Preliminary Assessment of Cognitive Function in Older Adults by Clock Drawing, Box Copying and Narrative Writing

Abstract

Objective: To examine 3 cognitive screening tasks in older adults and determine which demographic, medical and lifestyle variables obtained 13–17 years previously predicted cognitive performance. Design: Population-based longitudinal cohort study begun in 1981. Setting: Leisure World Laguna Hills, a California retirement community. Participants: The cohort is predominantly white, well educated, upper-middle class; two thirds are women. Data from 1,744 participants (mean age 83 years, range 57–103) were analyzed. Measurements: The baseline postal survey (1981–1985) asked demographic information, medical history, selected drug use and personal habits. The 1998 follow-up included 3 tasks to assess cognitive function: (1) clock drawing, (2) copying a box drawing and (3) narrative writing to describe a pictured scene. Results: Within age groups, women tended to have lower scores than men on box copying, similar scores on clock drawing, but higher informational and complexity scores on narrative writing. Performance decreased with increasing age and was poorest in individuals ≧90 years old. In addition to age, sex and education, significant predictors of performance were: exercise, body mass index and cataract surgery (clock); vitamin E supplements, daily aspirin use and gallbladder surgery (box); vitamin A supplements (narrative writing). Mean scores on all tasks were lower in those later identified as demented compared with nondemented but significantly different only for clock and box drawings. Conclusion: Performance on 3 cognitive tests showed significant effects of age, sex and education and was related to later dementia. Better performance by users of antioxidant supplements and aspirin suggests that these drugs may maintain cognitive function in later life.

Clock-Drawing Test Performance in the Normal Elderly and Its Dependence on Age and Education

“…However, these studies do not, unlike ours, demonstrate the influence of age or education on CDT scores. Interestingly, there was a positive association between CDT performance and education in the above-mentioned studies on subjects belonging mainly to the well-educated upper-middle class [5, 8]. In another study, higher sensitivities and lower specificities were observed among those 85 years old and above as compared with those in other age groups [1], which is in keeping with our findings.…”

“…Our study suggests that this procedure is an oversimplification, and that CDT performance should be interpreted according to a subject’s age and education. This suggestion corroborates the finding that both age and education emerged as predictors of CDT performance in studies on both the normal elderly [1, 2, 5] and impaired populations, such as those hospitalized for rehabilitation following stroke or a hip fracture [18]. However, these studies do not, unlike ours, demonstrate the influence of age or education on CDT scores.…”

“…The high prevalence of these error types suggests that they are normal variations of CDT performance in the normal elderly, and that they do not necessarily indicate significant cognitive impairment. Although these results are not directly comparable, they are qualitatively corroborated by those of a large epidemiological study with 1,629 normal subjects taking the CDT, which revealed that only 42% of them had a perfect performance on a 7-item score and 12% placed the numbers 1–12 with omissions or additions on the clock face [5]. A large mail-survey study by the same authors also showed that a high number of subjects did not correctly position all the numerals [8].…”

“…A large mail-survey study by the same authors also showed that a high number of subjects did not correctly position all the numerals [8]. However, in the 2 latter studies, no accurate diagnostic procedures seemed to have been carried out during the CDT administration, with cognitive impairment being somehow tautologically considered to be present if a subject did not have 3 numbers drawn in the upper-left quadrant of the clock face [5], and the subjects tested may have been different from those in our study. Another study found 48 out of 139 subjects considered as cognitively unimpaired to have pathological CDT performance (8 points or less) according to the Manos 10-point scoring instructions [2].…”

“…Similarly, it is not recommended as a single tool to screen for possible prodromal stages of dementia [2, 3]. Although some suggest that normal CDT performance most likely excludes the presence of even very mild Alzheimer’s disease [4], the opposite may not hold true and recent studies have found a high number of CDT errors in normal elderly subjects [2, 5]. …”