New guidelines from 3 professional organisations (EORTC, ASCO and NCCN) highlight the importance of identifying patients at risk of chemotherapy-induced neutropenia (CIN) and febrile neutropenia (FN) and providing preventative therapy to avert infection-related consequences and impaired treatment delivery. This prospective observational study was conducted to assess the incidence and predictors of CIN, FN and reduced chemotherapy administration in routine practice in Western Europe. Lymphoma patients starting new chemotherapy regimens were enrolled from 34 centres in Belgium, France, Germany, Spain and the UK. Treatment was as per usual clinical practice, except for one additional blood count at cycle 1 neutrophil nadir. Of the total 305 patients, 65 (21%) had Hodgkin lymphoma (HL) and 240 (79%) had non-Hodgkin lymphoma (NHL). Mean age at diagnosis ± SD was 39 ± 17 years for HL and 63 ± 13 years for NHL. Ann Arbour stages were distributed, I 9%; II 52%; III 19%; IV 20% in the HL group. In the NHL group, stages were distributed I 18%; II 26%; III 16%; IV 40%. Chemotherapy regimens for HL patients were mainly ABVD-like (72%), BEACOPP-like (12%) and Stanford V (8%); regimens for NHL patients were 3-weekly CHOP-like (74%), followed by 2-weekly CHOP-like (17%), ACVBP-like (4%) and NHL other (5%). Primary colony stimulating factor (CSF) prophylaxis was used in 18% of HL and 28% of NHL patients; whereas 37% of HL and 29% of NHL patients received secondary CSF prophylaxis. Primary CSF use by regimen type was: ABVD-like 9%; BEACOPP-like 63%; Stanford V 20%; 3-weekly CHOP-like 12%; 2-weekly CHOP-like 76%; ACVBP-like 56%; and NHL other 67%. FN occurred in 15% (95% CI 8–27%) of HL and 22% (CI 17–28%) of NHL patients. FN occurrence was ≥ 20% for BEACOPP-like; Stanford V; 3-weekly CHOP-like; ACVBP-like; and NHL other regimens. Grade 4 CIN was observed in 40% (CI 28–53%) of HL and 54% (CI 47–60%) of NHL patients. Neutropenia-related hospitalisations were reported for 14% of HL and 18% of NHL patients. Mean relative chemotherapy dose intensity (RDI) ± SD compared to plan, taking into account non-administered cycles, was 88 ± 13% for HL and 86 ± 21% for NHL. Low RDI (≤ 85%) was observed in 31% and 32% of HL and NHL patients, respectively. Multivariate logistic regression confirmed first cycle FN and fewer cycles with CSF use as predictors of low RDI (table). In line with new guidelines for CSF support during myelosuppressive chemotherapy, routine European practice should be revised to include primary prophylaxis with CSF for regimens where an FN incidence of ≥ 20% was seen (e.g. 3-weekly CHOP), and where less than optimal RDI was attained. This study was supported by Amgen (Europe) GmbH.Predictors of RDI ≤ 85%Odds ratio (95% CI)Age (per additional 10 years of age)1.38 (1.13–1.70)ECOG performance status ≥ 22.39 (1.12–5.11)Type of chemotherapy regimen †:2-weekly CHOP-like2.04 (0.90–4.65)ACVBP-like9.48 (2.12–42.40)NHL other1.98 (0.56–6.91)ABVD-like2.71 (1.09–6.76)HL other3.19 (0.96–10.65)Cycles with CSF administration (per additional cycle with CSF)0.85 (0.76–0.96)Cycle 1 FN occurrence2.82 (1.16–6.86)† 3-weekly CHOP-like, reference category, p = 0.018 for this set of covariates.