In this paper, cost and effectiveness data for six clinical interventions are applied simultaneously to a hypothetical population of 100,000 patients to show how selecting guidelines to maximize overall population benefit compares with selecting the best guidelines for individual patients. By entering effectiveness (added survival) and cost information from recent prevention, screening, diagnostic, and therapeutic guidelines into a computer-based optimization model, the options that maximized overall population effectiveness while keeping additional cost within varying specified constraints were identified. In 57% of selection opportunities, the clusters of guidelines that yielded maximum population benefit differed from those that maximized benefit for individual patients. Some choices were more stable than others over ranges of cost constraints. Clinical practice guidelines chosen to maximize cost-effectiveness for individual patients often do not maximize cost-effectiveness for populations of patients. To allocate resources as efficiently as possible, decision makers should consider other sources of information in addition to the recommendations of specific practice guidelines. "Robust" guidelines that simultaneously address both individual and societal health benefit should be sought.