Participants

Participants were aged ≥ 18 years, citizens of the UK and had no previous diagnosis of a vascular condition (self-assessed). The justification for sampling members of the public rather than patients lies in the context of how the values are likely to be used, namely to inform national health-care priority-setting. In this context, NICE states a preference for QALYs to be based on general population values. The total sample size was 821 (four telephone surveys were administered, with approximately 200 participants in each).

Recruitment and consent

Participants were recruited through various NHS trusts across England. The study covered a wide geographical area including South Yorkshire, Greater London, Kent, Staffordshire, Derbyshire, Lincolnshire, Lancashire, Leicestershire, Cambridgeshire, Essex, Cumbria, North Devon, Tyne and Wear, and Northumberland. A range of approaches were used to identify potential participants. These included approaching hospital visitors on site, having recruitment stands in dining halls, using posters around the trusts with study details and recruiting staff via blanket e-mails. When approaching potential participants, efforts were made to ensure that the sample was representative of the general population with regard to age and sex. Contact details of potential participants were securely transferred to the research team, who then contacted these participants (between September 2017 and January 2018) to arrange a date and time for telephone interviews. Participants who agreed to be interviewed were sent an interview booklet approximately 1 week before the interview took place. They were advised to read the interview booklet prior to the interview to familiarise themselves with the information enclosed. Verbal consent was obtained over the telephone prior to commencing the interview and the interview was audio-recorded. The use of verbal consent was accepted and approved by the South East Coast – Brighton and Sussex Research Ethics Committee, Health Research Authority (REC number 16/LO/0943), on 20 April 2017 along with all of the study documents. An amendment to the interview documents was submitted and approved on 2 May 2017.

Abdominal aortic aneurysm treatment method

A total of 209 participants completed the interview, giving a response rate of 64%. Missing data rates were low and did not exceed 2%.

When considering simple direction of preference, 167 (79.9%) participants stated that they would prefer EVAR, 40 (19.1%) participants indicated that they had a preference for open surgery and two (1.0%) participants said that they had no preference for either treatment. Factors that influenced treatment preferences for EVAR included the less invasive nature of the surgery and the quicker recovery times associated with the procedure. For open surgery, factors included having to have only one follow-up appointment and the feeling that the open procedure felt more permanent.

When participants were asked to make a sacrifice (trade-off) to have their preferred treatment, 45 and 22 respondents who had a stated a preference for EVAR and open surgery, respectively, indicated that they were not willing to sacrifice anything to support their preference and, thus, were deemed to value both treatments equally. Therefore, when strength of preference was taken into account, 122 (58.4%) participants preferred EVAR, 18 (8.6%) participants preferred open surgery and 69 (33%) participants had no preference. In the first year following treatment, those preferring EVAR were willing to give up a mean of 0.135 expected QALYs to have EVAR, whereas those preferring open surgery were willing to give up 0.033 expected QALYs to have open repair. These results indicate a clear preference for EVAR over open surgery among the sample.

Travel distance to hospital

A total of 608 participants completed the interviews (200 for AAA, 202 for CAD and 206 for PAD). Each respondent was asked to consider travelling one of four distances (5, 15, 30 or 60 miles) that were distributed equally within samples. The overall response rate was 64.4%. Missing data rates were low and did not exceed 5%.

The proportions of people willing to travel for AAA (open surgery), AAA (EVAR), CAD and PAD were 89%, 86.5%, 79.7% and 96.1%, respectively. Of these, 56.2%, 55.5%, 64.0% and 66.2% stated that they required compensation to travel for AAA (open surgery), AAA (EVAR), CAD and PAD, respectively.

Among the remaining respondents, the main reason cited for not being willing to travel was a preference for local services and/or a belief that all services should be available locally. Other reasons stated were transport difficulties with making the journey, not wanting to burden family and friends, and a concern among participants that they would feel isolated without having family and friends around them if there was no local provision.

Owing to the constraints of the study design, the maximum compensation that could be demanded by those people not willing to travel was limited to that associated with the success rate at the specialist hospital being 100%. Despite the amount of compensation required to travel by this group being unknown, to include them in the analysis they have been assigned this maximum value. This means that the value of the disutility of travelling reported below for the sample as a whole should be regarded as a minimum value.

For a typical patient aged < 65 years with one of the vascular conditions considered, the expected discounted lifetime QALYs demanded as compensation for having to travel an additional 30 miles for treatment and follow-up are as follows: 0.3541 for AAA (open surgery), 0.3869 for AAA (EVAR), 0.2041 for CAD and 0.6310 for PAD (results for the full range of distances considered are in Report Supplementary Material 13).

Taking AAA (open surgery) as an example, these results suggest that, if vascular services are reorganised so that the individual must travel an extra 30 miles for treatment and follow-up, their disutility of travelling is equal to 0.3541 QALYs. This should be netted off from any potential gain in QALYs that is expected to arise from treatment being at a specialist rather than a local hospital.