Subject enrollment procedures

Controls were selected using stratified random sampling from voter registration rolls of the 29 counties in the 5^th Congressional District. A frequency matching method was used to match control and case in 4:1 ratio by gender and age (± 5 years). Controls were selected prior to the knowledge of gender and age of recruited cases, in order to facilitate recruitment and these estimates were based on the gender and age distribution of lung cancer cases in the 5^th District in year 2009 and 2010. Following the first six months of recruitment, the Cancer Prevention and Control (CPC) staff compared the distributions of recruited controls and cases and the response rates of control and cases, and randomly selected further cohorts of controls to properly match the distribution of recruited cases. It should be noted that App KY is not racially or ethnically diverse, with over 96% of residents identifying as White, non-Hispanic; however, consistent with the Belmont Report [17], all races and or ethnicities were eligible for inclusion. Control subjects were contacted via phone by CPC staff and screened for eligibility and consent for home visit.

All newly diagnosed and histologically or cytologically confirmed lung cancer cases (International Classification of Diseases-9^th revision, 162.2–162.9) [18] were identified by the KCR for inclusion. Because the median survival time for all lung cancer patients is very short, recruitment procedures utilized the innovative rapid case ascertainment system developed by the KCR. All Kentucky health care facilities are required to report all cancer cases using the Cancer Patient Data Management System (CPDMS) developed by KCR [19]. This method uses the electronic pathology (e-path) system to provide the KCR with immediate notification of a cancer diagnosis each time a pathology report is generated. KCR served as the “honest broker” of the private health information for these patients. To identify each newly diagnosed lung cancer patient in the 5th District, a certified tumor registrar (CTR) at KCR weekly reviewed all of the incoming e-path reports indicating a diagnosis of lung cancer. KCR then sent a letter to the physicians for each eligible lung cancer patient to determine whether any issues (i.e. dementia, frailty) would preclude the patient from study participation (passive consent). Eligible patients were mailed a letter asking for their assent for contact by Kentucky Homeplace, a well-established lay health worker program in App KY. Patients who did not respond to the letter within two weeks were called by KCR staff to obtain their assent.

KCR staff created eligible study lists for contact by trained Kentucky Homeplace field research associates (FRA) to administer the questionnaire and collect biospecimens and drinking water samples from cases and controls at their homes. Additional screening of cancer patients was performed by Patient Navigators/FRAs at the Center for Rural Health in Hazard, Kentucky; St. Claire Regional Medical Center, Morehead, Kentucky; and the University of Kentucky. The FRAs were trained phlebotomists who performed all study procedures in subjects’ homes. Prior to a home visit by the research team, subjects gave verbal assent by phone and before data and sample collection began, study volunteers were provided a description of the study. FRAs obtained written informed consent and Health Insurance Portability and Accountability Act (HIPAA) authorization approved by the Institutional Review Board (IRB) of the University of Kentucky and regional partner institutions.