Objective: This study quantifies the direct and indirect costs of obesity within a cohort of commercially-insured employees between 2003-2005. Methods: A database of health plan claims, self-reported health risk assessments (HRAs), and productivity data (Thomson MarketScan) from 2003-2005 was used to identify individuals for analysis. The study cohort included individuals reporting a valid body mass index (BMI) value on an HRA and 12 months of continuous eligibility. Patients were divided into five groups according to the WHO classification of BMI: severely obese (BMI=35), obese (30=BMI<35), overweight (25=BMI<30), normal weight (18.5=BMI<25), and underweight (BMI<18.5). Mean direct medical costs and the costs associated with employee absence were calculated for each BMI group. Two-part regression models were used to estimate the incremental direct and indirect costs, conditional upon expenditure, associated with elevated BMI. Results: Of 88,984 employees (45% female, mean age=42.3), 9.9% were severely obese, 16.7% obese, 38.5% overweight, 34.0% normal weight, and <1% were underweight. With the exception of the underweight BMI group, univariate analysis reveals a graded pattern of expenditures for emergency room and outpatient pharmacy. As BMI increases, expenditures for these services increase. Further, compared to those with a normal BMI, the obese and severely obese have higher costs for inpatient hospital and outpatient visits and services. Regression-adjusted incremental direct medical costs associated with being overweight, obese, and severely obese were estimated to be $147.11, $712.34, and $1977.43, respectively. Regression-adjusted incremental indirect costs due to paid time off associated with being overweight, obese, and severely obese were estimated at $1403.81 $1511.24, and $1414.09, respectively. Conclusion: Overall health care costs were higher for workers who were obese or severely obese than for those of normal weight or who were overweight. Indirect costs incurred from paid time off were higher for workers in all categories of elevated BMI relative to those of normal weight.