The purpose of the study is to describe oncologists' perceptions and opinions about patient eligibility, guidelines, and barriers for use of granulocyte colony-stimulating factor (G-CSF), overall and stratified by their affiliation with the Oncology Care Model (OCM). In May 2018, we invited and recruited practicing US oncologists from a national database for an online survey. Level of agreement was identified using a seven-point scale, ranging from strongly disagree to strongly agree. Of 200 participating oncologists, 70 were OCM-affiliated. Overall, 65% of oncologists agreed or strongly agreed that all patients at high risk of febrile neutropenia (FN) should receive prophylactic G-CSF, and half agreed or strongly agreed that benefits of G-CSF outweigh the potential adverse effects. The most common barriers to G-CSF use for patients at high risk of FN included patient refusal (37.1% of OCM-affiliated oncologists vs. 21.5% of non-OCM-affiliated oncologists), not on protocol/not supported by guidelines (32.9% vs. 23.1%), lack of reimbursement to practice (30.0% vs. 15.4%), and concerns about insurance coverage (22.9% vs. 26.9%). More OCM-affiliated oncologists reported that their practices offer and strongly encourage adherence to a specific protocol for G-CSF use (49.2%) versus non-OCM oncologists (31.3%). Despite recommendations from national guidelines and strong evidence from randomized, controlled clinical trials, only two thirds of oncologists agree or strongly agree that all patients at high risk of FN should receive primary G-CSF prophylaxis. Decisions about G-CSF prophylaxis may be affected by factors other than risk of FN, such as patient choice, practice protocols/guidelines, lack of reimbursement, and insurance coverage.

INTRODUCTION: Febrile neutropenia (FN) is a serious complication of myelosuppressive chemotherapy associated with significant morbidity and mortality. Prophylaxis with G-CSFs reduces the risk of FN. While G-CSFs have been used for many years, there is little information on the factors US oncologists consider when making G-CSF use decisions or the impact of participation in emerging payment and delivery models. STUDY OBJECTIVE: To describe oncologist's perceptions and opinions on national or institutional guidelines for G-CSF use, eligibility of patients for, and barriers to, G-CSF use stratified by physicians' institutional affiliation (i.e., Oncology Care Model [OCM] vs. non-OCM practice). METHODS: In May 2018, we recruited 200 randomly selected US oncologists from a national database to participate in a 30-minute, independent review board approved, online survey. Physicians from Vermont, Maine, and Massachusetts were not included because of the state regulations regarding survey participation. We specifically targeted recruiting at least 125 community-based oncologists and at least 50 oncologists primarily affiliated with an OCM practice. Survey questions were developed to include topics such as risk factors and criteria considered to determine a patient's FN risk level; drivers and barriers to G-CSF use; and perceptions and experience with G-CSFs. The survey was pilot tested by nine oncologists in pen and paper (n=5) and online (n=4) version to improve its content and clarity. Oncologists participating in the pilot were considered ineligible to participate in the main study. We calculated proportions (%) and 95% confidence intervals (CI) for categorical variables collected via the survey. RESULTS: Of the 200 oncologists, 114 (57%) were hematologist-oncologists (vs. 86 oncologists), 125 (63%) were community-based (vs. 75 academic), and 70 (35%) were affiliated with an OCM practice (vs. 130 non-OCM). Most of survey participants (71%) followed National Comprehensive Cancer Network (NCCN) guidelines for G-CSF use. More OCM affiliated oncologists reported that their practices offer, and strongly encourage them to follow, a specific treatment protocol for G-CSF use (49% [95% CI: 37%, 61%]) compared to non-OCM practices (31% [95% CI: 24%, 40%]). Furthermore, 65% of oncologists overall agreed or strongly agreed that all patients with high risk for FN should receive prophylaxis with G-CSF; 60% [95% CI: 48%, 71%] in OCM affiliated versus 68% [95% CI: 59%, 75%] in non-OCM affiliated oncologists. Half of the oncologists agreed or strongly agreed that benefits of G-CSF therapy outweighed the potential adverse effects (Figure 1). Overall, the most common reported barriers to G-CSF use among patients at high risk for FN included patient refusal (27%), limitations to prescribing based on set protocols and guidelines (27%), and concerns about insurance coverage (26%) with the ranking of barriers differing by OCM affiliation (Figure 2). Oncologists reported that among all patients receiving chemotherapy that they treated in the past 6 months, 46% (95% CI: 42%, 49%) of patients in curative setting and 37% (95% CI: 33%, 41%) of patients in metastatic setting received G-CSF for primary prophylaxis. Oncologists affiliated with practices that offer or strongly encourage to follow specific treatment protocol were more likely to use G-CSF for primary prophylaxis in curative setting (54% [95% CI: 47%, 61%] vs. 42% [95% CI: 37%, 48%]). A majority (67%) of the oncologists reported an increase in use of pegfilgrastim on-body injector in the last 2 years and attributed it to convenient dosing (67%) and patients' need for active lifestyle (45%). CONCLUSIONS: Despite national guidelines and evidence from clinical trials, only two-thirds of oncologists agree or strongly agree that all patients at high risk for FN should receive primary prophylaxis, and only half agree or strongly agree that benefits of G-CSF therapy outweigh the potential adverse effects. Oncologists affiliated with OCM practices had different opinions on barriers to G-CSF use than oncologists not affiliated with OCM. Decisions for G-CSF use may be affected by factors other than risk of FN, such as patient convenience, lack of reimbursement to the practice, concerns regarding insurance coverage, and practice-specific treatment protocols. Disclosures McNamara: Adelphi Research: Consultancy. Gawade:Amgen: Employment, Equity Ownership. Belani:Amgen: Employment, Equity Ownership. Kelsh:Amgen: Employment, Equity Ownership.