Using administrative data to describe adherence to colon cancer follow-up testing in Manitoba, Canada.

Abstract

6079 Background: Relapse after treatment of early-stage colon cancer is common, but meta-analyses have demonstrated a 7%- 10% reduction in absolute mortality with follow-up testing using a combination of CEA testing, liver imaging and colonoscopy. This study uses administrative data to describe adherence to such testing at a population level, and examines patient, provider, disease and treatment factors that are associated with better adherence. Methods: Data was obtained by linking the Manitoba Population Health Research Data Repository with the Manitoba Cancer Registry and Laboratory Claims data. An expert committee recommended standards of test performance based on the 2005 ASCO guidelines. A cohort of 171 patients was identified, diagnosed with stage II or III colon cancer in 2004 who were alive 42 months after diagnosis. Data include physician contacts and performance of liver imaging, colonoscopy, and CEA testing over a 3-year period from 6 to 42 months after diagnosis. Multivariate logistic regression was used to assess the performance of each test at the recommended standard against variables including visits by physician specialty, treatment, stage, and patient characteristics. Data for CEA and liver imaging was available only for urban patients. Results: 84% of the cohort received at least one follow-up colonoscopy, with stage II patients and those with at least one visit yearly with a surgeon more likely to be adherent. 54% had at least one liver imaging test per year, with adherence predicted by receiving chemotherapy and at least one visit yearly with a medical oncologist. Only 19% received 3 or more CEA tests per year, with adherence predicted by higher stage and at least one visit yearly with a medical oncologist. Twelve per cent met the standards for all three tests over the 3 years. Conclusions: Adherence to CEA testing and liver imaging falls well below guidelines even in this “best” patient group. Colonoscopy is better provided. Regular contact with a medical oncologist, stage and provision of chemotherapy are important drivers of complete follow-up, but adherence is low for all groups. This quality of care gap needs to be addressed with interventions aimed at patients, physicians and at the system level. No significant financial relationships to disclose.