To define further the effectiveness of nonsurgical therapy for idiopathic scoliosis, predefined criteria were established for selection and data retrieval from studies of therapy and natural history, and the results were synthesized quantitatively. Only studies of patients with no more than a 50° Cobb angle scoliosis were considered. Twenty-four reports were eligible. There was a fivefold proportion of failures among patients with scoliosis> 30° at the start of therapy but no difference in progression between different kinds of nonsurgical therapies or between treated and untreated patients; these were the main findings of this quantitative analysis. These data cannot be used to prove the effectiveness or ineffectiveness of nonsurgical therapy for idiopathic scoliosis, and experimental controlled studies of different therapies seem to be justified both on ethical and scientific grounds. The findings of this overview can be used for their planning.