OBJECTIVES The purpose of this study is to examine the relationship between patient outcomes and conditions of the patients undergoing different treatments for lung cancer and to estimate the population burden, the cost of cancer, and to examine treatment choice in clinical decision-making. METHODS Lung Cancer data were extracted from the Medstat MarketScan Database based on ICD9 diagnosis codes. Kernel Density Estimation was used to examine the lung disease by Age, Length of Stay and Total Charges by patient conditions. Text Miner in Enterprise Miner was used to examine the data according to text strings of treatment procedures. Then we predict the occurrence of lung cancer according to patient age, gender, days of stay and total charge with the predictive modeling in Enterprise Miner. RESULTS There were 4718 observations related to lung cancer. There are more inpatient events starting at age 40, accelerating at age 50 and 55, and decreasing at 65. Patients with lung cancer had a higher probability of a stay of five days, which indicates that there was a higher probability of higher cost. We defined clusters of procedures with a frequency showing the effectiveness of treatment for patients. The Decision Tree is optimal with a 22.9% misclassification rate in the testing set compared with other models in Enterprise Miner. CONCLUSIONS Older patients are more likely to have lung cancers that would lead to a higher probability of longer stay and higher costs for the treatment procedure. With text analysis on the procedure codes and KDE, it shows that Levels IV and VI Surgical pathology, gross and microscopic examination are used for patients of higher risk with a higher cost compared to other procedures to diagnose lung cancer.