|Samfunnsokonomisk lonnsomhet av mammografiscreening i Norge [Cost-effectiveness of mammography screening in Norway]
|Karesen R, Bo J K, Haustveit S, Hervik A, Thoresen S O
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.
The use of mammography to screen for breast cancer.
Economic study type
The study population comprised all Norwegian women aged between 50 and 69 years.
The setting was the community. The economic study was conducted in the county of Oslo, Norway.
Dates to which data relate
The effectiveness estimates were derived from a publication in 1993 (see Other Publications of Related Interest). The resource use data related to the period from January to April 1997. The price year was 1997.
Source of effectiveness data
The effectiveness data were derived from a single observational study.
A model was used to synthesise the resource consumption data, recorded over the course of a population-based screening programme, with an estimate of the number of life-years saved from screening for breast cancer.
Methods used to derive estimates of effectiveness
The effectiveness of the screening programme was estimated from the results of a single study. The estimate of mortality in an unscreened population was assumed. The number of detected cancers and the subsequent mortality were the outcomes used in the model.
Estimates of effectiveness and key assumptions
Mammography was estimated to reduce mortality from breast cancer by 30%. This estimate was taken from a study based on a population of Swedish women. The authors compared some of the characteristics of the Swedish women in the primary study, with those of the Norwegian population undergoing the screening programme. For example, 40% of the cancers were less than 15 mm in the Swedish study, compared with 53% in Oslo. Further, the cancer had spread to the lymph system in 30% of the Swedish women, compared with 22% in the Norwegian women. The extrapolation of the study estimate to the population-based sample was assumed to be appropriate on the basis of these comparisons. The baseline mortality for breast cancer was assumed to be 50% in an unscreened population.
Measure of benefits used in the economic analysis
The outcome measure used in the economic analysis was the number of life-years saved. This was estimated from the number of cancers detected in the programme under evaluation, and was combined with the estimate of the reduction in mortality due to detection by screening. The benefits of mammography, in terms of the number of life-years saved, were aggregated over a period of 25 years.
The quantities of resource use for the procedures involved in the screening process were recorded for those patients in the screening programme. These were not reported separately in the paper. The analysis of the costs per procedure included the time spent by the personnel (radiographers, doctors, nurses and bioengineers), the cost of the bandages and medication, and the capital costs. The costs of resource use were estimated using hourly salary rates and reimbursement charges. The future costs were discounted at a rate of 4.5%. The prices were from 1996 to 1997. The total costs for mammography were aggregated over a 25-year period.
Statistical analysis of costs
No statistical analysis of costs was reported.
The indirect costs were not included in the analysis.
A sensitivity analysis was used to investigate variability in the cost data and the positive predictive value from the screening. The authors assumed that the average cost of screening per patient in the base-case, was the same all over Norway as in the population sample from the county of Oslo. However, other cost-comparisons have indicated that the costs of the screening programme may be higher in other counties. Therefore, the authors expressed the cost per screening over four alternative cost levels, ranging from an additional cost of 10% to an additional cost of 40% above the base-case. The positive predictive value in the population sample from Oslo was 0.19. The authors estimated the cost per life-year saved under alternative values ranging from 0.125 to 0.20.
The authors also presented an alternative scenario in which 80% of the women attended the screening, the additional costs of capital and diagnosing were included, and the costs of the patients' time were included. The result of this scenario did not deviate substantially from the base-case results of the model.
Estimated benefits used in the economic analysis
Using baseline parameters, the authors estimated that population-based mammography in Norway would save 4,500 life-years over a period of 25 years, compared with no screening.
The cost of screening one woman was estimated to be 395 NOK. The total cost of population-based screening for Norwegian women aged between 50 and 69 years was estimated to be 1,222,353,192 NOK. (Reviewer's note: the population of Norway is 4.5 million.)
Synthesis of costs and benefits
The aggregate programme costs were divided by the estimated number of life-years saved over the 25-year period of the study. The average cost per life-year saved for a population-based screening programme was 29,612 NOK. This increased to 41,457 NOK under the assumption that the total cost per screening was 40% higher, and to 31,048 NOK under the assumption that the positive predictive value was reduced to 0.125.
Population-based screening for breast cancer would save 4,500 life-years in Norway over a 25-year period. The use of mammography is cost-effective at a cost of approximately 30,000 NOK per life-year saved.
CRD COMMENTARY - Selection of comparators
Screening was compared with no screening, which may be relevant in populations where the procedure is not currently available. However, this is not the case in Norway. An incremental analysis of population-based screening programmes with varying degrees of screening frequency may, therefore, have been more appropriate. For example, biannual screening compared with screening every third or fifth year.
Validity of estimate of measure of effectiveness
The estimates were derived from one primary study. However, the design, conduct and validity of this study, on which the estimate of 30% reduction in mortality from breast cancer was based, was unclear. The validity of the estimate was therefore difficult to assess.
Validity of estimate of measure of benefit
The benefit was estimated by comparison with an assumption of an approximately 50% mortality from breast cancer under the no-screening alternative. This assumption was not derived from the clinical literature in spite of being key to the analysis. This should be taken into account when attempting to generalise the results from this study to other populations.
Validity of estimate of costs
The cost estimates for the analysis were comprehensive. However, they were derived from one specific screening centre and may, therefore, not be representative of institutions elsewhere in the country. This was recognised by the authors and reflected in the sensitivity analysis of the costs. The results for the costs for the clinical procedures were compared with the hospital reimbursement rates. This comparison provided further indication that the costs were realistic estimates.
The impact of screening attendance could influence the cost-effectiveness of the screening programme. This was not explored in the analysis.
Implications of the study
The authors did not make specific recommendations for policy or further research. However, their results quantify the health benefits and related costs, as well as the total cost of the screening programme in Norway.
Karesen R, Bo J K, Haustveit S, Hervik A, Thoresen S O. Samfunnsokonomisk lonnsomhet av mammografiscreening i Norge. [Cost-effectiveness of mammography screening in Norway] Tidsskrift for den Norske Legeforening 1999; 119(24): 3553-3559
Other publications of related interest
Nystrom L, Rutqvist LE, Wall S, Lindgren A, Lindqvist M, Ryden S, et al. Breast cancer screening with mammography: overview of Swedish randomized trials. Lancet 1993;341:973-8.
Subject indexing assigned by NLM
Breast Neoplasms /economics /mortality /prevention & Cost of Illness; Cost-Benefit Analysis; Female; Humans; Mammography /economics; Mass Screening /economics; Norway /epidemiology; Prevalence; Prognosis; control
Date bibliographic record published
Date abstract record published