To compare two groups of deceased individuals diagnosed 2008-2011 with lung or colorectal cancer, one comprising people with a survival less than 2 months (cases) from diagnosis and the other with survival of between 6 months and 2 years (controls).
To compare cases and controls within each cancer site, using GP note review and other clinical records to identify:
Patient factors (demographic and other patient characteristics including socio-economic status, rurality, marital status, smoking status, alcohol use and other lifestyle factors). Indicators of patient engagement with primary healthcare services including routine appointments, attendance for flu vaccinations, uptake of screening and mean consultations over the five years prior to diagnosis.
Disease factors (markers of disease severity, tumour grade, cell type and disease stage broken down by TNM/ DUKES, and symptoms at presentation).
Clinical factors (patient co-morbidities Treatment received including decisions not to treat and rationale).
Service factors (intervals on the patient pathway from first symptoms to diagnosis, to treatment, the referral mode and the number and pattern of service contacts in the year preceding diagnosis).
To identify potential survival time artefacts. For example, definition of diagnosis date (clinical versus pathological verification of tumour
To determine the role of each factor and their interplay in predicting early death to identify groups that may benefit from early diagnosis.
To publish results and any recommendations in peer-reviewed publications and a report.
To provide information on the patients with the poorest survival that is complementary to the ongoing work in the International Cancer Benchmarking Partnership (ICBP) in one UK region as well as a methodology for use in other ICBP jurisdictions.
To feedback findings to service managers and healthcare providers locally with a view to promoting service improvements.