|
Evaluation of the optimum cut-off point in immunochemical occult blood testing in screening for colorectal cancer |
Nakama H, Zhang B, Zhang X |
|
|
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 use of an immunochemical faecal occult blood test for the early detection of colorectal cancer. Three cut-off levels of faecal haemoglobin were considered for the determination of the sensitivity and specificity of the test. These were 50, 150, and 300 ng/mL.
Economic study type Cost-effectiveness analysis.
Study population The study population comprised asymptomatic people aged over 40 years.
Setting The setting was the community. The economic study was carried out in Japan.
Dates to which data relate The effectiveness and resource use data were gathered from April 1990 to March 1999. No price year was reported.
Source of effectiveness data The effectiveness evidence was derived from a single study.
Link between effectiveness and cost data The costing was carried out prospectively on the same patient sample as that used in the effectiveness analysis.
Study sample Power calculations to determine the sample size were not reported. A sample of 4,260 eligible patients participating in a medical check-up for colorectal cancer was included in the analysis. There were 1,156 patients in the age class 40 to 49 years, 1,527 in the age class 50 to 59 years, 930 in the age class 60 to 69 years, and 647 in the age class 70 years or older. The number of women was 2,281. The percentage of patients with a family history of colorectal cancer was 15.5%.
Study design This was a prospective, cross-sectional screening test evaluation study. The number of centres in which it was conducted was not reported. The patients were followed until colonoscopy was performed, which was conducted on all patients to evaluate the results of the immunochemical faecal occult blood test. No patient was lost to follow-up.
Analysis of effectiveness All patients included in the study were accounted for in the analysis. The primary health outcomes were the number of cases of colorectal cancer detected, and the sensitivity and specificity of the tests. The number of adenomatous polyps detected was also reported. The comparability of the study groups was irrelevant, as the same sample of patients provided the effectiveness evidence for both interventions.
Effectiveness results Out of the 27 cases of colorectal cancer detected by colonoscopy, 24 were detected with the cut-off point of 50 ng/mL, 22 with the cut-off point of 150 ng/mL, and 15 with the cut-off point of 300 ng/mL.
The sensitivity was 89% with the cut-off point of 50 ng/mL, 81% with the cut-off point of 150 ng/mL, and 56% with the cut-off point of 300 ng/mL. The specificity was 94% with the cut-off point of 50 ng/mL, 96% with the cut-off point of 150 ng/mL, and 97% with the cut-off point of 300 ng/mL.
There was a statistically significant difference in the sensitivity between the 50 and 300 ng/mL levels, (p<0.05), and between the 150 and 300 ng/mL levels, (p<0.05). There was a statistically significant difference in the specificity between the 50 and 300 ng/mL levels, (p<0.05).
No statistically significant difference was observed between the 50 and 150 ng/mL levels in terms of specificity. No comment was made on the comparison between the 50 and 150 ng/mL levels for sensitivity.
In terms of the number of adenomatous polyps equal to 1 cm or greater, 56 cases were detected with colonoscopy. Of these, 33 (59%) were detected with the cut-off point of 50 ng/mL, 30 (54%) with the cut-off point of 150 ng/mL, and 12 (21%) with the cut-off point of 300 ng/mL. For adenomatous polyps smaller than 1 cm, 162 cases were detected with colonoscopy. Of these, 36 (22%) were detected with the cut-off point of 50 ng/mL, 29 (18%) with the cut-off point of 150 ng/mL, and 18 (11%) with the cut-off point of 300 ng/mL.
Clinical conclusions The 50 and 150 ng/mL cut-off points proved to offer similar sensitivity and specificity rates, and performed better than the 300 ng/mL level.
Measure of benefits used in the economic analysis The measure of health benefit used in the economic analysis was the number of cancers detected, as assessed in the effectiveness analysis.
Direct costs Discounting was irrelevant as the costs for each patient were incurred over a short period of time. The economic analysis included the costs of screening all patients and the examination of those who were further examined. The unit costs were reported separately from the quantities of resources. The cost/resource boundary was unclear. The source of the cost data was only given for the price per slide for each test. The resource use was estimated using actual data derived from the study. No price year was reported. Only the average total costs were presented.
Statistical analysis of costs No statistical analysis of the costs was carried out.
Indirect Costs The indirect costs were not included in the analysis.
Currency US dollars ($). Some costs were also reported in Japanese yen.
Sensitivity analysis No sensitivity analyses were conducted.
Estimated benefits used in the economic analysis Of the 27 cases of cancer detected by colonoscopy, 24 were detected with the cut-off point of 50 ng/mL, 22 with the cut-off point of 150 ng/mL, and 15 with the cut-off point of 300 ng/mL.
Cost results The total screening costs were $30,981.82 with all the cut-off points, as all the patients were screened.
The costs of the total examination (colonoscopy for faecal occult blood test positives) were $37,909.09 with the cut-off point of 50 ng/mL, $23,863.64 with the cut-off point of 150 ng/mL, and $18,954.55 with the cut-off point of 300 ng/mL.
Therefore, the total costs were $68,890.91 with the cut-off point of 50 ng/mL, $54,845.46 with the cut-off point of 150 ng/mL, and $49,936.37 with the cut-off point of 300 ng/mL.
Synthesis of costs and benefits An average cost-effectiveness analysis was carried out to combine the costs and benefits of each cut-off level. An incremental analysis was not conducted. The average cost per cancer detected was $2,870.45 with the cut-off point of 50 ng/mL, $2,492.98 with the cut-off point of 150 ng/mL, and $3,329.09 with the cut-off point of 300 ng/mL.
Authors' conclusions Immunochemical faecal occult blood testing using a cut-off point of 150 ng/mL was the most cost-effective approach for the detection of colorectal cancer.
CRD COMMENTARY - Selection of comparators The rationale for the choice of the comparators was clear. The three cut-off levels were chosen from a set of five by empirical analysis, although the evidence was not shown. Colonoscopy was chosen as it represented the 'gold' standard for the detection of colorectal cancer. You should assess whether they represent widely used interventions in your own setting. While it might seem reasonable not to include no-screening as a comparator, it was not entirely clear how the sample used was chosen for attendance at the clinic. The authors point out that the costs and benefits of screening depend on the incidence or prevalence in the screened population. However, we do not know how high this is in comparison with the general population, or any population with a defined level of risk.
Validity of estimate of measure of effectiveness The analysis of effectiveness used a cross-sectional study. This, as the authors noted, was appropriate for the study question given the difficulties in carrying out a longitudinal population-based study. The baseline characteristics were presented but we do not know how representative the sample was of any population, other than in terms of their age. There will have been no selection bias as a single cohort of patients was used for the analysis. All of the patients included in the study were accounted for in the analysis. However, the number of centres in which the study was carried out was not reported.
Validity of estimate of measure of benefit The benefit measure used in the economic analysis was the number of cases of cancer detected. This appears to represent an appropriate measure of the success of the screening intervention. However, the use of more complex measures, such as life-years saved or quality-adjusted life-years saved, would have been useful since screening for colorectal cancer is likely to affect both the quantity and quality of patient life.
Validity of estimate of costs The perspective adopted in the cost analysis was not stated. The economic only included those costs strictly related to the screening test and examination. The unit costs and quantities of resources used were reported separately, thus increasing the transparency and generalisability. Appropriate currency conversions were performed and the costs were reported in US dollars. However, no price year was given, thus presenting difficulties in terms of reflation exercises to other settings. In addition, the costs were treated deterministically as statistical analyses were not conducted. Finally, the source of the cost data was not given for all items.
No incremental analysis was conducted. This means that we were not told what the increase in cost would be for an increase in benefit when moving from one technology to the more beneficial technology. This makes the claims regarding cost-effectiveness misleading. In fact, it can be calculated from the cost and benefit results that the increase in cost per case detected when going from 150 to 50 ng/mL is greater than when going from 300 to 150 ng/mL. However, it might be that this cost is worth paying. This can only be determined by reference to the opportunity cost, which is the loss in benefit by the diversion of resources from the current funding required to fund this change in screening practice. Beyond this, it was unclear, given that colonoscopy is the 'gold' standard, why this might not be offered to all for screening. One might assume that it is too costly or has too high a risk of complication. However, this latter point highlights the need to account more fully for the costs and benefits, as the authors acknowledge.
Other issues The authors did not compare their findings with those from other studies. In terms of the generalisability of the study results to other settings, the authors noted that caution is necessary when generalising the study results. This is because the estimated cost-effectiveness ratios depend on the incidence of colorectal cancer and the costs of endoscopic procedures, which may vary widely across countries. The authors noted some limitations of their analysis. These were mainly related to the exclusion of indirect and intangible costs, due to measurement difficulties. The study results were reported clearly.
Implications of the study The authors note that the type of screening test used in the analysis provided a standardised procedure for the detection of colorectal cancer, thus avoiding all problems related to the intra- and inter-observer variability of test results. Due to the simplicity of the method (no requirement for dietary restriction and short period for faecal collection), the proposed screening test is likely to offer high compliance rates in comparison with other tests used for the screening of colorectal cancer. This represents a key variable in the assessment of the total costs of the intervention. Future studies should focus on the inclusion of indirect and intangible costs. For example, the costs of treating complications of screening, the costs of initial treatment for polyps and cancer, terminal costs of care, and the costs of follow-up for cancer patients. The claim relating to the cost-effectiveness should be viewed in the light of the caveats highlighted.
Source of funding Supported in part by Grants-in-Aid for Scientific Research (No. 09670384) from the Ministry of Education, Science and Culture of Japan, and by Cancer Research (No. 8-2) from the Ministry of Health and Welfare of Japan.
Bibliographic details Nakama H, Zhang B, Zhang X. Evaluation of the optimum cut-off point in immunochemical occult blood testing in screening for colorectal cancer. European Journal of Cancer 2001; 37(3): 398-401 Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Colorectal Neoplasms /diagnosis /economics; Cost-Benefit Analysis; Female; Humans; Immunohistochemistry /economics /methods /standards; Male; Mass Screening /economics /methods /standards; Middle Aged; Occult Blood; Sensitivity and Specificity AccessionNumber 22001000612 Date bibliographic record published 28/02/2003 Date abstract record published 28/02/2003 |
|
|
|