Abstract
Background/Aims The colony stimulating factors (CSFs), filgrastim and its long-acting form, pegfilgrastim, are indicated by the Food and Drug Administration to decrease infections in patients with non-myeloid malignancies receiving myelosuppressive chemotherapy. Roughly 25–40% of treatment-naive patients receiving common chemotherapy regimens develop febrile neutropenia (FN). FN is associated with treatment delays, dose reductions, hospitalizations, and a high cost burden. CSFs decrease the incidence, length and severity of chemotherapy-related neutropenia in several solid tumors and prophylactically, decrease infection rates and neutropenia, infection-related mortality, and early deaths associated with chemotherapy. A reduction in absolute and relative risk for all-cause mortality is associated with CSF use and in combination with antibiotics for the treatment of FN, CSFs decrease the length of hospitalization; however, recent studies have shown these agents are frequently administered in a manner inconsistent with the recommended guidelines. The high costs associated with FN treatment, the high cost of CSFs and administration of CSFs in a manner inconsistent with scientific evidence creates both a clinical and economic challenge for health plans. To date, CSF data within the Virtual Data Warehouse (VDW) has not been evaluated for accuracy.
Methods We are conducting a validation study using tumor registry data and medical record abstraction (gold standard) to evaluate 100 patients within our Cancer Center who received a new chemotherapy and initial treatment with a CSF from 01/01/2012–12/31/2012 with 100 patients matched on age, diagnosis, stage and treatment who did not receive CSFs to verify the VDW CSF data. We will compute the sensitivity, specificity, and positive predictive value of the VDW data to determine the concordance between the gold standard tumor registry and chart review data and the VDW data.
Results Chart review is currently ongoing and the VDW patient population is being assembled for comparison with abstracted data.
Conclusions Ultimately, we plan to evaluate CSF use among HMORN sites and use CSF data within the VDW for studies among cancer patients within the HMORN.
