Mutations or exon deletions of the epsilon-sarcoglycan (SGCE) gene cause myoclonus-dystonia (M-D), but a subset of M-D patients are mutation-negative and the sensitivity and specificity of current genetic testing criteria are unknown. We screened 46 newly enrolled M-D patients for SGCE mutations and deletions; moreover, 24 subjects previously testing negative for SGCE mutations underwent gene dosage analysis. In our combined cohorts, we calculated sensitivity, specificity, positive and negative predictive values, and area under the curve of 2 published sets of M-D diagnostic criteria. A stepwise logistic regression was used to assess which patients' characteristics best discriminated mutation carriers and to calculate a new mutation predictive score ("new score"), which we validated in previously published cohorts. Nine of 46 (19.5%) patients of the new cohort carried SCGE mutations, including 5 novel point mutations and 1 whole-gene deletion; in the old cohort, 1 patient with a complex phenotype carried a 5.9-Mb deletion encompassing SGCE. Current diagnostic criteria had a poor ability to discriminate SGCE-positive from SGCE-negative patients in our cohort; conversely, age of onset, especially if associated with psychiatric features (as included in the new score), showed the best discriminatory power to individuate SGCE mutation carriers, both in our cohort and in the validation cohort. Our results suggest that young age at onset of motor symptoms, especially in association with psychiatric disturbance, are strongly predictive for SGCE positivity. We suggest performing gene dosage analysis by multiple ligation-dependent probe amplification (MLPA) to individuate large SGCE deletions that can be responsible for complex phenotypes. © 2013 Movement Disorder Society.