Case management programs are expensive and therefore require careful screening of enrollees to ensure cost-effectiveness. Screening tools, however, are imperfect, with positive predictive values usually below 50%. This article examines the relationship between the accuracy of the screening tools and the cost-effectiveness of case management. Using data from a Medicare health maintenance organization (HMO), we develop an optimized 5-question screening tool. We then simulate the use of this screening tool and its impact on the cost-effectiveness of several hypothetical case management programs. The article demonstrates that even screening tools with only 20-30% positive predictive value could turn a case management program into a cost-effective program.