Validation of the Turkish version of the Lewy body composite risk score.

Abstract

The Lewy Body Composite Risk Score (LBCRS) has been developed to increase the sensitivity and specificity of the diagnosis attributable to Lewy body pathology outside of specialty centers. We aimed to assess the validity and reliability of the Turkish version of the LBCRS in patients with dementia with Lewy Bodies (DLB) and investigate the discriminative power of the test in Turkish patients with Alzheimer's disease (AD) and DLB, and control group. The sample population (n = 512) comprised DLB (n = 113), DLB-Mild Cognitive Impairment (MCI)(n = 12), AD (n = 42), AD-MCI (n = 21), and control group (n = 324). A significant group difference was observed in the Turkish version of the LBCRS scores of the five groups (p < .001). The Cronbach's α value was 0.82 (95% CI: 0.799-0.868). The test-retest reliability score of the scale was r = 0.94 and p < .001. The subscales of the LBCRS (motor and nonmotor subdomains of the disease) were determined to explain 65.961% of the total variance with an eigenvalue >1. In patients with DLB, the cutoff score of ≥3 showed sensitivity (92%) and specificity (81%) (area under the curve [AUC] = 0.883, 95% CI: 0.815-0.951), p < .001) compared with the AD. Compared to the control group, the cutoff score of ≥3 showed a sensitivity of 98% and specificity of 97% (AUC = 0.994, 95% CI: 0.989-0.999, p < .001). The Turkish version of LBCRS permits accurate diagnosis of DLB with high sensitivity and specificity. Also, it can be useful to inform the caregivers regarding the course of the disease during the follow-up.