Validation of a new REM sleep behavior disorder questionnaire (RBDQ-HK)

Abstract

OBJECTIVES There are limited screening instruments for diagnosis of REM sleep behavior disorder (RBD) and none for quantifying the severity of disease. We aimed to validate a 13-item self-reported RBD questionnaire (RBDQ-HK) for diagnostic and monitoring purposes. METHODS Based on ICSD-II and our previous clinical and empirical work, the RBDQ-HK questionnaire was designed and administered in patients attending university-affiliated sleep clinic and psychiatric out-patient clinic, and subjects from the general population. ROC curve and exploratory factor analysis were employed to evaluate the scale, which had a score ranging from 0 to 100. RESULTS One hundred and seven RBD patients [mean age 62.6 (15.5) years; male 70.1%] and 107 control subjects [mean age 55.3 (9.0) years, male 57.9%] completed the questionnaire. The diagnoses of all the study subjects were independently ascertained by clinical interview and PSG. RBD patients had a significantly higher total RBDQ-HK score [mean (s.d.): 32.1 (16.1), range 3-71] than the control group [9.5 (10.2), range 0-55] (p<0.005). The RBDQ-HK demonstrated robust psychometric properties with moderate sensitivity (82.2%), specificity (86.9%), positive predictive value (PPV; 86.3%), and negative predictive value (NPV; 83.0%), high internal consistency and test-retest reliability. Exploratory factor analysis revealed two components (dream-related and behavioral factors) that corresponded to the essential clinical features of RBD. The best cut-off for total score (range 0-100) was at 18/19 and the best cut-off for factor 2 (behavioral factors including sleep talking, shouting, limb movements and sleep-related injuries, range 0-70) was at 7/8. CONCLUSIONS The RBDQ-HK has satisfactory validity and reliability as a measure of clinical RBD symptoms and severity. It may serve as an effective tool for diagnosis and evaluation of the disease course to facilitate future clinical and research studies.