Abstract
BACKGROUND: Oesophagogastric (OG) cancer-associated symptoms are common in primary care, but most research has focused on patients with a confirmed OG cancer diagnosis, rather than those presenting with symptoms for the first time. AIM: To examine diagnostic testing patterns for upper gastrointestinal (GI) symptoms linked to OG cancer. DESIGN AND SETTING: A retrospective cohort study was undertaken, which used a linked primary care database. It included de-identified patients aged >55 years who presented with symptoms suggestive of OG cancer between 2008 and 2022. METHOD: The study analysed the proportion of patients who underwent pathology, imaging, referral, upper GI endoscopy, or a test of treatment. Differences across socioeconomic groups were also examined, along with the proportion of patients diagnosed with OG cancer. RESULTS: The study cohort consisted of 44 402 patients, of whom 126 (0.3%) were diagnosed with OG cancer within 12 months of presentation. Reflux was the most common symptom (57%), followed by nausea (11%). Patients aged ≥75 years were less likely to be investigated or referred than those aged 56-64 years (odds ratio [OR] 0.59, 95% confidence interval [CI] = 0.56 to 0.62, P<0.001). Those from less disadvantaged areas were 1.4 times more likely to be investigated than people from the most disadvantaged areas (OR 1.44, 95% CI = 1.36 to 1.53, P<0.001). Patients on test-of-treatment medications were less likely to receive further investigation (OR 0.66, 95% CI = 0.63 to 0.69, P<0.001). Multiple symptoms and visits increased the likelihood of investigation (OR 2.77, 95% CI = 2.55 to 3.00, P<0.001). CONCLUSION: Significant variations in diagnostic testing could contribute to disparities in OG cancer outcomes.