|Finasteride: clinical and economic impacts
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.
A new medical treatment for benign prostatic hypertrophy (BPH), finasteride, was compared with the two most common treatment options that is transurethral resection of the prostate (TURP) and watchful waiting, for patients with varying degrees of symptom severity.
Type of intervention
Treatment; secondary prevention.
Economic study type
Cost-effectiveness analysis and cost-utility analysis.
Males receiving either finasteride, TURP or watchful waiting for the treatment of BPH. Patients had either mild, moderate or severe symptoms according to the American Urological Association (AUA) Symptom Index. Patients were classed according to their life expectancies of less than 3 years,4-14 years and longer than 14 years.
The practice setting was the community, primary or secondary care, depending on the health intervention concerned. The economic study took place at the Canadian Coordinating Office of Health Technology Assessment (CCOHTA), Canada.
Dates to which data relate
Although it was not stated directly, it appears that 1992-1995 data were used for the effectiveness analysis and the resources. The price date was not specified.
Source of effectiveness data
The estimate for final outcomes was based on a review of previously completed studies. The clinical information was obtained from randomised controlled trials (RCTs) of finasteride, and Clinical Practice Guidelines published by the Public Health Service Agency for Health Care Policy and Research (AHCPR).
Decision trees were used based on results from two RCTs of finasteride. The Clinical Practice Guidelines published by the AHCPR were used to derive the probabilities of improvement and no improvement for the three interventions, as well as the probability of improvement with incontinence or impotence, incontinence or intraoperative death with TURP. Decision trees were used for the cost-effectiveness and cost-utility analyses.
Outcomes assessed in the review
Symptom improvement using symptom severity scores based on the AUA Symptom Index and Quality Adjusted Life Years (QALYs). Outcomes included improvement, and no improvement but no further therapy for the watchful waiting and finasteride approaches. For TURP, improvement, improvement with impotence or incontinence, no improvement, incontinence and intraoperative death outcomes were estimated.
Study designs and other criteria for inclusion in the review
Two randomised controlled trials of finasteride versus placebo in men with BPH.
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
The methods used to judge the relevance and validity of the primary studies were not described. In terms of the methodology used to extract data from the primary studies, the report mentioned only that the clinical information was derived from RCTs and clinical practice guidelines. This information was then used as input into a decision model.
Number of primary studies included
Methods of combining primary studies
Investigation of differences between primary studies
Results of the review
Based on a Clinical Practice Guideline Publication (see text for reference), approximately 67% of patients improved on the drug as compared to 42% who improved with watchful waiting. Of those who responded, improvement was maintained over the 3 year period. Finasteride versus watchful waiting was found to decrease prostate size and urinary symptoms and to improve the symptom score. Adverse effects from finasteride therapy were minimal, but concern was raised over the approximately 50% drop in prostate-specific antigen (PSA) which makes the diagnosis of prostate cancer more tentative.
Measure of benefits used in the economic analysis
QALYs were estimated. A decision tree was used to derive expected QALYs for TURP, watchful waiting and finasteride. The severity of patient symptoms was measured using the AUA Symptom Index: mild = 0-7 points; moderate = 8-19 points; severe = 20-35 points. The Health Utilities Index - Mark II System was chosen as the utility index (the index covers sensation, mobility, emotion, cognition, self-care, pain and fertility). It is not stated by whom the utilities were valued but presumably the valuation was patient-based.
Annual direct costs from the health care system perspective were determined for each intervention. The following costs were included: hospital and OR costs, diagnostic procedure and laboratory tests, professional fees, complication costs and drug costs. In the decision model, the costs for each outcome probability were determined and summed to produce the total expected costs. The probabilities of each outcome for the three interventions were taken from clinical practice guidelines, while the efficacy of finasteride was estimated from two RCTs. One would expect discounting to have been applied as the time horizon was one to 15 years, however none was reported. Estimations of costs/quantities were derived from a decision model. Quantities were not reported. The price date was not specified.
Statistical analysis of costs
A number of different scenarios were presented, based upon varying degrees of symptom severity and different life expectancies. Also, the probabilities of the outcomes were varied in the decision model, using high and low probabilities and costs to determine whether the results of the analysis would be impacted by the assumptions and clinical data available. The method used was probabilistic sensitivity analysis.
Estimated benefits used in the economic analysis
QALYs were used as the measure of benefit, butthey were not reported in this overview. For patients with moderate symptoms and a life expectancy of 3 years or less, finasteride produced a better quality of life than watchful waiting. Finasteride produced more QALYs than surgery, for men suffering from mild symptoms and whose life expectancy was 14 years of less. For men who have severe symptoms, surgery produced a better quality of life. For moderate symptoms, watchful waiting produced a better quality of life than surgery. The time horizon for the analysis was from 1 to 15 years.
The side effects of treatment were considered in terms of the complication costs for TURP, but other side effects such as those from drug treatment, were not considered in the economic analysis. Although the results from the two RCTs of finasteride versus placebo suggest that adverse effects from the drug were minimal, concerns were raised surrounding the approximately 50% drop in prostate-specific antigen (PSA) which makes the diagnosis of cancer more tentative.
The annual direct mean costs to the health care system were as follows. In the first year TURP costCan$5,408, finasteride cost Can$926, and the cost was Can$142 for watchful waiting. For subsequent years, the cost was Can$41 for TURP, Can$821 for finasteride, and Can$142 for watchful waiting. For patients with moderate symptoms and a life expectancy of 3 years or less, finasteride was cheaper than watchful waiting. From 3 years on, finasteride was more expensive than watchful waiting. Estimates found that it would cost the health care system Can$842,000 more to use finasteride over watchful waiting over a 15 year period (assuming that watchful waiting remains successful over this period). If, on the other hand, the assumption is that all these patients would have to be medically treated within a year, then the extra cost would only be Can$19,000.
Finasteride was less costly than surgery for mild symptoms and those patients with life expectancy of 14 years or less, but more expensive after 14 years. For severe symptoms, surgery was more expensive than finasteride. This cost difference decreased over time until a life expectancy of 14 years or more is anticipated and then surgery was less costly. Watchful waiting was always less costly than TURP. In terms of the aggregate impact, the addition of finasteride only for patients classified as moderate (the population most likely to benefit) would be Can$2.7 million for every 10,000 men aged 60 or older in the general population. Assuming 30% of mild patients would end up receiving the drug (where watchful waiting is the appropriate therapy), then an additional Can$2.9 million or Can$5.6 million in total/10,000 men 60 years or older would be required to fund this therapy in Canada.
Synthesis of costs and benefits
Estimated costs and benefits were combined in terms of costs per QALY. No discount rate was provided. For those with moderate symptoms and a life expectancy of less than 3 years, finasteride was cheaper than other options. Between 4-14 years of life expectancy, finasteride had a slightly improved quality of life compared to watchful waiting. Cost/QALY ranged from Can$19,000 at 4 years to Can$44,000 at 15 years. Finasteride was a cheaper alternative with better results than surgery. For life expectancy over 14 years, the cost/QALY for finasteride was around Can$45,000 compared to watchful waiting. Compared to TURP, the slightly improved quality of life of finasteride now began to cost the health care system at a relatively minor additional cost of Can$1,000/QALY. For severe symptoms and life expectancy under 3 years, less quality of life was expected with finasteride but it was also a less expensive option compared to the other two treatments. For life expectancy of 4-14 years, finasteride would not be considered as first-line therapy. For life expectancy of 14 years of more, finasteride cost more and provided worse results than TURP or watchful waiting. TURP offered the highest QALY at a relatively low cost of Can$4,000 to Can$15,000.
The choice of finasteride is dependent upon two factors: life expectancy and the severity of symptoms. For men with mild symptoms, watchful waiting is most appropriate. For moderate and severe symptoms, the choice of therapy generally depends upon life expectancy, although surgery is often the preferred choice when symptoms are severe.
Additional detail about the two RCTs on which the study was based is required to make a more comprehensive assessment. No patient characteristics were provided and no report of the search strategy on which the analysis was based was reported. Costs and quantities were not reported separately. The economic evaluation was good and the results were presented clearly. To analyse the validity issues, more information is required. The overview stated that it was based on data reported in detail in a CCOHTA study by Baladi (see below).
Implications of the study
With the significant world-wide incidence and prevalence of BPH and prostate cancer, the findings of any study related to this neoplasm are certain to have important implications. This is particularly the case because there is still no consensus on how best to treat BPH, especially for men with moderate symptoms. If ethical considerations can be overcome, then further trials are required to assess the qualitative and quantitative effects of different treatments over several years. Given the number of men who present with prostate complaints, the resource implications are very significant and so would need to be assessed in a UK setting in order to test the generalisability of previous findings.
Otten N. Finasteride: clinical and economic impacts. Ottawa, ON, Canada: Canadian Coordinating Office for Health Technology Assessment. Technology Overview: Pharmaceuticals, 2. 1996
Other publications of related interest
The full study by Baladi J-F (Cost-effectiveness and cost-utility analyses of finasteride therapy for the treatment of benign prostatic hyperplasia. Ottawa: CCOHTA 1995) is available from CCOHTA.
The two RCTs of finasteride were:
Gormley G J et al. The effect of finasteride in men with benign prostatic hyperplasia. New England Journal of Medicine 1992;327:1185-1191.
Stoner E. Finasteride Study Group. Finasteride (MK-906) in the treatment of benign prostatic hyperplasia. Prostate 1993;22:291-299.
There have also been recent comprehensive reviews carried out by Selley etal (1996) and Marchant (1996) on the effectiveness and cost-effectiveness of treatments for BPH.
Subject indexing assigned by CRD
Cost-Benefit Analysis; Costs and Cost Analysis; Finasteride /economics /therapeutic use; Prostate /surgery; Prostatectomy /economics /methods; Prostatic Hyperplasia /diagnosis /therapy /economics; Treatment Outcome
Date bibliographic record published
Date abstract record published