Use of Hematopoietic Colony-Stimulating Factors: Comparison of the 1994 and 1997 American Society of Clinical Oncology Surveys Regarding ASCO Clinical Practice Guidelines

Abstract

PURPOSE: The American Society of Clinical Oncology (ASCO) Health Services Research Committee sought to assess whether more appropriate patterns of colony-stimulating factor (CSF) use occurred after the publication of ASCO evidence-based practice guidelines in 1994 and 1996 for patients with solid tumors or lymphoma. METHODS: In 1994 and 1997, questionnaires describing clinical scenarios were mailed to ASCO members who practiced medical oncology. Physicians were asked the extent to which they preferred to use a CSF for primary prophylaxis, secondary prophylaxis, or treatment of neutropenic complications. Multiple regression analyses were used to determine predictors of overall propensity to use CSFs and, when using a CSF, propensity to support longer schedules of CSF use. RESULTS: Decreased use of CSFs was shown in the following situations: (1) treatment for febrile neutropenia without localizing signs (39% in 1994 v 29% in 1997) or with a right lower lobe infiltrate (54% v 46%); (2) primary prophylaxis with paclitaxel for ovarian cancer (20% v 11%) or cyclophosphamide, doxorubicin, and vincristine chemotherapy for small-cell lung cancer (8.4% v 4.6%); and (3) secondary prophylaxis after afebrile neutropenia following chemotherapy for germ cell tumors (44.5% v 36.0%). One third fewer physicians supported the extended use of CSFs until an absolute neutrophil count ≥ 10,000/mm3 or a WBC count ≥ 10,000/mm3 was reached, both counts serving as criteria for stopping CSF therapy. However, we observed high rates of CSF use despite ASCO guideline recommendations against use in the following clinical situations: (1) primary prophylaxis in patients at low risk of febrile neutropenia (6% v 16%); (2) secondary prophylaxis late in the course of curative and palliative therapy (80% v 53%); and (3) treatment of afebrile and uncomplicated febrile neutropenia (30% v 60%). In 1994 and 1997, fee-for-service physicians were more likely than other physicians to prefer use of CSF support while maintaining treatment dose and schedule instead of using dose-reduction strategies, and, when using a CSF, they were more likely to support longer CSF treatment schedules (P < .05 for both scenarios). CONCLUSION: Decreased use and more appropriate use of CSFs in accordance with ASCO guideline recommendations occurred from 1994 to 1997, but there remain many opportunities to reduce CSF use with no clinical harm. Many oncologists continue to support the use of CSFs in scenarios and with scheduling criteria that the guidelines and evidence do not support. ASCO's evidence-based guidelines should be linked with formal continuous quality improvement initiatives to substantially improve the quality of supportive oncology care.