EFFECT OF A NATIONAL-CANCER-INSTITUTE CLINICAL-ALERT ON BREAST-CANCERPRACTICE PATTERNS

Purpose: This study evaluates the effect of the 1988 National Cancer I nstitute Clinical Alert regarding treatment of early-stage breast canc er on the patterns of treatment provided to patients. Patients and Met hods: Data analyzed were collected from the hospital and outpatient re cords of 12,534 female patients with a primary diagnosis of breast can cer (stages I and II) initially diagnosed during the years 1983 throug h 1989. Results: Analyses revealed that the proportions of patients wi th a negative lymph node status diagnosed after the May 1988 Clinical Alert who received adjuvant treatment (tamoxifen and/or multidrug chem otherapy) were significantly greater than predicted from treatment tre nds established before the Alert's release. Proportions of patients wi th positive lymph node status receiving adjuvant therapy subsequent to the Alert's release, in contrast, did not fall outside the projected confidence intervals for that group. Additional analyses showed a sign ificant effect of the Clinical Alert among several subgroups of node-n egative patients. Conclusion: Findings suggest that the Clinical Alert mechanism, followed by publication in the peer-reviewed scientific li terature, is an effective way to communicate important research findin gs to practitioners in the community. However, the Alert mechanism is controversial and should be used judiciously to ensure its credibility (C) 1994 by American Society of Clinical Oncology.