|
Projected national impact of colorectal cancer screening on clinical and economic outcomes and health services demand |
Ladabaum U, Song K |
|
|
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 study compared eight different strategies for screening for colorectal cancer (CRC), four conventional strategies and four newly developed strategies. The conventional strategies were annual faecal occult blood testing (FOBT), flexible sigmoidoscopy (FS) every 5 years, a combination of FOBT and FS (FOBT/FS), and colonoscopy (COLO) every 10 years. The four newly developed strategies were faecal DNA testing every 5 years (F-DNA-base), F-DNA-optimised, virtual colonoscopy every 10 years (VC-base), VC-optimised every 10 years, and VC-Pickhardt every 10 years. All strategies were compared with the "natural history", where colonoscopy was performed after the development of symptoms. In all positively screened CRCs with COLO, polyps were taken out and biopsies were carried out. In the case of other screening tests, positively screened cases received COLO, followed by polypectomy and biopsy when required.
Economic study type Cost-effectiveness analysis.
Study population The study population was the population of the USA adopting the size and age distribution as described by year 2000 US census data. In particular, the study population referred to average-risk individuals aged from 50 years and older up to 80 years.
Setting As this was a modelling study, the setting was not explicitly stated at the outset. The economic study was carried out in the USA.
Dates to which data relate The effectiveness data were derived from studies published between 1964 and 2004. The cost data were derived from official sources and literature published between 1990 and 2004. All costs were appropriately adjusted and reported for the price year 2003.
Source of effectiveness data The effectiveness data were derived from a review and synthesis of published studies.
Modelling To assess the impact of a national-level screening strategy, the authors constructed a decision analytic Markov model describing the natural history of CRC in the USA. The model has been described in four published studies (Song et al. 2004, Ladabaum et al. 2001, 2003 and 2004; see ,Other Publications of Related Interest- below for bibliographic details). The health states included in the model were normal, small adenomatous polyp (< 10 mm), large adenomatous polyp (>/= 10 mm), localised, regional or distant CRC, and dead. Individuals aged 50 years progressed through different health states for 50 one-year cycles until the age of 100 years or death. It was assumed that if individuals were negatively screened, follow-up screening was performed after 10 years using the initial screening method. The assessment of the economic and clinical burden of widespread screening at the national level was based on authors' assumptions.
Outcomes assessed in the review The following input parameters were assessed in the model:
the prevalence of a polyp, small polyp and large polyp at 50 years of age;
the annual transition rate from normal to small polyp and from small to large polyp,
the annual transition rate to cancer without polypoid precursor and to cancer from large polyp; and
symptomatic presentation of localised and regional cancer.
All of these parameters were derived from data published in the literature and from epidemiological and autopsy data. Further parameters included:
the mortality rate from treated localised and regional cancer;
the mean survival from distant cancer;
the mortality rate from cancer treatment;
the sensitivity of FOBT, COLO, F-DNA-base, F-DNA-optimised, VC-base and VC-Pickhardt for cancer and for large and small polyps;
the specificity of FOBT, FDNA-base, F-DNA-optimised, VC-base and VC-Pickhardt;
the percentage of polyps or cancer within reach of a sigmoidoscope;
the sensitivity of sigmoidoscopy for cancer, and for large and small polyps within reach of a sigmoidoscope;
the specificity of sigmoidoscopy for lesions within reach of a sigmoidoscope;
the major complication rates of COLO and of sigmoidoscopy; and
the mortality rates due to COLO and to sigmoidoscopy.
Study designs and other criteria for inclusion in the review The study designs and further criteria for inclusion in the review were not reported in the current study. Such details are given in the published modelling studies (Song et al. 2004, Ladabaum et al. 2001, 2003 and 2004).
Sources searched to identify primary studies The sources searched to identify primary studies were not reported in the current study. Such details are given in the published modelling studies (Song et al. 2004, Ladabaum et al. 2001, 2003 and 2004).
Criteria used to ensure the validity of primary studies No criteria that ensured the validity of the primary studies were reported. Such details are given in the published modelling studies (Song et al. 2004, Ladabaum et al. 2001, 2003 and 2004).
Methods used to judge relevance and validity, and for extracting data The authors reported several observations, suggesting that the model had validity. For example, the clinical outcomes used in the model were consistent with epidemiologic data and predictions of the model were consistent with available data from clinical studies or Surveillance, Epidemiology, and End results (SEER) data. Further details are given in the published modelling studies (Song et al. 2004, Ladabaum et al. 2001, 2003 and 2004).
Number of primary studies included Overall, the authors used 27 studies as sources of effectiveness evidence.
Methods of combining primary studies Investigation of differences between primary studies In the current study the authors did not report any methods used to investigate differences between the primary studies. Relevant details are given in the published modelling studies (Song et al. 2004, Ladabaum et al. 2001, 2003 and 2004).
Results of the review The results of the review were too numerous to be reported here. The reader is referred to Table 1 of the current study.
Methods used to derive estimates of effectiveness Regarding national projections, some estimates of effectiveness were based on authors' assumptions.
Estimates of effectiveness and key assumptions The authors assumed a 75% screening uptake rate at the national level. Based on the 2000 US census data, a steady state in the population size and age distribution was assumed. At the time the study was conducted, 40% screening uptake was assumed. Of those screened, 25% were screened using FOBT, 20% using FS, 20% using FOBT/FS and 35% using COLO.
Measure of benefits used in the economic analysis The authors used life-years gained (LYG) from screening for each screening strategy. The benefits were derived from the model. The health benefits were discounted at an annual rate of 3%.
Direct costs From the perspective of the third-party payer, the direct costs used in the analysis were the costs of FOBT, FS, FS with biopsy, COLO, COLO with biopsy or lesion removal, F-DNA, VC, endoscopy complications, and CRC care in localised, regional and distant cancer. All the costs were derived from official sources and data published in the literature. They were appropriately adjusted to reflect US 2003 dollar prices using the medical component of the Consumer Price Index. The unit costs were reported. However, the costs and the quantities were not described separately and the authors only reported the total cost per screening strategy. It is therefore difficult to decide the specific resources used in each strategy. Given the lack of cost data on the VC screening test, the authors assumed an equal cost to diagnostic COLO. The total costs per screening strategy were appropriately discounted.
Statistical analysis of costs The costs were treated deterministically.
Indirect Costs The indirect costs were not included in the analysis.
Sensitivity analysis The authors conducted sensitivity analyses in which all CRC costs for each stage were reduced by 25% to assess the robustness of the results to variability in the data. The type of sensitivity analysis conducted was not reported.
Estimated benefits used in the economic analysis Life-years per person after the age of 50 years were appropriately discounted and were reported. The life-years per person were:
18.686 in the natural history (no screening) strategy,
18.742 in the FOBT strategy,
18.734 in the FS strategy,
18.749 in the FOBT/FS strategy,
18.748 in the COLO strategy,
18.720 in the F-DNA-base strategy,
18.748 in the F-DNA-optimised strategy,
18.741 in the VC-base strategy, and
18.747 in the VC-Pickhardt strategy.
The number of LYG with each strategy in comparison with natural history (no screening) was not reported.
Cost results The total discounted costs for each screening strategy were reported per each person after the age of 50 years. The total cost per person was:
$1,813 in the "natural history" (no screening) strategy,
$2,269 in the FOBT strategy,
$2,647 in the FS strategy,
$2,989 in the FOBT/FS strategy,
$2,974 in the COLO strategy,
$4,292 in the F-DNA-base strategy,
$3,745 in the F-DNA-optimised strategy,
$3,411 in the VC-base strategy, and
$3,433 in the VC-Pickhardt strategy.
Synthesis of costs and benefits When compared with natural history, the cost per LYG was:
$8,100 in the FOBT strategy,
$17,300 in the FS strategy,
$18,700 in the FOBT/FS strategy,
$18,800 in the COLO strategy,
$73,200 in the F-DNA-base strategy,
$31,000 in the F-DNA-optimised strategy,
$28,700 in the VC-base strategy, and
$26,600 in the VC-Pickhardt strategy.
The authors reported that the results of the analysis were robust in the sensitivity analysis.
When the model was projected at the national level the following results were demonstrated. When no screening was implemented disease-related expenditures would be $8.4 million. When screening uptake was estimated to be 75%, the incidence of CRC decreased by 17 to 54% and annual CRC deaths decreased by 28 to 60%, depending on the screening strategy. Although expenditures for CRC care were reduced by $1.5 to $4.4 million (depending on the screening strategy), screening expenditures increased substantially. This resulted in an overall increase of health care expenditures ranging from $9.2 billion (FOBT screening strategy) to $15.4 billion (F-DNA-base screening strategy).
Authors' conclusions Despite the potential for dramatic reductions in colorectal cancer (CRC) morbidity and mortality, widespread CRC screening increases health care spending and is therefore unlikely to be cost-saving. .The current national endoscopic capacity may be adequate to support widespread use of screening colonoscopy in the steady state. The impact of emerging tests on colonoscopy demand will depend on the extent to which they replace screening colonoscopy or increase screening uptake in the population./
CRD COMMENTARY - Selection of comparators The choice of the comparators was explicitly justified. You should decide if these represent widely used technologies in your own setting.
Validity of estimate of measure of effectiveness The authors referred to separate papers for details of the model and the conduct of the review and it is therefore not possible to comment on the conduct of the review. However, based on the limited data in the current study, it appears that the authors have used data from the available studies selectively and that they have not investigated differences between the available studies. Although the authors discussed the validity of the clinical parameters that were based on published studies or epidemiologic data, they did not undertake any sensitivity analysis to determine the impact on health benefit of varying these clinical parameters.
Validity of estimate of measure of benefit The authors used LYG as the measure of benefit. The LYG were derived directly from the model. The summary measure of benefit was appropriate because it captured the impact of the interventions on LYG, which is the most relevant dimension of health affected by CRC. In addition, the use of LYG enables comparisons with the benefits of other health care interventions. However, the impact of screening on quality of life was not investigated. The benefits were discounted.
Validity of estimate of costs From the perspective of the third-party payer, all relevant costs appear have been included in the economic analysis. The authors reported in most cases the total costs per screening strategy. Hence, it is not possible to know whether all relevant costs for all categories of costs were included in the analysis. The costs and the quantities were not reported separately, which will not enable the analysis to be easily reworked for other settings. All costs were derived from official published sources and, although they were treated deterministically, an appropriate sensitivity analysis was conducted to test the robustness of the results. Discounting was conducted and the price year was reported, which will aid any future reflation exercises.
Other issues The authors compared their estimates and results with those from published studies, finding consistency. The issue of generalisability of the results to other settings was not directly addressed. However, the authors gave an extensive analysis regarding the current capacity of US health services to respond to widespread screening uptake. The authors do not appear to have presented their results selectively. Certain limitations to the model were reported. For example, the authors mentioned that the transition probabilities at the population level were unknown. Also, the model did not distinguish between different patient sub-groups, and estimates were based on a steady state population size and population growth will affect all estimates. Finally, screening compliance was thought to be constant and did not vary over time.
Implications of the study The authors suggested that a reorganisation of low-yield surveillance procedures in health services in the USA would improve their screening capacity. In addition, they suggested that further research is needed to derive more robust estimates of higher and lower risk of adenoma prevalence.
Source of funding Supported in part by the National Institutes of Health.
Bibliographic details Ladabaum U, Song K. Projected national impact of colorectal cancer screening on clinical and economic outcomes and health services demand. Gastroenterology 2005; 129(4): 1151-1162 Other publications of related interest Song K, Fendrick AM, Ladabaum U. Faecal DNA testing compared to conventional colorectal cancer screening methods: a decision analysis. Gastroenterology 2004; 126:1270-9.
Ladabaum U, Song K, Fendrick AM. Colorectal neoplasia screening with virtual colonoscopy: when, at what cost, and with what national impact? Clin Gastroenterol Hepatol 2004;2:554-63.
Ladabaum U, Chopra CL, Huang G, et al. Aspirin as an adjunct to screening for prevention of sporadic colorectal cancer. A cost-effectiveness analysis. Ann Intern Med 2001;135:769-81.
Ladabaum U, Scheiman JM, Fendrick AM. Potential effect of cyclooxygenase-2-specific inhibitors on the prevention of colorectal cancer: a cost-effectiveness analysis. Am J Med 2003;114:546-54.
Indexing Status Subject indexing assigned by NLM MeSH Colorectal Neoplasms /diagnosis /economics /prevention & Costs and Cost Analysis; Health Services Needs and Demand /economics; Humans; Mass Screening /economics; Reproducibility of Results; United States; control AccessionNumber 22005001743 Date bibliographic record published 30/06/2006 Date abstract record published 30/06/2006 |
|
|
|