Recruitment to trials is known to be difficult. Previous research suggests that a crucial factor may be participants' difficulty with the concept of randomization. This study explored patients' perceptions of randomization and reasons for consent or refusal to participate in the ProtecT study (a randomized trial of surgery, radiotherapy, and monitoring for localized prostate cancer). In-depth interviews were conducted with 21 men diagnosed with localized prostate cancer who were invited to participate in the ProtecT treatment trial. Interviewees were selected purposefully from three U.K. clinical centers to ensure the inclusion of similar proportions of those agreeing or refusing random treatment allocation in each of the treatment groups. Interviews explored men's recall and understanding of chance, comparison, and equipoise, and reasons for consent/refusal of randomization and acceptance/rejection of treatment allocation. Data were analyzed methodically using the techniques of constant comparison. Checking of coding and interpretation was assured by four experienced qualitative researchers. Recall and understanding of the major principles of the randomized design were good and were similar for "chance" and "comparison" between those who consented to and refused randomization. Clinical equipoise, however, caused difficulty. Almost all recalled and understood it, but those who found it acceptable tended to consent to randomization and those who could not accept it tended to refuse to participate. Belief in clinical equipoise was key to participants' consent to randomization. Ensuring patients understand and accept equipoise may thus increase their readiness to consent to participate in trials. A priority for future research is to focus on the provision and presentation of suitable and effective trial information, concentrating in particular on the neglected concept of clinical equipoise.