Regular Research Articles
Mapping Scores Onto Stages: Mini-Mental State Examination and Clinical Dementia Rating

https://doi.org/10.1097/01.JGP.0000192478.82189.a8Get rights and content

Objective

Although the clinical course of Alzheimer disease (AD) is gradual, it is useful for a number of reasons to distinguish between different levels of severity. The Clinical Dementia Rating (CDR) has demonstrated high validity and reliability for this purpose, but it requires a considerable amount of data to be collected both from the patient and from an informant. In the present study, the authors mapped Mini-Mental State Examination (MMSE) scores onto CDR categories to determine how well the MMSE performs as a surrogate of the CDR as a timesaving method of staging dementia.

Method

Eight hundred sixty-three probands, including 524 patients with probable AD, 92 patients with questionable dementia, and 247 with memory complaints but no objective cognitive impairment, were included. Cutoff scores were identified on one-half of the sample using a receiver operating characteristic analysis. The cutoff values were then applied to the other half of the sample, and the agreement between MMSE score ranges and CDR stages was determined by calculating Cohen's kappa.

Results

The MMSE discriminated well between CDR stages 0.5, 1, 2, and 3 but performed poorly in the separation between CDR stages zero and 0.5. The MMSE ranges were 30 for no, 26–29 for questionable, 21–25 for mild, 11–20 for moderate, and 0–10 for severe dementia. Substantial agreement between the two instruments was obtained for the categories mild (κ = 0.62, p <0.001, N = 115), moderate (κ = 0.69, p <0.001, N = 114), and severe dementia (κ = 0.76, p <0.001, N = 39), whereas the agreement was moderate for no (κ = 0.44, p <0.001, N = 120) and only fair for questionable dementia (κ = 0.28, p <0.001, N = 42).

Conclusion

The MMSE can be used as a surrogate measure for the CDR for the staging of dementia in AD.

Section snippets

Sample

The study was conducted in a university outpatient clinic for cognitive disorders. Eight hundred sixty-three patients, who sought or were referred for cognitive evaluation, were enrolled, including 524 patients who were diagnosed with probable Alzheimer disease, 92 patients with questionable dementia, and 247 subjects who reported memory loss but performed within age norms on cognitive tests and were unimpaired in activities of daily living. Diagnostic evaluation included informant interview,

RESULTS

The ROC analysis in the training sample showed that the MMSE discriminated well between all CDR stages. MMSE scores of 30, 29–26, 25–21, 20–11, and 10–0 mapped onto CDR categories of no, questionable, mild, moderate, and severe dementia, respectively. Cohen's kappa for the agreement of each MMSE score range and its corresponding CDR stage was 0.44 (N = 126) for MMSE 30/CDR 0, 0.23 (N = 50) for MMSE/29–26/CDR 0.5, 0.47 (N = 93) for MMSE 25–21/CDR 1, 0.67 (N = 129) for MMSE 20–11/CDR 2, and 0.70

DISCUSSION

We correlated MMSE scores with CDR ratings of dementia severity in a large group of subjects who were referred to a university memory clinic for cognitive evaluation. Our findings were relatively consistent with two previous studies that have examined the relationship between the two measures for the categories of no, mild, and moderate dementia (28.9 ± 1.3 standard deviations [SDs] for no, 20.0 ± 3.9 SDs for mild, and 16.1 ± 4.7 SDs for moderate dementia in the first study8; 19.7 ± 3.5 SDs for

References (20)

There are more references available in the full text version of this article.

Cited by (509)

View all citing articles on Scopus
View full text