|
An actuarial analysis shows that offering lung cancer screening as an insurance benefit would save lives at relatively low cost |
Pyenson BS, Sander MS, Jiang Y, Kahn H, Mulshine JL |
|
|
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. CRD summary This study evaluated the cost-effectiveness of lung cancer screening in a high-risk US population of privately insured people aged 50 to 64 years. The authors concluded that commercial insurers should consider providing coverage for lung cancer screening, for high-risk individuals. The reporting was transparent, but the focus on actuarial methods means that elements of the cost-effectiveness model were weak. The uncertainty in the cost-effectiveness results should be considered to be high. Type of economic evaluation Cost-effectiveness analysis Study objective This study evaluated the cost-effectiveness of screening for lung cancer in a high-risk US population of privately insured people, aged 50 to 64 years. High-risk was defined as a smoker or former smoker, with at least 30 pack-years of smoking. The focus of the analysis was to determine if annual lung cancer screening should be offered, as a commercial insurance benefit, to the defined high-risk population. Interventions Screening, using low-dose spiral computed tomography (CT), based on the International Early Lung Cancer Action Program (I-ELCAP) screening protocol (Henschke. 2011, see Other Publications of Related Interest), was compared against no screening. Methods Analytical approach:Two decision-tree models were constructed, based on the findings from the I-ELCAP study. One was used to estimate the extra monthly cost per person, across the insured population, if screening was made available - the rider price. The other was a stage-shift model, designed to simulate the costs and benefits of screening, over a lifetime. The authors stated that they took a payer perspective. Effectiveness data:The model cohort consisted of patients who were newly diagnosed, with lung cancer, in 2012, and those who were diagnosed in the 15 years before this and survived to 2012. Screening was applied to the 30% of 50- to 64-year-olds who were at high risk. Patient characteristics were based on data from the US Census Bureau and Social Security Administration, and data from the Surveillance Epidemiology and End Results (SEER) database. Mortality was from a published actuarial model of patients with lung cancer. The screening effectiveness was measured by early cancer detection, based on I-ELCAP data. For the cost-effectiveness model, it was assumed that the whole target population was screened each year. It was assumed that screening led to a diagnosis two years earlier, which saved life-years. Diagnosis rates were from the New York part of the I-ELCAP study. The distribution of detected lung cancer stages, without screening, was from the SEER database. Monetary benefit and utility valuations:Not relevant. Measure of benefit:The summary measure of benefit was life-years saved with screening. Cost data:The costs for patients at each cancer stage were from a large private-sector health care database. The prevalence of each stage, in the model, was combined with Medicare and Medicaid current procedural terminology codes (CPT) and International Classification of Diseases, Ninth Revision, codes (ICD-9), to calculate the treatment modes and patterns for the patients. The costs were direct medical costs, such as procedures, consultations, and tests. The costs, by cancer stage, were calculated for years one, two, and five and later of screening and cancer treatment, with an interpolation for the costs between years two and five. In years five and later, the cost per patient was adjusted upwards, using published medical index figures and assumed additional costs, for smokers. All costs were reflated to 2012 US $, using the consumer price index. Analysis of uncertainty:The authors conducted scenario analyses by adjusting the screening effectiveness, incorporating false positives and their treatment, varying the percentage of patients screened, and modifying the costs. Results Screening was estimated to produce 130,195 additional lung cancer survivors, who lived 1.7 million more years, at an additional cost of $4.4 billion annually. The incremental cost-effectiveness ratio (ICER) with screening, was $18,862 per life-year saved. The sensitivity analyses produced ICERs between $11,708 and $26,016. The authors stated that this cost-effectiveness of lung cancer screening compared favourably with the cost-effectiveness of cervical, colorectal, and breast cancer screening programmes. Authors' conclusions The authors concluded that commercial insurers should consider covering lung cancer screening for high-risk individuals, as its cost-effectiveness compared favourably with that of screening for other cancers. CRD commentary Interventions:The comparison of no intervention and screening was appropriately justified by the context of the analysis. The individual components of the screening programme were from a well-referenced international lung cancer screening protocol. Effectiveness/benefits:The methods were exhaustively reported and, where not listed in the appendix, references were given. The assumptions made in the model were usually conservative, with appropriate justifications provided. The model used established actuarial methods, some of which differ to those used for economic evaluations. Static percentages were used for the number of cases of each diagnosis that occurred with screening. Stochastic methods, which are well established in economic evaluations, could have produced a more thorough analysis of uncertainty, to give some idea of variance within the data. No variances were reported for any study data. The data appear to have been from valid sources, but the main clinical source was not described in detail, so a full assessment cannot be undertaken. Costs:The perspective was that of the insurer or payer. The cost codes were comprehensively reported, and were from appropriate industry and governmental sources, with good justifications given for the source selection and costing methods. The sources appear to have been appropriate, but may be of limited generalisability to the UK NHS. The unit costs were not reported for the different elements of treatment, and the original sources need to be consulted to validate the cost estimates. The costs were only presented as aggregates of the resources and prices, and it was not clear how the total costs were calculated for the ICER. Analysis and results:The focus of the analysis was to assess if lung cancer screening should be offered, as an insurance benefit, to the target high-risk population. The assumptions and analysis in the first model, to establish the additional cost per insured person, with screening, appear to have been appropriate. In the second model, to establish cost-effectiveness, the assumption of full compliance in the target population was highlighted as being clinically unrealistic. The impact of looking at cost trends until 2012, and the potential life-years saved after 2012, is unclear. The costs may not have required discounting, but it is likely that future life-years gained would require discounting. This potential mismatch could cause exaggerated benefits for patients and have an increasing effect on the ICER presented. The scenario analyses assessed changes in the populations detected and referred to treatment by screening, but did not assess changes to transition probabilities within the decision trees. There was no reporting of variance in the costs, outcomes, and transition probabilities within the model. It is unclear what effect any uncertainty could have had on the results. Concluding remarks:The reporting was fairly transparent, but the focus on actuarial methods means that elements of the cost-effectiveness model were weak. The uncertainty surrounding the ICER results should be considered to be high. Funding Funding received from the Lung Cancer Alliance, and the American Legacy Foundation. Bibliographic details Pyenson BS, Sander MS, Jiang Y, Kahn H, Mulshine JL. An actuarial analysis shows that offering lung cancer screening as an insurance benefit would save lives at relatively low cost. Health Affairs 2012; 31(4): 770-779 Other publications of related interest Henschke CI. International Early Lung Cancer Action Program: enrollment and screening protocol. New York, NY, USA. I-ELCAP. 2011. Indexing Status Subject indexing assigned by NLM MeSH Actuarial Analysis; Cost Savings; Humans; Insurance Benefits /economics; Lung Neoplasms /diagnosis; Mass Screening /utilization; Middle Aged; United States AccessionNumber 22012024531 Date bibliographic record published 22/01/2013 Date abstract record published 29/05/2013 |
|
|
|