Research and analysis

UK AAA screening programmes: 10-year effectiveness review

Published 6 May 2025

The aim of this report was to assess the effectiveness of population screening for AAA in men aged 65, by reviewing the AAA screening programmes across the UK between 2013 and 2023.

Executive summary

In 2005, the UK National Screening Committee (UK NSC) recommended screening for abdominal aortic aneurysm (AAA) in men aged 65. Although death rates had been reducing, there were still as many as 3,000 deaths each year from ruptured AAA (rAAA) in men aged over 65. The NHS AAA screening programmes in the UK were designed to reduce AAA-related deaths in men. Population-based ultrasound screening of men aged 65 commenced in 2009, and by 2013 was fully operational across the UK. This review considers whether the programmes have been effective in their first decade: April 2013 to March 2023.

The aim was to review the data collected by the 4 UK programmes in England, Scotland, Wales and Northern Ireland. Additionally, national statistics were obtained from the 4 nations on admissions to hospital for rAAA and deaths recorded with an underlying cause of rAAA. This was supplemented with activity data from the National Vascular Registry (NVR), the database of the Vascular Society of Great Britain and Ireland (VSGBI), and other international registry data. An updated cost-effectiveness analysis, qualitative quality of life research, and quantitative analysis of inequality were also included.

During the decade of the review, just over 3.5 million men aged 65 were eligible and invited for screening. Just under 80% of these men attended and had a definitive scan. Just over 1% of men had an abnormal scan result (a AAA measuring 3.0cm or more). Men with small AAAs were monitored with regular scans (surveillance) to make sure they did not grow to a dangerous size.

The prevalence of AAA fell about one third during the decade from around 1.2% to 0.8%. During the decade, 7,500 men with a large AAA were referred for treatment (1,558 from their first scan and 5,614 from surveillance monitoring). Most men had routine treatment and 98.5% of them survived, with their AAA repaired. The risk-adjusted elective mortality for intervention was 0.5% for endovascular aneurysm repair (EVAR) and 3.0% for open repair.

During the decade, the number of operations for rAAA recorded on the NVR in men reduced by around half, from around 1,000 per year down to just under 500. In men aged 65 to 74 years (the age group offered screening), the reduction was over two thirds, from around 350 per year to just over 100. Nationally, hospital admissions for rAAA in men (recorded in Hospital Episode Statistics (HES)) also reduced by around 50% from about 2,600 to 1,300 per year across the UK.

In all men aged 65 and over in England and Wales, the proportion of deaths recorded as being from rAAA fell from around 1.0% to 0.6%. Finally, international registry data examining rates of rAAA from 2014 to 2019 showed that the only consistent reduction in rAAA deaths were in Sweden and the UK. These are the only 2 countries in the international collaboration of vascular registries (VASCUNET) that have offered population screening for over a decade.  

Other measures we examined confirmed previous work that uptake was lowest in the most deprived socioeconomic quintile where the prevalence of AAA was also highest. AAA screening remains cost-effective using updated cost measures, and although screening does impair quality of life for some, the effect is small, and mitigation may be possible.

It can be concluded that the AAA screening programmes in the UK are effective and contribute substantially to ongoing reductions in AAA-related mortality (deaths).

Background

An AAA is a swelling of the aorta, the main artery supplying blood to the lower part of the body. AAA is more common in men and in some people, particularly smokers, the aorta can weaken and start to stretch. In a small number of people, the AAA grows so large it bursts; this is termed a rAAA. There is a high fatality rate among people who have a rAAA. Ultrasound screening of men aged 65 has been shown scientifically to detect AAA, which can then be treated before it ruptures. A population screening programme was recommended by the UK NSC in 2005. In that year, there were 2,622 deaths in men from rAAA in the UK, although this number was reducing over time as fewer people smoked cigarettes.

The AAA screening programmes in the UK aim to reduce AAA-related deaths in men by identifying aneurysms at a stage before rupture is likely to have occurred. Surveillance and treatment can then be offered at the appropriate stage to reduce the risk of AAA rupture, emergency hospital treatment, and death.

In 2005, the UK NSC made a recommendation for an AAA screening programme after a review of the evidence from randomised control trials (RCTs). The test is a one-off ultrasound scan offered to men in their 65th year. Men with an aorta below 3.0cm in diameter are reassured and discharged. Men found to have a small (3.0cm to 4.4cm) or medium (4.5cm to 5.4cm) aneurysm are offered surveillance annually and quarterly, respectively. Men with an aneurysm measuring 5.5cm or more are referred to a specialised vascular service for consideration of intervention.

Gordon Brown, the Prime Minister at the time, announced the AAA programme in 2008 and a phased implementation of the programme began across the UK in 2009. Full coverage of the population was achieved in Northern Ireland in 2012, and in England, Wales and Scotland in 2013.

All the UK screening programmes had completed 10 full years of screening by the end of March 2023. The following effectiveness review aims to determine if the AAA screening programmes in the UK can be considered effective.

Burden of AAA disease

AAA is a degenerative condition of the abdominal aorta, the main artery supplying blood to the abdomen, pelvis and legs. It results in progressive dilatation of the aorta over several years, with increasing risk of rupture as the aorta increases in diameter. Once the aorta reaches 5.5cm in diameter in men, it is usually considered that the risk of intervention to prevent rupture is lower than the risk of rupture for most people. The death rate due to AAA in men is around 1.5 times the rate in women. Ruptured AAA is a significant cause of death in elderly men; the risk increased during the 1990s, in parallel with the risk of cigarette smoking 20 years before (Figure 1).

It is known that deaths from AAA have been falling in the UK since around 2000. It is considered this is the result of changing smoking habits of the population, illustrated from Office for National Statistics (ONS) death rates from AAA collected over 30 years (Figure 1). However, in 2005, when the screening recommendation was made, there were still nearly 3,000 deaths in men from rAAA per year in the UK.

Figure 1: Mortality from all AAA (elective repair deaths) and ruptured AAA, 1979 to 2009 in England and Wales

The graph in Figure 1 shows there was a steady increase in the rate of mortality from AAA in England and Wales (measured in deaths per 100,000 population) from 1979 to the mid-1990s when it peaked at more than 60 per 100,000 in men and more than 20 per 100,000 in women. The rate of mortality then declined steadily.

The ONS has been publishing deaths from rAAA (International Classification of Diseases (ICD) code I71.3) by 5-year age groups since 2001 as part of the annual mortality statistics in England and Wales. In 2001, rAAA accounted for 1.46% of deaths in men aged 65 years and over in England and Wales. In 2005, there were 2,622 deaths (1.38%) in men aged 65 and over from rAAA in England and Wales.

A systematic review of available literature from 1977 to 2012 found that the mortality from AAA rupture was high, with nearly a third dying in the community before reaching hospital. Of those who underwent emergency AAA surgery, the postoperative mortality rate was 53% (95% confidence interval (CI) 48% to 59%), making the case fatality after rupture 82% (95% CI 78% to 86%) in high quality studies. This is compared with an elective postoperative mortality rate in high quality vascular services of 3% to 8% at that time, as stated in the second vascular surgery database report from the European Society of Vascular Surgery (ESVS) .

The evidence for population screening in men

In the 1990s, when deaths from rAAA were at a peak in England and Wales, there were a series of RCTs looking at the potential benefit of population screening for AAA in men.

The Chichester study included 6,431 men aged 65 to 80 years. There was a 42% reduction in AAA-related mortality at 5 years, 21% reduction at 10 years and 11% at 15 years with a 74% uptake of screening in the intervention group.

The Vibourg County study in Denmark included 12,639 men aged 64 to 73. There was a 67% reduction in AAA-related mortality over 52 months with a 76.6% uptake.

The Multicentre Aneurysm Screening Study (MASS) trial was the largest and most comprehensive RCT and included 67,800 men aged 65 to 74 years. It demonstrated a reduction of AAA-related mortality of 42% over 4 years with 80% uptake of screening in the intervention group. The reduction in deaths increased to 47% at 7 years and 48% at 10 years.

The Western Australia study included 41,000 men aged 65 to 83 years. This was the only trial in which screening did not reduce AAA-related mortality significantly (mortality ratio 0.61, 95% CI 0.33 to 1.11).

A cost-effectiveness analysis nested within the MASS trial suggested that screening men aged 65 would reduce AAA-related mortality within the thresholds of the National Institute for Health and Care Excellence (NICE) ‘willingness to pay’ guidance.

The 2018 review of all the screening trials (after an updated meta-analysis of the late results of all the AAA screening trials, and after the screening programmes commenced) concluded that screening reduced AAA-related mortality (odds ratio 0.66, 95% CI 0.47 to 0.93, p=0.02) but also had a small effect in reducing all-cause mortality (hazard ratio 0.99, 95% CI 0.96 to 0.99, p=0.03).

In 2013, when the screening programmes were fully rolled out, rAAA accounted for 1.02% of deaths in men aged 65 and over in England and Wales. Deaths were most common in the 75 to 79-year age group, and rare under 65 years. The number of deaths due to rAAA has decreased year on year in data from the ONS (Figure 2) and in 2023 accounted for 0.55% of deaths in men aged 65 years and over.

Figure 2: 2003 to 2023 data on deaths due to ruptured AAA by age group, England and Wales

The graph in Figure 2 shows there is an increase in the number of AAA deaths that occur between the age groups 65 to 69 and 75 to 79, before a decline between the 80 to 84 and 90 and over age groups. The graph also shows there has been an overall decrease in the number of AAA deaths every 5 years between 2003 and 2023.

Outcomes from elective AAA repair

Finally, a screening programme is only maximally effective if the treatment of the disease in question is managed effectively. At the time of the decision to commence AAA screening, it was known that the UK results for elective AAA surgery were among the worst in Europe. A 2008 European Registry group report showed a much higher reported mortality rate for elective AAA repair compared to 9 other European countries (7.9% in the UK compared with 3.5% in the rest of Europe). In response, in 2009 the VSGBI set standards for treatment of AAA in a quality improvement framework that effectively resulted in centralisation of vascular services into a hub and spoke model.

The VSGBI quality improvement framework introduced standards for care delivery and set a target of less than or equal to 3.5% mortality after elective AAA repair by 2013. This work coincided with the rollout of the AAA screening programme. Prior to the introduction of the screening programme, a pre-implementation quality assurance programme was set up in England and Wales to check and approve sites. Men diagnosed with AAA in the screening programme could only be referred to vascular services formally accredited to those standards. As a result, elective AAA mortality fell substantially during the next few years and in the latest NVR report is presently 3.0% for open aortic repair and 0.5% for EVAR.

In Northern Ireland, similar work was undertaken to review the wider configuration of vascular services in light of the VSGBI standards. When the screening programme was established in 2012, guidelines stated that all screen-detected AAAs should be managed in compliance with VSGBI guidance, at a regional vascular centre. Scotland also rolled out a quality assurance programme, however, there was no selection process or accreditation which allowed or prevented NHS health boards from setting up a screening programme. All boards were meant to adhere to the standards set out in the national screening programme.

Methods

The review focuses on the years 2013 to 2023 inclusive, which was the first decade where population screening was offered to each man in their 65th year.

Fundamentally, the AAA screening programme could be regarded as effective if it reduced the number of deaths from rAAA in the UK during 2013 to 2023. National data on AAA rupture was sought from a number of sources and triangulated. The main national resources were:

• government data on hospital admissions coded for ruptured AAA
• ONS data on death certificates recording ruptured AAA as the underlying cause of death

Data were also sought from the NVR, the database of aortic procedures held by the Clinical Effectiveness Unit at the Royal College of Surgeons of England. The NVR is commissioned by the Healthcare Quality Improvement Partnership (HQIP) and works in partnership with the VSGBI and the British Society for Interventional Radiology (BSIR). The NVR is the database of arterial vascular procedures conducted by vascular surgeons and vascular radiologists in the UK. Data on interventions for AAA were sought for the interval of interest.

Although the number of deaths from rAAA provides headline outcomes analysis, the effectiveness review working group agreed that this alone would be an inadequate assessment of the AAA screening programme. A number of other measures were also included for full study:

• programme standards were proposed by the AAA steering group in England before screening commenced, with targets to ensure that the programme was delivered consistently across the country and to a high standard - the other UK nations followed similar processes to put standards in place
• a cost-effectiveness analysis was conducted before the screening programme commenced and revisited in 2014 using updated cost estimates for newer aortic procedures; for the present review, the cost-effectiveness analysis was revisited once more, again using contemporary cost estimates, and a similar strategy
• qualitative measures including assessment of equity of delivery of screening, patient reported outcome measures and other patient assessments

The data in this report either cover screening years (financial year 1 April to 31 March) or calendar years (1 January to 31 December).

National data

Government databases were interrogated, and the data downloaded for analysis, at least for the years 2013 to 2023 inclusive. These included HES data that record hospital admissions, including a diagnostic ICD code of I71.3 for ruptured AAA and ONS data from death certificates, where the underlying cause of death was ruptured AAA. Data from HQIP and VSGBI was requested on rAAA treated surgically during the same time interval, also by treatment method and outcomes. We also requested data on AAA repairs for non-ruptured AAA to determine the effect on activity as a result of commencing the populations screening programme for AAA in men.

Programme data

Programme standards reports for England were published from screening year 2013 to 2014 to screening year 2022 to 2023 and were collated. Some programme reports for Wales, Scotland and Northern Ireland were also published and were collated. Extra data or data for years that were not published were requested directly from the programmes. The focus was on the number of men invited for screening, the uptake or coverage, and the number of AAAs detected over time.

Cost-effectiveness analysis

An updated cost-effectiveness analysis was run in 2024 using contemporary cost information, but similar to discrete event simulation methodology used before in the screening women for abdominal aortic aneurysm (SWAN) study.

Equality data

Information from the programmes on uptake or coverage according to deprivation was collected.

Qualitative data

Forms and surveys are available from within programmes, but a literature review was needed to determine whether there are other published information resources showing the effect of screening on quality of life (QoL).

Results

Cohort offered screening

Between April 2013 and March 2023, just over 3.5 million men in their 65th year were eligible and were offered AAA screening in England, Wales, Scotland and Northern Ireland. Just under 80% of men attended, with small variations in uptakes and coverage noted between the 4 nations (Table 1). All annual figures in the following sections refer to screening years (1 April to 31 March).

Table 1: Overview of AAA screening in 10 years (2013 to 2023 inclusive) by UK country

Country	Number eligible	Uptake (%)	Coverage (%)	Non-visualised screen rate (%)
England	2,968,569	79.6	75.8	1.25
Wales	177,777	77.8	Not measured	1.16
Scotland	353,473	84.5	77.8	2.36
Northern Ireland	97,060	82.8	65.9	1.77

In table 1, the figure in the number eligible column is the number offered screening for Wales. Wales do not measure coverage. In Northern Ireland, coverage was significantly reduced following the COVID-19 pandemic. Initial screening was paused for 11 months and there was a significant catch-up period.

Generic definitions

Coverage is the number of eligible cohort men who had a conclusive screen result within a specified time period.

Uptake is the number of eligible cohort men offered screening who had a conclusive screen result within a specified time period.

Non-visualised screens are the proportion of screens where the aorta could not be visualised.

There are slight differences between definitions for coverage, uptake and non-visualised screens for each of the UK countries. Specific definitions can be found at the beginning of the supplementary data tables.

Over the decade, uptake and coverage varied between countries, but also by year (Table A3 and Table A4 in the Annual data section of the supplementary data tables).

Impact of COVID-19

It is notable that there was a dip in coverage during the COVID-19 pandemic, which has been documented in the AAA standards report 2020 to 2021. Initial screening was either paused or programmes made the decision to invite fewer men for varying lengths of time in each of the four nations, and there were significant catch-up intervals as a result. Whereas coverage was reduced during the COVID-19 pandemic as programmes made the decision to invite fewer men, uptake was maintained by ensuring the men who were invited were actually screened.

Non-visualised scans

The majority of men had a definitive scan, but a small number of scans were incomplete because the aorta could not be visualised (non-visualised scan), usually due to intestinal gas, obesity, or high body mass index (BMI). The number of non-visualised scans varied between countries and local programmes, but ranged, on average, between 1% to 2% (Figure 3). More details are given in the Non-visualised screens section of the supplementary data tables.

Figure 3: Proportion of AAA scans resulting in a non-visualised aorta by UK country

In Figure 3, Wales includes self-referrals and Scotland includes all screening encounters.

The graph in Figure 3 shows that England and Wales had a similar non-visualised screen rate between the screening years 2013 to 2014 and 2022 to 2023. The rate in Northern Ireland started high at about 3.7% in screening year 2013 to 2014, before falling to a similar rate as England and Wales. The rate in Scotland increased between screening years 2013 to 2014 and 2019 to 2020 and decreased between screening years 2019 to 2020 and 2022 to 2023.

For England, the proportion of non-visualised scans has declined from 1.54% in screening year 2013 to 2014 to 1.00% in screening year 2022 to 2023. The proportion of non-visualised scans in Wales was 1.96% in screening year 2013 to 2014 and followed a similar pattern to England throughout the decade, falling to 0.76% in screening year 2022 to 2023. Scotland saw an increase in the proportion of scans that were non-visualised between screening years 2014 to 2015 and 2019 to 2020. There has been a decrease in screening year 2020 to 2021, however, fewer scans were undertaken due to the COVID-19 pandemic. This increase in Scotland is thought to be due to an increase in men with a high BMI and detailed quality assurance images in some NHS boards, as described in the 2021 Scotland AAA screening programme report. The proportion of non-visualised scans in Northern Ireland started at 3.68% and 3.71% in screening years 2013 to 2014 and 2014 to 2015 respectively, before dropping to 1.24% in screening year 2016 to 2017. The proportion has remained comparable to the English and Welsh screening programmes since.

Detection rate of aneurysms for screened men

A screened man is said to have an aneurysm if his aorta measures 3.0cm or more. The prevalence of aneurysms has declined in all 4 nations (Figure 4) between screening years 2013 to 2014 and 2022 to 2023. The prevalence was over 1% in all nations in 2013, but declined to under 1% in England and Wales by 2022. In England, the prevalence was around 0.8% in 2022. This is consistent with the declining prevalence of AAA in men aged 65 in the UK. In a study of 25 years of AAA screening in Gloucestershire, the mean aortic diameter in 65-year-old men declined by 12% over that period.

There has been some variation in prevalence in each year (Table A6 in the Prevalence of AAA section in the supplementary data tables), but the trend is generally down. The decline in the prevalence of AAA is thought to be due to reduced smoking, improved general fitness and improved cardiovascular risk management, with more men taking aspirin and/or statins.

Figure 4: Prevalence of AAA in screened men by UK country, April 2013 to March 2023

The graph in Figure 4 shows that the detection rate, or prevalence, of AAAs has consistently declined in all 4 nations between screening years 2013 to 2014 and 2022 to 2023. Wales, Scotland, and Northern Ireland detection rates are less stable than the rate in England due to smaller numbers. Northern Ireland peaked at about 1.8% in the screening year 2015 to 2016.

Most aneurysms detected at initial screen are classed as small (3.0cm to 4.4cm). About 10% of aneurysms are classed as medium (4.5cm to 5.4cm). The remaining 5% to 10% are classed as large aneurysms (5.5cm or more) and these men are referred for possible intervention (Figure 5).

Figure 5: Size of AAA detected on initial screen by UK country, screening year 2013 to 2014 to screening year 2022 to 2023

Figure 5 shows that most aneurysms detected through the screening programme are small (about 80%), 10% are medium and 10% are large. The proportion of aneurysms that are small has increased slightly over the decade, and the proportion of medium and large aneurysms has decreased slightly over the decade (2013 to 2023).

Referral and treatment

AAA screening programme data

Once an AAA has reached 5.5cm in diameter, in general, the risk from rupture is greater than the risk from intervention to repair the AAA. Men are then referred to a vascular service for consideration of treatment. Each man is evaluated individually to assess their fitness for intervention, and then counselled about their options. Some men decline intervention, particularly if their individual risks are above average.

There are 2 surgical treatment options: open repair or EVAR. EVAR is less invasive than an open repair and has shorter recovery times but is more costly and the stent grafts are less durable. Men may be required to have reintervention if their grafts leak or fail. Open repair is invasive and carries a higher risk of mortality but is more durable (Powell JT, 2010).  

In England, Wales and Northern Ireland, each of the AAA screening providers is linked with a specialised vascular service that was part of a vascular provider network as part of a quality assurance assessment prior to implementation of the screening provider service. AAA screening providers are required to refer patients to vascular services that are part of a vascular provider network and which meet the VSGBI guidance on service provision.

During the decade from 2013 to 2023 inclusive, 1,963 men were found to have an aorta measuring 5.5cm or more at their initial scan. A further 5,814 men had AAA that grew to 5.5cm or more during surveillance, and they too were referred to an accredited vascular service for consideration of treatment (Table 2).

Table 2: Number of men referred to vascular services by UK country

Year	Large AAA at initial scan (England)	Referred from surveillance (England)	Large AAA at initial scan (Wales)	Referred from surveillance (Wales)	Large AAA at initial scan (Scotland)	Referred from surveillance (Scotland)	Large AAA at initial scan (Northern Ireland)	Referred from surveillance (Northern Ireland)
2013 to 2014	238	336	9	Data unavailable	10	Data unavailable	12	3
2014 to 2015	188	385	18	2	15	59	9	13
2015 to 2016	193	399	12	1	26	51	10	11
2016 to 2017	170	484	13	21	34	43	5	19
2017 to 2018	173	506	12	22	25	43	4	13
2018 to 2019	144	522	7	30	16	64	6	25
2019 to 2020	148	508	13	27	23	64	6	23
2020 to 2021	99	535	4	22	28	29	0	28
2021 to 2022	94	633	10	35	25	56	11	32
2022 to 2023	111	666	14	33	20	38	8	33
Total	1,558	4,974	112	193	222	447	71	200

In table 2, the number of men referred from surveillance was calculated by subtracting the number of large aneurysms at initial scan from the number of referrals.

The programme standard for men found to have an aorta of 5.5cm or more and who are fit for intervention after assessment is that they should receive intervention within 8 weeks of their last conclusive scan. These times were arbitrary but considered achievable; their aim is to avoid any man having a rAAA whilst awaiting treatment and to minimise their anxiety whilst waiting as mortality from rAAA remains high. Vascular units have found this target challenging in the present NHS environment and these timelines are reported upon by the NVR (Figure 6, Table 3, and Table 4).

Figure 6: Men operated on within 8 weeks of their last conclusive ultrasound scan, by UK country

In Figure 6, the acceptable and achievable levels are based on the current English national programme standards and may not reflect the standards in the other UK nations. The graph shows that the proportion of men operated on within 8 weeks of their last conclusive scan has been declining in all 4 nations over the decade. The proportion of men operated on within 8 weeks in all 4 of the UK nations has been below 50% since the screening year 2019 to 2020.

Between 1 April 2013 and 31 March 2023, a total of 6,532 men in England, 300 men in Wales, 669 men in Scotland, and 250 men in Northern Ireland were referred for surgery, were suitable for intervention and did not decline treatment. A total of 2,929 men in England, 106 men in Wales, 205 men in Scotland and 99 men in Northern Ireland with a large aneurysm had aneurysm repair surgery within 8 weeks of their last conclusive ultrasound scan (43.1%) (see Table A14 in the Hospital admissions or finished consultant episodes section of the supplementary data tables).

Like all hospital services, the COVID-19 pandemic had a profound effect on vascular services with elective surgery being put on hold during the first and second waves. Men were stratified by the size of their aneurysm and then their wait time. Restoration of surgical services has also been hampered by staffing availability and an increase in men choosing to defer their aortic surgery.

Table 3: Timeframe for men who have AAA repair surgery, England

Year	Suitable for surgery	8 weeks or less	More than 8 weeks but less than or equal to 12 weeks	More than 12 weeks	Unknown
2013 to 2014	574	261 (45.5%)	99 (17.2%)	171 (29.8%)	43 (7.5%)
2014 to 2015	573	298 (52.0%)	112 (19.5%)	145 (25.3%)	18 (3.1%)
2015 to 2016	592	353 (59.6%)	103 (17.4%)	132 (22.3%)	4 (0.7%)
2016 to 2017	654	352 (53.8%)	119 (18.2%)	172 (26.3%)	11 (1.7%)
2017 to 2018	679	362 (53.3%)	119 (17.5%)	193 (28.4%)	5 (0.7%)
2018 to 2019	666	392 (58.9%)	117 (17.6%)	153 (23.0%)	4 (0.6%)
2019 to 2020	656	304 (46.3%)	97 (14.8%)	240 (36.6%)	15 (2.3%)
2020 to 2021	633	158 (25.0%)	126 (19.9%)	337 (53.2%)	12 (1.9%)
2021 to 2022	726	201 (27.7%)	159 (21.9%)	354 (48.8%)	12 (1.7%)
2022 to 2023	775	242 (31.2%)	154 (19.9%)	350 (45.2%)	129 (3.7%)

Data in table 3 were extracted in June 2024 through an ad-hoc query and are unvalidated. They therefore may differ from other published data.

Table 4: Timeframe for men who have AAA repair surgery, Wales

Year	Suitable for surgery	8 weeks or less	More than 8 weeks but less than or equal to 12 weeks	More than 12 weeks	Unknown
2013 to 2014	6	6 (100.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
2014 to 2015	27	27 (100.0%)	0 (0.0%)	0 (0.0%)	0 (0.0%)
2015 to 2016	31	26 (83.9%)	0 (0.0%)	3 (9.7%)	2 (6.5%)
2016 to 2017	41	17 (41.5%)	9 (22.0%)	13 (31.7%)	2 (4.9%)
2017 to 2018	37	14 (37.8%)	6 (16.2%)	14 (37.8%)	3 (8.1%)
2018 to 2019	49	12 (24.5%)	9 (18.4%)	18 (36.7%)	10 (20.4%)
2019 to 2020	46	13 (28.3%)	9 (19.6%)	20 (43.5%)	4 (8.7%)
2020 to 2021	33	2 (6.1%)	5 (15.2%)	25 (7.8%)	1 (3.0%)
2021 to 2022	56	13 (23.2%)	3 (5.4%)	35 (62.5%)	5 (8.9%)
2022 to 2023	47	6 (12.8%)	5 (10.6%)	28 (59.6%)	8 (17.0%)

In England, a total of 6,253 men underwent planned elective AAA repair by EVAR or open repair during the decade of this review. There were 3,113 EVAR procedures (49.7%) and 3,140 open repair procedures (50.3%). Respective perioperative mortality rates were 0.19% for EVAR and 1.08% for open AAA repair (Table 5).

Table 5: Mortality rate for men in the screening cohort who have elective surgery, England

Year	Number of open elective surgeries	Number of EVAR elective surgeries	Number of deaths from open elective surgeries	Number of deaths from EVAR elective surgeries	Open surgery mortality rate (%)	EVAR surgery mortality rate (%)
2013 to 2014	270	253	0	0	0.00	0.00
2014 to 2015	272	280	0	0	0.00	0.00
2015 to 2016	250	323	3	1	1.20	0.31
2016 to 2017	306	327	3	1	0.98	0.31
2017 to 2018	328	338	4	0	1.22	0.00
2018 to 2019	354	300	5	1	1.41	0.33
2019 to 2020	316	321	6	2	1.90	0.62
2020 to 2021	315	292	5	0	1.59	0.00
2021 to 2022	371	321	5	1	1.35	0.31
2022 to 2023	358	358	3	0	0.84	0.00

In table 5, only open or EVAR surgeries were included. Not known and Other were excluded.

National Vascular Registry data

Men who have an intervention in the UK are entered into the NVR, where their clinical details and outcomes are collated. Data submission is a mandatory requirement for all vascular services in England, and case ascertainment levels were high (over 90%) in the 2023 NVR report. Although recommended, data submission to the NVR is not mandated in Scotland, hence case ascertainment is lower (66%). A Vascular Services Quality Improvement Programme (VSQIP) annual report is produced on the outcomes of surgery for all major vascular conditions and available online. Annual figures from NVR data refer to calendar years.

Referral source

The proportion of men who have surgery for their AAA that are referred from AAA screening is increasing (Figure 7). In 2022, just under one third of AAA treated in the UK had been referred from an AAA screening programme.

Figure 7: Proportion of elective interventions by presenting problem, UK (NVR data)

In Figure 7, the presenting problem question was not mandatory in the national vascular dataset before January 2014, which is why the figure contains a lot of missing values. The reason for the increase in the missing data from 2020 onwards is because the NVR was modified to better capture those patients undergoing an aortic procedure who have already had an aortic intervention. These patients would not have been asked the same presenting problem question. The graph shows that the proportion of elective interventions recorded on the NVR as coming from the AAA screening programmes in the UK has been increasing since 2015. The proportion of elective interventions coming from a non-screening route has been decreasing since 2014. There were few unknowns between 2014 and 2019, but this has increased between 2020 and 2022.

Intervention

It is thought that an increasing number of men with a screen-detected AAA have open AAA repair, rather than EVAR. This is due to the durability of the procedure and acceptance of the higher perioperative risks. This change to more open surgery in men fit for open repair was endorsed in NICE AAA guidance in 2020. Men from screening programmes tend to be younger than men who have their AAA detected incidentally. This is borne out in Figure 8 which compares the intervention used for all men between 2013 and 2022 inclusive, and men aged 65 to 74, the cohort which includes men who have been offered screening. By 2022, for all elective AAA interventions, about two thirds of men have an EVAR, whereas for men in the screened cohort, many of whose AAA will be screen-detected, approximately half have an EVAR and the other half have an open AAA repair (Figure 8).

Figure 8: Proportion of non-ruptured (elective) AAA repairs by age group and type of procedure, UK (NVR data)

The graph in Figure 8 shows the proportion of open repair surgeries increased in all ages and the 65 to 74 year age group since 2016. Open repair surgeries have increased more in the 65 to 74 year age group compared to all ages. There is a higher proportion of open repair surgeries than EVAR surgeries in the 65 to 74 year age group.

Ruptured aortic aneurysms

In 2013 there were just under 1,000 procedures for rAAA recorded on the NVR. By the end of the decade, this number fell to just under 500, a reduction of around half (Figure 9). In the screened age group (65 to 74 years), the reduction was from 353 to 101 procedures per year, a reduction of over two thirds. This suggests that although rates of rAAA were falling anyway, the reduction was greater in the cohort offered screening.

Figure 9: Number of ruptured AAA repairs by age group, UK (NVR data)

Figure 9 shows there were approximately 400 to 450 procedures per year that are recorded in the NVR as a non-elective admission but not with a confirmed rupture. These patients are excluded from all NVR data tables and figures, and in forthcoming reports will be analysed and reported upon separately. The graph shows a decrease in the number of rAAA repairs in all ages. There was a greater decline in the 65 to 74 year age group compared to the other ages.

Short-term post-operative mortality

The NVR has recorded 225 in-hospital deaths in men undergoing elective EVAR and 450 in-hospital deaths in men undergoing elective open AAA repair between 2013 and 2022 (Figure 10). The risk-adjusted in-hospital mortality rate for elective AAA treatment recorded on the NVR has remained at around 1.5% since 2012. The 2023 VSQIP report showed that elective open repair has a higher mortality rate than elective EVAR (3.0% compared to 0.5%) (see Table A7 in the Elective and emergency surgeries section of the supplementary data tables).

The NVR has recorded 480 in-hospital deaths in men undergoing emergency EVAR and 2,096 deaths in men undergoing emergency open AAA repair between 2013 and 2022 (Figure 10). There was a reduction in the number of in-hospital deaths after rAAA repairs in the UK from a peak of 363 in 2014 down to around 150 in 2022. This equates to an approximate 50% reduction in the number of deaths after rAAA repair in the UK over the last decade recorded on the NVR.

Figure 10: Number of in-hospital deaths in men undergoing non-ruptured (elective) AAA repairs and ruptured (emergency) AAA repairs by type of procedure, UK (NVR data)

Figure 10 shows in-hospital deaths in all men undergoing elective or emergency surgery by type of procedure. Some of these men will not have been offered screening. The graph shows more deaths following emergency surgery compared to elective surgery. The number of in-hospital deaths after surgery for rAAA has reduced from around 300 per year to around 150 per year over the decade, a fall of around 50%.

National data

The aim of AAA screening is to reduce mortality from rAAA. National data are collected on two main outcomes:

• HES for admission to hospital with a diagnostic code of ICD I71.3 for rAAA
• ONS data on death certificates mentioning rAAA

HES data

The rates of rAAA in men from HES data demonstrate a reduction of around 50% during the last decade from 2,640 to 1,333 admissions per year in the UK (Figure 11). This reduction is most obvious in England due to the larger population but is evident in all 4 nations (see tables A7 to A10 in the Hospital admissions or finished consultant episodes section of the supplementary data tables).

Figure 11: Finished consultant episodes in men (all ages) for ruptured AAA, by UK country

The graph in Figure 11 shows a declining number of finished consultant episodes for rAAA for men of all ages in England since the screening year 2011 to 2012. The numbers are smaller for Wales, Scotland and Northern Ireland, but there has also been a slight decrease in numbers of finished consultant episodes or hospital admissions in these nations over the decade 2013 to 2023. The data for Scotland are only for people aged 60 years and over.

The same data are available over a longer interval in England (April 1998 to March 2023). This puts the reduction in hospital admissions in men with rAAA into perspective before and after the introduction of AAA screening (Figure 12).

Figure 12: Finished consultant episodes in men for ruptured AAA (ICD code I71.3), England

The graph in Figure 12 shows a decline in finished consultant episodes in men for ruptured AAA in England between the financial years 1998 to 1999 and 2022 to 2023. There was a slightly steeper decline between the financial years 2013 to 2014 and 2022 to 2023 compared to between the financial years 1998 to 1999 and 2011 to 2012.

ONS data

The proportion of deaths in men aged 65 years and over recorded as due to rAAA has reduced in all 4 nations in the past decade (Figure 13). It has almost halved in England from 1.02% in 2013 to 0.55% in 2023. The reduction in the other countries is proportionally less but in all 4 nations by 2022, rAAA accounted for only 0.6% of the deaths in men over 65 years.

Figure 13: Proportion of all deaths in men that are due to ruptured AAA (ICD code I71.3) by UK country

The graph in Figure 13 shows a decline in the proportion of all deaths that are from ruptured AAA in all 4 nations of the UK between 2013 and 2023. There are smaller numbers in Wales, Scotland and Northern Ireland compared to England so the rate of decline is more variable in those countries. Northern Ireland has the smallest proportion of deaths from rAAA although this was a similar proportion to the other nations in 2022 and 2023.

Data for England and Wales show a similar reduction, but over a longer interval (Figure 14).

Figure 14: Proportion of all deaths in men aged 65 and over that are due to ruptured AAA, England and Wales

The graph in Figure 14 shows an increase in the proportion of all deaths from ruptured AAA in England and Wales between 1991 and 2000. The rate was fairly stable between 2000 and 2003, and then declined from 2003 to 2023.

International data

Vascular registries internationally who belong to the European Society of Vascular Surgery, and who collect national data, join together to produce reports on vascular interventions across Europe and Oceania. This is called the VASCUNET Collaboration. The 2024 report on data from the VASCUNET Collaboration covering 2014 to 2019 analysed by the UK NVR team contains information about rates of rAAA seen in 9 countries over the last 6 years (Figure 15). The only 2 countries that have population screening programmes for AAA in men over the past decade are Sweden and the UK, which both show consistent reductions in rAAA over the last 6 years.

Figure 15: Data from VASCUNET: International Vascular Registry data on rates of rAAA over 6 calendar years (2014 to 2019) by country

The graph in Figure 15 shows the number of procedures for ruptured AAA between 2014 and 2019 in countries that are part of the VASCUNET collaboration. Most countries increased or stayed relatively consistent over the 5 years, but the number of procedures in Sweden and the UK decreased over the 5 years.

Table 6: Data from VASCUNET: International Vascular Registry data on rates of rAAA over 6 calendar years (2014 to 2019)

Year	Australia	Denmark	Finland	Hungary	NZ	Norway	Sweden	Switzerland	UK
2014	143	161	29	76	59	65	196	Data unavailable	1,074
2015	153	149	19	60	59	93	211	Data unavailable	923
2016	153	157	23	44	46	89	168	Data unavailable	960
2017	126	156	33	40	55	103	142	43	840
2018	140	138	26	63	49	89	145	57	722
2019	138	121	35	55	60	124	114	49	680

Cost-effectiveness of AAA screening in men

Predicted cost-effectiveness of AAA screening in men

UK AAA screening policy in men is based principally on evidence generated by the Multicentre Aneurysm Screening Study (MASS) trial. Between 1997 and 1999, MASS randomised 67,770 men aged 65 to 74 to receive an invitation to AAA screening (intervention group) or no invitation (control group). Final follow-up was completed in 2011, a mean follow-up of 13.1 years, and demonstrated a 42% reduction in AAA-related mortality in the group of men invited for AAA screening. Cost-effectiveness of AAA screening for men was established using 10-year follow-up data from MASS in 2009, demonstrating a cost per life-year gained of £7,600 in the screening arm.

Updated cost-effectiveness of AAA screening in men

In 2014 the National Institute for Health and Care Research (NIHR) commissioned the University of Cambridge to undertake a research study to determine the effectiveness of AAA screening for women (the SWAN study). Rather than undertaking a new clinical trial of AAA screening for women, in SWAN an alternative approach using a statistical model for AAA screening was used, with the project team developing a computer simulated model of AAA screening. The model was first set up to determine the effectiveness of AAA screening in men and validated against the real clinical trial data from the MASS trial.

The SWAN model has previously been used to estimate the updated cost-effectiveness of the AAA screening programme. Based on data from 2011, when AAA screening for men was initiated in the UK, economic modelling identified an incremental cost-effectiveness ratio of £7,370 per quality-adjusted life year gained from AAA screening. This is a similar estimate to that observed in MASS, albeit for non-quality adjusted life years. Due to the observed change in AAA prevalence between MASS (4.9% AAA prevalence in men aged 65 to 74) and data from the English screening programme in 2011 (1.5% AAA prevalence in men aged 65), the researchers went on to determine the effect of any further future fall in AAA prevalence on the economic effectiveness of AAA screening. Their analysis determined that the cost-effectiveness ratio of AAA screening for men would rise above £20,000 per quality-adjusted life year (the NICE willingness to pay threshold at the time) if AAA prevalence fell to 0.35% or less.

Most recently the SWAN model has been used to examine the impact of delayed invitation and reduced attendances observed in AAA screening during the COVID-19 pandemic.

Current estimate of the effectiveness of AAA screening strategy

As part of this review of the AAA screening programmes, a re-estimation of the contemporaneous effectiveness of AAA screening is desirable. This was undertaken using the SWAN discrete event simulation in January 2024.

Several key model parameters such as AAA prevalence and costs vary over time. The aim of this work was to provide an up-to-date estimate of screening effectiveness. The delivery of AAA screening was significantly affected between 2020 and 2023 by the COVID-19 pandemic. This means that data from these years may not be representative of the true current status of AAA screening. To allow for this, data from the most recent complete screening prior to the pandemic was used to repopulate the SWAN model. These data had been gathered by the English screening programme to inform their assessment of the impact of the COVID-19 pandemic on AAA screening. Analyses are based on the scenario of 300,000 men invited to AAA screening.

Clinical effectiveness

Table 7 shows the number of predicted events for the scenario of A) inviting men to AAA screening as per current screening practice and B) no invitation to screening. Inviting men for screening increases the number of elective AAA repairs and reduces the number of AAA ruptures and emergency repairs. AAA-related mortality is reduced by 10.3%. This is lower than the AAA mortality reduction seen in clinical trials of AAA screening due to the increased duration of follow-up included in the model.

Table 7: Predicted number of AAA-related events from computer modelling of 300,000 men invited and 300,000 men not invited for AAA screening in the year of their 65th birthday

Measure	A) Invited	B) Not invited	Change in number of events if invited for AAA screening (%)
Elective AAA repair	3,741	3,231	510 (15.8)
Non-intervention	550	477	73 (15.3)
AAA rupture	2,584	2,974	-390 (-13.1)
Emergency AAA repair	955	1,100	-145 (-13.2)
AAA-related deaths	2,324	2,591	-267 (-10.3)

Non-intervention in table 7 is the number of men with an AAA measuring more than 5.4cm who do not undergo surgical repair of their AAA when referred to a surgeon at a threshold of 5.5cm or more.

The AAA discrete event simulation model also provides estimates of the age when men experience AAA-related events (Table 8). Invitation to screening marginally reduces the mean age at which men undergo elective AAA repair and marginally increases the age at which AAA ruptures, deaths and emergency repairs occur.

Table 8: Predicted mean age at which AAA-related events occur in men invited or not invited for AAA screening in the year of their 65th birthday

Measure	A) Invited	B) Not invited
Elective AAA repair	79.7	81.3
Consultation for elective repair	79.5	81.1
AAA rupture	82.5	81.6
Emergency AAA repair	82.5	81.6
AAA-related deaths	82.8	82.1

Economic effectiveness

Based on the prevalence of AAA and size distribution of aortic diameters observed in the latest available pre-COVID screening programme data (screening year 2019 to 2020, AAA prevalence of 0.92%) and other input parameters updated in 2020, the AAA screening strategy is cost-effective at a willingness to pay threshold of £20,000 per quality-adjusted life-year (QALY) gained (Table 9).

Table 9: Predicted mean age at which AAA-related events occur in men invited or not invited for AAA screening in the year of their 65th birthday

Measure	Invited	Not invited	Difference
Life-years	12.866118	12.860847	0.005271
Cost	£219.26	£168.11	£51.15

Quality-adjusted life years: 0.003974 Incremental cost-effectiveness ratio (ICER) per life-year: 9,705 ICER per QALY: 12,870 Net benefit: £28.34

Net benefit is based on a threshold of £20,000 per QALY. Costs and life-years are discounted at 3.5% per year.

Conclusion of economic data

The AAA screening programme was commenced after prediction that it would be cost effective, using NICE thresholds from that time. The present investigation confirms that screening men aged 65 years for AAA remains cost effective at current willingness to pay thresholds. It is, however, incumbent upon the UK NSC to continue to keep this under review with changing AAA prevalence. The present model has been an effective tool for this process.

Quality of life

Review of the available evidence

Research on the psychological consequences of AAA screening shows that men may experience anxiety when invited for AAA screening and found not to have an AAA - although this is likely to be a temporary state. There is more concern about the potential for men with screening-detected AAA having raised anxiety levels that affect their health-related quality of life. Anxiety may be related to having AAA or attending surveillance to monitor AAA growth.

A 2017 systematic review of the psychological harms of having a screen-detected AAA found moderate levels after recent diagnosis. Qualitative research in 2018 on the consequences of AAA screening found that men suffer shock and anxiety even though they are pleased they are being monitored. Later qualitative studies in 2020 also found heightened anxiety and impacts on men’s health-related quality of life and that of their partner. A narrative review of 11 qualitative studies found that men in AAA surveillance had lower-health related quality of life compared to those with negative scans and the general population. A 2018 quantitative study in the UK showed agreement with this review - that mental health was worse in those diagnosed with AAA compared with those with negative screens, although this difference disappeared after 12 months. However, the most recent systematic review (2020), with meta-analysis, based on a generic quality of life measure, concluded that current evidence did not support a negative impact on quality of life from being in AAA surveillance.

Conclusions on quality of life

It is difficult to draw definite conclusions from the current evidence because studies have considered quality of life at different time points, and problems may vary by time and size of AAA. In addition, research looking at the evidence on psychological harms of screening has shown a tendency to use generic rather than specific measures of health-related quality of life. These may not be sensitive to screening-related harms and may also detect anxiety related to other issues. More recent research has used AAA-specific measures. A survey of 250 men with screening-detected AAA found more negative psychosocial consequences of screening. A Swedish study of 158 men with screening-detected AAA compared with 275 with normal screens found that 96% of men with AAA did not regret being screened. Men with AAA were more likely to have psychosocial consequences of screening, but the authors concluded that a minority had serious problems.

Mitigation

The psychosocial consequences of screening should be considered so that actions can be taken to reduce them. An ongoing research project, the Patient-Centred Abdominal Aortic Aneurism Study (PCAAAS) (funded by the NIHR) focuses on men in AAA surveillance in England. The research is measuring AAA-related anxiety, AAA-related effect on activities of daily living, and the psychological effects of attending for surveillance using measures suitable for AAA screening. A survey of over 700 men has found that they mainly have no or occasional anxiety about their AAA, but that some men in 3-monthly surveillance report being anxious most, or all, of the time. Ongoing qualitative interviews with men are showing how they value the information on AAA growth provided by surveillance even when they have heightened anxiety. The team are now developing an intervention to help men manage AAA-related anxiety, informed by the characteristics of men who are more likely to report psychosocial harm, qualitative interviews with men, and a survey of AAA screening staff. Staff providing AAA screening and surveillance in England are supportive of such an intervention and have clear ideas about what could help men manage anxiety. This focus on an intervention is also supported by authors of recent evidence reviews on reported experiences in patients under AAA surveillance and psychosocial consequences after AAA screening, which have identified the need for information to help men manage anxiety.

Equality

Quality assurance is part of the ongoing process of programme review to enable optimisation of uptake and outcomes. The beneficial effects of screening are maximised when uptake is high, and the test is acceptable to invitees. Screening programmes have a responsibility to ensure that information about the programme is easily and widely available to all the target population.

Social deprivation impacts on many aspects of health and access to healthcare services. There has been little research on the influence of social deprivation on the uptake of AAA screening.

With the availability of the NHS postcode directory, it is possible to assign men to lower super output areas (LSOA) and local authority (LA) geographies and therefore link them to a deprivation score. LSOA is a means of dividing local populations into small cohorts containing from 1,000 to 3,000 people. This means outcomes of screening can be linked to deprivation at local population level. A 2017 study on the effects of social deprivation and ethnicity on AAA screening inequalities identified lower screening uptake in areas of deprivation.

Current results

The results that follow below, together with the above-mentioned 2017 study, show that in general uptake is reduced in the most deprived areas of the UK (see Tables 10 to 12 and Figures 16 to 18 below). Figure 22 below is important, because it shows both the uptake and the aneurysm detection rate by deprivation quintile for England. Uptake is lower in the most deprived quintile, but men who live in the most deprived quintiles are more likely to have an AAA. The tables and figures in this section are by screening year (financial year 1 April to 31 March).

Table 10: Screening uptake by deprivation quintile, England

Year	Deprivation quintile 1 (most deprived)	Deprivation quintile 2	Deprivation quintile 3	Deprivation quintile 4	Deprivation quintile 5 (least deprived)
2013 to 2014	69.5	75.6	79.8	82.7	84.7
2014 to 2015	69.8	76.0	80.0	82.6	84.5
2015 to 2016	70.6	76.6	80.7	83.0	85.2
2016 to 2017	72.5	77.9	81.7	83.9	86.1
2017 to 2018	72.4	77.3	81.5	84.4	86.8
2018 to 2019	72.3	77.3	81.8	84.7	86.7
2019 to 2020	68.1	73.6	78.3	81.5	83.7
2020 to 2021	68.4	74.4	79.0	81.6	84.2
2021 to 2022	69.1	75.6	80.4	82.8	85.1
2022 to 2023	71.4	77.7	82.1	84.7	86.6

Table 11: Screening uptake by deprivation quintile, Wales

Year	Deprivation quintile 1 (most deprived)	Deprivation quintile 2	Deprivation quintile 3	Deprivation quintile 4	Deprivation quintile 5 (least deprived)
2013 to 2014	67.2	71.2	74.2	76.1	80.5
2014 to 2015	67.7	73.3	74.4	75.2	79.5
2015 to 2016	70.4	77.3	78.4	81.5	85.2
2016 to 2017	74.8	78.6	80.2	83.0	85.9
2017 to 2018	72.4	76.3	78.7	81.8	84.3
2018 to 2019	73.3	79.1	80.3	83.0	86.1
2019 to 2020	64.7	71.4	72.3	71.9	77.4
2020 to 2021	78.5	82.2	85.3	87.5	87.2
2021 to 2022	74.9	81.0	83.0	85.7	87.3
2022 to 2023	67.2	75.3	76.9	81.7	84.4

Table 12: Screening uptake by deprivation quintile, Scotland

Year	Deprivation quintile 1 (most deprived)	Deprivation quintile 2	Deprivation quintile 3	Deprivation quintile 4	Deprivation quintile 5 (least deprived)
2013 to 2014	77.4	83.1	86.6	88.4	90.4
2014 to 2015	75.3	82.8	85.2	87.3	88.5
2015 to 2016	76.0	81.7	84.3	87.4	88.8
2016 to 2017	76.0	82.2	85.7	87.6	89.0
2017 to 2018	76.2	82.5	85.7	87.8	89.1
2018 to 2019	76.6	82.0	85.0	87.3	90.0
2019 to 2020	76.9	81.6	85.4	87.5	89.6
2020 to 2021	74.9	80.3	83.8	86.6	89.9
2021 to 2022	76.4	81.3	85.6	88.6	88.9
2022 to 2023	76.1	81.5	85.3	88.6	89.9

Figure 16: Uptake by deprivation quintile, England, April 2013 to March 2023

The graph in Figure 16 shows that uptake in England follows the same pattern in quintile 1 and quintile 5 but uptake is about 15 percentage points lower in Quintile 1, the most deprived quintile, compared to Quintile 5, the least deprived quintile.

Figure 17: Uptake by deprivation quintile, Wales, April 2013 to March 2023

The graph in Figure 17 shows that uptake in Wales follows the same pattern in Quintile 1 (the most deprived quintile) and Quintile 5 (the least deprived) but uptake is about 10 to 15 percentage points lower in the most deprived quintile compared to the least deprived quintile. The gap between quintiles widened in screening year 2022 to 2023, to about 20 percentage points difference.

Figure 18: Uptake by deprivation quintile, Scotland, April 2013 to March 2023

The graph in Figure 18 shows that uptake in Scotland follows the same pattern in Quintile 1 (the most deprived) and Quintile 5 (the least deprived) but uptake is about 12 percentage points lower in the most deprived quintile compared to the least deprived quintile.

Figure 19: Aneurysm detection rate by deprivation quintile, England, April 2013 to March 2023

The graph in Figure 19 shows that the aneurysm detection rate in England has declined in both Quintile 1 (the most deprived) and Quintile 5 (the least deprived). There is about a 0.6 percentage point difference between quintile 1 and quintile 5, with more aneurysms being detected in men living in the most deprived quintile compared to the least deprived quintile.

Figure 20: Aneurysm detection rate by deprivation quintile, Wales, April 2013 to March 2023

The graph in Figure 20 shows a varying aneurysm detection rate in Wales over the decade due to smaller numbers. There was more of a decline in the aneurysm detection rate in Quintile 5 (the least deprived) compared to Quintile 1 (the most deprived). A smaller number of aneurysms were detected in the screening year 2020 to 2021 due to the COVID-19 pandemic.

Figure 21: Aneurysm detection rate by deprivation quintile, Scotland, April 2013 to March 2023

The graph in Figure 21 shows a decrease in the aneurysm detection rate in Scotland over the decade. There was a larger decrease in Quintile 1 (the most deprived) compared to Quintile 5 (the least deprived). There was a smaller number of aneurysms detected in screening year 2020 to 2021 due to the COVID-19 pandemic.

The numbers screened and the number of aneurysms detected were much smaller in Wales and Scotland compared to England therefore there was more variation in the detection rates.

Figure 22: Uptake and detection rate by deprivation quintile, England, April 2013 to March 2023

Figure 22 shows data from England over the duration of the effectiveness review. Detection rates on the right axis of the graph are compared with uptake on the left axis in the same men. For every year of the study, uptake is lower, and detection rates higher in the most deprived LSOAs.

As a consequence of these findings, in 2017, Public Health England (PHE) released a toolkit for local AAA screening providers to use to help address inequalities in the AAA screening programme. The AAA screening inequalities resource and guidance was developed to support screening programmes in each of the 4 nations, and lists examples of best practice from AAA screening services across the UK. Reducing barriers to initial screening attendance, improving access for underserved groups, and engagement with the public were all identified as interventions that could reduce inequalities in screening uptake and, as a result, AAA detection rate.

Discussion

Discussion (general)

The main finding of this review is the steady decline in deaths in men from rAAA during the decade of study. The number of hospital admissions for rAAA in the UK roughly halved and the proportion of men who died from rAAA was reduced from around 1.0% to 0.6%. Overall, the number of men admitted to hospital with a diagnosis of rAAA declined from over 2,400 per year to around 1,300 per year over the decade. Whereas part of this may have been due to ongoing reduction in the prevalence of AAA since the mid-1990s (and the detection of AAA by other imaging before age 65), it remains likely that a substantial proportion of the reduction is due to the AAA screening programme delivered by the NHS in each of the 4 nations. Evidence of the screening effect comes from the NVR where the number of rAAA treated by UK vascular surgeons fell by around half in men outside the cohort offered screening, but by over two thirds in men aged 65 to 74 years.

The screening programme itself has been shown to be effective. Almost 3.5 million men have been invited for screening, with an uptake of around 80% - higher than almost all the adult national screening programmes. Ultrasound is a highly efficient way of diagnosing AAA with very few failures (1% to 2%). Although not assessed in this review, the surveillance programme for small and medium AAAs has previously been shown to be effective. A total of 7,506 men had detection of an AAA of 5.5cm or greater either on initial screen, or surveillance during the decade of study. A total of 6,532 underwent intervention, almost equal numbers had open aortic surgery and EVAR. Perioperative mortality rates were considerably lower for screen-detected AAA (0.2% compared with 0.5% on the NVR for EVAR, and 1.1% compared with 3.0% after open AAA repair). These results are a tremendous improvement on the risks of elective AAA treatment in the 1990s.

The question is, how much of the reduction in death from rAAA is due to screening and how much is due to changes in disease prevalence and other improvements in healthcare?

The prevalence of AAA in 65-year-old men has reduced from 4% in the MASS study (which included men over 65) to around 0.8% in 2022. It declined significantly even over the interval covered by this review: prevalence of AAA in 65-year-old men reduced from 1.2% to 0.8%. A 2018 study in Gloucestershire on AAA growth rates showed that mean aortic diameter reduced by 12% over 25 years in 65-year-old men (smokers and non-smokers). This extraordinary change is universally believed to be due to changes in smoking habits in men, but the fact that it has occurred across all men means there must be other reasons such as improved healthcare, lifestyle changes and medication including antiplatelets, statins and antihypertensive drugs.

The changes in prevalence of rAAA could be considered a triumph of public health via smoking cessation and improved healthcare. This is why many countries have held off following Sweden and the UK in adopting population screening for AAA and are considering targeted screening instead, often based on people’s smoking habits. At the very least, screening for AAA has contributed to, and accelerated, the existing decline in aortic disease. It is highly unlikely that a public health campaign alone (including smoking cessation and a cardiovascular risk strategy), could have reduced mortality from AAA to the level seen currently. There is no sign that the decline in mortality from rAAA is slowing.

Untangling how much of the reduction is due to screening and how much due to improved public health is beyond the scope of this review, and a simple statistical analysis may be confounded by the multiple variables and changes over time. The best evidence comes from the NVR, where the reduction in the number of rAAA was by approximately two thirds in the screened cohort and only a half in the non-screened cohort.

Screening for AAA in men is offered in many countries around the world, but the UK and Sweden are the only 2 countries to run national screening programmes for over a decade and invite all men in their 65th year. It is noteworthy that these countries are the only ones in the most recent VASCUNET publication (August 2024) showing sustained reductions in rAAA. Other countries have different models of AAA screening. For example, in Germany, where many people are covered by private insurance, there is reimbursement for a scan arranged by the family doctor for men aged 65 years. Similarly in the USA, men are offered AAA screening if they take up a new Medicare policy at the appropriate age.

The Swedish AAA screening programme is co-ordinated locally in their 3 health boards but data are collated nationally. In 2024, the Swedish National Board of Health and Welfare (SNBHW - Swedish: Socialstyrelsen or SoS) conducted and published the first audit of the Swedish AAA screening programme. The years 2018 to 2022 were analysed. In summary, in the 5 years analysed:

• an average of 90% of the designated population was targeted by an invitation
• the attendance rate ranged from 79% to 83%
• the prevalence rate was 0.88%
• the 5-year surgery rate of a screening-detected AAA was 18%
• a range of approximately 80% to 100% adhered to follow-up
• there were few recorded cases of rupture with negative screening (less than 2 in 10,000 screened negative)
• there was a low occurrence of ruptures among those with screening-detected AAAs; approximately 0.2% after 5 years.

These results are very similar to the AAA screening programmes in the UK and suggests the model could be applied with confidence in other countries. For example, a recent pilot study of AAA screening undertaken in Serbia showed a high overall incidence of detection of AAA at 4.8% both in men (8.2%) and in women (1.3%), with smoking being highlighted as a major risk.  

Discussion (specific)

This effectiveness review has important messages which are different for the various provider and consumer groups.

For national healthcare providers

The UK AAA screening programmes have resulted in more benefits than just the simple reduction in AAA mortality. The screening programmes:

• identified a cohort of men with a 5.5cm or greater AAA, who were referred for elective surgery who might otherwise have died from rAAA
• stimulated a reorganisation of vascular services which has reduced elective perioperative mortality
• identified a cohort of men with small and medium AAA who receive medical treatment in surveillance which should reduce the number of heart attacks and strokes
• provided detailed data to inform commissioning and monitoring of AAA pathways
• provide data on natural history of small and medium sized AAAs (surveillance data)

For local screening programme commissioners

This report has provided programme data that can be used to further improve programme effectiveness with:

• increased focus on the 20% of men who do not attend screening who have an increased risk of AAA (equality)
• increased focus on those in the more socially deprived deciles as they have a higher incidence of AAA and increased vascular risk in general
• improved efficiency of the treatment pathway which will mitigate the risk of death due to rAAA following referral
• optimization of the surveillance pathway and medical care

For men at risk of AAA and their families

This report has shown that:

• they can have confidence that the programmes will identify 65-year-old men, offer a screening appointment with a high likelihood of a result, and that most screened men can be reassured that they have a low risk of rAAA in their lifetime
• if a man does have an AAA, whether large or small, the UK AAA screening programmes will manage him effectively
• the programmes have identified areas for further improvement, which will be addressed in forthcoming years

For countries outside the UK

For countries considering commencing an AAA screening programme in men:

• the outcomes of this study pertain to countries with similar smoking habits (and thus AAA prevalence) to those in the UK; countries with similar smoking habits and documented low AAA prevalence may consider looking at targeted screening for AAA as an alternative to population screening - there is of course the disadvantage that targeted screening can never hope to find all AAAs but might be a cost-effective alternative (research shows a number of potentially viable models exist)
• any country with a high prevalence of smoking is likely to have much higher prevalence of AAA and cardiovascular disease and would be better advised to consider population screening for AAA rather than targeted screening, since cost effectiveness should be better

For the UK NSC

The UK NSC should note the results of this effectiveness review, but may also examine the following considerations on process:

• the model of planning and roll-out of the NHS AAA screening programmes could be used in the implementation of other screening programmes
• the benefits of a bespoke data collection system for a national screening programme to enable reporting are clear to see, and the inclusion of clinical outcome data is important
• harmonisation of the 4 nations into a single group working together as a 4 nations team with shared standards also has advantages for programme construction and strategy, and report generation
• collaboration with national specialist societies, such as the VSGBI, also has advantages, for example looking at clinical outcome data and establishing agreement on standards
• this model for effectiveness review could be employed to review the AAA screening programmes again in 5 to 10 years to ensure continued effectiveness
• the basic structure of this model, with multisource input and triangulation of the results, provides a mode to evaluate the effectiveness of other screening programmes

Conclusion

In the last decade, there has been an approximate halving of the number of men treated in hospital with rAAA, and a similar reduction in the death rate from rAAA in the UK. It can be concluded a substantial proportion of this reduction is due to the NHS AAA screening programmes.

NHS AAA screening programmes are effective at identifying men with AAA and referring men for treatment.

Risks of intervention for screen-detected AAA are low, and within the target range (less than 3.5% elective mortality).

NHS AAA screening programmes remain cost effective at NICE willingness to pay thresholds, despite falling prevalence in 65-year-old men. This will need reviewing every 5 to 10 years to monitor AAA prevalence.

There remain quality of life and inequality issues (small reductions in quality of life, higher prevalence of AAA in lowest socio-economic groups), which are potential targets for improvement.

Acknowledgements

Authors

Writing team

Jonothan Earnshaw, Phil Gardner, Sophie Mitra and Harriet Strachan

Cost-effectiveness analysis

Katy Saxby and Matt Bown

Quality of life section

A O’Cathain, J Hughes, E Lumley, A Nasim, G Stansby and A Elstone, on behalf of the Patient Centred AAA Screening Study (PCAAAS)

The AAA screening quality of life research is funded by the NIHR Health and Social Care Delivery Research (HSDR) programme (NIHR135031). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care (DHSC).

Collaborators

DHSC

Mike Harris, Jo Jacomelli, Rad Latinovic, Anne Mackie, John Marshall, Anne Stevenson

NHS England

Morag Armer, Alan Elstone, Chris Hutchins, Katherine Jordan, Liz Luckett, Akhtar Nasim, Gerry Stansby, Lisa Summers, Ashley Thomas

Wales AAA screening programme

Louis Fligelstone, Heather Lewis, Heather Payne, Helen Tutt

Scotland AAA screening programme

Karen Hotopp, Aoife McCarthy, Douglas Orr, John Quinn, Tasmin Sommerfield, Garrick Wagner

Northern Ireland AAA screening programme

Claire Black, Louis Lau, Christine McKee, Gemma Reid

National Vascular Registry (NVR)

David Cromwell, Amundeep Johal, Arun Pherwani, Sam Waton, Robin Williams

University of Leicester

Matt Bown, Liam Musto

Swedish AAA screening programme update

Rebecka Hultgren, Sverker Svensjo, Anders Wanhainen,

Organisation acknowledgements

NHS AAA Screening Programme; Wales AAA Screening Programme; Scottish AAA Screening Programme; Northern Ireland AAA Screening Programme; National Vascular Registry, Clinical Effectiveness Unit at Royal College of Surgeons, England; Digital Health and Care Wales; Public Health Scotland; National Records of Scotland; The Viral Statistics Unit, Northern Ireland Statistics Research Agency (NISRA); Department of Health Statistics, Northern Ireland; University of Leicester; University of Sheffield.

General acknowledgements

The working group wishes to commend the efforts of all those involved in the design and roll out of the NHS AAA screening programme. In addition, the results in this review are the culmination of the efforts of all those who work in the programme to maintain and improve its standards.

The wealth of data collected on AAA in men over the last decade by the national screening programmes, and an NVR that reports on procedures and outcomes, has immense value. The data have been integral to the development of this report and will continue to inform improvements to the programme for years to come.

Finally, it is right to acknowledge the foresight and dedication of the two pioneers of population screening for AAA – Brian Heather and Alan Scott; vascular surgeons who saw a potential benefit that has now been realised.