Studies linking medical costs to behavioral risk and risk-lowering often use means for comparisons, although claims data are highly skewed. The result overestimates and obscures the case for work-site health promotion. In this study, high-cost analysis is illustrated in a sample of university employees. Five risk factors were examined: cholesterol, blood pressure, cardiovascular fitness, body fat, and smoking status. Screened employees who released their claims (n=367) were examined against a random sample of employees (n=587). Linear regression was used to determine the risk of having high claims costs within four gender-specific age groups. A formula was then applied to determine that more than 43% of the cost of medical claims was associated with elevated risk. High-cost analysis accounts for the skewness in claims data and presents a clear case for work-site health promotion.