Lung cancer is the leading cause of cancer mortality in the US. Early stage lung cancer can be cured with surgery, and more early stage tumors are expected to be detected with screening programs. Minimally invasive approaches (video assisted thoracic surgery, VATS) and limited pulmonary resection (sublobar) techniques can reduce procedural morbidity for lung cancer surgery, allowing more patients to have surgery. However, important questions remain about the completeness of cancer removal with these surgical strategies and, therefore, survival outcomes. This represents a critical barrier to matching lung cancer patients with the optimal surgical treatment. A long term goal of The Society of Thoracic Surgeons (STS) is to improve quality of surgical therapy for lung cancer. The STS General Thoracic Surgery Database (GTSD) captures unique patient-level clinical detail not found in any other cancer database. However, the STS GTSD does not provide longitudinal outcomes data, a limitation that can be overcome by linkage to Centers for Medicare and Medicaid Services (CMS) data. Our objective is to identify predictors of long-term outcomes following lung cancer resection, including the optimal strategies with respect to surgical approach and extent of resection. We hypothesize that: 1) long-term survival following lung cancer surgery varies according to individual patient clinical and treatment variables, 2) VATS approaches and sublobar resections are not associated with inferior long-term survival compared to more invasive approaches (thoracotomy), and standard resections (lobectomy), and [3) VATS approaches and sublobar resections are associated with more favorable economic outcomes than standard thoracotomy and lobectomy.] The rationale of this research is that knowledge of risk-adjusted long-term out- comes will facilitate selection of optimal therapy for patients with lung cancer. Our hypotheses will be tested by pursuing three specific aims: 1) Create a risk model for long-term survival following lung cancer resection. 2) Compare survival based on surgical approach, VATS vs. thoracotomy, and extent of resection, sublobar resection vs. lobectomy, for early stage lung cancer. [3) Compare resource use and costs according to surgical approach and extent of resection for early stage lung cancer.] In Aim 1, established algorithms will link the GTSD with CMS data to create a risk model for long-term survival following lung cancer resection.
Under Aim 2, VATS approaches and sublobar resections will be compared to thoracotomy approaches and lobectomies with respect to long-term survival with propensity analyses. [Lastly, under Aim 3, costs and resource use associated with VATS and sublobar resections will be compared with thoracotomy and lobectomy, respectively.] This re- search is innovative because linkage of the GTSD with CMS data will create a longitudinal data source uniquely suited for comparative effectiveness studies in lung cancer surgery. The contribution will be significant, as the most effective surgical strategies wil be determined based on individual patient characteristics. These findings will be disseminated by STS nationally and internationally to drive quality improvement in lung cancer care.
This study will link two large national patient databases, The Society of Thoracic Surgeons General Thoracic Surgery Database and Centers for Medicare and Medicaid Services administrative data, to examine longitudinal outcomes following lung cancer surgery. The proposed research is relevant to public health because characterization of variability in outcomes following surgical therapy of lung cancer is expected to enhance shared decision making and selection of optimal therapies for lung cancer patients. Thus, the proposed research is relevant to the AHRQ's mission to improve the quality, safety, efficiency, and effectiveness of health care services for all Americans.