Gs. Cooper et al., Geographic and patient variation among medicare beneficiaries in the use of follow-up testing after surgery for nonmetastatic colorectal carcinoma, CANCER, 85(10), 1999, pp. 2124-2131
BACKGROUND. There are a paucity of data supporting the routine use of follo
w-up testing to detect recurrent disease after potentially curative initial
surgery in patients with nonmetastatic colorectal carcinoma.
METHODS. Using the population-based Surveillance, Epidemiology, and End Res
ults (SEER) registry, all patients age greater than or equal to 65 years wi
th local or regional colorectal carcinoma who were diagnosed in 1991, under
went surgical resection, and survived at least 6 months after diagnosis wer
e identified. All inpatient, hospital outpatient, and physician/supplier Me
dicare claims from 6 months after diagnosis through 1994 were examined for
follow-up procedures of interest. procedure use during follow-up was compar
ed across patient groups using both bivariate and multivariate analyses.
RESULTS. A total of 5716 patients were identified, with 1.3% found to have
developed subsequent primary tumors of the colon or rectum, and 74% survivi
ng through 1994. One or more procedures of interest were performed in 88% o
f patients; the most commonly performed tests were liver enzymes, chest X-r
ays, colonoscopy, and computed tomography scans. Lower rates of testing gen
erally were observed with older age groups, patients with fewer comorbiditi
es, and patients who did not survive through the follow-up period. Among al
l procedures studied, there also was significant variation in the rates of
testing across the 9 SEER areas, varying from 1.5-fold to 3.6-fold. The geo
graphic variation persisted in multivariate models adjusting for potentiall
y confounding factors.
CONCLUSIONS. The current study found significant variability in the use of
follow-up procedures, with the most striking differences apparent across ge
ographic regions. Further studies are needed to determine the underlying re
asons for the disparities, as well as the impact of surveillance on patient
outcomes. Cancer 1999; 85:2124-31. (C) 1999 American Cancer Society.