Skip to main content
. Author manuscript; available in PMC: 2010 Mar 16.
Published in final edited form as: J Neurol. 2009 Oct 1;257(3):359–366. doi: 10.1007/s00415-009-5324-y

Table 3.

Tests of agreement between clinical diagnoses and pathological diagnoses

Event Sensitivity (%) Specificity (%) kappa* (95% CI)
AD versus others (control or AD + DLB or DLB) 85.0 51.1 0.36 (0.33, 0.39)
AD + DLB versus others (pure AD, pure DLB, control) 12.1 96.0 0.10 (0.05, 0.15)
DLB versus others (control, AD + DLB, pure AD) 32.1 98.3 0.37 (0.29, 0.45)

Statistical comparison of clinical and pathological diagnoses using data from Table 2. The results indicate poor sensitivity for diagnoses involving dementia with Lewy bodies (DLB), but high specificity in the clinical diagnoses. Sensitivity is much higher for clinical diagnoses of pure Alzheimer's disease (AD) whereas specificity is much lower. The kappa coefficients indicate relatively poor agreement levels

*

All kappa coefficients are significant at the p < 0.0001 level