There is conflicting information regarding the recommendations of bracing, physical therapy and cessation of sports for young athletes with symptomatic spondylolysis. The purpose of this study was to identify factors affecting the prognosis and to find the optimal method of non-operative treatment. The patients in our study were athletes who visited our children's hospital for low back pain with lumbar spondylolysis and were treated non-operatively from 1990 to 2002. Clinical and radiological outcomes were reviewed retrospectively. The effects of bracing, physical therapy, cessation of sports, duration of symptoms before the first hospital visit, lateralisation of spondylolysis, age, gender, onset of low back pain after lumbar trauma during sports, bone scan uptake, vertebral level of the lesion, associated scoliosis or spina bifida and radiological bony healing were analysed using univariate and multivariate analysis with logistic regression. The mean age of patients was 13 years (range 7 to 18 years). The mean follow-up was 4.2 years (range 1.2 to 12 years). Of 132 patients, 48 patients had excellent results with no pain during sports, 76 good, 6 fair, and 4 poor. Cessation of sports, early non-operative intervention, and a unilateral spondylolysis appeared to be factors associated with excellent outcomes. However, bracing, physical therapy, age, gender, level of lesion, history of trauma, increased uptake on bone scan, or associated scoliosis or spina bifida were not factors. Bony healing was not related to the clinical outcome. The non-operative treatment of spondylolysis in children can yield excellent clinical outcomes, and the absence of bony healing has no influence on clinical outcome. Factors in this study found to correlate with an excellent outcome include unilateral spondylolysis, acute spondylolysis, and treatment with cessation of sports for 12 weeks