Background Many authors have reported about the high prevalence rates of self-reported back pain in children. Nevertheless, little is known about the diagnosis of back disorders - regardless of whether the diagnosis is associated with back pain or not. Therefore, the aim of this study was to analyse the prevalence rates and costs of diagnosis of back disorders in childhood and youth. Methods We conducted a secondary data analysis of a large, population based German data set (2,300,980 insurants of statutory health insurance funds) which allowed for identification of prevalence rates of diagnoses of back disorders in children (age group 0-14 years) and youths (age group 15-24 years) using three digit ICD-10 codes for dorsopathies (M40 - M54: kyphosis and lordosis; scoliosis; spinal osteochondrosis; other deforming dorsopathies; ankylosing spondylitis; other inflammatory spondylopathies; spondylosis; other spondylopathies; spondylopathies in diseases classified elsewhere; cervical disc disorders; other intervertebral disc disorders; other dorsopathies, not elsewhere classified; dorsalgia). Direct treatment costs were calculated based on the real incurred costs for cases with a singular diagnosis of a back disorder. Wherever possible, the results of the random sample were extrapolated to all insurants of statutory health insurance funds (i. e., about 90% of the German population). Results We found prevalence rates for the diagnosis of back disorders to range between 0.01 - 12.5%. "Scoliosis" (M41) and "dorsalgia" (M54) were the most frequent diagnoses in both age groups. Based on these results, it was calculated that in 2002 alone, approximately 1.4 million children/youths in Germany were diagnosed with "dorsalgia" (M54), and that the direct costs for back disorders in childhood and youth accounted for at least 100 million Euros. Conclusions Instead of focusing on the individual, and self-reported disorder or disability, this analysis allowed for the detailed evaluation of medical experts' opinion on back disorders in childhood and youth and for a more objective or public health oriented insight in the topic of diagnosis of back pain and other back disorders. However, due to the methodological limitations by using ICD-10 coding, standardized random validity checks of population based data sets should be mandatory.