended diagnostic criteria for Parkinson's disease with dementia. Mov Disord 2012;27:248–253. [DOI] [PubMed] [Google Scholar]
- 37. Barton BR, Bernard B, Czernecki V, et al. Comparison of the Movement Disorder Society Parkinson's disease dementia criteria with neuropsychological testing. Mov Disord 2014;29:1252–1257. [DOI] [PubMed] [Google Scholar]
- 38. Isella V, Mapelli C, Siri C, et al. Validation and attempts of revision of the MDS‐recommended tests for the screening of Parkinson's disease dementia. Parkinsoniosm Rel Disord 2014;20:32–36. [DOI] [PubMed] [Google Scholar]
- 39. Kiesmann M, Chanson J‐B, Godet J, et al. The Movement Disorders Society criteria for the diagnosis of Parkinson's disease dementia: their usefulness and limitations in elderly patients. J Neurol 2013;260:2569–2579. [DOI] [PubMed] [Google Scholar]
- 40. Di Battista ME, Giustini P, Bernardi S, Stirpe P, Vanacore N, Meco G. A simplified algorithm may lead to overestimate dementia in PD. A clinical and epidemiological study using criteria for PD‐D proposed by the Movement Disorders Task Force. J Neural Transm 2011;118:1609–1612. [DOI] [PubMed] [Google Scholar]