Patient and Physician Preferences for Nonoperative Management for Low Rectal Cancer: Is It a Reasonable Treatment Option?

Abstract

Background: Although the body of evidence supporting nonoperative management for rectal cancer has been accumulating, there has been little systematic investigation to explore how physicians and patients value the tradeoffs between oncologic and functional outcomes after abdominal perineal resection and nonoperative management.

Objective: The purpose of this study was to elicit patient and physician preferences for nonoperative management relative to abdominal perineal resection in the setting of low rectal cancer.

Design: We conducted a standardized interviews of patients and a cross-sectional survey of physicians.

Settings: Patients from 1 tertiary care center and physicians from across Canada were included.

Patients: The study involved 50 patients who were previously treated for rectal cancer and 363 physicians who treat rectal cancer.

Interventions: Interventions included standardized interviews using the threshold technique with patients and surveys mailed to physicians.

Main outcomes measures: We measured absolute increase risk in local regrowth and absolute decrease in overall survival that patients and physicians would accept with nonoperative management relative to abdominal perineal resection.

Results: Patients were willing to accept a 20% absolute increase for local regrowth (ie, from 0% to 20%) and a 20% absolute decrease in overall survival (ie, from 80% to 60%) with nonoperative management relative to abdominal perineal resection, whereas physicians were willing to accept a 5% absolute increase for local regrowth (ie, from 0% to 5%) and a 5% absolute decrease in overall survival (ie, from 80% to 75%) with nonoperative management relative to abdominal perineal resection.

Limitations: Data were subject to response bias and generalizable to only a select group of patients with low rectal cancer.

Conclusions: Offering nonoperative management as an option to patients, even if oncologic outcomes are not equivalent, may be more consistent with the values of patients in this setting. See Video Abstract at http://links.lww.com/DCR/A688.