|
Adjuvant chemotherapy (5-fluorouracil and levamisole) in Dukes' B and C colorectal carcinoma: a cost-effectiveness analysis |
Norum J, Vonen B, Olsen J A, Revhaug 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 Adjuvant chemotherapy (ACT), consisting of 5-fluorouracil (5-FU) and levamisole, following radical surgery.
Economic study type Cost-effectiveness analysis.
Study population Patients radically operated on for colorectal cancer (CRC) Dukes' B or C who were under 75 years old, had no concomitant malignancy, ulcerative colitis, Crohn's disease, renal, heart or liver failure and who were capable of undergoing one year of adjuvant therapy, were invited to participate. 95 patients were recruited, one patient proved to have Dukes' A and was excluded, so 94 (53 females, 41 males) were included. Their median age at diagnosis was 62 years (range: 36-76 years), 72% had colonic carcinoma and 64% had Dukes' B and 34% Dukes' C CRC. 47 were randomised to surgery plus ACT and 47 to surgery alone.
Setting Patients were recruited from the three counties of northern Norway: Finmark (23), Troms (32) and Nordland (39). Patients were treated at the University Hospital of Tromso (UHT) and 8 other local hospitals.
Dates to which data relate Patients were recruited from January 1993 to February 1996. Patient records were analysed in April 1996 with respect to sex, age at diagnosis, tumour location (rectal or colonic carcinoma), Dukes' stage (B or C), treatment modality (adjuvant chemotherapy or follow-up only) and time of follow-up. Questionnaires (Quality of Life (QoL)) were mailed between April and May 1996. All costs including those for chemotherapy, hospital stay and travelling were calculated in 1995 prices.
Source of effectiveness data The evidence was derived from a single study.
Link between effectiveness and cost data Costing was undertaken prospectively on the same patient sample as that used in the effectiveness study.
Study sample Power calculations were not undertaken. 95 patients were recruited, 1 patient proved to have Dukes' A and was excluded, and therefore 94 were included. 21 patients (24%) failed to return the questionnaire, but no significant difference was detected between these patients and those who responded with respect to age, sex, tumor location, Dukes' stage or treatment modality (apparently 1 of the deceased patients either responded or failed to return the questionnaire).
Study design This was a single-centre randomised controlled trial (RCT). Patients were followed for a median of 16 months (range: 0-37 months); 82 (87%) were still alive at follow-up, 62 (76%) patients responded to the questionnaire, all of the questions were answered by all but two of the patients, who had not marked their scores in the simple QoL scale. No information was provided on the randomisation method.
Analysis of effectiveness The analysis was based on available data from completers only, as described above. The primary health outcome was QoL (EuroQol, a simple QoL scale and the global QoL measure of the EORTC QLQ-C 30).
Effectiveness results No significant differences in QoL (combining all 3 QoL assessments) were revealed between the follow-up and adjuvant chemotherapy group and there were no differences between withdrawals and the others, neither did follow-up time influence the QoL. The median QoL in all patients and measures was 0.83 (0-1 scale).
Clinical conclusions Adjuvant chemotherapy does not affect short-term QoL.
Modelling No modelling was undertaken.
Measure of benefits used in the economic analysis Quality Adjusted Life Years (QALYs) were used. Life expectancy in Norway at age 62 is 19.6 years; and with radical surgery this patient has a 0.53 probability of a 5-year survival and a 0.47 probability of living less than 5 years. The authors assumed that if the patient survives 5 years, he/she will go on to live for the number of years of their normal expectancy with no relapse (19.6 years). Thus the life expectancy of a patient undergoing surgery is calculated as 12.8 ((0.53x19.6)+(0.47x5)). The quality weight was derived from the mean QoL in all patients and measures in this study.
Direct costs Direct costs included cost of hospital stay and costs of chemotherapy. The cost of hospital stay was based on 1995 annual accounts of the Departments of Surgery (234 per patient per day)) and Oncology (219 per patient per day), medication costs were calculated separately. The total account of the two departments included four areas of costs: staff costs, overheads, consumables and capital. The median hospital stay of patients treated at the Department of Surgery was used for all patients when calculating the total costs of hospitalisation. The cost of chemotherapy was based on the 1995 UHT pharmacy costs. The cost of chemotherapy was 2,188 per patient for 12 months, which includes pharmacy administration costs. The cost of chemotherapy administration in an outpatient setting was calculated according to the tariff of the Norwegian Medical Association (768 per patient per year). In a sensitivity analysis three different discount rates were used: 0%, 5% and 10%.
Statistical analysis of costs Indirect Costs Indirect costs included travel costs due to ACT which were estimated according to the tariff of the National Insurance Association and a qualified guess (480/patient treated with ACT). Costs due to follow-up were considered almost equal in the 2 arms and were not included in the analysis.
Currency UK pounds sterling (), with 1 = 10.00 NOK
Sensitivity analysis Three levels of survival benefit were chosen: 5%, 10% and 15% improvement in 5-year overall survival.
Estimated benefits used in the economic analysis QALYs gained from surgery alone were 8.6 without relapse and 2.0 with relapse. QALYs gained from adjuvant chemotherapy after surgery were 9.5 (1.7 with relapse) assuming 5% raised 5-year survival; 10.2 (1.6 with relapse) assuming 10% raised 5-year survival; and 11.0 (1.3 with relapse) assuming 15% raised 5-year survival. The undiscounted QALYs gained by chemotherapy ranged from 0.6 to 1.7. Employing a discount rate of 5% the minor gain in QoL was in the range of 0.2-0.7.
Cost results ACT in Dukes' B and C CRC raised costs in colon carcinoma by 3,360 from 2,106 to 5,466 and in rectal carcinoma from 3,276 to 6,632. The authors assumed 63% would go through one year of adjuvant chemotherapy with those who withdrew doing so after one month of therapy. Calculation of raised costs is reported in detail in the paper.
Synthesis of costs and benefits The incremental costs per QALY gained at a 0% discount rate were 5,600 for 5% improvement in 5-year survival, 2,800 for 10% improvement in 5-year survival and 1,976 for 15% improvement in 5-year survival.
The incremental costs per QALY gained at a 5% discount rate were 16,800 for 5% improvement in 5-year survival, 6,720 for 10% improvement in 5-year survival and 4,800 for 15% improvement in 5-year survival.
The incremental costs per QALY gained at a 10% discount rate were 33,600 for 5% improvement in 5-year survival, 16,800 for 10% improvement in 5-year survival and 16,800 for 15% improvement in 5-year survival.
Authors' conclusions Using a cut-off point of 20,000 per QALY, adjuvant chemotherapy in CRC appears to be cost-effective only when the improvement in 5-year survival is greater than 5%. Adjuvant chemotherapy does not affect short-term QoL.
CRD COMMENTARY - Selection of comparators The choice of comparator, surgery alone, was clear and sound.
Validity of estimate of measure of benefit The analysis was based on an RCT, which was appropriate for the study question. The study sample seems to be representative of the study population. No details were provided on the comparability of treatment groups at analysis. Although an RCT was conducted, outcomes were analysed for treatment completers only; however, no significant differences were detected between drop-outs and completers. Estimation of benefits was obtained directly from the effectiveness analysis. This choice of estimate seems justified. The validity of the study's design is uncertain, as there were no details of randomisation procedure.
Validity of estimate of costs Direct costs as well as indirect costs (travel costs) were included. Costs due to follow-up were considered almost equal in the two arms and were not included in the analysis. Costs and quantities were reported separately. The authors performed appropriate currency conversions; the date to which prices relate was reported and three different discount rates for the costs per QALY gained (0%, 5% and 10%) were used.
Other issues The authors made appropriate comparisons of their findings with those from other studies. The issue of generalisability to other settings was not addressed. The study enrolled patients in Northern Norway and this was reflected in the author's conclusion.
Implications of the study If a threshold of 20,000 per QALY gained is accepted, ACT is cost-effective as long as a benefit greater than 5% prevails.
Source of funding Research grant from the Aakre Foundation, Faculty of Medicine, UHT, Norway.
Bibliographic details Norum J, Vonen B, Olsen J A, Revhaug A. Adjuvant chemotherapy (5-fluorouracil and levamisole) in Dukes' B and C colorectal carcinoma: a cost-effectiveness analysis. Annals of Oncology 1997; 8(1): 65-70 Indexing Status Subject indexing assigned by NLM MeSH Adult; Aged; Antineoplastic Combined Chemotherapy Protocols /economics /therapeutic use; Chemotherapy, Adjuvant /economics; Colorectal Neoplasms /drug therapy /economics /mortality /pathology /surgery; Colostomy; Cost-Benefit Analysis; Female; Fluorouracil /administration & Follow-Up Studies; Health Care Costs; Humans; Length of Stay /economics; Levamisole /administration & Male; Middle Aged; Norway /epidemiology; Quality of Life; Survival Analysis; Survival Rate; Travel /economics; dosage /economics; dosage /economics AccessionNumber 21997000504 Date bibliographic record published 31/10/2000 Date abstract record published 31/10/2000 |
|
|
|