Research has indicated the ability of the Theory of Planned Behavior to predict blood donation. This study tested an extended version of the Theory of Planned Behavior (attitude, subjective norm, and perceived behavioral control with the addition of past behavior, moral norm, self-identity, and anticipated regret) in predicting donation intention and behavior among donors and nondonors and if forming implementation intentions improved attendance at a mobile blood donation clinic. This study used a cross-sectional follow-up design. A questionnaire was distributed to 600 staff and students at the National University of Ireland, Galway, before the arrival of a mobile clinic to Galway city. Half of these participants were invited to make implementation intentions specifying how, where, and when they planned to donate blood. A second questionnaire measuring reported attendance at the clinic was distributed 2 weeks after the first questionnaire. A total of 172 eligible donors returned questionnaires (29% response rate). The extended Theory of Planned Behavior accounted for 51 percent of the variance in intention to donate in the future: anticipated regret, attitude, perceived behavioral control, self-identity, and subjective norm significantly predicted intention. Donors differed from nondonors: they possessed more favorable attitudes toward blood donation, had a greater sense of donor identity, and believed more strongly in a moral obligation to donate blood than nondonors. Those who made implementation intentions were no more likely to donate that those who had not. Owing to the different factors influencing the donation decision for donors and nondonors, separate strategies to promote donation should be designed for these subgroups, targeting the elements most pertinent to each group.