Terrence Steyer, M.D.
Chair, Department of Family Medicine
Alexopoulos, GS, RC Abrams, RC Young, and CA Shamoian (1988)
The CSDD is a 19-item instrument specifically designed for the rating of symptoms of depression in demented patients. Items were constructed so that they can be rated primarily on the basis of observation. To simplify the use of the scale, the severity of each item is rated according to three explicitly defined grades: absent, mild or intermittent, and severe. The scale is administered in two steps. First the clinician interviews the patient's caregiver on each of the 19 items, and then briefly interviews the patient. The caregiver is instructed to base his report on observations of the patient's behavior during the week prior to the interview. After interviewing the patient, if there are any large discrepancies between the clinician's observations and the caregiver's report, then the clinician will again interview the caregiver to clarify the reason for disagreement. The CSDD is scored on the basis of the clinician's final judgment. Total time for the administration is approximately 30 minutes: 20 minutes with the caregiver and 10 minutes with the patient.
The clinician must be a professional trained in the current concepts of the phenomenology of depression. Further training to use the CSDD is minimal.
The CSDD is specifically designed for elderly patients. In the initial study the median age of the 26 patients was 81 years with a range of 63 to 93. Interrater reliability was judged by comparing scores given by two psychiatrists. Because patients with more severe dementia may be more difficult to score, the patient population was divided into two groups based upon their Mini-Mental State Examination scores (above or below the median). The concordance of the total CSDD scores obtained independently by the two psychiatrists was the same for each group (k=0.6). The measurement of internal consistency yielded a Cronbach alpha of 0.84 suggesting the the CSDD is internally consistent. There was a significant correlation between the total CSDD score and the rank order of the Research Diagnostic Criteria (RDC) measure of depression (r=0.83, p<0.001). In another measure of validity, CSDD scores for demented patients hospitalized for major depression were significantly lower upon discharge than those obtained on admission. Differences in the CSDD score are sensitive to differences in the RDC categories of depression.
The CSDD has not been specifically tested with minority patients. However, since the majority of the assessment is based on the clinician's and the caregiver's observations, and not on the patient's responses to questions, racial differences to the items are likely to be minimal.
Sharp and Lipsky (2002) recommend the CSDD for patients with cognitive deficits.
The CSDD requires more time to complete than many other scales for determining depression. It requires the participation of a caregiver who has thorough knowledge of the patient's status over the previous week. It is more likely to yield meaningful results than measures which rely only on patient responses to questions due to the inability of cognitively deficient patients to respond adequately to other surveys.
The CSDD is available in full as an appendix to the original publication.
Alexopoulos, GS, RC Abrams, RC Young, and CA Shamoian (1988) Cornell Scale for Depression in Dementia. Biol Psychiatry 23, 271-284.
Sharp, LK, and MS Lipsky (2002) Screening for Depression across the lifespan: A review of measures for use in primary care settings. Am. Fam. Physician 66 (6): 1001-1008.