OBJECTIVES: The incidence of febrile neutropenia (FN) depends on the cancer type and the chemotherapy regimen used. In Belgium, reimbursement of granulocyte-colony stimulating factors (G-CSF) in primary prophylaxis against FN is limited to 4 indications. This study aimed to provide real-life information on the incidence and impact of FN in chemotherapy-cancer combinations excluded from G-CSF primary prophylaxis reimbursement. METHODS: Based on ICD-9 code and drug name all chemotherapy-cancer combinations with at least one patient having an ICD-9 code corresponding to neutropenia (288.0) and/or fever (780.6) and where G-CSF primary prophylaxis was not reimbursed, were retrieved from the IMS Hospital Disease database for the period 2005-2008. This database includes longitudinal (per calendar year) information on diagnoses and drugs prescribed in about 34% of all Belgian hospital beds. Incidence of FN (cases of FN with chemo-cancer combination divided by total number of patients with this chemo-cancer combination), mortality in patients with and without FN and impact of FN on subsequent chemotherapy treatment decisions were assessed. RESULTS: Among the 25,544 patients at risk, 3,191 (13%) had at least one FN episode. Highest incidence rates were found in combinations of cisplatin-containing regimens with head and neck (71/287, 25%), stomach (24/110, 22%) and esophagus (36/202, 18%) cancers, lung cancers treated with cisplatin-etoposide (52/292, 18%) or carboplatin-etoposide (102/659, 16%) regimen and multiple myeloma treated with doxorubicin-vincristine regimen (26/152, 17%). Overall, 50% of first FN episodes occurred during cycle 1. Of the 3191 FN patients 11% died, 24% switched chemotherapy regimen and 22% stopped treatment during the cycle with FN. FN occurred subsequently in 27% of 1367 patients continuing the same regimen. CONCLUSIONS: This study suggests clinically significant FN-incidence is associated with chemotherapy regimens where G-CSF primary prophylaxis is not reimbursed in Belgium, which may lead to negative outcomes in terms of mortality and treatment disruption.