Cardiovascular disease results in more deaths and higher medical costs than any other medical problem. Cardiac patients may be transported to centers for specialized care. We evaluated historical, current event, and physiologic items (n = 32) for ability to predict use of specialized care and hospital costs. For 199 patients studied, seven items were prognostic. A model classifying patients by presence of predictors was developed. For the group without predictors and the group with multiple predictors, sensitivity and specificity were respectively very good. For 125 (63%) of the patients in middle categories, the model was not sufficiently sensitive to be prognostic. A scoring system for all cardiac patients could not be developed. Patients requiring only diagnostics were responsible for financial deficits, as were all Medicare patient groups. Patients staying longer than seven days, or having surgery, or both, were responsible for the largest deficits (if on Medicare) and the highest profits (if not on Medicare). Advance validation of the need to transport is difficult, with far-reaching medical and financial implications.