|
A cost-effectiveness approach to the Norwegian follow-up programme in colorectal cancer |
Norum J, Olsen J A |
|
|
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 a follow-up programme in colorectal cancer (CRC). This consisted of following the patients for four years with regular carcinoembryonic antigen (CEA) monitoring, ultrasound of the liver, chest radiograph, rectoscopy and colonoscopy.
Type of intervention Other (follow-up surveillance after surgery).
Economic study type Cost-effectiveness and cost-utility analysis.
Study population The study population comprised patients that had undergone a curative resection of CRC.
Setting The setting appears to have been that of a hospital. The study was performed at the University Hospital of Tromso, Norway.
Dates to which data relate The effectiveness data related to studies published between 1973 and 1996. The dates for the resource use, costs and the price year were not provided.
Source of effectiveness data The effectiveness data were derived from a review of completed studies and from assumptions made by the authors.
Modelling A decision analysis model appears to have been used to estimate the health benefits and costs of the programme proposed.
Outcomes assessed in the review The outcomes assessed in the review were:
the median time from surgery to suspicion of recurrence;
the percentage of suspected recurrences;
the percentage of patients with false positive relapse;
the percentage of recurrent patients;
the percentage of recurrent operable patients;
the percentage of patients with operable recurrence who would not experience recurrence again;
the percentage of patients with operable recurrence who would experience progressive disease; and
the median survival gain for operable patients undergoing salvage surgery compared with those who, for a variety of reasons, did not undergo re-operation.
The median age at diagnosis of CRC, and the life expectancy at that age in Norway in 1993, were also assessed. These outcomes were included as input parameters in the model in order to estimate the number of life-years gained with the programme.
Study designs and other criteria for inclusion in the review The authors stated that clinical data from the existing English literature were used. They did not report what types of studies were included in the review.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies The authors reported that the data used from one of the studies included in the review were supported by three other studies. However, they did not state any other criteria that validated the other studies included in the review.
Methods used to judge relevance and validity, and for extracting data The outcomes assessed in the review were included as input parameters in the model in order to estimate survival of CRC. The authors compared CRC survival obtained from the model with the national survival data of CRC (Statistics Norway), to show that the model overestimated the number of deaths during the second year of follow-up. However, this overestimation was compensated for by the underestimation of deaths during the first ten months and the fourth year of follow-up.
Number of primary studies included At least eight studies provided effectiveness data and were included in the review.
Methods of combining primary studies The results of the individual primary studies were reported using a narrative method.
Investigation of differences between primary studies The results of one of the primary studies included in the review were supported by three other studies. The authors did not report any further investigation of differences between the other studies reviewed.
Results of the review The median time from surgery to suspicion of recurrence was reported to be 10 months.
Forty-five per cent of the patients had suspected recurrences, therefore, 55% of the patients did not have suspected recurrence.
Since 5% of the patients had a false positive relapse, the percentage of recurrent patients was, therefore, 40.
The percentage of recurrent operable patients was 10.
The percentage of patients with operable recurrence who would not experience recurrence again was 2.
The percentage of patients with operable recurrence who would experience progressive disease was 8.
The median gain in survival experienced by operable patients undergoing salvage, compared with those who did not undergo re-operation, was about 14 months. Survival was 36 months for salvage patients versus 22 months for those not undergoing a re-operation.
The median age at diagnosis of CRC in Norway was assessed to be 73 years, with a life expectancy of 12 years. Therefore, the life expectancy for those patients who underwent re-surgery without posterior recurrence was considered to be 12 years.
Methods used to derive estimates of effectiveness From the results of the review, the authors made some assumptions about the period of follow-up.
Estimates of effectiveness and key assumptions It was assumed that 100% of the patients experiencing curative resection CRC would go through the follow-up programme. The authors also assumed that all patients were followed up for 10 months, and those with no suspected relapse, those with false positive relapse, and those with relapse and salvage surgery were followed from 10 to 48 months.
Measure of benefits used in the economic analysis The benefit measures were the life-years gained (LYG) and quality-adjusted life-years (QALYs). The authors reported that the quality of life was measured as 0.83, according to a published study (Norum et al; see Other Publications of Related Interest).
Direct costs The resource quantities and the costs were reported separately. The direct costs included in the analysis were those of the health service. The authors considered the costs associated with the follow-up during the first 10 months for all the patients. They also considered the costs from 10 to 48 months for 62% of the patients (those with no suspected relapse, those with false positive relapse, and those with relapse and salvage surgery). The costs included as follow-up costs were for CEA monitoring, ultrasound of the liver, visit to a radiologist, chest radiograph, visit to a general practitioner (GP) or surgeon, and travel to the GP, surgeon or radiologist. The authors also included those costs associated with suspected relapses and re-surgery. The costs related to the former (suspected relapses) were for computed tomography of the abdomen and pelvis, chest radiograph, colonoscopy, bone scan, blood test, visit to surgeon, visit to radiologist, hospitalisation and travel.
The costs were calculated according to the tariffs given by the Norwegian Medical Association and the National Insurance Administration. The travelling costs were based on a qualified guess, although the authors did not report any further information about how this qualified guess was reached. Discounting was relevant as some of the costs were incurred during a period of at least 2 years. A 5% discount rate was used to discount the costs. The costs reported appear to have been the average costs. The costs were converted to UK pounds sterling. The price year was not given.
Statistical analysis of costs No statistical analysis of the costs was reported.
Indirect Costs No indirect costs were reported.
Currency UK pounds sterling (). The conversion rate was 1 = NOK 10.00.
Sensitivity analysis From the results of one of the reviewed studies, the authors estimated the cost per LYG when it was considered that 10% of operable patients would be rescued by the salvage surgery with a total of 3 years gained per patient. Moreover, the authors performed one-way sensitivity analyses. The parameters varied were the frequency of salvage surgery, the gain in survival and the quality of life. The area of uncertainty investigated was variability in the data. The authors considered a cut-off level of $30,000 (about 20,000) to determine the limits of the parameters for which the intervention was still cost-effective.
Estimated benefits used in the economic analysis Operable patients with recurrent CRC under the follow-up programme receiving salvage surgery would have lived 12 years. Those patients with recurrent CRC who were not followed up and who, therefore, did not receive salvage surgery, would have lived 1.8 life-years.
Cost results The cost per patient for 10 months' follow-up was 269.
The cost per patient for investigation due to suspected relapse in 45% of the patients was 517.
The cost of re-surgery was 707.
The cost of follow-up from month 10 to month 48 was 450.
The cost of follow-up was estimated to be 1,232 per patient.
When the costs of further investigation because of suspected recurrent disease and those of surgical intervention were included, the total cost per patient was 1,943.
Synthesis of costs and benefits The authors reported cost-effectiveness ratios showing the cost per LYG and the cost per QALY. The cost per LYG was 9,525, while the cost per QALY was 11,476.
When alternative survival data were considered, the cost per LYG increased to 16,192 and the cost per QALY was 19,508. The one-way sensitivity analyses showed that, when considering a limit of approximately 20,000 per QALY as cost-effective, the frequency of re-surgery could increase from 10% to 11 to 30% depending on the survival gain calculated (either 10.2 years in 2% of the patients or 1.2 years in 10% of the patients). The limit for the mean survival gain was 0.12 years in all patients, 5.9 years in 2% or 1.2 years in 10% of the patients. The corresponding limit in the quality of life was 0.46 to 0.81 depending on the calculated gain in survival.
Authors' conclusions The follow-up programme in colorectal cancer (CRC) appears to have been cost-effective. The authors also stated that, if carcinoembryonic antigen (CEA) monitoring was abandoned as an indicator for second-look surgery, the cost-effectiveness might even improve.
CRD COMMENTARY - Selection of comparators The authors did not justify the choice of the comparator. However, given the intervention under study (a follow-up programme to detect patients with potentially recurrent CRC), the comparator used seems to have been reasonable. Moreover, do-nothing is a comparator widely used in most of the economic evaluations of interventions, because it allows the active value of the strategy to be evaluated. You should decide whether there is an alternative health technology in your own setting to follow-up patients with potentially recurrent CRC.
Validity of estimate of measure of effectiveness The authors did not provide evidence that a systematic review of the literature had been undertaken, although the criteria they gave were "English studies". To make an objective judgement about the validity of the retrieved studies would require more details of the review, as the authors did not report the inclusion criteria used to select the studies or the sources searched to identify them. These factors hinder the validity of the effectiveness results, although plausible sensitivity analyses were conducted to assess variability in the data. The dates to which the effectiveness data related were not reported.
Validity of estimate of measure of benefit The benefits were estimated directly from the effectiveness results derived from the review, because they were based on the Norwegian statistics. A positive aspect of this estimation was that the authors also reported the benefits obtained using alternative survival data.
Validity of estimate of costs It would appear that all the categories relevant to the perspective adopted were accounted for in the analysis, and that no relevant costs were omitted. The resource quantities and the costs were reported separately. However, the dates to which the costs and quantities related and the price year were not given, thus hindering reflation exercises to other settings. No sensitivity analyses of the quantities or costs were performed, which limits the interpretation of the study findings. Discounting was appropriately performed. The authors reported that they did not include the production gains derived from the follow-up programme in the analysis, due to the characteristics of the study population (people usually retired). However, it was unnecessary to consider these given the perspective adopted and the age of the typical patient (post-retirement patients).
Other issues The results were compared appropriately with those from other studies. The authors warned about the difficulties of generalising the results to other countries. However, they stated that the cost-effectiveness may be estimated using data from each country to run the model. A further limitation reported by the authors was that they combined Norwegian cost data with non-Norwegian effectiveness data, which may not be correct for the Norwegian programme, although this was investigated using sensitivity analyses. The authors also highlighted the point that the costs and psychological burden associated with morbidity and mortality in both false positive and true positive results should not be ignored, and they may compensate for the benefits derived from the follow-up programme.
Implications of the study When the costs and distress produced on patients having a false positive result are not taken into consideration, the follow-up programme appears to be cost-effective (considering a cut-off of $20,000 for decision-makers). However, caution should be exercised given the fact that a systematic review may not have been performed and the study presents some limitations. The authors recommend considering the exclusion of CEA monitoring as an indicator for second-look surgery, which can improve the cost-effectiveness of the follow-up programme.
Bibliographic details Norum J, Olsen J A. A cost-effectiveness approach to the Norwegian follow-up programme in colorectal cancer. Annals of Oncology 1997; 8: 1081-1087 Other publications of related interest Berge I, Ekelund C, Mellner BP. Carcinoma of the colon and rectum in a defined population, an epidemiological, clinical and post-mortem investigation of colorectal cancer and co-existing benign polyps in Malmo, Sweden. Acta Chirurgica Scandinavica 1973;438:1-86.
Cancer in Norway 1993. Oslo: The Cancer Registry of Norway, Institute for Epidemiological Cancer Research; 1996.
Camunas J, Enriquez JM, Devesa JM, et al. Value of follow-up in the management of recurrent colorectal cancer. European Journal of Surgical Oncology 1991;17:530-5.
Norum J, Vonen B, Olsen JA, Revhaug A. Adjuvant chemotherapy (5-fluorouracil and levamisole) in Dukes' B and C colorectal carcinoma: a cost-effectiveness analysis. Annals of Oncology 1997;8:65-70.
Pihl E, Hughes ES, McDermot FT, et al. Disease free survival and recurrence after resection of colorectal cancer. Journal of Surgical Oncology 1981;16:333-41.
Indexing Status Subject indexing assigned by NLM MeSH Carcinoembryonic Antigen /analysis /economics; Colonic Neoplasms /economics /mortality /therapy; Colonoscopy; Cost-Benefit Analysis; Disease-Free Survival; Female; Follow-Up Studies; Humans; Male; Norway; Quality-Adjusted Life Years; Rectal Neoplasms /economics /mortality /therapy; Treatment Outcome AccessionNumber 21998008053 Date bibliographic record published 30/04/2003 Date abstract record published 30/04/2003 |
|
|
|