|
Could chest X-ray screening for lung cancer be cost-effective? |
Caro J J, Klittich W S, Strauss G |
|
|
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 A programme of annual chest X-ray screening for lung cancer was examined in male smokers.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised male smokers aged 45 to 80 years.
Setting The setting of the clinical study was unclear (chest X-ray screening could be performed in primary or secondary care). The economic study was performed in Massachusetts, USA.
Dates to which data relate The effectiveness data were taken from studies published between 1984 and 1986. No dates for resource use were reported explicitly. The price year was 1998.
Source of effectiveness data The effectiveness data were gathered from a non-systematic review of the literature.
Modelling A Markov model was used to estimate the cost-effectiveness of annual chest X-ray screening over a long time horizon (lifetime). The cycle length was one year, but no further details of the health states or other characteristics of the model were given. The authors stated that further details of the economic model were reported in a separate manuscript, but they did not provide bibliographic details.
Outcomes assessed in the review The primary health outcomes assessed in the review were the stage distributions of tumours detected by screening and survival rates.
Study designs and other criteria for inclusion in the review A formal review of the literature was not carried out. The included studies from which the effectiveness data were gathered, were all randomised controlled trials.
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 Four primary studies were included in the analysis. These were the Memorial Sloan-Kettering Lung Project (MSKLP), John Hopkins Lung Project (JHLP), Mayo Lung Project (MLP) and the Czechoslovakia Study.
Methods of combining primary studies The method use to combine the primary studies was not explicitly stated.
Investigation of differences between primary studies The authors mentioned that the comparators used in the primary studies were different. The MSKLP and JHLP compared annual chest X-ray screening alone to sputum cytology plus chest X-ray screening. The MLP compared annual chest X-ray screening plus cytology to chest X-ray screening plus cytology performed every 4 months. The Czechoslovakia study compared annual chest X-ray screening to no screening. However, no differences in mortality rates between the technology under study and the comparators were found in the primary studies. The authors focused, therefore, on the differences in stage distributions of tumours detected by the screening programme.
Results of the review In the MSKLP, 41% of the tumours detected by screening programmes were Stage I, 6% were Stage II, and 53% were Stage III or VI. In the JHLP, 57% of all participants were Stage I. In the MLP, 38% of the tumours detected were Stage I, 3% were Stage II, and 59% were stage III and IV. In the Czechoslovakia study, 51% of the cancers in the screened population were Stage I or Stage II.
All of the trials showed higher survival rates for the screened population, but the details were not reported. Stage distributions and survival rates were used to calculate the mortality rate due to lung cancer in the model.
Methods used to derive estimates of effectiveness The authors made assumptions about the reduction in mortality rate with the screening programme and the patients' participation rate.
Estimates of effectiveness and key assumptions In the base-case, the authors assumed an 18% reduction in the mortality rate due to chest X-ray screening and a 100% participation rate.
Measure of benefits used in the economic analysis The health benefit used in the economic analysis was the life-years gained (LYG). This was estimated using the Markov model. The key variable used to calculate the LYG was an 18% reduction in the annual mortality rate due to chest X-ray screening.
Direct costs The costs were discounted at 3% per annum. The unit costs and the quantities of resources used were not reported separately. The direct costs were for the annual chest X-ray, with a professional fee and a programme fee for operating and overhead expenses, and diagnostic consultations for individuals with positive X-ray tests. The start-up costs for the screening programme were not included because they were considered to be relatively low. The costs of potential unnecessary biopsies due to false-positive tests were also not included. The quantity/cost boundary was that of the US health care system. The source of the unit costs was not explicitly reported. One chest X-ray screening per year was assumed. The future costs were estimated using the economic model. The time horizon of the analysis was lifetime. The price year was 1998.
Statistical analysis of costs Statistical analyses of the costs, if performed, were not reported.
Indirect Costs The indirect costs were not included in the analysis.
Sensitivity analysis Univariate sensitivity analyses were performed on three key parameters. The parameters were reduction in mortality rate, participation rate and costs of the screening programme. Sensitivity analyses on the costs are likely to have been performed to investigate the generalisability of the results. Sensitivity analyses on reduction in mortality rate and participation rate have been carried out because they were based on assumptions. Also, a threshold analysis was performed to estimate the cost of screening that would lead to incremental cost-effectiveness ratios higher than $100,000.
Estimated benefits used in the economic analysis In a population of 100,000 male smokers aged 45 to 80 years, an annual chest X-ray screening programme would avoid more than 2,700 deaths, assuming a participation rate of 100% and a reduction in the mortality rate of 18% per year. This would lead to an increase of 38,290 LYG for this cohort of patients in comparison with non-screened individuals.
Cost results The incremental cost per person of an annual chest X-ray, compared with no screening, was approximately $3,000 at a 3% discount rate.
Synthesis of costs and benefits In the base-case, compared with no screening, the incremental cost of annual chest X-ray screening was $8,772 per LYG when the benefits were not discounted. This increased to $19,874 per LYG when a discount rate of 3% was applied.
When doubling the cost estimates, the incremental cost of the annual screening ranged from $17,543 per LYG to $39,747 per discounted LYG.
When varying the reduction in mortality from 6 to 14%, the incremental cost of the screening programme ranged from $26,315 to $11,278 per non-discounted LYG, and from $59,621 to $25,552 per discounted LYG.
When varying the participation rate from 25 to 75%, the incremental cost of chest X-ray screening varied from $13,608 to $9,308 per non-discounted LYG, and from $30,831 to $21,091 per discounted LYG.
The authors stated that all the incremental ratios obtained in the sensitivity analyses, although relatively different, could be considered as good value for money from a health economic policy perspective.
The threshold value for the cost of the screening that would result in cost-effectiveness ratios higher than $100,000 was $1,800.
Authors' conclusions The results of the model suggested that an annual chest X-ray programme is a cost-effective strategy, even without high rates of effectiveness, when the cost of the test does not exceed $1,800.
CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparator was clear. The annual chest X-ray screening was compared with no screening. However, it would have been interesting had the authors compared annual chest X-ray screening with other types of tests, with different frequencies, or if they had considered a different target population. You should decide whether it represents a valid comparator in your own setting.
Validity of estimate of measure of effectiveness The studies included in the analysis were not chosen on the basis of a systematic review of the literature. In addition, the authors did not report the rationale for the choice of the articles included. The effectiveness results of the studies included in the analysis were not combined. All of the studies included were randomised controlled trials, thus giving a high internal validity to the effectiveness results. However, only limited details of the results of these trials were given, and no quantitative information on survival rates was reported. This limits the overall internal validity of the results obtained.
Validity of estimate of measure of benefit The main measure of benefits (LYG) was obtained using a Markov model. No information on the structure of the model was reported. In addition, the key parameter used to calculate the LYG was the reduction in the mortality rate, which was derived from an assumption made by the authors. However, sensitivity analyses were performed to address the issue of the uncertainty around this key parameter. The use of LYG ensures the comparability with the health benefits of other programmes and studies.
Validity of estimate of costs The categories of costs included were relevant to the perspective used. The reasons for excluding some cost categories were given. However, the unit costs were not reported. It would have been interesting to know the cost of the chest X-ray assumed in the base-case analysis, in order to compare it with the cost found in the threshold analysis. Sensitivity analyses were performed to assess the impact of different cost estimates on the final incremental ratios.
Other issues The issue of generalisability was partly addressed through the sensitivity analyses. However, the results of the study were not compared with those from other cost-effectiveness studies published in the literature. The authors stated that the results of this study should be treated with caution given the numerous assumptions used to derive the estimates in the model. The results of the sensitivity analyses for different levels of participation rate are particularly important, given the relatively unrealistic 100% participation rate used in the base-case.
Implications of the study Further studies are necessary to measure the efficacy of annual chest X-ray screening for lung cancer.
Bibliographic details Caro J J, Klittich W S, Strauss G. Could chest X-ray screening for lung cancer be cost-effective? Cancer 2000; 89(11 Supplement S): 2502-2505 Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Aged, 80 and over; Cost-Benefit Analysis; Humans; Lung Neoplasms /economics /mortality /prevention & Male; Markov Chains; Mass Chest X-Ray /economics; Middle Aged; Models, Economic; Smoking /adverse effects; control /radiography AccessionNumber 22001000069 Date bibliographic record published 31/01/2004 Date abstract record published 31/01/2004 |
|
|
|