Cancer patients' willingness to pay for blood transfusions at home: results from a contingent valuation study in a French cancer network.

Home blood transfusion may be an interesting alternative to hospital transfusion, especially when given with curative or palliative intent or for terminal care in advanced-stage cancer patients. However, there is limited information about patients' attitude toward this type of care. The purpose of this study was to measure French cancer patients' willingness to pay (WTP) for home blood transfusion and to analyze determinants of their choice. A contingent valuation survey was administered to 139 patients receiving transfusions in the framework of a regional home care network or in the hospital outpatient department. Participation was high (90%). Most patients (65%) had received home care, including 43% blood transfusions. Just under half of the patients gave a zero WTP, among which we identified 8 protest bidders. The median WTP for home blood transfusion was 26.5 per patient. In multivariate analysis, long home-hospital distance, poor quality of life, and previous experience of home care were identified as important factors in determining how much more patients would be willing to pay for transfusion at home. These results demonstrate the benefits of developing domiciliary services to improve patient well-being, notably for the weakest among them. The significant impact of previous home care experience on WTP is probably related to the strong involvement of physicians from the blood center and to their active contribution to a high-level homecare network. Some of our findings could be useful for policy decision-making regarding home care.

Assessing short-term effects and costs at an early stage of innovation: the use of positron emission tomography on radiotherapy treatment decision making.

OBJECTIVES: Positron emission tomography (PET) is an innovative imaging tool. Associated with computed tomography (CT), it allows a better definition for the tumor volume for radiotherapy, compared with CT only. The aim of this study was to assess the effects of PET on resource allocation (costs and savings) and on the choice of the following treatment in radiotherapy. METHODS: In 2004 and 2005, 209 patients were enrolled (97 patients with Hodgkin's disease and 112 with non-small cell lung cancer) in a national study conducted in eight hospitals. Two treatment decisions made on the basis of CT only or CT associated with PET, were compared in a prospective study where each subject was his/her own control. The direct medical cost of using PET was assessed by microcosting, using data collected from specific questionnaires. The costs of new tests and the costs and savings associated with changes in the chosen treatment were calculated on the basis of reimbursement rates. RESULTS: The mean cost of using PET was approximately euro 800 per patient (50 percent for the radionuclide 18F-FDG [2-[18F]fluoro-2-deoxy-D-glucose]). Radiotherapy treatments were modified for 10 percent of patients with Hodgkin's disease versus 40 percent of patients with lung cancer. Overall, the use of PET induced both increases and decreases in the mean cost per patient: the net effect was a euro 425 and euro 931 cost increase in lung cancer and Hodgkin's disease, respectively. CONCLUSIONS: The use of PET for radiotherapy decision making seems more valuable for lung cancer than for Hodgkin's disease, both in terms of costs and changes in radiotherapy treatment. This result might help policy makers for prioritization.

Cost-effectiveness analysis of strategies for HER2 testing of breast cancer patients in France.

OBJECTIVES: A cost-effectiveness analysis was conducted comparing diagnostic strategies for determining the HER2 status of invasive breast carcinomas, as an indication for trastuzumab at metastatic relapse. METHODS: A decision tree compared five strategies distinguished by (i) the use of immunohistochemical (IHC) and/or fluorescent in situ hybridization (FISH) techniques, and (ii) the test schedule (at initial diagnosis or metastatic relapse). Most cost and effectiveness data came from a French multicentric study of 2,045 patients from eight hospitals. We were not able to select final criteria for trastuzumab effectiveness, because published data rely on IHC techniques not used in France (i.e., HercepTest). We, therefore, selected two intermediate criteria for inappropriate treatment at relapse, that is, patients with HER2-amplified tumors not receiving trastuzumab (Criterion 1) and HER2-nonamplified tumors improperly treated with trastuzumab (Criterion 2). Sensitivity analyses were then performed to assess the robustness of the results to (i) discount rate, (ii) cost of FISH, and (iii) tissue fixation technique. RESULTS: The strategy using IHC at diagnosis was dominated by the four other strategies. Among these approaches, the only efficient strategy for both criteria was IHC used alone at metastatic relapse; strategies using FISH, or IHC followed by FISH on IHC2+ cases were efficient for Criterion 1, whereas IHC followed by FISH on IHC2+ and 3+ cases was efficient for Criterion 2. CONCLUSIONS: Determining HER2 status at diagnosis, as an indication for trastuzumab at metastatic relapse, incurs substantial incremental costs, which do not appear to be justified. No other strategy can be excluded at first.

Cost comparison of two surgical strategies in the treatment of breast cancer: sentinel lymph node biopsy versus axillary lymph node dissection.

OBJECTIVES: The feasibility and accuracy of sentinel lymph node biopsy (SLNB) in the treatment of breast cancer is widely acknowledged today. The aim of our study was to compare the hospital-related costs of this strategy with those of conventional axillary lymph node dissection (ALND). METHODS: A retrospective study was carried out to determine the total direct medical costs for each of the two medical strategies. Two patient samples (n = 43 for ALND; n = 48 for SLNB) were selected at random among breast cancer patients at the Centre Leon Bérard, a comprehensive cancer treatment center in Lyon, France. Costs related to ALND carried out after SLNB (either immediately or at a later date) were included in SLNB costs (n = 18 of 48 patients). RESULTS: Total direct medical costs were significantly different in the two groups (median 1965.86 Euro versus 1429.93 Euro, p = 0.0076, Mann-Whitney U-test). The total cost for SLNB decreased even further for patients who underwent SLNB alone (median, 1,301Euro). Despite the high cost of anatomic pathology examinations and nuclear medicine (both favorable to ALND), the difference in direct medical costs for the two strategies was primarily due to the length of hospitalization, which differs significantly depending on the technique used (9-day median for ALND versus 3 days for SLNB, p < 0.0001). CONCLUSIONS: A lower morbidity rate is favorable to the generalization of SLNB, when the patient's clinical state allows for it. From an economic point of view, SLNB also seems to be preferred, particularly because our results confirm those found in two published studies concerning the cost of SLNB.

A qualitative approach to the use of economic data in the selection of medicines for hospital formularies: a French survey.

OBJECTIVE: Qualitative interviews were conducted with pharmacists in hospitals and clinics in the Rhône-Alpes region of France to determine the role of economic data when selecting medicines for formularies, to identify barriers to the use of this information and to study to what degree a healthcare establishment's financing system influences the use of this data. METHOD: A stratified sample of healthcare establishments with over 100 short-stay beds were included: (1) thirteen public and semi-private hospitals financed through annual global budgets and (2) six private clinics financed on a fee-for-service basis. Interviews were carried out between October 1999 and January 2000, and coded independently by two researchers. MAIN OUTCOME MEASURE: A multiple correspondence analysis was performed to compare the two groups of healthcare establishments. RESULTS: The influence of economic data in the decision-making process is limited, for other factors appear to have greater weight: (1) efficacy and safety of medicines (2) relations between decision-makers and the pharmaceutical industry and (3) patient quality of life. Economic data used was mainly related to medication prices and quantities consumed. This data was used in a large number of decisions and seemed to have more importance in hospitals than in clinics. Information related to resources that could be saved by the inclusion of a new medicine on formularies was seldom used and apparently considered less important in hospitals than in clinics. Pharmacoeconomic evaluations were very rarely used. Six barriers to the use of economic data were raised by the pharmacists, including: lack of time, which limits the collection and analysis of such information; insufficient health economics training, an obstacle to decision-makers' analytical capacity; and closed budgets within hospitals. CONCLUSION: Economic data concerning 'medication budgets' appears to have a greater impact in public and semi-private hospitals than in private clinics. Obstacles linked to the decision-making context itself were particularly highlighted, and it can be concluded that in order to increase the use of economic data, it is first necessary to create an environment that is more favourable to its application.

Economic evaluation of antineoplasic chemotherapy administered at home or in hospitals.

OBJECTIVES: Comparative economic evaluations of chemotherapy administered in hospital day-care units or in the home are relatively scarce. Furthermore, most existing evaluations do not include methodologic studies. This study seeks to compare the costs of anticancer chemotherapy with hospital at-home care versus a hospital day-care unit in the Rhône-Alpes region of France. METHODS: This study is based on a randomized controlled crossover trial that included 42 patients, to whom chemotherapy courses were alternatively given in both settings. All cost categories were taken into account according to microcosting methods. A detailed assessment was performed on coordination and health care in both structures (marginal costs and average costs), from the viewpoint of society. RESULTS: The marginal cost for one chemotherapy administration was significantly higher with hospital at-home care than in the hospital day-care unit ($232.5 vs. $157, p < .0001). Conversely, the average cost was significantly lower with home care than at the hospital ($252.6 vs. $277.3, p = .0002). CONCLUSIONS: The results show that the interest of developing home care in anticancer chemotherapy is questionable regarding costs. In the French healthcare system, where there is a surplus of hospital beds, marginal costs seem to be more relevant indicators in most cases than average costs.