An abdominal aortic aneurysm (AAA) is a swelling of the main blood vessel in the body, the aorta. If an AAA gets too large it can burst (rupture) and cause fatal internal bleeding. The NHS offers AAA screening to men at the age of 65.
In 2010, 1.50% of men screened by the NHS had an AAA. This fell to 0.97% in 2019. As AAAs become less common, AAA screening costs more per person found to have an AAA. Eventually the NHS will not be able to justify spending money on AAA screening.
An alternative, more cost-effective approach is to only invite men for AAA screening if they are at high risk of having an AAA. It is not known if this approach misses many men with AAAs in the group who are not offered screening.
In this research we will analyse results from the NHS AAA Screening Programme from 2013-2022. General practice records will be obtained for men invited for screening. We will work out what would have happened if only men with risk factors for AAA had been invited for screening. We will also see if there are other details in general practice records that could be used to invite men for AAA screening in a more efficient way. We will see if AAA screening can be targeted at groups of men who are at a high risk and, if so, whether such a targeted screening programme will still find the majority of men with AAAs.
Can the effectiveness of Abdominal Aortic Aneurysm (AAA) screening be improved by targeting screening at individuals most likely to have an AAA, whilst ensuring that AAA detection rates remain acceptable to patients and the public?
Background: Screening for AAA is both clinically and economically effective. The main determinant of this effectiveness is disease prevalence. AAA prevalence is decreasing over time, steadily reducing the efficiency of the current NHS AAA Screening Programme (NAAASP) screening policy. One alternative to whole population screening is targeted screening of high-risk groups such as smokers. Whether this would detect a clinically and publicly acceptable proportion of disease, and whether it would improve cost-effectiveness is unknown.
Aim: To determine the clinical outcomes and cost-effectiveness of targeted AAA screening.
Objectives: 1) Link individual mens’ NAAASP screening records to primary care records and prepare the linked dataset for analysis; 2) Determine the primary care record risk factors for screen-detected AAA and establish targeted AAA screening criteria. This will include the development of a multivariable AAA risk prediction model for screen-detected AAA. 3) Use the linked dataset to undertake in-silico trials of targeted AAA screening including long-term clinical and economic effectiveness modelling.
Methods: To perform the in-silico trials individual mens’ outcomes from the NAAASP (2013 to 2022, ≈2,300,000 men, 1% with AAA) will be linked to primary care data from the Clinical Practice Research Datalink (CPRD) (20% of records). Risk factors for AAA will be used as targeted screening criteria in in-silico trials, with diagnostic accuracy as the primary outcome. Trial results will be used to re-parameterise a discrete event simulation model of AAA screening to estimate the long-term clinical and economic effectiveness of the targeted screening.
Outcomes are considered for the overall research project and for each specific objective:
Overall outcomes:
Diagnostic accuracy of targeted screening for AAA - The current NAAASP strategy of whole population screening will be considered as the gold standard. The historical outcomes of screening (AAA detection at screening) will be compared with the outcomes for targeted screening strategies. Men with AAA detected in the current strategy who would not be included in any particular targeted strategy will be considered to have had their diagnosis missed by the new strategy. Standard approaches to determining diagnostic accuracy (sensitivity, specificity, true/false postive/negative) will be used and reported in this outcome;
Primary care clinical risk model for screen-detected AAA - development of this risk model will be an outcome for the study, irrespective of performance;
Clinical effectiveness of targeted AAA screening - the outcomes above will be used to reparameterise a discrete event simulation model of AAA screening. This model considers clinical outcomes for individuals invited for AAA screening to a point 30 years in the future. The model estimates clinical outcomes of screening such as all-cause mortality, aneurysm-related mortality, numbers of AAA ruptures, number of AAA repairs (elective and emergency). These clinical outcomes will be used to assess clinical effectiveness;
Economic effectiveness of targeted AAA screening - the discrete event simulation model also generates economic effectiveness estimates for AAA screening. For this outcome, the base case scenario of the current AAA screening strategy will be compared with the new strategies generated in this research. This will primarily be incremental cost effectiveness ratio and cost per quality-adjusted life year;
Objective 1) Link individual mens’ NAAASP screening records to primary care records and prepare the linked dataset for analysis.
Outcomes:
Establish a linkage between CPRD and NAAASP datasets;
Obtain linked data;
Prepare data for analysis;
Objective 2) Determine the primary care record risk factors for screen-detected AAA and establish targeted AAA screening criteria. This will include the development of a multivariable AAA risk prediction model for screen-detected AAA.
Outcomes:
Identify risk factors for screen-detected AAA in primary care data;
Generate clinical risk algorithms for screen-detected AAA that combine individual risk factors;
Comparative performance of clinical risk algorithms;
Objective 3) Use the linked dataset to undertake in-silico trials of targeted AAA screening including long-term clinical and economic effectiveness modelling.
Outcomes:
Clinical effectiveness of targeted AAA screening;
Economic effectiveness of targeted AAA screening;
Matthew Bown - Chief Investigator - University of Leicester
Matthew Bown - Corresponding Applicant - University of Leicester
Aiden Smith - Collaborator - University of Leicester
Claire Lawson - Collaborator - University of Leicester
Guiqing Yao - Collaborator - University of Leicester
Liam Musto - Collaborator - University of Leicester
Sylwia Bujkiewicz - Collaborator - University of Leicester
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation