Evaluation of a screening interview for restless legs syndrome

Abstract

We evaluated a fully structured interview for restless legs syndrome (RLS) for potential use in primary care settings and in epidemiological research. Seventy-four veterans were recruited at Veterans Affairs outpatient clinics. The interview was administered telephonically by trained non-clinicians (time 1) and readministered face to face (time 2). A physician conducted gold standard examinations. We calculated sensitivity, specificity and reproducibility. Reproducibility was low (kappa = 0.34, P < 0.01), but was higher for interviews repeated within 1 year (kappa = 0.55, P < 0.01). Including those reporting > or =3 symptoms as cases, sensitivity ranged from 63% (time 1) to 75% (time 2). Specificity ranged from 88% to 71%. The sensitivity and specificity reported here are lower than previously reported in specialty care. This interview for RLS might be useful for preliminary screening of patients with related complaints if followed by additional diagnostic maneuvers or might be used in observational epidemiological research.