Temporal trends in colorectal procedure use after colorectal cancer resection.

BACKGROUND

After curative cancer resection, routine colon surveillance is recommended. It is not known whether trends over time in cancer survivors parallel that of the general population.

OBJECTIVE

Our purpose was to describe temporal changes in the use of posttreatment procedures.

DESIGN

Retrospective cohort study.

SETTING

Linked tumor registry and Medicare claims data.

PATIENTS

Medicare beneficiaries >65 years old who were diagnosed with local or regional stage colorectal cancer from 1992-2002 and who underwent surgical resection.

MAIN OUTCOME MEASUREMENTS

Use of colonoscopy, sigmoidoscopy, or barium enema within 1 year, 18 months, or 3 years of diagnosis.

RESULTS

A total of 62,882 patients were followed up for 1 year and 35,784 for 3 years. Colonoscopy within 1 year was performed in 25.9%, within 18 months in 53.8%, and within 3 years in 70.3%. Corresponding rates for sigmoidoscopy were 7.4%, 10.2%, and 14.9%, respectively, and were 3.4%, 5.1%, and 7.9%, respectively, for barium enema. There was a decrease over time in the receipt of colonoscopy within 1 year of diagnosis (31.3% in 1992 to 20.6% in 2002), no change in 18-month rates, and a smaller increase in colonoscopy use within 3 years (66.5% to 72.3%). The use of sigmoidoscopy and barium enema declined over time. Overall procedure use within 1 year and 18 months also decreased and 3-year rates were essentially unchanged. These differences were maintained in multivariate analyses.

LIMITATIONS

Accuracy of procedure coding and indications for tests could not be measured.

CONCLUSIONS

Temporal trends in procedure use in cancer survivors were consistent with the general population. Importantly, despite guideline recommendations and Medicare reimbursement, 25% of patients who undergo curative treatment do not receive surveillance examinations and this was unchanged over time.