Although several procedures for treating scoliosis have been developed, the most effective treatment is still based on early detection. For early diagnosis of idiopathic scoliosis, many authors have proposed methods of school screening; however, there is still no standardized screening program. The aim of this study was to evaluate a school screening method and the prevalence and distribution of scoliosis in Italian school children, aged 9-14 years, and to determine if the screening method can reduce morbidity in an Italian territory. The screening program consisted of three steps: the first step was a clinical examination carried out by the school physician and two specialists. In the second step, doubtful cases (presence of a hump between the two sides of the torso, in the thoracic or thoracolumbar region, measured using a hump meter) were evaluated by an orthopedic specialist and subsequently controlled every 6 months either clinically or by radiographic examination. The third step was the classification of the scoliosis and procedures for treatment. All patients were scheduled for a follow-up program and were evaluated during the subsequent 3 years. Statistical analyses were performed with GraphPad Prism 6. A total of 8995 children were screened for scoliosis. Of these, 487 showed clinical signs of scoliosis, and 181 were referred for anteroposterior radiographs because of a positive result on the forward-bending test (hump>5 mm). No significant statistical difference was observed by the three clinical examiners. Of the 181 patients who were referred, 69 were radiographed, and the clinical diagnosis was confirmed in 94.2% of the cases. The prevalence of scoliosis (defined as a curve of ≥10°) was 0.76% (65 of 8995 children), and most of the curves (44; prevalence 67.69%) were small (<20°). The overall ratio of boys to girls was 1 : 3.3, but varied according to the magnitude of the curve (1 : 3 for curves of <20°, 1 : 3.25 for curves of 20-29°, and 1 : 4 for curves of ≥30°). Double curves were the most common type identified, followed by thoracolumbar curves; specifically, of the 65 children who had a curve, 21 (32.30%) had a double curve, 18 (27.6%) had a thoracolumbar curve, 17 (26.1%) had a lumbar curve, and nine (13.84%) had a thoracic curve. In the following 3 years, only four patients were found to have curves more than 20° and none more than 30°. Our results show that the school screening program was accurate and repeatable. Moreover, screening children for scoliosis using a simple test appears to be an effective means of early detection. Above all, the screening process effectively decreased morbidity in the territory at a negligible cost.