Many snoring children present obstructive symptoms according to their parents. The seriousness of the symptoms and the possibility of the obstructive sleep apnea syndrome (OSAS) in these children may be difficult to judge on the basis of the clinical findings and the patient's history only. In order to evaluate snoring children's relative risk (RR) to have OSAS, their symptoms and signs, as reported by the parents, and clinical findings were compared with the results of overnight polysomnography (PSG). An obstructive apnea index (AI) ≥ 1 in PSG was regarded as the criterion for OSAS. The mean AI was 1.55 (range 0-15), and 29 children had a pathological AI, while 49 had a normal PSG recording. Apneic episodes every night detected by the parents was the most important single risk factor for OSAS (RR 3.6, 95% confidence interval (CI) 1.7-7.7). The RR ratio decreased when apneas appeared less frequently, but any detected apnea was still a single risk factor (RR 1.4, CI 1.2-1.8). The other risk factors of night-time symptoms were constant snoring (RR 1.5, CI 1.0-2.1) and restless sleep (RR 2.1, CI 1.1-4.0). Of the daytime symptoms, absence of excessive sleepiness was a protective factor against OSAS (RR 0.3, CI 0.1-1.0). Previous adenoidectomy was found to be a risk factor (RR 1.7, CI 1.1-2.7), as was tonsillar enlargement (RR 1.4, CI 1.1-1.8). These two findings suggest that the epipharyngeal space does not play a central role in the development of OSAS in children. OSAS cannot be reliably diagnosed without PSG, which is the most important examination for snoring children with obstructive symptoms. For clinical decisions, the consideration of risk factors is essential.