The quality of CT surveillance in resected colorectal cancer survivors at Veterans Health Administration facilities.
525 Background: Annual CT scan for three years is the current standard for post-treatment surveillance of colorectal cancer (CRC) survivors following curative treatment. We conducted a retrospective study with the primary aim to assess the patient, physician & organizational characteristics associated with quality of CT surveillance at Veterans Administration (VA) facilities. Methods: The Department of Veterans Affairs Central Cancer Registry was used to identify patients with stage I-III CRC between 2001 & 2009. Patient (age, race, marital status, income, priority status, year of diagnosis, stage, comorbidity), provider (primary care, specialist, other) & facility (region & organizational survey) characteristics were measured. In the primary analysis, we analyzed the impact of the 2005 ASCO update in CRC surveillance guidelines. In the secondary analysis, we included only patients diagnosed after 2005 to avoid any bias due to change in guidelines. Overuse (OU) was defined as > 1 CT/12 months, whereas underuse (UU) was defined as < 1 CT/12 months. The Wald test was used for univariate analysis & logistic regression models, for multivariate analysis. Results: The primary analysis demonstrated no change in the receipt of CT scans before & after the 2005 guideline update, as there was an increasing trend favoring more CT scans in general.In the secondary analysis, there were a total of 2,263 patients. Overall, UU (19.44% patients) occurred more often than OU (15% patients). In univariate analyses, age, stage, insurance status, provider type & facility region were significant in both OU & UU analyses (p <.001). Comorbidity was associated with OU only (p =.02). Different organizational characteristics were significant for OU & UU cohorts. Multivariate analyses showed younger age [AOR = 2.17 (1.65 - 2.86)] & higher stage [AOR =.71 (.59 -.85)] were associated with OU at p <.001. Older age [AOR = 1.89 (1.54 - 3.10)], lower stage [AOR =.63 (.49 -.81)] at p <.001 & facility region [AOR = 1.84 (1.01 - 3.35)] at p =.04 were associated with UU. Conclusions: While present, there did not appear to be substantial rates of UU or OU among CRC survivors in the VA. Age & cancer stage were associated with the quality of CT surveillance.