Diagnostic criteria in dementia
Diagnosing Dementia- ICD not so bad after all
Background: The discrepancy between results of diagnosing dementia with ICD-10 and DSM-IV has been shown by several studies. Our aim was to show that the two diagnostic systems are more or less alike if ICD-10 is interpreted in the way we believe is in the intention of the ICD-10 authors.
Diagnosing dementia - ICD-10 not so bad after all
A comparison between dementia criteria according to DSM-IV and ICD-10
Mala Naik MD1,2
Harald A. Nygaard MD, PhD1,3
The study was performed in the Geriatric out-patient department Haraldsplass Deaconess Hospital, Bergen, Norway. All the data was collected during routine examination of patients who were referred for dementia assessment.
207 consecutively referred patients and their caregivers were interviewed and the patients were clinically examined. The patients included in the study were recruited from the assessments of 3 doctors, two consultants in geriatric medicine and one experienced resident. In addition the necessary blood tests, psychometric tests and a CT scan of the brain were performed on all patients. For routine purposes we use an algorithm based on the research criteria of the ICD-10 (WHO,1993) where the various symptom categories are listed. Based on the interviews, the clinical examination and results from psychometric tests the physician ticks a box for the presence of the individual criteria. A dementia diagnosis automatically emerges when sufficient criteria are fulfilled. In conjunction with this study a similar algorithm was constructed based on DSM-IV(American psychiatric Association , 1994 ). There were separate forms for each of the two diagnostic systems. By going through the algorithms for each patient we were able to conclude whether the patient had a dementia or not. The following psychometric tests are routinely used : Mini Mental State Examination Folstein et al., 1975) Kendrick object learning test (Kendrick et al, 1979) and the Logical Memory subtest of the Wechsler Memory Scale (Wecshler, 1945) , Trail making A and B (Reitan, 1958), Clock Drawing Test (Sunderland et al., 1989) and part tests of the CAMDEX-battery(Roth et al.,1988) .Experienced nurses perform most of the tests. The criteria for diagnosing dementia according to DSM-IV(American psychiatric Association , 1994 ) and ICD-10 (WHO,1993) respectively are depicted in Table 1.
Descriptive statistics were applied. Agreement between ICD-10 and DSM-IV was assessed by Cohen's kappa. Interpretation of kappa was made according to Daly & Bourke (Daly et al.,2000) : kappa 0.2 = poor, 0.2 < k < 0.4 = fair, 0.4 < k < 0.6 = moderate, 0.6 < k < 0.8 = good, 0.8 < k = excellent agreement. P < 0.05 was considered statistically significant.
A total of 207 patients were eligible for the study. One patient in whom no informant was available was excluded from the analysis. A total of 206 patients, 71 men and 135 women were included in the statistical analysis. The mean age of the study group was 78.3 (SD 6.5), men 78.0 (SD 7.7) and women 78.5 (SD 5.9), p = 0.57). A diagnosis of dementia was made for 198 patients and there was 100% agreement (kappa = 1.0) between ICD-10 and DSM-IV diagnosis. According to ICD-10, 105 patients were mildly, 85 moderately and eight severely demented.