LOCATION: Nova Scotia,
a province in Eastern Canada with a population of 954 000. 17.8% of Nova Scotian residents are current smokers.
DATABASE: All primary lung cancer cases diagnosed in Nova Scotia,
Canada in 2014 (identified through the provincial cancer registry). Lung cancer is a reportable disease in Nova Scotia and all cases are identified using the SEER (surveillance,
epidemiology and end results) based standards of the North American Association of Central Cancer Registries and the Canadian Council of Cancer Registries (16).
PATIENT DATA: Age,
gender,
stage at diagnosis,
histology,
biopsy date and PET date were obtained from the registry. Additional data including smoking history,
indication for testing,
investigation history (number,
date and results of tests including CT) and treatment history were obtained through review of electronic and paper charts.
CASE IDENTIFICATION: Test requisitions and physician notes for each case were reviewed to identify the clinical indication for testing. Clinical indication was classified as either:
1) Symptomatic (lung cancer detected due to testing performed in a patient with signs and symptoms possibly related to lung cancer (e.g.,
hemoptysis,
dyspnea,
chest pain,
cough) or signs and symptoms of disease later proven to represent metastatic lung cancer;
2) Coincidental (lung cancer detected due to testing performed for another reason (initial CT considered diagnostic);
3) Screening;
4) Surveillance (lung cancer diagnosed after CT surveillance of a low risk or questionable lung abnormality).
IMAGING DATA: All cancers detected through surveillance were identified on the most recent CT and all prior CTs were reviewed. For all CTs in which the cancer could be identified,
the cancer was measured in 2 dimensions on a single axial image using the longest dimension and the longest perpendicular dimension. Both the mean total nodule/mass size and the solid portion were measured. Nodules/masses were classified as either solid,
part solid or ground glass. All radiography,
computed tomography (CT) and PET performed in the province is included on a single Picture Archiving and Communication System (PACS). There is no private radiology or CT imaging performed outside of this system.
DIAGNOSTIC INTERVAL: All CT reports were reviewed. Three time points were identified for each patient: date lesion first visible,
date of definitive CT and date of treatment initiation. The first definitive CT was determined by the level of certainty conveyed by the report.
For example,
“highly suspicious for lung cancer” or “consistent with lung cancer” accompanied by a recommendation for referral to a lung cancer specialist were considered definitive. At the time of the study,
there were no reporting guidelines in place apart from Fleischner guidelines (3),
and the majority of thoracic CTs were reported by non-thoracic radiologists. CTs could be reported by fully credentialled radiologists (Royal College of Physicians and Surgeons of Canada) or by other radiologists (e.g.,
radiologists working under a defined/limited license in areas of need).
Institutional Research Ethics Board Approval was obtained. The study adhered to the Personal Information Protection and Electronic Documents Act.