The efficacy of faecal occult blood test as primary screening for colorectal cancer has been demonstrated. Screening programs, to be effective, should guarantee high compliance in the target population. The aim of this paper is to describe the design of three connected studies aimed at obtaining precise indications for planning a colorectal cancer screening program with high compliance. We designed a survey, with a randomised controlled trial nested within it, and a case-control study nested within that and defined by the results of the trial. The complex interconnection of studies reflects the aim to produce indications for an evidence-based planning of a public health program, which is itself, a complex phenomenon. The trial was designed to evaluate two different types of tests, Immunochemical and Guaiac, and two different providers, general practitioner and hospital, with a 2 x 2 factorial design. The randomization was performed at two different levels to minimize the loss of power: at the practice level for test type (cluster randomisation) and individual level for provider type.