|
Cost-effective prostate cancer detection: reduction of low-yield biopsies |
Littrup P J, Kane R A, Mettlin C J, Murphy G P, Lee F, Toi A, Badalament R, Babaian R |
|
|
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 Digital rectal examination (DRE), transrectal ultrasound (TRUS) and prostate specific antigen (PSA)-related tests in the detection of prostate cancer.
Economic study type Cost-effectiveness study.
Study population Men undergoing testing for prostate cancer detection.
Setting The practice setting was US hospitals involved in the American Cancer Society National Prostate Cancer Detection Project.
Dates to which data relate The Prostate cancer detection project was conducted between 1987-1994. There is no indication of the years to which prices relate.
Source of effectiveness data Effectiveness data were derived from a single study
Link between effectiveness and cost data Costing was undertaken retrospectively, on the same patient sample as that used in the effectiveness study.
Study sample 2,930 men with complete data sets formed the study population. The men were aged 55-70 years upon entry, had no history of prostate cancer and provided informed consent for participation. No power calculations related to the sample size were stated, but the authors referred to earlier studies in which methodological details are given.
Study design Retrospective cohort study based on a multicentre trial. The period of follow-up was 5 years. Loss to follow-up was 69 out of an initial cohort of 2,999.
Analysis of effectiveness The analysis of the clinical study was based on treatment completers only. The primary health outcomes used in the analysis were the percentage reduction in biopsy and the percentage increase in cancer detected.
Effectiveness results Systematic biopsy detected 80.7% of cancers in the first year whilst the tailored-biopsy approach (biopsy of patients with volume disproportionate PSA elevations or suspicious DRE) detected 71.9% of cancers but resulted in a 34% reduction in the number of biopsies performed.
Clinical conclusions Rather than perform systematic biopsy on all patients with PSA levels greater than 4ng/ml, a 16-55% reduction in biopsies could be achieved with a cancer loss of 4-25% by limiting biopsy to patients with an increased PSAD (PSA density) level and/or abnormal results of DRE.
Modelling Decision trees and receiver-operator characteristic curve analysis (ROC) were used to assess (retrospectively) the differential effects of biopsy decisions in Year 1 for patients with PSA levels greater than 4ng/ml or with a suspicious DRE.
Measure of benefits used in the economic analysis The outcome measures used in the economic evaluation were the percentage reduction in biopsies and the percentage increase or decrease in cancer detected. Clinicians' values were used in assessing health states.
Direct costs Direct health service costs were considered: costs of performing DRE, PSA, TRUS and cost of biopsy. Quantities and costs were not considered separately. The costs were not discounted.
Sensitivity analysis No sensitivity analysis was performed but the authors suggest that their ranking of cost-effectiveness would be robust to different biopsy costs.
Estimated benefits used in the economic analysis The cancer detection rates for the different screening and biopsy strategies analysed were:
2.5% for DRE alone (tailored),
3.7% for PSA3 alone (PSA at 3ng/ml, tailored),
3.4% for PSA4 alone (PSA at 4ng/ml, tailored),
4.1% for PSA4+DRE (tailored),
4.6% for PSA4+DRE (systematic),
3.2% for age PSA (tailored),
3.3% for age PSA (systematic),
3.9% for age PSA +DRE (tailored) and
4.4% for age PSA+DRE (systematic).
Systematic biopsy consistently resulted in a 50% increase in the number of biopsies performed, but with only a 3-13% increase in cancer yield. The tailored biopsy approach (i.e. biopsy of all patients with disproportionate PSA elevations or concurrently suspicious TRUS and DRE) resulted in a 34% (16-55%) reduction in biopsy with only an 11% (4-25%) loss in cancer detection. The smallest decrease in cancer detection per biopsy avoided (4.5%) was obtained by including all DRE abnormalities or an elevated PSAD level (p<0.05).
Cost results The costs for the screening and biopsy strategies analysed were:
$114,320 for DRE alone (tailored),
$134,860 for PSA3 alone (PSA at 3ng/ml, tailored),
$116,080 for PSA4 alone (PSA at 4ng/ml, tailored),
$177,100 for PSA4+DRE (tailored),
$218,600 for PSA4+DRE (systematic),
$100,600 for age PSA (tailored),
$111,600 for age PSA (systematic),
$166,100 for age PSA +DRE (tailored) and
$199,100 for age PSA+DRE (systematic).
Synthesis of costs and benefits The cost per cancer for the screening and biopsy strategies analysed was:
$4,511 for DRE alone (tailored),
$3,645 for PSA3 alone (PSA at 3ng/ml, tailored),
$3,414 for PSA4 alone (PSA at 4ng/ml, tailored),
$4,320 for PSA4+DRE (tailored),
$4,752 for PSA4+DRE (systematic),
$3,144for age PSA (tailored),
$3,382 for age PSA (systematic),
$4,259 for age PSA +DRE (tailored) and
$4,525 for age PSA+DRE (systematic).
A tailored approach produced a reduction of 17.5% in total cost.
Authors' conclusions Cost-effective prostate cancer detection with PSA as a parameter is better achieved if screening and biopsy decisions are not linked intimately. A tailored-biopsy approach for patients with disproportionately elevated PSA levels or suspicious DRE results in the greatest biopsy reduction by selecting lower risk groups for more conservative follow-up. As a rough estimate, the 17.5% reduction in total cost by using the tailored-biopsy approach for PSA elevations of 2.5-10 ng/ml could translate into national savings of approximately $1 billion per year for the estimated 28 million US men aged 50-70 years. At current detection levels of 200,000 cancers per year, $67 million could be saved each year with a tailored-biopsy approach.
CRD Commentary This was a very well conducted multicentre trial. However, the potential bias of the study, acknowledged by the authors, was towards detecting either visible or palpable cancers. In addition, average costs (for example, biopsy costs) were used in estimating the cost-effectiveness and there is no clear indication of whether or not such figures are good approximations of marginal costs.
Implications of the study Given the lack of any conclusive mortality reduction benefit from early detection of prostate cancer, it seems appropriate that informed consent to an invasive procedure may include a cancer-risk estimate in order to involve the patient in biopsy decisions. Increased patient involvement empowers not only the patient but may help to assuage the growing medicolegal implications in this area. Of course, how the risks are presented to patients in practice is another issue.
Source of funding Supported by a Cancer Control Grant from the American Cancer Society.
Bibliographic details Littrup P J, Kane R A, Mettlin C J, Murphy G P, Lee F, Toi A, Badalament R, Babaian R. Cost-effective prostate cancer detection: reduction of low-yield biopsies. Cancer 1994; 74(12): 3146-3158 Other publications of related interest Mettlin C J et al. Findings on the detection of early prostate cancer in 2425 men. Cancer 1991;67:2949-58.
Babaian R J et al. The relationship of PSA to DRE and TRUS in detecting adenocarcinoma of the prostate. Cancer 1992;69:1195-200.
Kane R A et al. PSA levels in 1695 men without evidence of prostate cancer. Cancer 1992;69:1201-7.
Mettlin C J et al. Characteristics of prostate cancers detected in multi-modality early detection. Cancer 1993;72:1701-8.
Mettlin C J et al. Relative sensitivity and specificity of serum PSA level compared to age-referenced PSA, PSA densityand PSA change: data from the American Cancer Society National Prostate Cancer Detection Project. Cancer 1994;74:1615-20.
Comment in: Cancer 1994;74(12):3077-9.
Indexing Status Subject indexing assigned by NLM MeSH Age Factors; Aged; Biopsy /economics; Cost-Benefit Analysis; Decision Support Techniques; Humans; Male; Middle Aged; Physical Examination /economics; Predictive Value of Tests; Prostate /pathology; Prostate-Specific Antigen /analysis; Prostatic Neoplasms /diagnosis /ultrasonography; ROC Curve; Rectum; Retrospective Studies AccessionNumber 21995000069 Date bibliographic record published 30/09/1998 Date abstract record published 30/09/1998 |
|
|
|