Original Research Article Accepted: December 1, 2010 Published online: January 15, 2011
Dement Geriatr Cogn Disord 2011;31:89–97 DOI: 10.1159/000323317
Does the Clock Drawing Test Predict Dementia? Results of the Leipzig Longitudinal Study of the Aged (LEILA 75+) Lena Ehreke a Melanie Luppa a Hans-Helmut König b Arno Villringer c Steffi G. Riedel-Heller d
a
Public Health Research Unit, Department of Social Medicine, Occupational Health and Public Health, University of Leipzig, Leipzig, b Department of Medical Sociology and Health Economics, University Medical Center Hamburg-Eppendorf, Hamburg, c Max Planck Institute for Human Cognitive and Brain Sciences and Day Clinic of Cognitive Neurology, University of Leipzig, and d Department of Social Medicine, Occupational Health and Public Health, University of Leipzig, Leipzig, Germany
Key Words Clock Drawing Test ⴢ Dementia ⴢ Predictive validity ⴢ Mild cognitive impairment
Abstract Background/Aims: Conversion rates to dementia are known to be high for patients with mild cognitive impairment (MCI), but the diagnosis of MCI is very time-consuming. Since the Clock Drawing Test (CDT) is quick to administer, it will be of interest to compare the predictive validity of the CDT and of an MCI diagnosis for the diagnosis of dementia. Methods: In a sample of 384 individuals, CDT scores and the presence of MCI were assessed at baseline and then compared between individuals with an incident dementia diagnosis at follow-up and those without. Multivariate analyses, receiver operating characteristic analyses and values of sensitivity and specificity of the CDT were performed. Results: Individuals with incident dementia had significantly higher CDT scores at baseline than those without dementia. CDT was a significant predictor of incident dementia after adjusting for other factors. CDT reached a sensitivity of 68% and specificity of 65%. The
© 2011 S. Karger AG, Basel 1420–8008/11/0312–0089$38.00/0 Fax +41 61 306 12 34 E-Mail
[email protected] www.karger.com
Accessible online at: www.karger.com/dem
area under the receiver operating characteristic curve of CDT was 0.70 and therefore slightly lower than for MCI diagnosis (0.78). Conclusions: Because of the only slightly lower predictive value of the CDT, its quick application and scoring compared to the MCI concept applied, it will be worthwhile to improve the CDT scoring system in order to increase the predictive validity in dementia. Copyright © 2011 S. Karger AG, Basel
Introduction
The prevalence and incidence of dementia are high in the elderly worldwide [1]. Being diagnosed with dementia has an enormous impact on patients themselves, their caregivers and society [2]. Yet potential therapy strategies and several promising methods for prevention exist, which are expected to reduce the risk or postpone the onset of dementia [3, 4]. Therefore research aims to identify individuals at risk prior to diagnosis of dementia in order to maximize preventive and treatment efficacy.
Dipl.-Psych. Lena Ehreke, Department of Social Medicine, Occupational Health and Public Health, Public Health Research Unit, University of Leipzig Philipp-Rosenthal-Strasse 55, DE–04103 Leipzig (Germany) Tel. +49 341 972 4591, Fax +49 341 972 4569 E-Mail Lena.Ehreke @ medizin.uni-leipzig.de
Research on the preclinical detection of dementia focused in recent years on the concept of mild cognitive impairment (MCI) [5, 6], a transitional phase between normal aging and dementia. Individuals diagnosed with MCI are more likely to develop dementia than normal control groups [7, 8], showing annual conversion rates between 10 and 40% [9] compared to cognitively healthy controls with a risk of 2% per year [10] and therefore representing a high-risk group for developing dementia. However, there is no general agreement on how MCI should be defined and also be generally operationalized [11]. Moreover, a diagnosis of MCI is mostly time-consuming, because it requires a comprehensive neuropsychological assessment also including neuroimaging and partial investigation of biomarkers. The Clock Drawing Test (CDT) has been widely used to screen for dementia [12], especially because of its easy, quick and simple administration and good acceptance among elderly individuals. To perform the CDT task, different cognitive skills are required, e.g. auditory and visual comprehension, concentration, visuospatial abilities, abstract conceptualization, and also executive control [12, 13]. There are a number of versions, differing with respect to application and evaluation; however, the CDT essentially includes drawing a clock face, adding all numbers and setting a specific time. Recent studies have evaluated the value of the CDT in order to screen for MCI. With the different existing CDT scoring systems the quality of differentiation between individuals with and without MCI has not been sufficient in recent studies [14]. It is still questionable and controversially discussed though whether the CDT is able to identify persons at an early stage of dementia or at high risk of onset of dementia [15–21]. We therefore want to test the predictive validity of the CDT compared to the predictive validity of MCI, on the basis of a representative German population-based sample aged 75 years and older.
Subjects and Methods Sample The data was derived from the Leipzig Longitudinal Study of the Aged (LEILA 75+), a population-based study on the epidemiology of dementia and MCI. At baseline a total of 1,692 individuals aged 75 years and over were included in the sample: 1,500 of these were identified by systematic random sampling from an age-ordered list from the local registry office. Additionally, institutionalized individuals were included in the study by proportion (n = 192), by systematic random sampling from an age-ordered list provided by the four institutions in the study area. The baseline assessment was conducted between January 1997 and June 1998. The study design of the LEILA 75+ and recruitment issues with their influence on
90
Dement Geriatr Cogn Disord 2011;31:89–97
the outcome of the study are described in detail elsewhere [22, 23]. All individuals who were assessed at baseline were requested to take part in up to five follow-up assessments, which were conducted between July 1998 and April 2005, on average every 1.4 years. The sample consisted of all study subjects interviewed face to face in the second follow-up assessment of the LEILA 75+ study, which was conducted between January 2000 and July 2001. We started with the second follow-up, because the CDT was applied first at the second follow-up in the course of the study. Of the 741 study participants interviewed face to face at this time point, the data of 357 study participants were excluded from the following analyses, because 138 subjects suffered from dementia, 167 subjects had incomplete assessments, and 52 study participants had no information at the following assessment (third follow-up). Therefore, results are based on the data of the remaining 384 study participants, representing the baseline sample for this analysis. Development of dementia was investigated at the following assessment (third follow-up) (fig. 1). Instruments Participants were interviewed in their home environment by trained psychologists and physicians. Structured clinical interviews were conducted with the participants. If it was not possible to administer the Structured Interview for the Diagnosis of Dementia of the Alzheimer Type, Multi-Infarct Dementia and Dementias of Other Etiology according to ICD-10 and DSM-IV (SIDAM) [24, 25] at the assessment (e.g. because of death or severe weakness or because relatives of participants refused the administration of the SIDAM on behalf of the elderly person in their care), we offered a fully structured proxy interview in order to obtain information on cognitive and psychosocial functioning as well as subjective memory impairment. Dates of death were obtained from structured proxy interviews and verified with data from the official registry office. SIDAM. The SIDAM consists of a neuropsychological test battery, third-party information on psychosocial impairment and a section for clinical judgment including severity rating of dementia. The neuropsychological test battery covers six areas of cognitive functioning: orientation, memory, intellectual abilities, verbal abilities and calculation, constructional abilities (visual-spatial) as well as aphasia and apraxia. The SISCO score is calculated based on 55 items, including 30 items from the Mini-Mental State Examination (MMSE) [26]. Clinical evaluation and diagnosis were provided by surveying the individuals and, if necessary, informants. Furthermore, a 14-item scale for the assessment of activities of daily living was included (SIDAM-ADL/IADL scale). Data on sociodemographic variables (age, gender, education and marital status), on subjective memory impairment (‘Do you feel like your memory is getting worse?’) and on comorbidity (diabetes mellitus, stroke, myocardial infarction, rheumatism, impairments in hearing and vision) were collected based on a standardized questionnaire by surveying individuals. Clock Drawing Test. More than a dozen versions of the CDT can be found in the literature; they differ in terms of administration and scoring systems [12], however, no version has gained general acceptance. In this study the CDT version of Shulman et al. [27] was used: (1) Individuals were presented a sheet of paper with a predrawn circle 9 cm in diameter, and they were given the following instruction: ‘This is supposed to be a clock. Please put the missing numbers on the clock. Then set the time at 10 past 11.’
Ehreke /Luppa /König /Villringer / Riedel-Heller
Total sample at baseline (n = 1,692)
Investigated at baseline (n = 1,378)
Died before interview (n = 57)
Refused (n = 242)
Not located (n = 15)
Interview at second follow-up (n = 741) + CDT assessment
Died before interview (n = 354)
Refused (n = 163)
Not located (n = 20)
Population at risk of developing dementia (n = 384)
Dementia at second follow-up (n = 138)
Incomplete asssessment (n = 167)
No information at follow-up (n = 52)
Deceased (n = 8)
Refused (n = 40)
Incident dementia cases (n = 28)
Dementia-free survival (n = 356)
Proxy interview (n = 100)
Not located (n = 4)
Fig. 1. Sampling flowchart of the study.
(2) The clock was rated by the investigator according to a modified scoring system of Shulman et al. [27], with scores ranging from 1 to 6: the higher the score, the greater the number of errors and the more severe the impairment. To facilitate the scoring for the interviewer, pictures of clocks for each score were provided. This scoring system was empirically derived and guaranteed a high level of inter-rater reliability [27].
were assessed with the SIDAM-ADL/IADL scale. Subjects with only one impairment or with no impairments on the 14-item SIDAM-ADL/IADL scale were regarded as functionally unimpaired.
Diagnosis of MCI MCI was diagnosed according to the criteria of Winblad et al. [6]. These consensus criteria proposed by the International Working Group on MCI include: (a) absence of dementia according to DSM-IV or ICD-10; (b) evidence of cognitive decline – subjective cognitive impairment (measured by self-rating or informant report) and impairment on objective cognitive tasks and/or evidence of decline over time on objective cognitive tasks, and (c) preserved baseline activities of daily living or only minimal impairment in complex instrumental functions. Dementia according to DSM-IV was excluded with the SIDAM. The criterion of subjective cognitive complaints was fulfilled when the question on subjective memory impairment was positively answered. Data on objective cognitive decline – were obtained from the SIDAM neuropsychological test battery. Impairment in all four cognitive domains was defined as test performance of more than 1 standard deviation below the mean value for age- and education-specific norms [29]. Functional activities
Analysis Differences in sociodemographic characteristics, CDT scores and diagnosis of MCI between individuals with developing dementia and without dementia during both time points were investigated using the two-sided t test, Mann-Whitney U test and corrected 2-test as appropriate. Multiple Cox proportional hazards models were calculated with dementia diagnosis as the outcome variable. Time until dementia diagnosis was therefore calculated in days between the time of baseline assessment and the next follow-up assessment. Diagnosis of dementia was made on average 511 days (standard deviation = 115 days) or 17 months/1.4 years after the initial assessment. For Cox proportional hazards regression, the time of the onset of the disorder was assumed to be midway between baseline and the follow-up when dementia was diagnosed. Besides the CDT score or the MCI diagnosis, the explanatory variables for the multiple Cox proportional hazards models were age, gender, education [30], marital status and living situation. For each variable, hazard ratios and 95% confidence intervals (CIs) were calculated. We provided two kinds of regression models: the first model, using the ‘enter’ method, included all variables (also variables with no significant effect on incident dementia diagnosis, full model), the other model used the ‘forward stepwise’ method and included only variables with a statistically significant contribution to the model (parsimonious model). In addition, the receiver operating characteristic (ROC) of the CDT and the MCI diagnosis was calculated. The sensitivity, specificity and Youden index [31] for optimal cutoff point differentiation were reported. The significance level was set at 0.05 for all analyses.
CDT: Predictive Validity for Dementia
Dement Geriatr Cogn Disord 2011;31:89–97
Diagnosis of Dementia Consensus conferences of physicians and psychologists were held for each subject. The clinical diagnosis was made according to DSM-IV criteria. The cognitive criteria for a dementia diagnosis were based either on cognitive testing by SIDAM [24] or – with proxy interviews – on Clinical Dementia Rating scale data [28].
91
Table 1. Sample characteristics at baseline (n = 384)
Dementia- Test free cases statistic (n = 356, 93%)
Variables
Dementia cases (n = 28, 7%)
Mean age 8 SD years
85.168 83.038 5.03 4.13
Gender, n Female Male
27 (96) 1 (4) 15 (54) 12 (43) 1 (4)
229 (64) 79 (22) 48 (14)
Marital status, n Single Married Divorced Widowed
1 (4) 1 (4) 2 (7) 24 (86)
37 (10) 97 (27) 30 (8) 192 (54)
CDT, MR 8 SD Diagnosis of MCI, n Yes No
t = –2.182
