The incidences of neuroblastoma in screened cohorts and unscreened cohorts were compared to evaluate the usefulness of neuroblastoma mass screening. Data for screened and unscreened populations were collected from annual reports from the Neuroblastoma Committee and the number of cases of neuroblastoma with information on the screening history was provided by the Japan Children's Cancer Registry, whose registration rate has been estimated to be 0.71. In the age category of 6-11 months, the ratios of incidences per 10⁶ person-years for unscreened to screened cohorts were 6.58 [95% confidence intervals (CI) 4.51-9.60] during the early period, 1985-1989, and 11.15 (95% CI 7.65-16.25) during the later period, 1990-1994. However, they were 0.67 (95% CI 0.31-1.44) and 0.81 (95% CI 0.33-2.03) for the age categories of 1-4 y and 2-4 y during the early period and 1.32 (95% CI 0.63-2.76) and 1.17 (95% CI 0.48-2.89) for the age categories of 1-4 y and 2-4 y, respectively. The incidences of neuroblastoma per 106 person-years were not different among screened and unscreened cohorts. Neuroblastoma mass screening has not contributed to any decrease in the incidence of neuroblastoma in children over 1 y of age.