Validation of the self-completed Cambridge-Hopkins questionnaire (CH-RLSq) for ascertainment of restless legs syndrome (RLS) in a population survey

Abstract

Background and purpose: Epidemiological studies of restless legs syndrome (RLS) have been limited by lack of a well validated patient-completed diagnostic questionnaire that has a high enough specificity to provide a reasonable positive predictive value. Most of the currently used patient completed diagnostic questionnaires have neither been validated nor included items facilitating the differential diagnosis of RLS from conditions producing similar symptoms. The Cambridge-Hopkins diagnostic questionnaire for RLS (CH-RLSq) was developed with several iterations to include items covering the basic diagnostic features of RLS and to provide some basic differential diagnosis. This validation study sought to determine the sensitivity and specificity of the RLS diagnosis based on this questionnaire.

Patients and methods: The CH-RLSq was completed by 2005 blood donors who were asked to consent to being contacted for a telephone diagnostic interview. A scoring criterion was established for ascertainment of RLS based on the clinical definition of the disorder and the exclusion of "mimic" conditions. A weighted sample (N=185) of all completed questionnaires was selected for expert clinical diagnosis of RLS using the validated Hopkins Telephone Diagnostic Interview (HDTI). The telephone interviewers were blinded to all questionnaire responses.

Results: A telephone diagnosis was obtained on 183 of the sample's 185 questionnaires. The questionnaire's normalized sensitivity and specificity were 87.2% and 94.4%, respectively, for RLS compared to not RLS. The positive predictive values in this sample were 85.5%.

Conclusions: The Cambridge-Hopkins RLS questionnaire provides a reasonable level of sensitivity and specificity for ascertainment of RLS in population-based studies.