A pilot study has been carried out to assess the feasibility of screening for neuroblastoma in 6-month-old infants in the north of England. A total of 20,829 infants were screened. Two true-positive cases were found, along with eight false-positive and three false-negative cases. It was shown that the concentrations of catecholamine metabolites in the urine are dependent on the creatinine content; centiles have been established to allow this relationship to be taken into account. Five of the eight false-positive cases would have been correctly assessed as normal if the new centiles had been used. Preliminary results lend support to the need for a well-designed study of neuroblastoma screening to be carried out, with death from this disease as the only end point.