Date of Award

Fall 2017

Document Type

Open Access Dissertation


Epidemiology and Biostatistics


The Norman J. Arnold School of Public Health

First Advisor

Jan M. Eberth


Despite progress in detection and treatment, lung cancer remains the leading cause of cancer-related death in the United States. The United States Preventive Services Task Force (USPSTF) recommends adults at high risk for lung cancer undergo annual low-dose computed tomography (LDCT) screening, however, lung cancer screening (LCS) uptake remains low. Qualitative research on family physician (FP) perceptions and experiences with LCS has been limited since USPSTF publication and Centers for Medicare and Medicaid Services (CMS) decision memo. We conducted a qualitative study to assess FP knowledge and perceptions of LCS and gain insight into their current experiences with LDCT. A convenience sample of FPs were asked to participate in Skype audio interviews. A semi-structured interview guide was used to navigate the interviews. A theme codebook was developed using the constant comparison technique. All interviews were coded by two reviewers

We found that FP knowledge about the scientific evidence and patient eligibility criteria for LDCT was suboptimal. Age and smoking history were the primary drivers of a FPs decision to discuss LCS. Most FPs knew that they should initiate LDCT discussions with high risk patients, however, they indicated that they would be willing to screen patients outside of the specified criteria. LDCT cost and lack of time were cited as barriers. Facilitators included screening tools in the clinic waiting room and electronic medical record notifications. These results indicate a need for FP education about LCS, as well as tools to assist providers in the clinic.

As LCS becomes more widely adopted, more lung cancers will be detected at an earlier stage. While tumor molecular testing (MT) is currently recommended for patients with metastatic disease, MT could increasingly be used in early stage patients to guide initial treatment decisions. Disparities in MT and targeted therapy utilization may exist. We quantitatively evaluated factors related to MT and erlotinib utilization and the impact of these on overall survival (OS).

Stage IIIB/IV non-small cell lung cancer (NSCLC) cases diagnosed between January 1, 2002 and December 31, 2012 and available through the South Carolina Central Cancer Registry were linked to SC State Employee Health Plan (SCSEHP) and Medicaid administrative claims data. MT and erlotinib utilization were independently categorized as “yes” or “no” based on claims data. We found several characteristics associated with MT, including younger age, having an out-of-state provider, being diagnosed in 2010 or later, adenocarcinoma histology, and low tumor grade. Risk of death was reduced and OS was longer for patients with MT. Younger age, female sex, SCSHEP insurance, having an out-of-state provider, adenocarcinoma histology, and having molecular testing were associated with erlotinib utilization. Risk of death was lower for patients treated with erlotinib and OS was longer. These results suggest that tumor MT and erlotinib utilization lead to improved patient survival. Additional research should evaluate these important factors in nationally representative datasets.

Included in

Epidemiology Commons