|
Role of ultrasonography in detecting mammographically occult breast carcinoma in women with dense breasts |
Corsetti V, Ferrari A, Ghirardi M, Bergonzini R, Bellarosa S, Angelini O, Bani C, Ciatto S |
|
|
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 authors assessed ultrasonography carried out using an Aloka Pro Sound SSD-5500 echograph with a linear multifrequency 5- to 10-MHz probe in addition to mammography in the detection of breast cancer. They compared this with mammography alone.
Type of intervention Diagnosis of breast cancer.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised women aged between 25 and 96 years who were undergoing breast palpation and mammography. There were no specific inclusion or exclusion criteria.
Setting The setting was tertiary care. The economic study was carried out in Italy.
Dates to which data relate The effectiveness and resource use data were collated for patients diagnosed between January 2000 and September 2004. A price year was not reported.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was carried out on the same sample of patients as that used in the effectiveness study.
Study sample The study sample comprised women presenting consecutively at the study setting to undergo breast palpation and mammography between the dates of the study. There was no report that power calculations were carried out to estimate how large the sample would need to be to ensure statistically significant results and to reduce the possibility that the results were due to chance alone. Nevertheless, the authors had a very large sample size (a consecutive series of 17,883 women were included in the study).
Study design The authors designed a diagnostic study that aimed to determine the difference between mammography and ultrasonography in their ability to detect breast carcinoma. All participants underwent mammography. If mammography was positive or even minimally suspicious, then participants did not go forward for ultrasonography for the purposes of this study. Only if mammography showed a clear negative result did participants then receive ultrasonography. Ultrasonography was treated as the 'gold' standard diagnostic technique for the purpose of this study, owing to it not being affected by the masking effects of dense breasts. The analysis was conducted at a single centre. Cancer cases detected by ultrasound alone were reviewed first by four internal reviewers and then by an external reviewer with extensive experience. The reviewers were blinded, with positive cases being mixed with negative controls (the case-to-control ratio was 1:1 for internal review and 1:4 for external review).
Analysis of effectiveness The primary health outcomes were the number of cancers detected using mammography, the number of suspicious mammographies and diagnosed cancers, and the number of negative mammographies. Following this, the authors noted the number of ultrasonographies performed and the number of additional cancers detected using ultrasonography (i.e. false-negative cases with mammography). To assess possible clinical and ultrasonographic abnormalities, other investigations such as targeted radiological examination, fine-needle aspiration cytology and core biopsy were conducted. Therefore, the potential false-positive cases with ultrasonography were reported. No statistical analysis was reported.
Effectiveness results Out of a total of 17,883 mammographies, the number of cancers detected was 167 (detection rate 0.93%), the number of suspicious mammographies was 257 and diagnosed cancers 138, and the number of negative mammographies was 11,177.
A total of 6,449 ultrasonographies performed on mammogram participants were classified as not suspicious.
The number of additional cancers detected using ultrasonography was 29 (detection rate 0.44%, 29/6,449), accounting for 17.3% of all cancers detected.
The false-positive rate with ultrasonography was reported to be 1.25%.
The notes for 25 of these 29 cases were mixed with negative controls and examined by internal and external reviewers. After excluding symptomatic cases and cases reviewed as suspicious at mammography, the number of diagnoses of additional cancers that could be attributed to ultrasonography alone was 15 (detection rate 0.23%, 15/6,449).
The authors also provided a breakdown of these additional cancers detected by age, stage, morphology and radiological density.
The detection rate was 0.11% in the under 40 age group, 0.22% in the 40 - 49 age group, 0.32% in the 50 - 59 age group, and 0.14% in the over 59 age group.
Clinical conclusions The authors concluded that "present experience confirms the possibility of ultrasonography detecting mammographically occult cancer in dense breasts".
Measure of benefits used in the economic analysis The authors estimated the number of women examined and the number of additional cancer cases detected by ultrasonography as summary measures of health benefit. The estimation of benefits was obtained directly from the effectiveness analysis.
Direct costs A very basic cost analysis was carried out in which the cost of diagnosing participants was estimated. Although a perspective was not reported, the unit costs estimated seem to suggest that the perspective of the health care provider (tertiary care clinic) was adopted. The costs of palpation, ultrasonography, ultrasonography-guided percutaneous sampling, cytological examination, histological examination and surgical biopsy were estimated. The authors considered two scenarios for surgical biopsy: outpatient procedure under local anaesthesia and hospitalisation. The unit costs were based on 'current charges' at the study setting for each diagnostic procedure required, and the unit costs and the quantities were reported separately. Quantities were observed from the effectiveness study. Discounting was not required as the authors were interested in the immediate costs of diagnosis. Reflation does not appear to have been necessary as the unit costs seem to have remained unchanged over the period of the study. A price year would have completed the report.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not estimated and were not relevant to the apparent perspective.
Sensitivity analysis The authors explored the impact of differing surgical biopsy scenarios.
Estimated benefits used in the economic analysis From the effectiveness analysis, 17,883 women were examined and 15 detected cancers were attributable to ultrasonography.
Cost results The total cost for all participants was EUR 387,717.78 (outpatient surgical biopsy) or EUR 399,041.40 (inpatient surgical biopsy).
Synthesis of costs and benefits The authors did not report the extra cost per additional cancer detected (i.e. the incremental cost-effectiveness ratio) for ultrasonography compared with mammography alone. They reported the mean cost per additional cancer detected using both mammography and ultrasonography (based on the 6,449 participants classified as having a non suspicious mammography).
The cost per additional cancer detected was EUR 25,847.85 (outpatient surgical biopsy) or EUR 26,602.76 (inpatient surgical biopsy).
The cost per woman examined was EUR 21.68 (outpatient surgical biopsy) or EUR 22.31 (inpatient surgical biopsy).
Authors' conclusions The addition of ultrasonography to mammography in dense breasts may be a useful policy. However, it should be carefully monitored, as it is associated with substantial costs.
CRD COMMENTARY - Selection of comparators The authors compared ultrasonography with mammography. Mammography is current practice in the authors' setting, while ultrasonography represented a more sensitive diagnostic tool that was associated with greater cost. The authors aim, therefore, was to achieve greater understanding of the usefulness of ultrasonography in their setting.
Validity of estimate of measure of effectiveness The authors designed a diagnostic study that centred on a review of potential cancer cases originally categorised as negative at mammography. The use of both internal and external reviewers and the case-mix of positive and negative mammographies improved the internal validity of the results. The study population comprised individuals self-referred to the tertiary clinic setting and the study sample was representative of this. Unfortunately, as the authors acknowledged, a limitation of the study was that this population does not represent the broader screening population in which individuals do not self-refer for screening. Therefore, the effectiveness results are not directly generalisable to the screening population and this limits the usefulness of the results. Some sensitivity analyses could have been carried out to explore differences between this population and the general screening population, and thus estimate the impact. Despite this, the results are indicative and the very large sample size improves the validity of the results for the study population.
Validity of estimate of measure of benefit The number of women examined and the number of additional cancers detected were used as summary measures of health benefit. These are sensible choices given the diagnostic nature of the study. They also enable comparisons to be made with other diagnostic analyses.
Validity of estimate of costs The authors did not set out to conduct a cost-effectiveness analysis. Instead, their aim was to understand the usefulness of ultrasonography. Nevertheless, to improve the readers understanding of the issues they carried out a cost analysis which, although basic, was very thoroughly reported. The analysis focused on the cost of each diagnostic technique. The unit costs and the quantities were reported separately, and the authors were very clear about which factors were included in the analysis. Charges were used as a proxy of diagnostic costs. The use of charges to proxy costs has the limitation of not reflecting true opportunity costs, thus restricting the external validity of the results. The costing would have been improved by a clear statement of the perspective from which the costs were estimated, and the inclusion of a price year. Further, the additional cost of ultrasonography per case detected was not investigated, which might have introduced a misunderstanding when comparing mammography and ultrasonography strategies: the additional cost of the ultrasonography strategy per case detected would have been lower than the cost per additional case detected reported in the present paper.
Other issues Several comparisons were carried out with findings from other studies, although the authors reported that such comparisons are difficult "due to differences in age distribution, criteria to define a dense breast, ultrasonographic equipment and accuracy". Despite these potential problems, the authors reported that the diagnostic contribution of ultrasonography was found to be in agreement with previous reports. The issue of generalisability was considered at length, with the authors commenting on the self-selected nature of the sample (as already mentioned). The results were presented clearly and without evidence of selective reporting. Several limitations were discussed, such as the fact that the reviewers knew they were reanalysing results. This might have maximised their mammographic suspicions and introduced a bias into the results. In addition, the authors stressed that their cost analysis was only indicative and did not include working times or allow for organisations to differ between settings. Overall, the conclusions were an accurate reflection of the results presented and related well to the scope of the study.
Implications of the study Owing to the limited generalisability of the results, the authors acknowledged that their current results "do not constitute sufficient evidence to support the introduction of this policy in current screening practice". They highlighted current ongoing prospective controlled trials that might improve the body of evidence in this area.
Bibliographic details Corsetti V, Ferrari A, Ghirardi M, Bergonzini R, Bellarosa S, Angelini O, Bani C, Ciatto S. Role of ultrasonography in detecting mammographically occult breast carcinoma in women with dense breasts. Radiologia Medica 2006; 111(3): 440-448 Other publications of related interest Zappa M, Spagnolo G, Ciatto S. Measurement of the costs in two mammographic screening programmes in the province of Florence. J Med Screen 1995;2:191-4.
Indexing Status Subject indexing assigned by NLM MeSH Adult; Age Factors; Aged; Aged, 80 and over; Breast /pathology; Breast Neoplasms /radiography /ultrasonography; Carcinoma /radiography /ultrasonography; Case-Control Studies; Costs and Cost Analysis; Female; Humans; Mammography /economics; Mass Screening /economics; Middle Aged; Predictive Value of Tests; Ultrasonography, Mammary /economics AccessionNumber 22006000863 Date bibliographic record published 31/12/2006 Date abstract record published 31/12/2006 |
|
|
|