Study Objectives: To evaluate leg movement activity during sleep (LMS) in normal school-age children and adolescents, to eventually establish age-specific periodic LMS (PLMS) index thresholds that support the diagnosis of restless legs syndrome (RLS) and to evaluate the utility of other LMS indices. Methods: Polysomnographic recordings from 61 controls, 46 children with RLS, and 44 children with narcolepsy type 1 (NT1) were analyzed for total leg movements (LMS), PLMS, and isolated leg movements (ISOLMS) duration and indices, separately for school-age children and adolescents. Moreover, intermovement interval (IMI) graphs and time-of-night distribution of LMS were analyzed and cut-off thresholds for PLMS index and total LMS index were assessed for the separation of RLS from controls. Results: All indices tended to decrease from school-age children to adolescents in normal controls and in NT1, while in RLS, only PLMS index increased. All school-age children had a similar IMI distribution with a single peak at IMI 2-4s followed by gradual decline. In adolescents with RLS a second peak at IMI 10-50s, was seen. Time-of-night distribution of most indices decreased in RLS while most tended to increase in NT1. A PLMS index cutoff of 2/h best differentiated RLS from controls in school-age children (accuracy 70.0%) and in adolescents (accuracy 70.8%); however, most NT1 subjects also showed PLMS indices higher than this threshold. Conclusions: PLMS index alone does not reliably predict the diagnosis of RLS in children and adolescents. However, analyses of IMI distribution and time-of-night distribution provide additional elements to support a diagnosis of RLS.