|
Cost utility analysis of prophylactic pamidronate for the prevention of skeletal related events in patients with advanced breast cancer |
Dranitsaris G, Hsu T |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The use of pamidronate infusions of 90 mg each, given monthly for the prevention of skeletal-related events (SREs) in patients with advanced breast cancer.
Study population The study population comprised women with advanced breast cancer who developed nonvertebral bone fractures.
Setting The setting was secondary care. The economic study was carried out in Toronto, Canada.
Dates to which data relate The effectiveness data were obtained from a clinical trial that was reported in 1996 (see Other Publications of Related Interest). The resource use data related to the years 1988 to 1997. Some unit costs related to 1992 and 1997; the dates for the other unit costs were not reported.
Source of effectiveness data The effectiveness data were derived from a review of published clinical studies plus estimates made by the authors.
Modelling A decision-analytic model was used to integrate the effectiveness, cost and health state preference data.
Outcomes assessed in the review The outcomes assessed in the review included the incidence of bone fractures, radiation treatment, hypercalcaemia and surgery to the bone. These were assessed in women on pamidronate therapy, compared with placebo.
Study designs and other criteria for inclusion in the review A review of the literature was carried out to find randomised, double-blind, placebo-controlled trials with bone fractures as end points.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Number of primary studies included One primary study was included in the review.
Methods of combining primary studies Not applicable since only one study was used.
Investigation of differences between primary studies Not applicable since only one study was used.
Results of the review Twenty per cent of the women in the pamidronate group developed at least one nonvertebral fracture, compared with 30% of those in the placebo group, (p=0.02).
Combining all skeletal-related complications, 46% (95% confidence interval, CI: 39 - 53) of the patients in the pamidronate group versus 62% (95% CI: 55 - 69) of those in the placebo group had at least one event.
When hypercalcaemia was excluded, the SRE rates were 43% (95% CI: 36 - 50) in the pamidronate group versus 56% (95% CI: 49 - 63) in the placebo group.
Methods used to derive estimates of effectiveness The authors made assumptions in the construction of the model.
Estimates of effectiveness and key assumptions It was assumed that 100 percent of patients who developed bony complications (e.g. SREs) would be admitted to hospital for treatment. (Correspondence with the author, subsequent to this abstract being written, indicates that this assumption was not realistic as, in the light of more recent data, not all patients who develop SREs become symptomatic, and those that do can often be treated as outpatients with low cost analgesics. A more recent trial has indicated that at least 60 percent of breast cancer patients with SREs did not require hospital admission due to these factors.
Measure of benefits used in the economic analysis This study used quality-adjusted life-years (QALYs) as an outcome measure. The baseline quality of life values were elicited from a sample of 25 women aged at least 18 years, using the time trade-off technique. Potential interviewees were screened by telephone and the 25 identified women were interviewed in person. Additionally, quality of life values were elicited from a non-random sample of 25 female health care professionals.
Direct costs The resource use quantities and the unit costs were analysed separately, though not all were reported in the paper. The cost analysis was relevant to a hospital perspective. It included the monthly 90 mg dose of pamidronate for the median 10 treatments, along with associated "drug preparation, administration, supplies and ambulatory care admission". Other costs were for hospitalisations, outpatient admissions, analgesics and other prescriptions, radiation treatments, blood products, diagnostic imaging, laboratory tests, paramedical services (such as physiotherapy and occupational therapy) and physician costs. The paper provided details on which of these cost elements were included when estimating the cost of each SRE.
The unit costs related to the two hospitals involved in the study. The resource use data related to the years 1988 to 1997. Some of the unit costs dated back as far as 1992, but no price year was given for the overall costs. The average costs per patient were used in the analysis. Discounting was irrelevant given the short timeframe of this study (12 months).
Statistical analysis of costs The authors reported the 95% CIs around the mean cost for each type of SRE. The mean costs were treated deterministically.
Indirect Costs The indirect costs were not included in this study.
Sensitivity analysis One-way sensitivity analyses were conducted using the 95% CIs as ranges for the SRE rates, treatment costs and quality of life values (from the general public). These analyses were carried out to investigate the robustness of the results to general uncertainty surrounding the model parameters. To evaluate the impact of the most resource intensive event (bone surgery), the data were reanalysed excluding this cost, and again, using a year-long cost of the surgery and rehabilitation obtained from another study. Finally, all of these analyses were repeated using the quality of life valuations elicited from the health care professionals.
Estimated benefits used in the economic analysis The average number of QALYs gained by patients on pamidronate, compared with placebo, was 0.15 over the 12-month duration of the model.
Cost results The annual costs per patient (weighted by the probability of SREs) were Can$9,180 on pamidronate and Can$6,380 on placebo.
Synthesis of costs and benefits The incremental costs and benefits were combined to yield an estimate of Can$18,700 per QALY gained. The most important factor in this result was the cost of treating SREs, the frequency of which was reduced with pamidronate treatment. The authors reported that, even assuming a worst-case scenario using the lower bound of the 95% CI for the treatment costs, the maximum incremental cost per QALY gained was Can$31,600.
Authors' conclusions These findings suggest that prophylactic pamidronate provides benefit to patients with advanced breast cancer at a "reasonable cost". A key economic benefit appears to be the reduction in the number of patients requiring costly bone surgery and ensuing rehabilitation.
CRD COMMENTARY - Selection of comparators This study used placebo as a comparator. The authors implicitly justified this, as they stated that advanced breast cancer patients would not normally be given pamidronate as a prophylactic therapy. This must be compared with current practice in your own setting for the prevention of SREs in this patient group.
Validity of estimate of measure of effectiveness The study from which the effectiveness data were derived appears to have had high internal validity. However, the external validity was more uncertain, as no other effectiveness studies were mentioned or employed in this study.
Validity of estimate of measure of benefit The authors referred to well-respected economic evaluation guidelines to justify their use of the general public to derive the quality of life measures. They restricted this sample to those who supported the health care system through tax contributions. The reasons for this criterion were not given. The authors also demonstrated that their results were fairly robust when the quality of life values elicited from health care professionals were substituted for those of the general public.
Validity of estimate of costs The authors presented detailed cost estimates for the treatments involved in this study. They acknowledged that the cost estimates related to the two hospitals involved in this study, and these may not be generalisable to other institutions. However, the sensitivity analysis that was conducted over a wide range of cost estimates suggested that this intervention was cost-effective over this range. Further, the authors pointed out that the economic appeal of pamidronate would have been further improved by the inclusion of physiotherapy, chronic care and indirect costs arising from SREs.
Hospital resource use following a fracture, radiation and hypercalcaemia were obtained from a retrospective chart audit. While this method has the advantage of reflecting actual use (as opposed to use within a clinical trial), the medical charts are not designed to collect study data. Thus, the accuracy of the data will be dependent on the clarity and comprehensiveness of the records.
It would appear that the resource use data were obtained from the charts of patients who were bisphosphonate (for example, pamidronate)-naive. In the model, these estimated costs were applied to a SRE (such as a fracture) for patients in both the placebo and pamidronate groups. The implicit assumption was that pamidronate treatment can affect the frequency, but not the resource-intensity, of SREs in this population. If pamidronate was associated with less severe SREs, then the incremental cost-effectiveness would be even more favourable. Excluding this possibility biases the study against pamidronate. Correspondence received from the authors indicates that the assumption made of 100 percent admission for SREs has been shown to be too high (see estimates of effectiveness section above) which would have the effect of reducing overall costs if a more realistic figure (approximately 40 percent) were used.
Other issues The decision analytic model appears to have limited the number of SREs a patient can have within the 12-month study period to one. If patients can have more than one event during this time, the model will be biased against pamidronate, that is, pamidronate is actually more cost-effective than suggested by the model results.
No comparison of these results was made with any other studies. It would have been interesting to know how this study contributes to, and compares with, the existing literature on the cost-effectiveness of prophylactic pamidronate in advanced breast cancer patients.
Implications of the study The authors conclude that, although prophylactic pamidronate appears to be a cost-effective intervention, the actual cost of providing it (in Ontario) would be large given the prevalence of metastatic breast cancer.
Source of funding Supported by a grant from Novartis Pharma Canada Inc.
Bibliographic details Dranitsaris G, Hsu T. Cost utility analysis of prophylactic pamidronate for the prevention of skeletal related events in patients with advanced breast cancer. Supportive Care in Cancer 1999; 7(4): 271-279 Other publications of related interest Hortobagyi GN, Theriault RL, Porter L, et al. Efficacy of pamidronate in reducing skeletal events in patients with breast cancer and lytic bone metastases. New England Journal of Medicine 1996;335:1785-91.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Antineoplastic Agents /administration & Bone Neoplasms /prevention & Breast Neoplasms /economics /pathology /therapy; Canada; Cancer Care Facilities /economics; Cost-Benefit Analysis; Diphosphonates /administration & Drug Administration Schedule; Female; Hospital Costs /statistics & Humans; Infusions, Intravenous; Middle Aged; Models, Econometric; Sensitivity and Specificity; Severity of Illness Index; control /secondary; dosage /economics; dosage /economics; numerical data AccessionNumber 21999001339 Date bibliographic record published 31/12/2002 Date abstract record published 31/12/2002 |
|
|
|