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 study examined ultrasonography screening for abdominal aortic aneurysm (AAA) in men aged 65 to 74 years, compared with no screening. After screening, men with an aortic diameter of 3.0 cm or greater were defined as having an AAA and were subsequently invited for recall scans to monitor the growth of the aneurysm. Men with an aortic diameter of 3.0 to 4.4 cm were re-screened every year, while those with an aortic diameter of 4.5 to 5.4 cm were re-screened every 3 months. Men were considered for elective surgery when the aortic diameter reached 5.5 cm, aortic expansion was 1.0 cm or more in 1 year, or when they experienced symptoms attributable to the aneurysm. Finally, men with an aortic diameter of less than 3.0 cm on the initial scan were not re-screened.

Screening.

Cost-effectiveness analysis.

The study population comprised men aged 65 to 74 years. Men with known AAA, previous AAA surgery, or terminal illness were not included.

The setting was secondary care. The economic study was carried out in the UK.

The effectiveness and resource use data were gathered from 1997 to March 2005. The costs were measured in 2004/05 prices.

The costing was performed prospectively on the same sample of patients as that used in the effectiveness analysis.

The design and results of the MASS are reported in a separate publication (see 'Other Publications of Related Interest' below for bibliographic details), thus limited information on the primary study was reported in the current paper. Overall, of the 70,495 patients who were initially identified, 2,725 were excluded because they did not meet the inclusion criteria. Thus, the final study sample included 67,770 men, of which 33,887 were in the control group and 33,883 in the screening group. The mean age at enrolment was 69.2 years in both groups.

This was a prospective, randomised controlled trial, which was carried out in four medical centres in the UK. Details of randomisation were not reported. The length of follow-up ranged from 5.9 to 8.2 years (mean 7.1 years). The loss to follow-up because of death was 2.1% overall (2.2% in the control group and 2.1% in the screening group). Loss to clinical follow-up (non-attendance at recall scans) was 19% at 4 years and 24% at 7 years. Blinding was not performed.

The primary outcome measure was AAA-related mortality. This was defined as all deaths within 30 days of any AAA surgery (elective or emergency) plus all deaths coded as ruptured AAA, AAA without mention of rupture, ruptured aortic aneurysm at unspecified site, or aortic aneurysm at unspecified site without mention of rupture. The secondary clinical end point was all-cause mortality. Mortality data were obtained after flagging on the national database. The baseline comparability of the study groups was not reported, but the groups were likely to have been similar given the large sample size and the randomised design. However, the authors stated that patients who accepted an invitation to screening were more likely to be a less-fit group. The analysis of the clinical study was conducted on an intention to treat basis. Both measures of mortality were compared using unadjusted Cox regression.

The total AAA-related deaths per 1,000 person-years were 196 in the control group and 105 in the screening group, giving a hazard ratio of 0.53 (95% confidence interval, CI: 0.42 to 0.68).

When adjusted for baseline differences, the hazard ratio was reduced to 0.42 (95% CI: 0.33 to 0.55).

The total deaths per 1,000 person-years for all-cause mortality were 7,119 in the control group and 6,882 in the screening group, giving a hazard ratio of 0.96 (95% CI: 0.93 to 1.00).

The effectiveness analysis showed that screening was effective in reducing AAA-related deaths and all-cause mortality in men aged 65 to 74 years.

The summary benefit measure used was the number of life-years (LYs) gained with screening over no screening. LYs were estimated as the area under the Kaplan-Meier curves for both groups, or calculated using two definitions of mortality (i.e. AAA-related deaths and all-cause deaths). An annual discount rate of 3% was applied.

The analysis of the costs was undertaken from the perspective of the health system. It included the costs of invitation to screening, re-invitation after nonresponse, initial scan, recall scan, consultation for elective surgery, elective AAA surgery and emergency AAA surgery. The unit costs were reported extensively, but quantities of resources used were given only for a few items. Resource use was determined prospectively from the sample of men included in the clinical trial. The source of the costs was not explicitly reported. Discounting was relevant given the time horizon of the analysis (7 years), and an annual rate of 3% was used. The costs were estimated using 2004/05 prices.

The costs and quantities of resources used appear to have been treated deterministically.

Productivity costs were not considered in the economic analysis.

US dollars ($). The costs were initially measured in UK pounds sterling (), then converted using a rate of 1.00 = $1.58.

The issue of uncertainty in the cost-effectiveness estimates was investigated in a deterministic sensitivity analysis in which the following scenarios were considered:

the costs retained at the 2000/01 financial-year level for comparison with previous publications,

quality-adjusted life-years based on age-related reductions,

US-based unit cost estimates for scans and surgeries,

an increase of 50% in the cost of a consultation, and

scenarios 3 and 4 combined.

Alternative values of utility weights were obtained from a published study, while US cost estimates were based on Medicare reimbursement. CIs for cost-effectiveness ratios were also calculated, probably using a bootstrapping technique, and cost-effectiveness acceptability curves were generated.

When AAA-related deaths only were considered, the total LYs over the 7-year period were 2,143.30 in the control group and 2,145.82 in the screening group (difference 2.52).

When all-cause deaths were considered, the total LYs over the 7-year period were 2,124.96 in the control group and 2,131.48 in the screening group (difference 6.52).

In both cases, differences in LYs were much higher in years 5 to 7 that in the first four years.

When only AAA-related deaths were considered, the total costs per patient over the 7-year period were $116.72 in the control group and $251.54 in the screening group (difference $134.82).

When all-cause deaths were considered, the total costs per patient over the 7-year period were $116.63 in the control group and $251.65 in the screening group (difference $135.02).

Incremental cost-effectiveness ratios were calculated in order to combine the costs and benefits of the alternative strategies.

When AAA-related deaths only were considered, the incremental cost per LY gained with screening over no screening was $19,500 (95% CI: 12,400 to 39,800).

When all-cause deaths were considered, the incremental cost per LY gained with screening over no screening was $7,600 (95% CI: 3,300 to no limit).

The cost-effectiveness acceptability curve showed that, at a willingness-to-pay of $40,000 per LY gained, the probability that AAA ultrasonography screening would be cost-effective was 98%.

The results of the sensitivity analysis showed that the incremental cost-effectiveness ratio was:

$16,400 (95% CI: 10,400 to 33,400) when 2000/01 prices were used,

$24,600 (95% CI: 15,700 to 49,700) with quality of life adjustment (cost-utility analysis),

$29,600 (95% CI: 18,900 to 60,200) with US costs,

$20,700 (95% CI: 10,400 to 33,400) with consultation costs increased by 50%, and

$30,800 (95% CI: 19,700 to 62,600) with consultation costs increased by 50% and US costs.

Screening ultrasonography for abdominal aortic aneurysm (AAA) was cost-effective and its value for money improved over time. Screening ultrasonography remained cost-effective even in the most unfavourable scenario.

The rationale for the choice of the comparators was clear in that no screening represents the current pattern of care in most settings. You should decide whether this is a valid comparator in your own setting.

The analysis of effectiveness was based on a large clinical trial, which was appropriate for the study question. Limited information on the design and other aspects of the study was provided as the trial itself was published in a companion paper. However, the length of follow-up, the use of intention to treat in the analysis of effectiveness, and the large sample of patients and institutions involved ensure a high internal validity of the clinical estimates.

The summary benefit measure (LYs) was appropriate in that it reflected the most relevant dimension of health for patients at risk of developing an AAA. Discounting was performed, as recommended by UK guidelines. Two alternative sources of mortality data were considered. The impact of screening on quality of life was investigated in the sensitivity analysis.

The analysis of the costs was consistent with the stated perspective, and all the relevant categories of costs appear to have been included. A breakdown of the cost items was given and the unit costs were reported. This might enable the analysis to be replicated in other settings. However, limited information on resource use was given. The source of the costs was not stated clearly, but it is likely to have reflected typical UK sources. The sensitivity analysis investigated the use of alternative sources of costs. The price year was reported, thus facilitating reflation exercises in other time periods.

The authors did not compare their findings with those from other studies. They did not address the issue of the generalisability of the study results to other settings and limited sensitivity analyses were carried out. This reduces the external validity of the analysis. However, variability in the clinical and economic estimates was adequately addressed using a bootstrapping technique and by presenting cost-effectiveness acceptability curves. The authors noted some limitations to their analysis such as the inclusion of deaths from aortic aneurysm at an unexpected site, which might include some thoracic aortic aneurysm. This could have led to an underestimation of treatment effectiveness.

The study results support the use of screening ultrasonography for AAA.

Supported by a grant from the UK Medical Research Council and a Raymond and Beverly Sackler Studentship Award.

Because readers are likely to encounter and assess individual publications, NHS EED abstracts reflect the original publication as it is written, as a stand-alone paper. Where NHS EED abstractors are able to identify positively that a publication is significantly linked to or informed by other publications, these will be referenced in the text of the abstract and their bibliographic details recorded here for information.

Multicentre Aneurysm Screening Study Group. The Multicentre Aneurysm Screening Study (MASS) into the effect of abdominal aortic aneurysm screening on mortality in men: a randomised controlled trial. Lancet 2002;360:1531-9.

Multicentre Aneurysm Screening Study Group. Multicentre Aneurysm Screening Study (MASS): cost effectiveness analysis of screening for abdominal aortic aneurysms based on four year results from randomised controlled trial. BMJ 2002;325:1135.

Henriksson M, Lundgren F. Decision-analytical model with lifetime estimation of costs and health outcomes for one-time screening for abdominal aortic aneurysm in 65-year-old men. Br J Surg 2005;92:976-83.

Lindholt JS, Juul S, Fasting H, Henneberg EW. Cost-effectiveness analysis of screening for abdominal aortic aneurysms based on five year results from a randomised hospital based mass screening trial. Eur J Vasc Endovasc Surg 2006;32:9-15.