|
Clinical Dementia Rating Scale (CDR)
Please note that this section is an archive (last updated in June 2006). [disclaimer]
Sections:
Overview |
Instrument Reviews |
Construct Overviews |
Book Compendium Reviews |
Internet Site Reviews
|
|
|
|
Practical Information
|
Instrument Name:
|
Clinical Dementia Rating Scale (CDR)
|
Instrument Description:
|
The CDR is an instrument for rating the severity of dementia. Ratings are based on separate interviews with the patient and a caregiver, with scores determined in six independently rated categories: memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal care. Also known as the Washington University CDR.
|
Price:
|
Free
|
Administration Time:
|
90 minutes (Ref: 1)
|
Publication Year:
|
1982 (Original)
|
Item Readability:
|
Conducted verbally. Items are slightly technical, but appropriate for clinicians.
|
Scale Format:
|
The original format directs the interviewer to use short, simple cognitive testing of the interviewer’s choice to help determine category scores.
|
Administration Technique:
|
Standardized structured interview with subject and collateral source. Clinical knowledge or training required for those who are not physicians or psychologists. (Ref: 1)
|
Scoring and Interpretation:
|
For each category, a subject is assigned a score of 0 (none) 0.5 (questionable), 1 (mild), 2 (moderate), or 3 (severe), determined by comparing the subject’s characteristics with descriptive guidelines for each score. A Global CDR is then computed from these six category scores using an algorithm in which memory is considered the primary score and all others are secondary. Several algorithms to compute the Global CDR have been developed. (Ref: 4-5) CDR scores computed only from caregiver information are adequately similar to CDR scores computed from both caregiver and patient interviews, and can be used as substitutes if necessary. (Ref: 11)
|
Forms:
|
English, French, Spanish, Italian, Portuguese, Dutch, Czech, Bulgarian, Greek, Hebrew, Chinese, Japanese, and Korean are available through the Consortium to Establish a Registry for Alzheimer’s Disease (CREAD). (Ref: 10) The extended version of CDR (4 – profound, 5 – terminal) (Ref: 8) and a chronic care facility adaptation of CDR (CDR-CC) (Ref: 13) were introduced.
|
|
|
Research Contacts
|
Instrument Developers:
|
Charles P. Hughes
Leonard Berg
Warren L. Danziger
Lawrence A. Coben
Ronald L. Martin
E-mail: morrisj@abraxas.wustl.edu
Phone: (314) 286-2683 or (314) 286-1967
Fax: (314) 286-2448 or (314) 286-1985
Mailing address:
Washington University School of Medicine
660 S. Euclid
Campus Box 8111- MAP
St. Louis, MO 63110
|
Instrument Development Location:
|
Department of Neurology and Neurological Surgery
Washington University School of Medicine
600 S. Euclid Avenue
St Louis, Missouri 63110
Washington University School of Medicine
Department of Psychiatry
600 S. Euclid Avenue
St Louis, Missouri 63110
|
Instrument Developer Email:
|
morrisj@abraxas.wustl.edu
|
Instrument Developer Website:
|
alzheimer.wustl.edu/adrc2/
|
|
|
Annotated Bibliography
|
1. Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. A new clinical scale for the staging of dementia. Br J Psychiatry 1982 Jun;140:566-72. [PMID: 7104545]
 |
Purpose: The authors set out to develop a global, clinical dementia staging and cognitive function assessment. |
|
Sample: 58 controls, 43 subjects with mild dementia, 16 subject with questionable dementia, 18 additional subjects with moderate and three severe dementia subjects. Ages ranged from 65 to 80 and were living in the community. Age, sex and social position were used to match cases to controls. |
|
Methods: Three neurologists and one psychiatrist conducted interviews separately of the subject and a collateral source as outpatients. The subject was also asked to perform verbal tasks from the Boston Diagnostic Aphasia Evaluation. The collateral source was asked to rate the subject’s memory, ability to solve problems, and household tasks, which were then tested on the subject. |
|
Implications: The CDR was reported to be reliable and valid in this population for the purposes of staging dementia. |
2. Burke WJ, Miller JP, Rubin EH, Morris JC, Cohen LA, Duchek J, Wittels IG, Berg L. Reliability of the Washington University clinical dementia rating. Arch Neurol 1988 Jan;45(1):31-2. [PMID: 3337672]
|
Purpose: The authors describe the inter-rater reliability of the CDR in dementia of the Alzheimer type (DAT) research conducted at the Washington University Alzheimer’s Disease Research Center. |
|
Sample: The researchers used 25 videotaped interviews from previous studies of DAT (Berg L et al. 1984, Berg et al. 1982, Storandt M et al., 1984). |
|
Methods: Three neurologists and two psychiatrists reviewed two tapes each first, which were viewed again by a different clinician. Statistical analyses was performed to determine the agreement between interviewers. |
|
Implications: The CDR was shown to be reliable for rating dementia of Alzheimer type by clinicians. |
3. McCulla MM, Coats M, Van Fleet N, Duchek J, Grant E, Morris JC. Reliability of clinical nurse specialists in the staging of dementia. Arch Neurol 1989 Nov;46(11):1210-1. [PMID: 2818255]
|
Purpose: The authors determined the inter-rater reliability of master’s trained nurses in identifying senile dementia of Alzheimer’s type (SDAT). |
|
Sample: Two groups were used that included 44 SDAT patients and 58 controls. The SDAT group consisted of 23 females and mean age (±SD) of 71.5 (± 4.9) years. The control group consisted of 30 females and mean age (±SD) of 71.7 (± 4.9) years. |
|
Methods: An experienced physician trained three nurses to use the Initial Subject Protocol (an assessment tool developed for a larger longitudinal study; Berg L et al. 1982) and employ the CDR scale. The nurses independently reviewed 25 videotaped assessments unaware of the physician-assigned CDR scores. |
|
Implications: The authors show that there was high agreement on CDR scores between physicians and clinical nurse specialists with training. |
4. Gelb DJ, St Laurent RT. Alternative calculation of the global clinical dementia rating. Alzheimer Dis Assoc Disord 1993 Winter;7(4):202-11. [PMID: 8305188]
|
Purpose: The authors claimed that previous algorithms developed by the original developers and the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) resulted in inconsistent ratings. Three scoring methods are evaluated. |
|
Sample: 252 unique CDR determinations were reviewed; 117 were of patients wih probable Alzheimer’s disease and 135 from other dementias. These were collected from the University of Michigan dementia registry data prior to 1 February 1992. |
|
Methods: The three algorithms was published by Hughes et al. (1982); the second was used by CERAD; the third was proposed by this group where the median of all six subscale scores is the global CDR. |
|
Implications: The authors contend that their proposed algorithm should be used over the others since it was free of anomalies, inconsistencies, and indeterminate cases. They point out that theirs is also a simple and straightforward calculation that has shown internally consistency. |
5. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology 1993 Nov;43(11):2412-4. [PMID: 8232972]
|
Purpose: The author describes the recent modifications that have been made to the measure and the scoring used for the CDR in Alzheimer’s-type dementia. |
|
Sample: No sample is mentioned. |
|
Methods: The scoring algorithm is explained. |
|
Implications: A tab-chart is included to explain significances in each domain and impairment. |
6. Juva K, Sulkava R, Erkinjuntti T, Ylikoski R, Valvanne J, Tilvis R. (1995). Usefulness of the Clinical Dementia Rating Scale in Screening for Dementia. Int Psychogeriatr 1995 Spring;7(1):17-24. [PMID: 7579017]
|
Purpose: The authors evaluate the CDR as a screening tool for dementia in a population of 75-, 80-, and 85-year old Finnish people. |
|
Sample: The three age groups were randomly selected from the community and included N=300 in each of the three age groups (75-, 80-, and 85-year-olds; N=900 total). At the end of the study there were 795 subjects due to attrition or death. The second phase involved a sub-sample of 174 subjects (158 following refusal or death). |
|
Methods: Evaluation included a set of questionnaires for the subject and a “close informant” (e.g. questions about memory and functional capacity, ADL, and IADL), interview by a nurse (MMSE), and an examination by a general practitioner who performed the CDR rating. The second phase of the study included a neurological exam of a subsample that included a dementia assessment based on DSM-III-R criteria, clinical history and MMSE. A one-year follow-up of the entire sample took place by a trained nurse, including a new CDR rating. |
|
Implications: The authors found the CDR to have sufficient sensitivity (92%) and specificity (94%) in this population, though the risk of false positives, they contend, needs to be evaluated further. The CDR cut off score was 1. |
7. Dooneief G, Marder K, tang MX, Stern Y. The clinical dementia rating scale: community-based validation of "profound' and "terminal' stages. Neurology 1996 Jun;46(6):1746-9. [PMID: 8649584]
|
Purpose: The authors present the Extended CDR Scale, which includes profound (CDR 4) and terminal (CDR 5) stages, and evidence for its validation. |
|
Sample: The sample was from a community setting of volunteers in the Washington Heights Columbia Aging Project (New York). There were 956 volunteers, whose mean age were 80.9 years, were 31% male, and whose average education was 7.3 years. |
|
Methods: Subjects were evaluated annually (220 were seen twice, 143 three times, etc.) and given a battery of tests to measure functional status (Schwab and England Activities of Daily Living, Blessed Dementia Rating Scale (cognitive and function subscales)) and dementia (CDR). Other neuropsychological tests were also administered. Neuropsychologists and neurologists performed the CDR ratings. |
|
Implications: The authors reported that the extended rating on the CDR was found to be “valid and clinically meaningful.” The subjects with scores at 4 or 5 had poorer prognoses and were more demented. |
8. Fillenbaum GG, Peterson B, Morris JC. Estimating the validity of the clinical dementia rating scale: the CERAD experience. Aging 1996 Dec;8(6):379-85. [PMID: 9061124]
|
Purpose: This is the only identified study in which concurrent validity evidence is provided by neuropsychological tests that were separate from those tests used to determine the CDR rating. |
|
Sample: There were 434 community-dwelling volunteers at entry and 235 of these were available for follow-up at one year. Inclusion criteria were based on scores from the Mini Mental State Exam (<25), the Orientation-Memory-Concentration test (>6), and a scorable response on the CERAD Neuropsychology Battery. |
|
Methods: The authors administered relevant cognitive functioning sections from the above-mentioned tests at baseline and follow-up at one year. |
|
Implications: The CDR was shown to have content and criterion validity, though the authors note that all components measure cognitive function and are thus closely related. |
9. Morris JC, Ernesto C, Schafer K, Coats M, Leon S, Sano M, Thal LJ, Woodbury P. (1997). Clinical dementia rating training and reliability in multicenter studies: the Alzheimer’s Disease Cooperative Study experience. Neurology. 1997 Jun;48(6):1508-10. [PMID: 9191756]
|
Purpose: Describes a training method to increase inter-rater reliability for the CDR, especially when scores are intended to be compared from multiple sites. |
|
Sample: There were 82 investigators from 30 sites. The composition of the raters, all health professionals, included 43% physicians, 17% nurses, 15% PhDs, and 24% other (social workers, psychometrists, or research assistants). |
|
Methods: All were randomly assigned into one of two training groups, either viewing videotapes or reading written interviews of the same three subjects (of CDR 0.5, 1, and 2), and then independently scored the CDR, after which a gold standard was viewed or discussed for those subjects. The participants were once again re-assigned randomly into two workshops where they were shown different videotaped subjects (Group 1: CDR 0, 0.5, 2 and Group 2: CDR 0.5, 2, 3), except that the CDR 2 was the same in each group, and each was scored independently. |
|
Implications: The study found agreement levels among investigators that are comparable to published reliability data. There was no statistically significant difference between MDs and non-MDs in assessing dementia, though CDR-inexperienced raters, the authors state, may benefit from feedback during training. They feel the instrument is capable of being used in multi-center trials as a standardized tool. |
10.
Morris JC. Clinical dementia rating: A reliable and valid diagnostic and staging measure for dementia of the Alzheimer type. Int Psychogeriatr 1997;9 Supl 1:173-6;discussion 177-8. [PMID: 9447441]
|
Purpose: The author describes the CDR, including its development, standardization for multicenter use, validity, recommendations for use, and limitations. |
|
Sample: There is no sample information. |
|
Methods: Only general methods are described. |
|
Implications: The author describes its use among dementia patients and in clinic settings. |
11.
Waite L, Grayson D, Jorm AF, Creasey H, Cullen J, Bennett H, Casey B, Broe
GA. (1999). Informant-based staging of dementia using the clinical dementia rating. Alzheimer Dis Assoc Disord. 1999 Jan;13(1):34-7. [PMID: 10192640]
|
Purpose: The authors examine the validity of the CDR rating when only informant data is utilized. |
|
Sample: Data from 360 subjects was used in this study (out of a possible 630 in the first wave and 449 in the second wave). The sample was drawn from a longitudinal study of older persons in community setting that included a war veterans and widow sample and a local area probability sample. Mean age of the useable sample was 83.8 years and range was 78-79 years. |
|
Methods: Physicians interviewed subjects, who were given a medical and neurological assessment that included medical history, and then completed the CDR using this information only. Social scientists, who were not clinically trained, interviewed the informant (chosen by the subject) with questions that covered the CDR’s domains, and then completed the CDR with only this information. |
|
Implications: There was moderate agreement between the informant and direct interview by a trained clinician. The CDR was found to be valid in this circumstance. |
12.
Oremus M, Perrault A, Demers L, Wolfson C. (2000). Review of outcome measurement instruments in Alzheimer's disease drug trials: psychometric properties of global scales. J Geriatr Psychiatry Neurol. 2000 Winter;13(4):197-205. [PMID: 11128059]
|
Purpose: This is a literature review on the CDR and other global outcome measures appropriate for use in drug trials for Alzheimer’s disease. |
|
Sample: Several study samples are discussed. |
|
Methods: Several study methods are discussed. |
|
Implications: Researchers in this area are asked to consider the psychometric properties of these instruments, as well as their sample sizes, and time between test and retest. |
13.
Marin DB, Flynn S, Mare M, Lantz M, Hsu MA, Laurans M, Paredes M, Shreve T,
Zaklad GR, Mohs RC. (2001). Reliability and validity of a chronic care facility adaptation of the Clinical Dementia Rating scale. Int J Geriatr Psychiatry. 2001 Aug;16(8):745-50. [PMID: 11536340]
|
Purpose: The authors introduce a chronic care facility adaptation of the CDR (CDR-CC) and determine its reliability and validity. |
|
Sample: The sample consisted of 62 residents of a chronic care facility who were already participating in a longitudinal aging and Alzheimer’s disease study. |
|
Methods: The interview was based on the CERAD battery. All raters were trained to be reliable using the Alzheimer’s Disease Clinical Studies group. The differences in the CDR-CC from the original are detailed in the article. |
|
Implications: The CDR-CC was found to be reliable in this sample. |
14.
Juva K, Sulkava R, Erkinjuntti T, Ylikoski R, Valvanne J, Tilvis R. Staging the severity of dementia: comparison of clinical (CDR, DSM-III-R), functional (ADL, IADL) and cognitive (MMSE) scales. Acta Neurol Scand 1994 Oct;90(4):293-8. [PMID: 7839817]
|
Purpose: Presents comparison between the CDR, MMSE, ADL/IADL (functional status measures), and DSM-III-R dementia staging criteria. |
|
Sample: From a random sample of 795 elderly individuals 83% (N=656) participated. Age groups included 75-year-olds (N=240), 80-year-olds (N=214), and 850-year-olds (N=202). |
|
Methods: The subjects were given the CDR, MMSE, Instrumental Activities of Daily Living-scale, and Index of Activities of Daily Living. |
|
Implications: Correlations between the CDR and the DSM-IIIR was described as moderate. The correlation between the CDR and MMSE was described as fair. CDR correlations were better with the ADL (and IADL) than the DSM-IIIR criteria.
|
|
|
top
|
Factors and Norms
|
Factor Analysis Work:
|
All six CDR components loaded on one factor, Lambda=0.82-0.91, explaining 76% of the variance. (Ref: 8)
|
Normative Information Availability:
|
No information found.
|
|
|
Reliability Evidence
|
Test-retest:
|
90 subjects were assigned CDR scores, then were re-interviewed by a second blinded interviewer 6-9 months later. All 50 subjects who received a 0 at baseline were rated as 0 at re-test. Of the 14 who received a 0.5, 9 were rated at the same level at re-test, 1 was rated as a 0, 4 were rated as 1. Of the 26 who received a 1, 19 were rated at the same level at re-test, 6 received a 2, and 1 received a 3. All of these results are consistent with the amount of progression that might be expected over this time period, except for the person moving from a 1 to a 3. (Ref: 1) CDR Memory component score was significantly associated with all six CDR component scores one year later (Spearman correlations ranged from 0.40 - 0.58, p < .0001). (Ref: 8)
|
Inter-rater:
|
All of the following statistics refer to agreement on the Global CDR score. Kendall Tau B Correlation = 0.91, and weighted K (Kappa) = 0.87 between 5 clinicians (psychiatrists and neurologists). (Ref: 2) K=0.77 between two nurses, K=0.75 between nurse and physician. (Ref: 3) K=0.83 between a variety of types of clinicians in a multi-site study. Conclusion is that physicians and non-physicians are equally adept at using the CDR. (Ref: 9)
|
Internal Consistency:
|
In support of the use of a Global CDR score, all six CDR components loaded on one factor, Lambda=0.82-0.91, explaining 76% of the variance. (Ref: 8)
|
Alternate Forms:
|
The extended version of CDR (4 – profound, 5 – terminal) (Ref: 7) and a chronic care facility adaptation of CDR (CDR-CC) (Ref: 13) were introduced.
|
|
|
Validity Evidence
|
Construct/ Convergent/ Discriminant:
|
Correlation among the CDR and other dementia assessments were studied: CDR & Blessed Dementia Scale (cognitive/functional impairment measure)= 0.74 (Ref: 1) and 0.67 (Ref: 7); CDR & Short Portable Mental Status Questionnaire (cognitive impairment measure) = 0.84 (Ref: 1); CDR & Face-Hand Test (cognitive impairment measure) = 0.57 (Ref: 1) ; CDR & Schwab and England Activities of Daily Living Scale = -0.64 (as CDR increases functional independence decreases) (Ref: 7); CDR & Blessed Information-Memory-Concentration Test (cognitive impairment measure) = 0.57 (Ref: 7); CDR & # words learned and # words recalled on CERAD memory test = -0.55, -0.30, respectively (Ref: 8); CDR & Mini-Mental State Examination (MMSE) orientation and non-orientation items = -0.56, -0.57, respectively (Ref: 8); CDR & CERAD Verbal Fluency task = -0.52 (Ref: 8); CDR & Boston Naming Test = -0.49 (Ref: 8). All correlations were significant (p < 0.001). After creating severity categories for the MMSE (none/mild/moderate/severe), the agreement with the CDR was K=0.33 (this fair level of agreement might be due to 17 of 93 subjects were too severely demented to be tested with the MMSE). Agreement with Diagnostic and Statistical Manual III-R (DSM-III-R) based dementia diagnosis was K=0.56. (Ref: 14) Sensitivity/specificity compared to DSM-IIR-R was 0.92/0.94. (Ref: 6)
In support of discriminant validity, CDR was not associated with gender, race, or cause of dementia. (Ref: 7)
|
Criterion-related/ Concurrent/ Predictive:
|
In a study in which patients were re-assessed with the CDR at yearly intervals, Cox proportional hazards models show that there is a linear progression through the levels of the CDR, with the relative risk of moving to the next level and the relative risk of death increasing with the value of the baseline level. Survival decreases with increases in CDR (Ref: 7).
|
Content:
|
No information found.
|
Responsiveness Evidence:
|
1.8-point (McLendon and Doraiswamy) and 2.4-point (Berg et al.) annual increase for untreated, mild to moderate Alzheimer disease patients have been reported. (Ref: 12)
|
Scale Application in VA Populations:
|
No information found; however, Waite et al drew a sample from Australian Veterans. (Ref: 12)
|
Scale Application in non-VA Populations:
|
Yes.
|
|
|
Comments
|
The CDR has been examined extensively with the same general finding that the instrument is valuable for staging the severity of dementia. The instrument has shown responsiveness and predictive validity, as well as reliability and concurrent/discriminant validity. Another significant advantage of this instrument is the support available from the developers. The web site offers a web-based interviewer training program, complete with videos for interviewers to watch and rate to establish inter-rater reliability.
|
Updates
No information found.
|
|
|
|
