|
Cost-effectiveness of colorectal cancer screening in the average risk population |
Leshno M, Halpern Z, Arber N |
|
|
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 Screening for colorectal cancer (CRC) in those at average risk of the disease was studied. Six screening strategies were considered:
no screening;
one-time colonoscopic screening (COL);
colonoscopy repeated at a 10-year interval (COL-10);
annual faecal occult blood testing (FOBT);
annual FOBT and flexible sigmoidoscopy in a 5-year interval (FOBT+SIG); and
annual detection of altered human DNA in a stool test.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised individuals aged 50 years who were living in Israel. These individuals were considered to have an average risk for CRC.
Setting The setting was secondary care in Israel.
Dates to which data relate The effectiveness data were obtained from studies published between 1961 and 2000. The dates to which the resource use data related were not reported. The price year was 2000.
Source of effectiveness data The effectiveness data were derived from a review of completed studies.
Modelling A POMDP was used to simulate the progression from normal colonic epithelium to adenomatous polyp to malignancy. Further details were provided elsewhere (see Other Publications of Related Interest). The POMDP was used to estimate the costs and benefits of the screening strategies. Unlike a Markov model, the POMDP allows for uncertainty regarding the previous health state. The authors argued that this was a more realistic representation of the natural history of screening for CRC, as the outcome of a screening test is only an indication of the true health state of the individual. The health states included in the POMDP were normal colonic epithelium, low-risk polyp, high-risk polyp, localised cancer, regional CRC, distant CRC, and death. The duration of a cycle was one year. The model was programmed using MATLAB.
Outcomes assessed in the review The outcomes of the review included:
the prevalence of localised CRC and polyps at 50 years of age;
the percentage of high-risk polyp out of all polyps and lesions (polyps) in the lower colon;
the sensitivity and specificity of the various screening strategies;
the rates of major complication and death;
the effectiveness of treatment for localised or regional CRC;
the annual transitional probabilities from one health state to another; and
the rates of annual mortality from CRC (localised, regional, distant).
Study designs and other criteria for inclusion in the review 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 Twenty-five studies were included in the review.
Methods of combining primary studies Investigation of differences between primary studies Results of the review At age 50 years, the prevalence of localised CRC was 0.5% and that of polyps was 10%.
The proportion of high-risk polyps out of all polyps was 33%, and the proportion of lesions (polyps) in the lower colon was 60%.
The sensitivity of colonoscopy was 95% and the specificity was 100%.
The sensitivity of detecting altered human DNA in a stool test was 91% for cancer, 70% for low-risk polyp and 82% for high-risk polyp. The specificity was 90%.
The sensitivity of FOBT was 50% for cancer and 10% for polyps. The specificity was 90%.
The sensitivity of SIG was 95% for high-risk polyps and 85% for low-risk polyps. The specificity was 100%.
The rate of major complications was 0.23% of colonoscopies.
The rate of death was 1 out of 5,000 colonoscopies.
The effectiveness of the treatment was 90% for localised CRC and 70% for regional CRC.
The rate of annual mortality was 0.2% for localised CRC, 3% for regional CRC and 55% for distant CRC.
Measure of benefits used in the economic analysis The outcome measure used in the economic analysis was the life-years gained. These were discounted at a rate of 3%.
Direct costs The costs were discounted at a rate of 3%. The quantities and the costs were not analysed separately. The costs of treating CRC included those incurred due to initial, continuing and terminal care. The cost data were obtained from the Israel health ministry official costs, and the actual average costs paid by a large health maintenance organisation (HMO) in Israel. The costs were for medical personnel and supplies to provide the service, and overheads (administration, charting and automated information systems). The costs of tests were also obtained from the HMO. The price year was 2000.
Statistical analysis of costs No statistical tests of the costs were undertaken.
Indirect Costs The indirect costs were not included.
Currency New Israeli shekels (NIS). The authors reported that one US dollar ($) was equivalent to 5 NIS, but they did not use this conversion rate to present the results in US dollars.
Sensitivity analysis The authors stated that sensitivity analyses were carried out on most parameters to investigate variability in the data. One- and two-way analyses were performed. The method used to select the ranges studied was unclear. Results were presented for the following:
the cost of colonoscopy with polypectomy;
the cost of complications of colonoscopy;
the cost of treating localised CRC;
the cost of treating regional CRC;
the effectiveness of the treatment for localised and regional CRC;
the cost of SIG;
the probability of a major complication due to colonoscopy;
compliance with follow-up colonoscopy;
the prevalence of polyps at age 50 years;
the percentage of lesions (polyps) in the lower colon; and
the time horizon.
Estimated benefits used in the economic analysis Under the assumption of 100% compliance for all screening strategies, life expectancy was 19.143 years under COL, 19.174 under FOBT+SIG, 19.152 under annual FOBT, 19.172 under COL-10, 18.992 under no screening, and 19.171 under annual detection of altered human DNA in stool test.
Compared with COL, FOBT+SIG resulted in 11.57 days of life gained.
The other strategies were dominated by FOBT+SIG.
The duration of these benefits was assumed to be over the lifetime of the individuals.
To be consistent with Israeli data, the authors assumed that the lifetime horizon would be until 79 years of age and that the screening programme continued until the age of 79.
The mortality rate was 3.10% under no screening, 1.25% under COL, 0.73% under FOBT, 0.58% under annual detection of altered human DNA in stool test, 0.50% under FOBT+SIG and 0.20% under COL-10. This implied that, relative to no screening, the reduction in the mortality rate was greatest for COL-10 (93.7%), followed by FOBT+SIG (83.8%), annual detection of altered human DNA in stool test (81.3%), FOBT (76.5%) and COL (59.5%).
Cost results The lifetime cost per person was NIS 1,407 under COL, NIS 1,447 under FOBT+SIG, NIS 2,023 under annual FOBT, NIS 2,481 under COL-10, NIS 2,573 under no screening, and NIS 2,942 under annual detection of altered human DNA in stool test. These costs were incurred over the lifetime horizon of the model.
Synthesis of costs and benefits The incremental cost-effectiveness ratios (ICERs) were estimated as the cost per life-year gained.
The FOBT+SIG strategy had an ICER of NIS 1,268 per life-year gained relative to the COL strategy.
The authors found that one of the main determinants of this result was the compliance rate with colonoscopy follow-up. If this compliance rate was reduced from the 100% assumed in the base-case to 60%, the ICER would be NIS 4,984 per life-year gained. Similarly, if the compliance rate was reduced further to 40%, the ICER would be NIS 5,780 per life-year gained. The authors concluded that compliance rates were an important determinant of the cost-effectiveness of screening strategies and, therefore, should be considered when determining the preferred strategy for screening for CRC.
Authors' conclusions It was highly cost-effective to commence screening for colorectal cancer (CRC) in asymptomatic individuals of average risk at the age of 50 years.
CRD COMMENTARY - Selection of comparators Although no explicit justification was given for the comparator used, it would appear to represent a do nothing strategy. You should decide if the comparator represents current practice in your own setting.
Validity of estimate of measure of effectiveness The authors did not state that a systematic review of the literature had been undertaken. The estimates of effectiveness from the primary studies were not combined. The authors did not consider the impact of differences between the primary studies when estimating the effectiveness.
Validity of estimate of measure of benefit The estimation of benefits was modelled. The instrument (POMDP) used to derive the measure of health benefit appears to have been appropriate since it accounts for uncertainty in previous health states, which is realistic for screening.
Validity of estimate of costs All the categories of costs relevant to the perspective adopted appear to have been included in the analysis. The costs and the quantities were not reported separately. A sensitivity analysis of the quantities was not conducted, which may limit the interpretation of the study findings. A sensitivity analysis of the prices was conducted.
Other issues The authors did not make direct comparisons of their findings with those from other studies. The issue of generalisability to other settings was not addressed. This may limit the validity of the results for two reasons. First, the results related to those who were 50 years of age and classified as being at average risk of the disease. Therefore, the results may not be applicable to those who are at increased or moderate risk. Second, the cost data and time horizon of the model were representative of the health system and population in Israel. The authors did not discuss the extent to which these results may continue to hold in other settings. However, the sensitivity analysis performed on the unit costs and the time horizon indicated that the results were not sensitive to these parameters. The authors decided to omit the screening strategy that entailed a flexible sigmoidoscopy every five years in line with evidence that such a strategy will become less successful.
The authors did not specify the size of the hypothetical cohort studied in the model. The authors did not appear to present their results selectively. However, they did not offer an explanation for the seemingly counterintuitive finding that life expectancy was optimised under FOBT+SIG, while the mortality rate was lowest under COL-10. The study was concerned with screening in individuals aged 50 years with average risk of the disease, and this was reflected in the authors' conclusions.
Implications of the study The authors did not make any recommendations for changes in policy or practice, or the need for further research.
Source of funding Supported in part by the Roter Fund at Maccabi Healthcare Services.
Bibliographic details Leshno M, Halpern Z, Arber N. Cost-effectiveness of colorectal cancer screening in the average risk population. Health Care Management Science 2003; 6: 165-174 Other publications of related interest Monahan GE. A survey of partially observable Markov decision processes: Theory, models, and algorithms. Management Science 1982;28:1-16.
Lovejoy WS. A survey of algorithmic methods for partially observed Markov decision processes. Annals of Operations Research 1991;28:47-66.
Indexing Status Subject indexing assigned by NLM MeSH Algorithms; Colonoscopy /economics /methods; Colorectal Neoplasms /diagnosis /economics /epidemiology; Cost-Benefit Analysis; Humans; Israel /epidemiology; Life Expectancy; Markov Chains; Mass Screening /economics; Occult Blood; Risk Factors AccessionNumber 22003008228 Date bibliographic record published 30/11/2004 Date abstract record published 30/11/2004 |
|
|
|