The authors examine the problems of planning and conducting a multicentric case-control study on diet and gastric cancer in Italy. The solutions chosen for the study design, cases and controls identification, dietary interview, production of a common protocol for the field work are discussed. Results on the evaluation of the quality and comparability of collected data are presented. Further, compliance of cases and controls to the interview and to the blood and urine sampling with reasons of non-response are shown. Finally, the phases of the study and the methods for improving and controlling homogeneity among Centers are summarized.