The clock drawing test as a cognitive screening tool for elderly patients in an acute-care hospital.

Abstract

An important part of CGA is the evaluation of cognitive status , which relates to , and may even predict, functional status (1). Research has indicated that 15-30% of elderly persons admitted to hospital suffer from cognitive impairment (2, 3). According to epidemiological surveys, this impairment often goes unrecognized. There are several tools available for evaluating cognitive status (4-6), and the most widely used is the well-validated Mini-mental State Examination (MMSE) (7-9). Unfortunately, many of these tools cannot be administered in acute-care hospitals due to the brevity of hospitalization, and lack of appropriate conditions for administration. The Clock Drawing Test (COT) is a quick, simple , easily-administered, low-cost tool for assessing cognitive status, which can be administered at the patient's bedside. It requires visuo-spatial orientation, concentration , short-term memory and planning ability, and is widely used as a sensitive marker of dementia of the Alzheimer 's type (10-18). The objective of our study was to validate the COT for use with elderly patients admitted to acute-care hospitals, and determine its suitability for use in this setting . in 1991 , a geriatric consultation team (GCT) was established in the Hadassah Medical Center, an acutecare university hospital in Jerusalem, Israel. Each patient aged 75+ admitted to the hospital is screened by a member of the GCT to determine whether he/she requires CGA. The study population comprised elderly patients admitted through the emergency room (the majority to internal medicine departments), and was representative of elderly patients given CGAs. Reasons for non-inclusion included coma, severe cognitive impairment, and an inability to write. The process of administering the COT has been described in the literature (12, 14). Briefly, the patient is given a pencil and paper, and instructed to "draw a clock" (this takes about one minute on average) . On completing the drawing , the patient is asked to indicate "ten past eleven " on the clock. In our study, the test was administered to patients when their acute condition had stabilized. An MMSE was then administered (taking approximately 25 minutes). For scpring, we employed the Clock Drawing interpretation Scale (CDIS) used by Mendez et al. (16) (maximum score = 20), defining three score categories : "severe impairment" (0-10), "moderate impairment" (11-16), and "normal" (17-20). We also defined three categories for the MMSE: "severe impairment" (0-15), "moderate impairment" (16-24), and "normal" (25-30). Pearson correlation coefficients were used to examine the relationship between the COT and the MMSE: CDIS scores were compared to MMSE scores for sensitivity and specificity, and positive and negative predictive values were calculated . The study population comprised 71 elderly patients (43 male). The average age was 81 .7 (SD±5.6) years , and 31 % were aged> 85. Forty-four were not married (the majority widowed). Prior to hospitalization, 30 lived alone, 37 with a spouse or relative, and 4 in sheltered housing projects. The average length of hospitalization was 16.5 (±19.1) days. Prior to hospitalization, 40% of the patients were fully independent in ADL. Positive correlations were found between CDIS and MMSE scores (r=0.76), and between the CDIS score categories and the MMSE score categories (r=0 .66) . We compared the "normal" and "moderate/ severe impairment" scores on the COT with those on the MMSE (Table 1). Discrepancies between the two scores were found in 11 patients: eight patients