You are here

Article Text

Article menu

Download PDFPDF

Cardiovascular medicine
Original research
Impact of patient sex on selection for abdominal aortic aneurysm repair: a discrete choice experiment
  1. Anna Louise Pouncey1,
  2. Luis Enrique Loría-Rebolledo2,
  3. Linda Sharples3,
  4. Colin Bicknell1,
  5. Mandy Ryan4,
  6. Janet Powell5
  1. 1 Department of Surgery and Cancer, Imperial College London, London, UK
  2. 2 Health Economics Research Unit, University of Aberdeen, Aberdeen, UK
  3. 3 Medical Statistics, London School of Hygiene and Tropical Medicine, London, UK
  4. 4 Institute of Applied Health Sciences, University of Aberdeen, Aberdeen, UK
  5. 5 Imperial College London, London, UK
  1. Correspondence to Dr Anna Louise Pouncey; a.pouncey{at}imperial.ac.uk

Abstract

Objectives Women with an abdominal aortic aneurysm (AAA) are less likely to receive elective repair than men. This study explored the effect of patient sex and other attributes on vascular surgeons’ decision-making for infrarenal AAA repair.

Design Discrete choice experiment.

Setting Simulated environment using case scenarios with varying patient attributes.

Participants Vascular surgeons.

Interventions Surgical decision-making.

Main outcome measures AAA repair versus no repair and endovascular versus open repair.

Results 182 surgeons completed 2987 scenarios. When all other attributes were equal, a woman was more likely to be offered an AAA repair (marginal rate of substitution (MRS) 3.86 (95% CI 2.93, 4.79)), while very high anaesthetic risk (MRS −4.33 (95% CI –5.63, –3.03)) and hostile anatomy (MRS −3.28 (95% CI –4.55, –2.01)) were deterrents. Increasing age did not adversely affect the likelihood of offering repair to men but decreased the likelihood for women, which negated women’s selection advantage from the age of 83 years. Women were also more likely to be offered endovascular repair (MRS 2.57 (95% CI 1.30, 3.84)).

Conclusions Patient sex alone did not account for real-world disparity observed in selection for surgery. Rather, being a woman was associated with a higher likelihood of being offered AAA repair but also a higher likelihood of being offered less invasive endovascular repair. Increased age decreased the likelihood of surgical selection for women but not men. Preference for less invasive repair, combined with inferior rates of anatomical suitability, and the comparably older age of women at the time of AAA repair selection may account for lower rates of repair for women observed.

  • Cardiovascular Disease
  • Vascular surgery
  • Health Equity

Data availability statement

Data are available upon reasonable request.

https://creativecommons.org/licenses/by/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/licenses/by/4.0/.

https://doi.org/10.1136/bmjopen-2024-091661

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    STRENGTHS AND LIMITATIONS OF THIS STUDY

    • The discrete choice experiment (DCE) design enabled evaluation of the relative importance of different attributes (eg, patient characteristics), which influence whether a surgeon decides to offer an abdominal aortic aneurysm (AAA) repair.

    • The effect of attributes on AAA repair choices was consistent with real-world observations.

    • Due to restrictions on the number of DCE attributes, not all variables which may be considered during selection for AAA repair could be included.

    • This DCE examined surgeons’ preferences, which form only one component of the shared decision-making process.

    Introduction

    Despite a fourfold higher risk of rupture at smaller abdominal aortic aneurysm (AAA) diameters for women compared with men,1 2 previous research indicates that women with an AAA are up to 25% less likely to be offered elective surgical repair.3 4 Women are also less likely to be deemed morphologically suitable for minimally invasive endovascular aneurysm repair (EVAR), which has a lower operative mortality risk and is often tolerated by patients considered too frail to withstand the perioperative insult of open aortic repair (OAR).3 5–7

    In the UK, this disparity in surgical selection is compounded by the availability of population-based AAA screening for men alone.8 As a result, although 1/5th of AAA diagnoses and 1/3rd of AAA ruptures are in women, only 1/10th of elective repairs are undertaken in women.3 9 10 Of the women who do receive surgery, an increased risk of mortality and complications, compared with men, for both elective OAR and EVAR is observed.11–29 This sex disparity in AAA repair and outcomes may arise from differential selection for surgery. For example, surgeons may elect to delay elective AAA repair for women until a later and more complex stage of their aneurysmal disease. Alternatively, it may reflect differences in comorbid status or anatomical features, which are associated with increased technical difficulty and adverse outcomes.3 7 30 31 While the NICE (National Institute for Health and Care Excellence) guidelines recommend the same diameter threshold for repair in men and women, other clinical practice guidelines recommend a lower threshold or acknowledge that the appropriate threshold for women is unknown.32 33

    To understand why women are less likely to receive elective AAA repair, it is essential to examine whether and how sex or sex-specific patient attributes influence surgical selection. The surgeon’s opinion is a crucial part of the clinical decision-making process, which ideally includes input from a multidisciplinary team and shared decision-making with the patient.

    This study employed a discrete choice experiment to investigate the factors influencing vascular surgeons’ treatment preferences for patients with AAA. Specifically, we examined two key decisions: (1) whether to recommend AAA repair or decline intervention (eg, ‘turn-down’) and (2) whether to opt for EVAR or open repair. The study aimed to explore whether patient sex impacts surgeons’ decision-making and to identify other factors that significantly influence their choices.

    Methods

    A discrete choice experiment (DCE) is a choice-based survey which quantifies preferences for different options, such as treatment strategies. In these experiments, participants are presented with a series of clinical scenarios (known as choice sets), each defined by several attributes with specific levels (eg, age and sex).34 When DCE participants select the option they believe is best, their decision reflects how much they value the different features (or attributes) of that option.35–37 It relies on the assumption that people make rational choices, and when they do not, it is due to random factors. DCEs are used to produce a measure of benefit (or utility) for each attribute and level, helping to assess their relative importance and calculate the likelihood of different choices.38

    Although the DCE method originated in economics,35–37 they are now increasingly used in healthcare to understand clinical decision-making.39 40 Relevant guidance from the ISPOR (International Society for Pharmacoeconomics and Outcomes Research) Conjoint Analysis Experimental Design Good Research Practices Task Force was used for this study.41 42 An extended description of methods and the distributed questionnaire are presented in online supplemental material, and a schematic of the DCE development process is available in online supplemental figure S1.

    Supplemental material

    Supplemental material

    Development of DCE design

    To evaluate vascular surgeons’ decisions regarding (1) AAA repair and (2) selection for EVAR, a two-tier DCE was designed. The experiment presented a series of case scenarios featuring patients with AAA with varying attributes. Surgeons were asked to make two forced choices for each scenario: (1) AAA repair decision: Would you offer this patient AAA repair? (Yes/No) and (2) repair method decision: assuming this patient has been deemed suitable for repair, and could have an infrarenal clamp, would you prefer to offer endovascular rather than open repair? (Yes/No).

    Selection of attributes and levels

    The role of sex in AAA repair decision-making cannot be studied in isolation, as other factors influencing these decisions may also vary between women and men. Therefore, additional relevant attributes needed to be selected for the DCE. A multistage approach was used, starting with a scoping review of the literature to identify factors commonly used for AAA repair risk assessment and a review of current EVAR anatomical ‘Instructions For Use’ (IFU) criteria.33 43–65 Next, interviews were conducted with consultant vascular surgeons,66 and qualitative analysis was performed to identify and prioritise attributes.67 This analysis followed the six-step process outlined by Clarke and Braun, using NVivo software (V.1.6.1, QSR International).68–72 Finally, a panel of six vascular surgeons reviewed the results to achieve consensus on attribute selection (defined as >75% agreement)73 74 (see online supplemental figure S1–S4).

    Questionnaire design, pilot testing and distribution

    A fractional factorial design with 16 scenarios was developed to create a choice set from which preferences could be estimated for all possible scenarios.75 The order of the scenarios was randomised to reduce the effect of participant fatigue. The design also included one additional scenario with clear dominance, to assess choice desirability, and one additional repeat scenario, to assess choice consistency.76 Participants provided informed consent and were blinded to the study’s primary objective. Additional data collected from participants included surgeon gender, level of experience (early: <10 years, mid-level: 10–20 years or experienced: >20 years), proportion of EVAR in clinical practice (categorised into 1st ‘low’, 2/3rd ‘mid-range’ and 4th ‘high’ quartiles) and geographical location (categorised as UK, Europe, North America or other).77–80 The survey was delivered using the Qualtrics XM Platform (Qualtrics, Provo, Utah, USA).

    A pilot was then conducted with 18 vascular surgeons to assess whether the volume of scenarios was appropriate, assess respondent efficiency and gauge the perceived realism of the design.81 Pilot test results were also used to determine the minimum sample size requirement: 9 respondents completing 16 choice scenarios were needed to assess the effect of patient sex on selection for AAA repair and 140 respondents were needed for EVAR (α=0.05, β=0.2).35 Therefore, a recruitment target of 168 respondents was chosen, accounting for a 20% rate of attrition. The final selected attributes and levels are presented in table 1. Attributes for the Pilot test and full DCE are also presented within online supplemental table S1.

    View this table:
    Table 1

    Final selected attributes and levels for the discrete choice experiment

    A multifaceted recruitment approach was employed, including international conferences, distribution via social media and collaboration through research networks. Participants were eligible if they were a consultant vascular surgeon with experience of both open and endovascular aortic repair and a good understanding of English. Prespecified withdrawal criteria included an incomplete survey response (eg, failure to consent or answering fewer than 50% of questions) and lack of clinical equipoise (eg, selecting the same alternative for all questions).

    Econometric analysis

    Separate models were constructed for each outcome: (1) the likelihood of selecting AAA repair versus turndown and (2) the likelihood of selecting EVAR versus open repair. Each model included all relevant exposure variables: patient sex, age, aneurysm size, anatomical suitability, anaesthetic risk and patient perspective. The following utility/benefit function was estimated for each analysis using an error components model.

    Vsurgical_repair/endovascular_repair = ASC0 + ß1female_sex + ß2age + ß3anaes_mod + ß4anaes_high + ß5anaes_vhigh + ß6patient_anxiety + ß7AAA_diameter + ß8anatomy_neck + ß9anatomy_access + ß10anatomy_neck+access

    Vturndown/open_repair = ASC1

    Within this utility function, V represents the systematic utility—the predictable, measurable component of a decision based on known factors. The reference levels are (1) ‘turndown’ for analysis and (2) ‘open repair’ for analysis. The error components model was chosen to relax the irrelevance of independent alternatives assumption, classifying alternative specific constants as random: ASC0 is set to 0, while ASC1 , which captures the unexplained variation in choice for the reference levels, is modelled as a normally distributed random parameter and estimated using 500 Modified Latin Hypercube Sampling interindividual draws.82

    Within the function, ß1-10 represents specific parameter coefficients associated with the defined attribute levels. Because utility is on a relative scale, the coefficients are interpreted in relation to each other rather than in absolute terms. A positive value indicates that the attribute level combination was preferred relative to the reference level, and a larger value indicates a stronger preference. To further aid interpretation, marginal rates of substitution (MRS), which enable assessment of trade-off between attributes, were calculated using AAA size (mm) as the denominator and the delta method for SE estimation.83 For instance, if a surgeon is deciding whether to offer a procedure based on a patient’s age and the size of their aneurysm, the MRS would indicate how much of an increase in AAA size would be needed to offset an increase in the patient’s age to keep the likelihood of offering the procedure the same. The mean percentage difference in the probability of selection for women and men was also calculated for a variety of scenarios.

    All analyses were conducted in RStudio, using the Apollo R package.84 85 To improve model performance and aid interpretation, continuous data were centred and scaled, using the median value for age (70 years) and the threshold value (55 mm) for AAA size. Parameter estimates were presented with robust SEs, and a two-sided p value of <0.05 was considered significant. For each error components model, interactions between patient sex and other attributes were also introduced. The performance of models with interaction terms was compared using likelihood ratio, χ2 statistic and Akaike information criterion, and goodness of fit was estimated using adjusted pseudo R2. The assessment of final model performance was conducted using share prediction tests and fits tests, which assessed how the model’s predictions align with the actual observed data. Prespecified subgroup analyses were also conducted to assess preference heterogeneity associated with surgeon characteristics (eg, surgeon gender, level of experience, preference for EVAR repair and geographical location) reporting the proportion of observed choices compared with model predictions.

    Consistency, validity and sensitivity analyses

    Sensitivity analyses were used to explore the effect of forced choice on evaluation of repair type selection by repeating the analysis for the subgroup of cases where ‘repair’ was selected in question A. The quality of choice data was also assessed using the following criteria: Choice desirability—a scenario where all patient-related features were set to their best creating a dominant choice, Choice stability—a repeated scenario to assess for consistency in responses and Clinical Equipoise—whether a participant selected the same choice for more than 90% of questions. Survey response time was used as an indicator of participant engagement. The stability of findings was evaluated through subgroup analyses excluding respondents who failed consistency and validity assessments.

    Results

    A total of 182 respondents completed the DCE. AAA repair was chosen in 64.93% (1881/2897) and EVAR in 62.35% (1772/2842) of cases. 42 (23.08%) respondents were women, 70 (38.46%) had <10 years of experience, 69 (37.91%) had 10–20 years of experience and 36 (19.78%) had >20 years of experience. The median percentage of EVAR performed was 70 (IQR: 40–100). Geographically, 60 (32.97%) practiced in the UK, 48 (26.37%) in Europe, 45 (24.73%) in North America and others practicing in Latin America, the Middle East, South Africa and Southeast Asia (see online supplemental table S2).

    Selection for AAA repair

    The error components model for selection for AAA repair (vs ‘turn-down’) is demonstrated in table 2. A woman patient and an increase in aneurysm size increased the likelihood of offering an AAA repair, while an increase in age, very high anaesthetic risk and hostile neck and access anatomy decreased the likelihood of offering an AAA repair. Patient-reported anxiety did not contribute to the surgeon’s decision-making. MRS in relation to AAA size indicated that being a woman had the greatest effect on increasing likelihood of repair selection (MRS 3.86 (95% CI 2.93, 4.79)), equivalent to a 3.86 mm increase in AAA size compared with men. Very high anaesthetic risk (compared with low anaesthetic risk) was the greatest deterrent (MRS −4.33 (95% CI –5.63, –3.03)), followed by the presence of hostile neck and access anatomy (compared with IFU anatomy, MRS −3.28 (95% CI −4.55, –2.01)). Increasing age was also associated with increased reluctance to offer repair (MRS −0.19 (95% CI −0.22, –0.15, per year)) (see figure 1).

    View this table:
    Table 2

    Error component model for AAA repair versus no repair

    Figure 1
    Figure 1

    Marginal rates of substitution to enable evaluation of trade-off between attributes in relation to abdominal aortic aneurysm (AAA) size (mm) for selection for AAA repair (vs no repair). Blue=positive effect on the likelihood of selection for AAA repair, red=negative effect on the likelihood of selection for AAA repair, grey=no significant effect.

    Selection for EVAR versus open repair

    The error components model for selection for AAA repair method decision (EVAR vs open) is demonstrated in table 3. A woman patient, increased age, AAA size and anaesthetic risk were associated with an increased likelihood of selection for EVAR. Hostile neck and access anatomy and patient-reported anxiety decreased the likelihood of selection for EVAR. MRS in relation to AAA size (mm) indicated that very high anaesthetic risk, compared with low anaesthetic risk, had the greatest effect on the likelihood of offering EVAR rather than open repair (MRS 16.31 (95% CI 12.74, 19.89)). A woman patient (MRS 2.57 (95% CI 1.30, 3.84)) and increasing age also increased the likelihood of offering EVAR (MRS 0.45 (95% CI 0.34, 0.56), per year), while hostile access and neck anatomy decreased the likelihood of offering EVAR (MRS −9.79 (95% CI –12.43, –7.14)) (see figure 2).

    View this table:
    Table 3

    Error component model for EVAR versus open repair

    Figure 2
    Figure 2

    Marginal rates of substitution to enable evaluation of trade-off between attributes in relation to abdominal aortic aneurysm (AAA) size (mm) for selection for endovascular aneurysm repair (vs open repair). Blue=positive effect on the likelihood of selection for AAA repair, red=negative effect on the likelihood of selection for AAA repair, grey=no significant effect.

    Attribute sex interactions

    A significant interaction was observed between age and sex for selection for AAA repair. This model, demonstrated in table 4, performed better than the model without interactions. Within this age-sex interaction model, age had no significant effect on selection for AAA repair for men but was associated with a reduction in the likelihood of AAA repair for women, such that, all else being equal, the positive selection effect associated with being a woman was negated from the age of 83 years (see figure 3.) No significant interaction between sex and other attributes was observed to influence selection for EVAR.

    View this table:
    Table 4

    Error component model for AAA repair versus no repair, with patient sex-age interaction term

    Figure 3
    Figure 3

    Percentage difference in the probability of selection for abdominal aortic aneurysm (AAA) for men and women using an error component model with an age-sex interaction term (see table 4). Blue=positive effect on the likelihood of selection for AAA repair, red=negative effect on the likelihood of selection for AAA repair, grey=no significant effect.

    Choice probabilities

    When all other attributes were equal, women were significantly more likely to be offered AAA repair (percentage difference (%) 25.40 (95% CI 3.84, 46.96)); however, adjusting for real-world differences such as an increase in age of 4 years for women (21.50 (95% CI −1.44, 44.45)) or for hostile anatomy (4.24 (95% CI −35.45, 43.93)), the selection advantage for women was no longer significant. Moreover, adjusting for high anaesthetic risk (−3.25 (95% CI −42.93, 36.44)) or for an aneurysm 5 mm smaller in diameter (−8.18 (95% CI −37.31, 20.96)), women were less likely to be offered repair. A woman who was 4 years older, with hostile anatomy, high anaesthetic risk and a smaller aneurysm size (−5 mm) was 43.88% (95% CI –99.83, 12.08) less likely to be offered repair, compared with a man, with low anaesthetic risk, and anatomy within IFU (see figure 4).

    Figure 4
    Figure 4

    Percentage difference in probability of being offered abdominal aortic aneurysm (AAA) repair for women patients compared with men, using a main effects error component model (see table 3). Error bars=95% CI. Blue=positive effect on the likelihood of selection for AAA repair, red=negative effect on the likelihood of selection for AAA repair, grey=no significant effect.

    Respondent heterogeneity

    No significant differences in selection for AAA repair, selection for EVAR or attribute importance were observed for men or women surgeons (132 men and 42 women). Early career surgeons offered AAA repair more frequently ((n=70) stated choice/model prediction (c/p): 777/725, p<0.001) than mid-level surgeons ((n=69) c/p: 646/710, p<0.001). Early career surgeons were more likely to offer EVAR to older patients than experienced surgeons (MRS age: early 0.62 (95% CI 0.44, 0.81), mid-level 0.40 (95% CI 0.24, 0.56) and experienced 0.21 (95% CI 0.02, 0.39)). No significant differences in the relative importance of other attributes were identified.

    Surgeons in the lowest quartile for use of EVAR (n=31) were less likely to offer an AAA repair (c/p: 301/319, p<0.036) and less likely to offer EVAR (EVAR c/p: 236/313, p<0.001). Surgeons in the highest quartile for use of EVAR (n=54) were more likely to offer an AAA repair (c/p: 582/559, p<0.041) and more likely to offer EVAR (EVAR c/p: 536/632, p<0.001). Surgeons operating in the UK were less likely to offer an AAA repair ((n=60) c/p: 547/620, p<0.036) than surgeons in North America ((n=45) c/p: 514/466 p<0.001). UK surgeons were also less likely to offer EVAR (EVAR c/p: 521/599, p<0.001) than surgeons in North America (EVAR c/p: 495/452, p<0.001) (see online supplemental tables S3–S10).

    Assessment of consistency and validity

    On assessment of choice desirability, 181/182 (99.45%) respondents chose to offer AAA repair for the dominant choice scenario. Most respondents demonstrated a degree of clinical equipoise: 6/182 (3.3%) offered AAA repair and 7/182 (3.9%) offered EVAR for over 90% of cases. Choice stability was superior for AAA repair for which 15/182 (8.24%) were inconsistent compared with EVAR for which 34/182 (18.68%) inconsistent. On subgroup analyses for those who met choice stability, choice desirability or equipoise or all three criteria, no significant difference in the direction of attribute effect or MRS was observed (see online supplemental tables 11 and 12).

    Discussion

    This DCE investigated the effect of patient sex and additional factors on the likelihood of selection for AAA repair. Findings suggest that patient sex alone does not explain the observed disparity in clinician’s decision-making for AAA repair. Rather, when all other attributes were equal, a woman patient was more likely to be offered an AAA repair. However, on interaction analyses, while increased age had no significant effect for men, it reduced the likelihood of surgical selection for women. This negated the positive association with AAA repair selection for women from the age of 83 years. This finding has relevance as women are on average 4 years older than men at the time of AAA repair, and in hospital episode statistics data are reported to have an increased turndown rate at older ages, compared with men.86 87

    Women were also more likely to be selected for less invasive EVAR. Preference for a less invasive approach, when combined with reduced anatomical suitability among women compared with men (34% vs 54%; OR 0·44 (95% CI 0.32, 0.62)), may therefore contribute to the lower rates of surgical selection for women reported.3 Indeed, the presence of hostile neck and access anatomy, when compared with IFU anatomy, was associated with lower likelihood of selection for both AAA repair and for EVAR and is demonstrated to be more common for women, even among patients selected for repair.3 88 Further, patient-related factors which were associated with reduced likelihood of selection for AAA repair and, conversely, increased likelihood of selection for less-invasive EVAR were age and increased anaesthetic risk. Of note, although women are often cited to have less comorbidity, they are more likely to have respiratory disease, which could translate into increased anaesthetic risk.86

    Directions of attribute effect and heterogeneity in choice preferences were consistent with real-world observations. However, several limitations of DCEs to the healthcare setting have been cited, including normative issues, cognitive burden and applicability in different healthcare settings.89 Constraints on the number and complexity of DCE attributes meant that not all variables, such as the results of physiological fitness testing, could be included.90 Therefore, factors which have not been explicitly described may also influence selection for AAA repair. This analysis was restricted to infra-renal AAA, whereas in clinical practice, patients with hostile aortic neck anatomy may be considered for more complex repair options.91 Interpretation of findings is also limited by a lack of comprehensive real-world data regarding comorbid status for those not offered repair.92 The recruitment strategy may be biased towards opinions of those who engage in research and quality improvement. Additionally, variation in resources and standard practice may influence how respondents interpreted DCE attributes. Although study hypotheses were not disclosed within the survey, on-going research efforts by the team may have alerted respondents to the purpose of the study, potentially introducing bias towards offering AAA repair for women.

    This DCE design examined surgeon’s preference, which forms only one part of the shared decision-making (SDM) process, and it is important to recognise that final treatment strategy also depends on patient and multidisciplinary opinions.93 The impact of SDM was examined in the OVIDIUS trial (Operative Vascular Intervention Decision-Making Improvement Using SDM tools) which included 73 men and 14 women with AAA. It found that patients were more likely to choose non-operative treatment (28.8% vs 21.4%) when SDM processes were used.94 However, whether and how priorities differ between women and men in SDM has been underreported. A recently published work to evaluate the effect of a decision aid on agreement between patient preference and AAA repair type included 234 men and only 1 woman.95 Further previous work, which sought to elicit AAA patient preferences, included only 5 women. However, 4 (80%) of these women expressed a preference for less invasive EVAR.96

    Given the lack of prognostic data for women with AAA, surgeons may rely on sex stereotypes or personal experience to make operative decisions. A similar tendency towards less invasive treatment has been reported in cardiology, where women are less likely to receive complete revascularisation and considerably less likely to receive a coronary artery bypass grafting.97 98 Due to the lack of AAA screening for women, their AAAs are more often detected incidentally and their risk factors are less likely to be controlled. Combined with increased elective mortality rates, this may lead vascular surgeons to favour less invasive options.86 99 Additionally, the sex-frailty paradox—where women more often meet frailty criteria but experience lower mortality rates than men—may further increase preference for less invasive repair.100 However, these tendencies may also arise from a form of ‘benevolent sexism’ or ‘protective paternalism’, where a desire to protect women inadvertently results in a lower standard of care.101 102

    Conclusion

    Patient sex alone did not account for real-world disparity observed in selection for surgery. Rather, being a woman was associated with a higher likelihood of being offered AAA repair but also a higher likelihood of being offered less invasive endovascular repair. Increased age decreased the likelihood of surgical selection for women but not men. Preference for less invasive repair, combined with inferior rates of anatomical suitability, and the comparably older age of women at the time of AAA repair selection may account for lower rates of repair for women observed.

    Data availability statement

    Data are available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    Study Investigators obtained a favourable opinion from Imperial College Research Ethics Committee (ICREC 21IC6875, IRAS number 293054) and Imperial College London acted as the sponsor. Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    The authors thank all volunteers who participated in this study and the study contributors.

    References

    1. ↵
      1. The RESCAN Collaborators
      . Surveillance Intervals for Small Abdominal Aortic Aneurysms. JAMA 2013;309:806. doi:10.1001/jama.2013.950
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      2. Sweeting MJ ,
      3. Thompson SG ,
      4. Brown LC , et al
      . Meta-analysis of individual patient data to examine factors affecting growth and rupture of small abdominal aortic aneurysms. Br J Surg 2012;99:655–65. doi:10.1002/bjs.8707
      OpenUrlCrossRefPubMed
    3. ↵
      2. Ulug P ,
      3. Sweeting MJ ,
      4. von Allmen RS , et al
      . Morphological suitability for endovascular repair, non-intervention rates, and operative mortality in women and men assessed for intact abdominal aortic aneurysm repair: systematic reviews with meta-analysis. Lancet 2017;389:2482–91. doi:10.1016/S0140-6736(17)30639-6
      OpenUrl
    4. ↵
      2. McGinigle KL ,
      3. Browder SE ,
      4. Strassle PD , et al
      . Sex-related disparities in intervention rates and type of intervention in patients with aortic and peripheral arterial diseases in the National Inpatient Sample Database. J Vasc Surg 2021;73:2081–9. doi:10.1016/j.jvs.2020.11.034
      OpenUrl
    5. ↵
      2. Mofidi R ,
      3. Goldie VJ ,
      4. Kelman J , et al
      . Influence of sex on expansion rate of abdominal aortic aneurysms. British Journal of Surgery 2007;94:310–4. doi:10.1002/bjs.5573
      OpenUrlCrossRefPubMed
    6. ↵
      2. Lo RC ,
      3. Lu B ,
      4. Fokkema MTM , et al
      . Relative importance of aneurysm diameter and body size for predicting abdominal aortic aneurysm rupture in men and women. J Vasc Surg 2014;59:1209–16. doi:10.1016/j.jvs.2013.10.104
      OpenUrlCrossRefPubMed
    7. ↵
      2. Ayo D ,
      3. Blumberg SN ,
      4. Gaing B , et al
      . Gender Differences in Aortic Neck Morphology in Patients with Abdominal Aortic Aneurysms Undergoing Elective Endovascular Aneurysm Repair. Ann Vasc Surg 2016;30:100–4. doi:10.1016/j.avsg.2015.09.002
      OpenUrl
    8. ↵
      2. Sweeting MJ ,
      3. Masconi KL ,
      4. Jones E , et al
      . Analysis of clinical benefit, harms, and cost-effectiveness of screening women for abdominal aortic aneurysm. Lancet 2018;392:487–95. doi:10.1016/S0140-6736(18)31222-4
      OpenUrl
    9. ↵
      2. Scott SWM ,
      3. Batchelder AJ ,
      4. Kirkbride D , et al
      . Late Survival in Nonoperated Patients with Infrarenal Abdominal Aortic Aneurysm. Eur J Vasc Endovasc Surg 2016;52:444–9. doi:10.1016/j.ejvs.2016.05.008
      OpenUrl
    10. ↵
      2. Waton S ,
      3. Johal A ,
      4. Heikkila K , et al
      . National vascular registry: 2015 annual report. London, 2015.
    11. ↵
      2. Powell JT ,
      3. Sweeting MJ ,
      4. Ulug P , et al
      . Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparing outcomes of endovascular or open repair for abdominal aortic aneurysm over 5 years. Br J Surg 2017;104:166–78. doi:10.1002/bjs.10430
      OpenUrlPubMed
    12. ↵
      2. Sidloff DA ,
      3. Saratzis A ,
      4. Sweeting MJ , et al
      . Sex differences in mortality after abdominal aortic aneurysm repair in the UK. Br J Surg 2017;104:1656–64. doi:10.1002/bjs.10600
      OpenUrl
    13. ↵
      2. Trenner M ,
      3. Kuehnl A ,
      4. Reutersberg B , et al
      . Nationwide analysis of risk factors for in-hospital mortality in patients undergoing abdominal aortic aneurysm repair. Br J Surg 2018;105:379–87. doi:10.1002/bjs.10714
      OpenUrl
    14. ↵
      2. Trenner M ,
      3. Salvermoser M ,
      4. Busch A , et al
      . Effect Modification of Sex and Age for the Hospital Volume-Outcome Relationship in Abdominal Aortic Aneurysm Treatment: Secondary Data Analysis of the Nationwide German Diagnosis Related Groups Statistics From 2005 to 2014. J Am Heart Assoc 2020;9:e014534. doi:10.1161/JAHA.119.014534
    15. ↵
      2. Lo RC ,
      3. Bensley RP ,
      4. Hamdan AD , et al
      . Gender differences in abdominal aortic aneurysm presentation, repair, and mortality in the Vascular Study Group of New England. J Vasc Surg 2013;57:1261–8. doi:10.1016/j.jvs.2012.11.039
      OpenUrlCrossRefPubMed
    16. ↵
      2. Indrakusuma R ,
      3. Jalalzadeh H ,
      4. Vahl AC , et al
      . Editor’s Choice – Sex Related Differences in Peri-operative Mortality after Elective Repair of an Asymptomatic Abdominal Aortic Aneurysm in the Netherlands: a Retrospective Analysis of 2013 to 2018. Eur J Vasc Endovasc Surg 2019;58:813–20. doi:10.1016/j.ejvs.2019.05.017
      OpenUrl
    17. ↵
      2. Locham S ,
      3. Shaaban A ,
      4. Wang L , et al
      . Impact of Gender on Outcomes Following Abdominal Aortic Aneurysm Repair. Vasc Endovascular Surg 2019;53:636–43. doi:10.1177/1538574419868040
      OpenUrl
    18. ↵
      2. Barbey SM ,
      3. Scali ST ,
      4. Kubilis P , et al
      . Interaction between frailty and sex on mortality after elective abdominal aortic aneurysm repair. J Vasc Surg 2019;70:1831–43. doi:10.1016/j.jvs.2019.01.086
      OpenUrl
    19. ↵
      2. de Guerre LEVM ,
      3. Varkevisser RRB ,
      4. Swerdlow NJ , et al
      . Sex differences in perioperative outcomes after complex abdominal aortic aneurysm repair. J Vasc Surg 2020;71:374–81. doi:10.1016/j.jvs.2019.04.479
      OpenUrl
    20. ↵
      2. Chung C ,
      3. Tadros R ,
      4. Torres M , et al
      . Evolution of gender-related differences in outcomes from two decades of endovascular aneurysm repair. J Vasc Surg 2015;61:843–52. doi:10.1016/j.jvs.2014.11.006
      OpenUrlCrossRefPubMed
    21. ↵
      2. Dubois L ,
      3. Novick TV ,
      4. Harris JR , et al
      . Outcomes after endovascular abdominal aortic aneurysm repair are equivalent between genders despite anatomic differences in women. J Vasc Surg 2013;57:382–9. doi:10.1016/j.jvs.2012.09.075
      OpenUrlCrossRefPubMed
    22. ↵
      2. Schermerhorn ML ,
      3. Bensley RP ,
      4. Giles KA , et al
      . Changes in Abdominal Aortic Aneurysm Rupture and Short Term Mortality 1995–2008. Ann Surg 2012;256:651–8. doi:10.1097/SLA.0b013e31826b4f91
      OpenUrlCrossRefPubMed
    23. ↵
      2. Mehta M ,
      3. Byrne WJ ,
      4. Robinson H , et al
      . Women derive less benefit from elective endovascular aneurysm repair than men. J Vasc Surg 2012;55:906–13. doi:10.1016/j.jvs.2011.11.047
      OpenUrlCrossRefPubMed
    24. ↵
      2. Mani K ,
      3. Lees T ,
      4. Beiles B , et al
      . Treatment of abdominal aortic aneurysm in nine countries 2005-2009: a vascunet report. Eur J Vasc Endovasc Surg 2011;42:598–607. doi:10.1016/j.ejvs.2011.06.043
      OpenUrlCrossRefPubMed
    25. ↵
      2. Ramanan B ,
      3. Gupta PK ,
      4. Sundaram A , et al
      . Development of a risk index for prediction of mortality after open aortic aneurysm repair. J Vasc Surg 2013;58:871–8. doi:10.1016/j.jvs.2013.03.024
      OpenUrlCrossRefPubMed
    26. ↵
      2. Deery SE ,
      3. Soden PA ,
      4. Zettervall SL , et al
      . Sex differences in mortality and morbidity following repair of intact abdominal aortic aneurysms. J Vasc Surg 2017;65:1006–13. doi:10.1016/j.jvs.2016.08.100
      OpenUrl
    27. ↵
      2. Laine MT ,
      3. Laukontaus SJ ,
      4. Sund R , et al
      . A Population-Based Study of Abdominal Aortic Aneurysm Treatment in Finland 2000 to 2014. Circulation 2017;136:1726–34. doi:10.1161/CIRCULATIONAHA.117.028259
      OpenUrlAbstract/FREE Full Text
    28. ↵
      2. Nevidomskyte D ,
      3. Shalhub S ,
      4. Singh N , et al
      . Influence of Gender on Abdominal Aortic Aneurysm Repair in the Community. Ann Vasc Surg 2017;39:128–36. doi:10.1016/j.avsg.2016.06.012
      OpenUrl
    29. ↵
      2. Abedi NN ,
      3. Davenport DL ,
      4. Xenos E , et al
      . Gender and 30-day outcome in patients undergoing endovascular aneurysm repair (EVAR): an analysis using the ACS NSQIP dataset. J Vasc Surg 2009;50:486–91. doi:10.1016/j.jvs.2009.04.047
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      2. Ahanchi SS ,
      3. Carroll M ,
      4. Almaroof B , et al
      . Anatomic severity grading score predicts technical difficulty, early outcomes, and hospital resource utilization of endovascular aortic aneurysm repair. J Vasc Surg 2011;54:1266–72. doi:10.1016/j.jvs.2011.05.019
      OpenUrlCrossRefPubMed
    31. ↵
      2. Rudarakanchana N ,
      3. Reeves BC ,
      4. Bicknell CD , et al
      . Editor’s Choice – Treatment Decisions for Descending Thoracic Aneurysm: Preferences for Thoracic Endovascular Aneurysm Repair or Surveillance in a Discrete Choice Experiment. Eur J Vasc Endovasc Surg 2014;48:13–22. doi:10.1016/j.ejvs.2014.03.015
      OpenUrlCrossRefPubMed
    32. ↵
      Abdominal aortic aneurysm: diagnosis and management [NG156]. NICE National Institute for Health and Care Excellence; 2020.49.
    33. ↵
      2. Wanhainen A ,
      3. Verzini F ,
      4. Van Herzeele I , et al
      . Editor’s Choice – European Society for Vascular Surgery (ESVS) 2019 Clinical Practice Guidelines on the Management of Abdominal Aorto-iliac Artery Aneurysms. Eur J Vasc Endovasc Surg 2019;57:8–93. doi:10.1016/j.ejvs.2018.09.020
      OpenUrlCrossRefPubMed
    34. ↵
      2. Bridges JFP ,
      3. Hauber AB ,
      4. Marshall D , et al
      . Conjoint Analysis Applications in Health—a Checklist: A Report of the ISPOR Good Research Practices for Conjoint Analysis Task Force. Value Health 2011;14:403–13. doi:10.1016/j.jval.2010.11.013
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      2. de Bekker-Grob EW ,
      3. Donkers B ,
      4. Jonker MF , et al
      . Sample Size Requirements for Discrete-Choice Experiments in Healthcare: a Practical Guide. Patient 2015;8:373–84. doi:10.1007/s40271-015-0118-z
      OpenUrlCrossRefPubMed
    36. ↵
      2. Lancaster KJ
      . A New Approach to Consumer Theory. Journal of Political Economy 1966;74:132–57. doi:10.1086/259131
      OpenUrlCrossRefWeb of Science
    37. ↵
      2. Ryan M ,
      3. Gerard K ,
      4. Amaya-Amaya M
      , eds. Using Discrete Choice Experiments to Value Health and Health Care. Dordrecht: Springer Netherlands, 2008.
    38. ↵
      2. Lancsar E ,
      3. Louviere J
      . Experiments to Inform Healthcare Decision Making A User’s Guide. Pharmacoeconomics 2008;26:661–77. doi:10.2165/00019053-200826080-00004
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      2. Nathan H ,
      3. Bridges JFP ,
      4. Schulick RD , et al
      . Understanding surgical decision making in early hepatocellular carcinoma. J Clin Oncol 2011;29:619–25. doi:10.1200/JCO.2010.30.8650
      OpenUrlAbstract/FREE Full Text
    40. ↵
      2. Bassford CR ,
      3. Krucien N ,
      4. Ryan M , et al
      . U.K. Intensivists’ Preferences for Patient Admission to ICU: Evidence From a Choice Experiment. Crit Care Med 2019;47:1522–30. doi:10.1097/CCM.0000000000003903
      OpenUrlCrossRefPubMed
    41. ↵
      2. Bridges JFP ,
      3. Hauber AB ,
      4. Marshall D , et al
      . Conjoint analysis applications in health--a checklist: a report of the ISPOR Good Research Practices for Conjoint Analysis Task Force. Value Health 2011;14:403–13. doi:10.1016/j.jval.2010.11.013
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      2. Lancsar E ,
      3. Louviere J
      . Conducting discrete choice experiments to inform healthcare decision making: a user’s guide. Pharmacoeconomics 2008;26:661–77. doi:10.2165/00019053-200826080-00004
      OpenUrlCrossRefPubMedWeb of Science
    43. ↵
      2. Potgeiter R ,
      3. Hindley H ,
      4. Mitchell D , et al
      . Delivering a national quality improvement programme for patients with abdominal aortic aneurysms. 2012.
    44. ↵
      Abdominal aortic aneurysm: diagnosis and management. Evidence review T: Effectiveness of endovascular aneurysm repair compared with open surgical repair of ruptured abdominal aortic aneurysms. NICE; 2020.
    45. ↵
      2. Khashram M ,
      3. Williman JA ,
      4. Hider PN , et al
      . Systematic Review and Meta-analysis of Factors Influencing Survival Following Abdominal Aortic Aneurysm Repair. Eur J Vasc Endovasc Surg 2016;51:203–15. doi:10.1016/j.ejvs.2015.09.007
      OpenUrlCrossRefPubMed
    46. ↵
      2. Sutton R ,
      3. Bann S ,
      4. Brooks M , et al
      . The Surgical Risk Scale as an improved tool for risk-adjusted analysis in comparative surgical audit. Br J Surg 2002;89:763–8. doi:10.1046/j.1365-2168.2002.02080.x
      OpenUrlCrossRefPubMedWeb of Science
    47. ↵
      2. Bertges DJ ,
      3. Goodney PP ,
      4. Zhao Y , et al
      . The Vascular Study Group of New England Cardiac Risk Index (VSG-CRI) predicts cardiac complications more accurately than the Revised Cardiac Risk Index in vascular surgery patients. J Vasc Surg 2010;52:674–83, . doi:10.1016/j.jvs.2010.03.031
      OpenUrlCrossRefPubMedWeb of Science
    48. ↵
      2. Braun K ,
      3. Brunkwall J ,
      4. Gawenda M
      . Risikoeinschätzung in der Aortenchirurgie - Evaluation des SVS / AAVS Comorbidity Severity Score. Zentralbl Chir 2007;132:477–84. doi:10.1055/s-2007-981358
      OpenUrlPubMed
    49. ↵
      2. Karthikesalingam A ,
      3. Vidal-Diez A ,
      4. De Bruin JL , et al
      . International validation of a risk score for complications and reinterventions after endovascular aneurysm repair. Br J Surg 2015;102:509–15. doi:10.1002/bjs.9758
      OpenUrl
    50. ↵
      2. Patterson BO ,
      3. Holt PJE ,
      4. Hinchliffe R , et al
      . Predicting risk in elective abdominal aortic aneurysm repair: a systematic review of current evidence. Eur J Vasc Endovasc Surg 2008;36:637–45. doi:10.1016/j.ejvs.2008.08.016
      OpenUrlCrossRefPubMed
    51. ↵
      2. Barnes M ,
      3. Boult M ,
      4. Maddern G , et al
      . A Model to Predict Outcomes for Endovascular Aneurysm Repair Using Preoperative Variables. Eur J Vasc Endovasc Surg 2008;35:571–9. doi:10.1016/j.ejvs.2007.12.003
      OpenUrlCrossRefPubMed
    52. ↵
      2. Kertai MD ,
      3. Boersma E ,
      4. Klein J , et al
      . Optimizing the prediction of perioperative mortality in vascular surgery by using a customized probability model. Arch Intern Med 2005;165:898–904. doi:10.1001/archinte.165.8.898
      OpenUrlCrossRefPubMedWeb of Science
    53. ↵
      2. Vanzetto G ,
      3. Machecourt J ,
      4. Blendea D , et al
      . Additive value of thallium single-photon emission computed tomography myocardial imaging for prediction of perioperative events in clinically selected high cardiac risk patients having abdominal aortic surgery. Am J Cardiol 1996;77:143–8. doi:10.1016/s0002-9149(96)90585-8
      OpenUrlCrossRefPubMedWeb of Science
    54. ↵
      2. Eagle KA ,
      3. Coley CM ,
      4. Newell JB , et al
      . Combining clinical and thallium data optimizes preoperative assessment of cardiac risk before major vascular surgery. Ann Intern Med 1989;110:859–66. doi:10.7326/0003-4819-110-11-859
      OpenUrlCrossRefPubMedWeb of Science
    55. ↵
      2. Tang T ,
      3. Walsh SR ,
      4. Fanshawe TR , et al
      . Estimation of physiologic ability and surgical stress (E-PASS) as a predictor of immediate outcome after elective abdominal aortic aneurysm surgery. The American Journal of Surgery 2007;194:176–82. doi:10.1016/j.amjsurg.2006.10.032
      OpenUrl
    56. ↵
      2. Chaikof EL ,
      3. Fillinger MF ,
      4. Matsumura JS , et al
      . Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1061–6. doi:10.1067/mva.2002.123991
      OpenUrlCrossRefPubMedWeb of Science
    57. ↵
      2. Steyerberg EW ,
      3. Kievit J ,
      4. Otterloo JC
      . Perioperative mortality of elective abdominal aortic aneurysm surgery. A clinical prediction rule based on literature and individual patient data. Arch Intern Med 1995;155:1998–2004. doi:10.1001/archinte.155.18.1998
      OpenUrlCrossRefPubMedWeb of Science
    58. ↵
      2. Carlisle JB ,
      3. Danjoux G ,
      4. Kerr K , et al
      . Validation of long‐term survival prediction for scheduled abdominal aortic aneurysm repair with an independent calculator using only pre‐operative variables. Anaesthesia 2015;70:654–65. doi:10.1111/anae.13061
      OpenUrl
    59. ↵
      2. Biancari F ,
      3. Hobo R ,
      4. Juvonen T , et al
      . Glasgow Aneurysm Score predicts survival after endovascular stenting of abdominal aortic aneurysm in patients from the EUROSTAR registry. Br J Surg 2006;93:191–4. doi:10.1002/bjs.5262
      OpenUrlCrossRefPubMed
    60. ↵
      2. Ambler GK ,
      3. Gohel MS ,
      4. Mitchell DC , et al
      . The Abdominal Aortic Aneurysm Statistically Corrected Operative Risk Evaluation (AAA SCORE) for predicting mortality after open and endovascular interventions. J Vasc Surg 2015;61:35–43. doi:10.1016/j.jvs.2014.06.002
      OpenUrl
    61. ↵
      2. Giles KA ,
      3. Schermerhorn ML ,
      4. O’Malley AJ , et al
      . Risk prediction for perioperative mortality of endovascular vs open repair of abdominal aortic aneurysms using the Medicare population. J Vasc Surg 2009;50:256–62. doi:10.1016/j.jvs.2009.01.044
      OpenUrlCrossRefPubMedWeb of Science
    62. ↵
      2. Grant SW ,
      3. Grayson AD ,
      4. Purkayastha D , et al
      . Logistic risk model for mortality following elective abdominal aortic aneurysm repair. Br J Surg 2011;98:652–8. doi:10.1002/bjs.7463
      OpenUrlPubMed
    63. ↵
      2. Prytherch DR ,
      3. Sutton GL ,
      4. Boyle JR
      . Portsmouth POSSUM models for abdominal aortic aneurysm surgery. Br J Surg 2001;88:958–63. doi:10.1046/j.0007-1323.2001.01820.x
      OpenUrlCrossRefPubMedWeb of Science
    64. ↵
      2. Tang T ,
      3. Walsh SR ,
      4. Prytherch DR , et al
      . VBHOM, a data economic model for predicting the outcome after open abdominal aortic aneurysm surgery. Br J Surg 2007;94:717–21. doi:10.1002/bjs.5808
      OpenUrlCrossRefPubMed
    65. ↵
      2. Grant SW ,
      3. Hickey GL ,
      4. Grayson AD , et al
      . National risk prediction model for elective abdominal aortic aneurysm repair. Br J Surg 2013;100:645–53. doi:10.1002/bjs.9047
      OpenUrlCrossRefPubMed
    66. ↵
      Qualitative methods for health research. Choice Reviews Online 2009;47:47–0901. doi:10.5860/CHOICE.47-0901
      OpenUrl
    67. ↵
      2. Coast J ,
      3. Horrocks S
      . Developing attributes and levels for discrete choice experiments using qualitative methods. J Health Serv Res Policy 2007;12:25–30. doi:10.1258/135581907779497602
      OpenUrlCrossRefPubMed
    68. ↵
      2. Braun V ,
      3. Clarke V
      . Thematic analysis. In: APA Handbook of Research Methods in Psychology, Vol 2: Research Designs: Quantitative, Qualitative, Neuropsychological, and Biological. Washington: American Psychological Association, 2012: 57–71. Available: http://content.apa.org/books/13620-000
    69. ↵
      2. Varpio L ,
      3. Paradis E ,
      4. Uijtdehaage S , et al
      . The Distinctions Between Theory, Theoretical Framework, and Conceptual Framework. Acad Med 2020;95:989–94. doi:10.1097/ACM.0000000000003075
      OpenUrlCrossRefPubMed
    70. ↵
      2. Braun V ,
      3. Clarke V
      . Using thematic analysis in psychology. Qual Res Psychol 2006;3:77–101. doi:10.1191/1478088706qp063oa
      OpenUrlCrossRefPubMed
    71. ↵
      2. Maxwell J
      . Understanding and Validity in Qualitative Research. Harv Educ Rev 1992;62:279–301. doi:10.17763/haer.62.3.8323320856251826
      OpenUrlCrossRefWeb of Science
    72. ↵
      2. Deterding NM ,
      3. Waters MC
      . Flexible Coding of In-depth Interviews: A Twenty-first-century Approach. Sociol Methods Res 2021;50:708–39. doi:10.1177/0049124118799377
      OpenUrlCrossRef
    73. ↵
      2. de Bekker-Grob EW ,
      3. Donkers B ,
      4. Bliemer MCJ , et al
      . Can healthcare choice be predicted using stated preference data? Soc Sci Med 2020;246:112736. doi:10.1016/j.socscimed.2019.112736
      OpenUrl
    74. ↵
      2. Abiiro GA ,
      3. Leppert G ,
      4. Mbera GB , et al
      . Developing attributes and attribute-levels for a discrete choice experiment on micro health insurance in rural Malawi. BMC Health Serv Res 2014;14:235. doi:10.1186/1472-6963-14-235
      OpenUrl
    75. ↵
      2. Hahn GJ ,
      3. Shapiro SS , General Electric Company
      . A Catalog and Computer Program for the Design and Analysis of Orthogonal Symmetric and Asymmetric Fractional Factorial Experiments. 1st edn. Schenectady, New York: General Electric, Research and Development Center, 1966.
    76. ↵
      2. Day B ,
      3. Bateman IJ ,
      4. Carson RT , et al
      . Ordering effects and choice set awareness in repeat-response stated preference studies. J Environ Econ Manage 2012;63:73–91. doi:10.1016/j.jeem.2011.09.001
      OpenUrl
    77. ↵
      2. Scali ST ,
      3. Arnaoutakis DJ ,
      4. Huber TS , et al
      . The Influence of Surgeon Experience on Patient Selection and Outcomes After Open Abdominal Aortic Aneurysm Repair. J Vasc Surg 2019;70:e48. doi:10.1016/j.jvs.2019.06.074
    78. ↵
      2. Wallis CJD ,
      3. Jerath A ,
      4. Coburn N , et al
      . Association of Surgeon-Patient Sex Concordance With Postoperative Outcomes. JAMA Surg 2022;157:146. doi:10.1001/jamasurg.2021.6339
      OpenUrl
    79. ↵
      2. Boyle JR ,
      3. Mao J ,
      4. Beck AW , et al
      . Editor’s Choice – Variation in Intact Abdominal Aortic Aneurysm Repair Outcomes by Country: Analysis of International Consortium of Vascular Registries 2010 – 2016. Eur J Vasc Endovasc Surg 2021;62:16–24. doi:10.1016/j.ejvs.2021.03.034
      OpenUrl
    80. ↵
      2. Beck AW ,
      3. Sedrakyan A ,
      4. Mao J , et al
      . Variations in Abdominal Aortic Aneurysm Care: A Report From the International Consortium of Vascular Registries. Circulation 2016;134:1948–58. doi:10.1161/CIRCULATIONAHA.116.024870
      OpenUrlAbstract/FREE Full Text
    81. ↵
      2. Mariel P ,
      3. Hoyos D ,
      4. Meyerhoff J , et al
      . Experimental Design. 2021:37–49.
    82. ↵
      2. Hess S ,
      3. Train KE ,
      4. Polak JW
      . On the use of a Modified Latin Hypercube Sampling (MLHS) method in the estimation of a Mixed Logit Model for vehicle choice. Transportation Research Part B: Methodological 2006;40:147–63. doi:10.1016/j.trb.2004.10.005
      OpenUrlCrossRefWeb of Science
    83. ↵
      2. Daly A ,
      3. Hess S ,
      4. de Jong G
      . Calculating errors for measures derived from choice modelling estimates. Transportation Research Part B: Methodological 2012;46:333–41. doi:10.1016/j.trb.2011.10.008
      OpenUrl
    84. ↵
      1. Team
      . RStudio: integrated development environment for R. PBC, Boston, MA RStudio; 2020.
    85. ↵
      2. Hess S ,
      3. Palma D
      . Apollo: A flexible, powerful and customisable freeware package for choice model estimation and application. Journal of Choice Modelling 2019;32:100170. doi:10.1016/j.jocm.2019.100170
      OpenUrlCrossRef
    86. ↵
      2. Pouncey AL ,
      3. David M ,
      4. Morris RI , et al
      . Editor’s Choice – Systematic Review and Meta-Analysis of Sex Specific Differences in Adverse Events After Open and Endovascular Intact Abdominal Aortic Aneurysm Repair: Consistently Worse Outcomes for Women. Eur J Vasc Endovasc Surg 2021;62:367–78. doi:10.1016/j.ejvs.2021.05.029
      OpenUrl
    87. ↵
      2. Maheswaran R ,
      3. Tong T ,
      4. Michaels J , et al
      . Socioeconomic disparities in abdominal aortic aneurysm repair rates and survival. Br J Surg 2022;109:958–67. doi:10.1093/bjs/znac222
      OpenUrl
    88. ↵
      2. Sweet MP ,
      3. Fillinger MF ,
      4. Morrison TM , et al
      . The influence of gender and aortic aneurysm size on eligibility for endovascular abdominal aortic aneurysm repair. J Vasc Surg 2011;54:931–7. doi:10.1016/j.jvs.2011.02.054
      OpenUrlCrossRefPubMed
    89. ↵
      2. Bryan S ,
      3. Dolan P
      . Discrete choice experiments in health economics. Eur J Health Econom 2004;5:199–202. doi:10.1007/s10198-004-0241-6
      OpenUrl
    90. ↵
      2. Moulakakis KG ,
      3. Dalainas I ,
      4. Kakisis J , et al
      . Endovascular Treatment versus Open Repair for Abdominal Aortic Aneurysms: The Influence of Fitness in Decision Making. Int J Angiol 2013;22:9–12. doi:10.1055/s-0033-1333868
      OpenUrl
    91. ↵
      2. Dillavou ED ,
      3. Muluk SC ,
      4. Rhee RY , et al
      . Does hostile neck anatomy preclude successful endovascular aortic aneurysm repair? J Vasc Surg 2003;38:657–63. doi:10.1016/s0741-5214(03)00738-9
      OpenUrlCrossRefPubMedWeb of Science
    92. ↵
      2. Whittaker JD ,
      3. Meecham L ,
      4. Summerour V , et al
      . Outcome after Turndown for Elective Abdominal Aortic Aneurysm Surgery. Eur J Vasc Endovasc Surg 2017;54:579–86. doi:10.1016/j.ejvs.2017.07.023
      OpenUrl
    93. ↵
      2. Swart M ,
      3. McCarthy R
      . Shared decision making for elective abdominal aortic aneurysm surgery. Clin Med (Lond) 2019;19:473–7. doi:10.7861/clinmed.2019-0352
      OpenUrl
    94. ↵
      2. Stubenrouch FE ,
      3. Peters LJ ,
      4. de Mik SML , et al
      . Improving Shared Decision Making in Vascular Surgery: A Stepped Wedge Cluster Randomised Trial. Eur J Vasc Endovasc Surg 2022;64:73–81. doi:10.1016/j.ejvs.2022.04.016
      OpenUrl
    95. ↵
      2. Eid MA ,
      3. Barry MJ ,
      4. Tang GL , et al
      . Effect of a Decision Aid on Agreement Between Patient Preferences and Repair Type for Abdominal Aortic Aneurysm: A Randomized Clinical Trial. JAMA Surg 2022;157:e222935. doi:10.1001/jamasurg.2022.2935
    96. ↵
      2. Winterborn RJ ,
      3. Amin I ,
      4. Lyratzopoulos G , et al
      . Preferences for endovascular (EVAR) or open surgical repair among patients with abdominal aortic aneurysms under surveillance. J Vasc Surg 2009;49:576–81. doi:10.1016/j.jvs.2008.09.012
      OpenUrlCrossRefPubMedWeb of Science
    97. ↵
      2. Bertoni AG ,
      3. Bonds DE ,
      4. Lovato J , et al
      . Sex disparities in procedure use for acute myocardial infarction in the United States, 1995 to 2001. Am Heart J 2004;147:1054–60. doi:10.1016/j.ahj.2003.11.019
      OpenUrlCrossRefPubMedWeb of Science
    98. ↵
      2. Khan E ,
      3. Brieger D ,
      4. Amerena J , et al
      . Differences in management and outcomes for men and women with ST-elevation myocardial infarction. Med J Aust 2018;209:118–23. doi:10.5694/mja17.01109
      OpenUrl
    99. ↵
      2. Pouncey AL ,
      3. Sweeting MJ ,
      4. Bicknell C , et al
      . Sex-specific differences in the standard of care for infrarenal abdominal aortic aneurysm repair, and risk of major adverse cardiovascular events and death. Br J Surg 2023;110:481–8. doi:10.1093/bjs/znad018
      OpenUrl
    100. ↵
      2. Park C ,
      3. Ko FC
      . The Science of Frailty. Clin Geriatr Med 2021;37:625–38. doi:10.1016/j.cger.2021.05.008
      OpenUrlPubMed
    101. ↵
      2. Glick P ,
      3. Fiske ST
      . The Ambivalent Sexism Inventory: Differentiating hostile and benevolent sexism. J Pers Soc Psychol 1996;70:491–512. doi:10.1037/0022-3514.70.3.491
      OpenUrlCrossRefWeb of Science
    102. ↵
      2. Glick P ,
      3. Fiske ST
      . An ambivalent alliance. Hostile and benevolent sexism as complementary justifications for gender inequality. Am Psychol 2001;56:109–18. doi:10.1037/0003-066X.56.2.109
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • X @pounce321

    • Contributors ALP, JP, CB and MR contributed to the conceptualisation of the project and funding application. All authors contributed to the design of the study. ALP developed the discrete choice experiment with experimental design support and statistical analysis support from LEL-R, LS and MR. All authors contributed to the interpretation of the data as well as the drafting and revision of the manuscript. All authors gave final approval. ALP is responsible for the overall content as guarantor.

    • Funding AP was supported by a National Institute for Health and Care Research doctoral fellowship for this study (NIHR DF 301767).

    • Competing interests All authors have completed the ICMJE uniform disclosure form at http://www.icmje.org/disclosure-of-interest/ and declare: ALP was supported by a National Institute for Health and Care Research doctoral fellowship for this work; no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work. All other authors have no competing interest to declare.

    • Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.