US screening for hip dysplasia was performed on 1421 full-term newborns. The study was aimed at: 1) evaluating US feasibility in an unselected maternity ward population; 2) determining the frequency of the different types of hips, and correlating ours with literature data; 3) evaluating the efficacy of both an early prevention and treatment. All US examinations were performed within the first week of life and the 2842 hips classified according to Graf. At birth, normal hips (Ia, Ib) were 2064 (72.6%) (group I); physiologically immature hips (IIa) were 721 (25.4%) (group II); pathological hips (group III) were: 43 (1.5%) IIc, 8 (0.3%) IId, 6 (0.2%) III. Hips in group B were checked at 3 months: 502 were normal, 51 were type IIb, and 1 was type IId: the latter 52 hips were treated and normalized in the following months. Hips in group C (types IIc, IId, and III = 57 hips) were treated and checked every seventh week until normalization. Our experience confirms US value in the evaluation of hip dysplasia in the newborn. In our opinion, the use of this method should be encouraged, although US screening of all newborns remains controversial on a cost-benefit ratio. Multicentric studies will better define US sensitivity, specificity and reliability. The correlation of our results with literature data was difficult, because study populations are not always homogeneous. As for therapy, we found the plastic splint very effective and easy to use. In 2 cases, Milgram devices were used for a few months.