STUDY DESIGN: Serial screening of Roma children for idiopathic scoliosis. OBJECTIVE: To confirm or reject the observation that the prevalence of scoliosis is reduced in the Roma population and possibly to explain it. MATERIALS AND METHODS: The authors conducted serial screening for idiopathic scoliosis of 1,034 indigenous Roma children (542 boys and 492 girls), aged 4-18 years (857 children were aged 8-18 years) from 1997 to 2011. Age, height, weight, body mass index, years of schooling, and menarche for girls were recorded. Children were clinically examined for body asymmetries and a standing posteroanterior spinal radiograph was obtained in selected cases. RESULTS: Sixty Roma children (6%) had clinical humps. Single humps, according to location, were mostly benign and not related to progressive scoliotic curves. In children with right thoracic humps a left lumbar component could be overlooked. Of 60 children, only 4 (3 girls and 1 boy) with right thoracic and left lumbar or thoracolumbar humps had true progressive scoliotic curves with greater than 10° Cobb angle (prevalence rate, 0.35%). This is 5 times less than the rate of 1.5% in Greek children. One of these 4 children was young and had possible congenital scoliosis; the other 3 were early adolescents. CONCLUSIONS: A substantial difference in the prevalence of scoliosis between Roma and Greek children was documented. The usual percentage of adolescent scoliosis found in the Greek population (approximately 15-17 cases/1,000 children) was not seen in this sample of Roma children.
