BACKGROUND: A clinical trial of patients with de novo acute myeloid leukaemia (AML) showed that haematopoietic support with filgrastim (granulocyte colony-stimulating factor, G-CSF) following induction and consolidation chemotherapy accelerated recovery from neutropenia. The clinical benefits included reductions in infections, anti-infective therapy and length of hospital stay. OBJECTIVE: The objective of this economic analysis is 2-fold. First, it aims to determine if the observed clinical benefits from the use of filgrastim would lead to cost savings from the perspective of a healthcare institution in the UK. Second, the analysis compares the results of two methods on collection of resource use data. DESIGN: A retrospective cost-minimisation analysis was undertaken based on the clinical results of all UK patients enrolled in the trial. Two cost models were developed: a model based only on the medical resource use collected in the case report forms (the CRF model); and a model based on all medical resources collected from patient medical files (the PF Model). Treatment costs of AML between filgrastim and the placebo arm were compared for the first induction cycle as well as the first induction and the first consolidation cycles combined. Results from the two models were compared. SETTING AND PATIENTS: The CRF model was applied to two samples of patients: all UK patients (n = 82) and patients enrolled at one centre [the Manchester Royal Infirmary (MRI) (n = 30)], whereas the PF model was applied to the MRI patient sample only. RESULTS: For all UK patients, using the CRF model, the filgrastim-treated arm produced cost savings of 747 pounds sterling (9.0%) and 2135 pounds sterling (14.4%) [1998 values] per patient in the first induction cycle and in the induction and consolidation cycles combined, respectively. For the patients at MRI the CRF model resulted in cost savings with filgrastim of 177 pounds sterling (2.2%) and 414 pounds sterling (3.2%) per patient respectively. Using the PF model the savings at MRI were 910 pounds sterling (8.6%) and 1285 pounds sterling (8.0%) per patient, respectively. CONCLUSION: Use of filgrastim in the treatment of AML in the UK may result in net cost savings. A retrospective analysis using total resources obtained through patient files produced higher cost savings estimates than that obtained by resources noted in the CRFs. The models based on PF resource data may be more reliable because they are more comprehensive. However, the cost estimates in this study may have been impacted by sample size, site characteristics, disease and treatment settings. Therefore, further evaluation on the methods for collecting resource use data in larger, multicentred studies is warranted.
RCT Entities:
BACKGROUND: A clinical trial of patients with de novo acute myeloid leukaemia (AML) showed that haematopoietic support with filgrastim (granulocyte colony-stimulating factor, G-CSF) following induction and consolidation chemotherapy accelerated recovery from neutropenia. The clinical benefits included reductions in infections, anti-infective therapy and length of hospital stay. OBJECTIVE: The objective of this economic analysis is 2-fold. First, it aims to determine if the observed clinical benefits from the use of filgrastim would lead to cost savings from the perspective of a healthcare institution in the UK. Second, the analysis compares the results of two methods on collection of resource use data. DESIGN: A retrospective cost-minimisation analysis was undertaken based on the clinical results of all UK patients enrolled in the trial. Two cost models were developed: a model based only on the medical resource use collected in the case report forms (the CRF model); and a model based on all medical resources collected from patient medical files (the PF Model). Treatment costs of AML between filgrastim and the placebo arm were compared for the first induction cycle as well as the first induction and the first consolidation cycles combined. Results from the two models were compared. SETTING AND PATIENTS: The CRF model was applied to two samples of patients: all UK patients (n = 82) and patients enrolled at one centre [the Manchester Royal Infirmary (MRI) (n = 30)], whereas the PF model was applied to the MRI patient sample only. RESULTS: For all UK patients, using the CRF model, the filgrastim-treated arm produced cost savings of 747 pounds sterling (9.0%) and 2135 pounds sterling (14.4%) [1998 values] per patient in the first induction cycle and in the induction and consolidation cycles combined, respectively. For the patients at MRI the CRF model resulted in cost savings with filgrastim of 177 pounds sterling (2.2%) and 414 pounds sterling (3.2%) per patient respectively. Using the PF model the savings at MRI were 910 pounds sterling (8.6%) and 1285 pounds sterling (8.0%) per patient, respectively. CONCLUSION: Use of filgrastim in the treatment of AML in the UK may result in net cost savings. A retrospective analysis using total resources obtained through patient files produced higher cost savings estimates than that obtained by resources noted in the CRFs. The models based on PF resource data may be more reliable because they are more comprehensive. However, the cost estimates in this study may have been impacted by sample size, site characteristics, disease and treatment settings. Therefore, further evaluation on the methods for collecting resource use data in larger, multicentred studies is warranted.
Authors: C H Pui; J M Boyett; W T Hughes; G K Rivera; M L Hancock; J T Sandlund; T Synold; M V Relling; R C Ribeiro; W M Crist; W E Evans Journal: N Engl J Med Date: 1997-06-19 Impact factor: 91.245
Authors: J E Godwin; K J Kopecky; D R Head; C L Willman; C P Leith; H E Hynes; S P Balcerzak; F R Appelbaum Journal: Blood Date: 1998-05-15 Impact factor: 22.113
Authors: J L Harousseau; J Y Cahn; B Pignon; F Witz; N Milpied; M Delain; B Lioure; T Lamy; B Desablens; F Guilhot; D Caillot; J F Abgrall; S Francois; J Briere; D Guyotat; P Casassus; B Audhuy; Z Tellier; P Hurteloup; P Herve Journal: Blood Date: 1997-10-15 Impact factor: 22.113
Authors: G Heil; D Hoelzer; M A Sanz; K Lechner; J A Liu Yin; G Papa; L Noens; J Szer; A Ganser; C O'Brien; J Matcham; A Barge Journal: Blood Date: 1997-12-15 Impact factor: 22.113
Authors: J L Harousseau; B Witz; B Lioure; M Hunault-Berger; B Desablens; M Delain; F Guilhot; P Y Le Prise; J F Abgrall; E Deconinck; D Guyotat; J P Vilque; P Casassus; O Tournilhac; B Audhuy; E Solary Journal: J Clin Oncol Date: 2000-02 Impact factor: 44.544
Authors: V Trillet-Lenoir; J Green; C Manegold; J Von Pawel; U Gatzemeier; B Lebeau; A Depierre; P Johnson; G Decoster; D Tomita Journal: Eur J Cancer Date: 1993 Impact factor: 9.162
Authors: J M Rowe; J W Andersen; J J Mazza; J M Bennett; E Paietta; F A Hayes; D Oette; P A Cassileth; E A Stadtmauer; P H Wiernik Journal: Blood Date: 1995-07-15 Impact factor: 22.113
Authors: R J Mayer; R B Davis; C A Schiffer; D T Berg; B L Powell; P Schulman; G A Omura; J O Moore; O R McIntyre; E Frei Journal: N Engl J Med Date: 1994-10-06 Impact factor: 91.245