Article Text

Download PDFPDF

Survey of ankle–brachial pressure index use and its perceived barriers by general practitioners in the UK
  1. R Yap Kannan1,2,
  2. N Dattani1,
  3. R D Sayers1,
  4. M J Bown1,2
  1. 1Department of Cardiovascular Sciences, University of Leicester, Leicester, UK
  2. 2NIHR Leicester Cardiovascular Biomedical Research Unit, Leicester, UK
  1. Correspondence to Dr R Yap Kannan, Department of Cardiovascular Sciences, University of Leicester, Robert Kilpatrick Clinical Sciences Building, Infirmary Road, Leicester LE2 7LX, UK; rameshykannan{at}


Background Peripheral arterial disease (PAD) is often undetected until complications arise, despite it being a major healthcare burden and an independent risk factor for cardiovascular death and systemic atherosclerosis. Appropriate diagnostic tools are as important as clinical knowledge and skill to investigate patients for PAD. Currently, the ankle–brachial pressure index (ABPI) is the recommended diagnostic tool for PAD.

Purpose We explore current opinions on ABPI by general practitioners (GPs) and the limitations to its implementation in primary care practice.

Methods GPs attending a regional 1-day study event, were surveyed in October 2014. Survey questionnaires were placed at the top of each conference pack for each attendee. The survey questionnaire was modelled from the ankle–brachial index (ABI) usage survey questionnaire used in the PAD Awareness, Risk and Treatment: New Resources for Survival (PARTNERS) preceptorship study.

Results All respondents were GPs, with a survey response rate of 77.1%. All respondents regarded ABPI as an important test, that is primarily performed by nursing staff (79.5%) in their respective GP surgeries. 70% and 97% of GPs found ABPI useful for the diagnosis of asymptomatic and symptomatic PAD, respectively. 69% of GPs regarded ABPI as a feasible test in primary care practice. Time constraints (84%), staff availability (89%) and staff training (72%) were cited as the main limitations to its use.

Conclusions Targeted training of nursing staff may improve ABPI usage, although a less time-consuming test for PAD may be another option.

View Full Text

Statistics from


Peripheral arterial disease (PAD) is an early indicator of systemic atherosclerosis1 and an independent predictor of cardiovascular mortality.2 ,3 It is a major health and financial burden.4 PAD affects about 20%–22% of people over the age of 45.5 Of these patients, about 10% have typical intermittent claudication.6 The severity of PAD symptoms generally depends on the severity of both large vessel stenosis/occlusions and the presence or absence of microvascular disease.7 Therefore, patients with complete arterial occlusion may remain asymptomatic.8 Patients with asymptomatic PAD have a threefold to fourfold excess risk of having coronary artery disease and cerebrovascular disease.5 ,9 Recent focus has been on the primary prevention of this disease,10 by improving the diagnostic practices in primary care.11 Ankle–brachial pressure index (ABPI) is the main diagnostic tool for PAD in primary care.12 ,13

Recent surveys on the use of ABPI have suggested that primary care practitioners should receive targeted training to perform and analyse ABPIs.14 Similar recommendations were made over a decade ago, following a large multicentre programme that assessed the practice and perceptions of primary care clinicians, with the PAD Awareness, Risk and Treatment: New Resources for Survival (PARTNERS) in the USA.11 Following this, a PARTNERS preceptorship programme enrolled and trained primary care staff in the technique of performing and interpreting ABPIs. An ABPI usage survey conducted on participants of both the PARTNERS and the PARTNERS preceptorship programme summarised that primary care clinicians accepted ABPI as a simple diagnostic tool and their role to diagnose PAD despite existing barriers.15 Nevertheless, evidence to support the uptake of ABPIs in primary care is lacking in the USA and worldwide.

The purpose of this survey was to assess if English general practitioners (GPs) regarded ABPI to be an important test and what were the potential limitations to its implementation in a primary care setting in the UK.


The Leicestershire, Northamptonshire and Rutland Regional Ethics Committee 1 was consulted prior to conducting the survey. Ethical approval was deemed unnecessary.

A survey was conducted on GPs attending a 1-day ‘GP masterclass regional study day’ at the University of Leicester conference centre in October 2014. The whole day event was organised by Spire Hospital Leicester and was open to all GPs in the UK. On the day of the event, participants were seated in clusters. For each seat, survey questionnaires were placed at the top of the conference packs. Following the opening address by the chairperson, the researcher (RYK) and his research was briefly introduced. The chairperson requested all attendees to complete the survey questionnaire and the event feedback form prior to the first lecture. The request was repeated at three break sessions and at the end of the event. Participants were requested to leave survey responses on their table at the end of the day. Participants were informed of the availability of the researcher throughout the day, if there were any queries.

The GP masterclass regional study day organised by the Spire Hospital Leicester was initiated in 2012. It is designed to assist GPs in meeting their revalidation needs by collecting continuing professional development credits, in addition to networking opportunities. The event has a good attendance record over the past 2 years (148 and 168, respectively). The event feedback form was introduced in the events in the second year (2013). The event feedback questionnaire response rate in 2013 was 61.9% (104 of 168 attendees).

A cross-sectional survey was conducted, using a survey questionnaire modelled from the ankle–brachial index (ABI) usage survey originally created by the steering committee of the PARTNERS preceptorship programme in the USA,15 with minor relevant modifications to suit the UK healthcare system and to allow comparisons to be made. Like the utility questionnaire, this was a multiple-choice close-ended questionnaire. The survey questionnaire comprised of eight questions (figure 1). The first and second question identified the participant’s profession and opinions on diagnostic options for PAD. The remaining questions focused on ABPI. These included who performs the test, how often it is used, how useful has it been in the participant's practice, how feasible it is to incorporate it into practice, whether it is a good screening tool for select patient groups and finally what are the perceived limitations to its use in primary care practice. The survey questionnaire was piloted on members of staff within the department and the departmental statistician with an interest in survey questionnaires. Modifications were made based on suggestions and feedback obtained.

Figure 1

Peripheral arterial disease survey questionnaire.

Raw data was double entered into the SPSS data entry software. All variables were transcribed with the help of a codebook. All data collected on the questionnaires were anonymous. Missing values were coded in the SPSS program and included in the data analysis.


A total of 118 GPs attended the study day. The survey questionnaire response rate was 77.1% (91 of 118 attendees) compared with the event feedback response rate which was 84.7% (100 of 118 attendees). All 91 respondents were GPs. The results attempt to answer five questions. These were whether ABPI is regarded as an important test (question 2), whether ABPI was used and if so, who is involved (questions 3 and 4), to gauge perceived importance of ABPI for PAD diagnosis (question 5 and 6), whether ABPI as a diagnostic test for PAD is feasible in primary care (question 7) and if so, what are the limitations to its use (question 8). Results are summarised in figure 2.

Figure 2

PAD survey responses. ABPI, ankle–brachial pressure index; GP, general practitioner; PAD, peripheral arterial disease; TBPI, toe–brachial pressure index.

Importance of diagnostic tests for PAD diagnosis

All respondents felt that risk assessment and ABPI played an important role in PAD diagnosis. Opinions on the role of questionnaires and ankle pressures were divided.

Similarly, most GPs were unsure of the role of toe–brachial pressure index and toe pressures.

ABPI usage and staff involvement in PAD diagnosis

The vast majority of ABPIs were performed by nurses (26.1% nurse practitioners, 36.4% registered nurses and 17% district nurses) according to the respondents with the majority of GP surgeries using ABPI on a monthly (42%) to annual (30%) basis.

Perceived utility of ABPI for the care of PAD

The majority of GPs found ABPI to be more useful for the diagnosis and clinical management of symptomatic patients compared with asymptomatic patients. As a screening tool, the respondents felt that ABPI was most useful in diabetes, followed by the elderly and patients with chronic renal failure, but was least useful for healthy patients.

Feasibility of ABPI in GP practice

Although the vast majority of respondents deemed ABPI to be either moderately (54%) or very feasible (15%), a third (31%) deemed it to be not feasible.

What are the perceived limitations to the use of ABPI in primary care?

The limitations were divided into seven broad categories. These included administrative (time and financial constraints), structural (availability of space), ABPI-related (clinical significance of ABPI and ABPI interpretation), patient-related (willingness), staff-related (availability, willingness and training), equipment-related (availability of Doppler and cuffs) and test performance-related issues (application of cuff, performance of walk test or handheld Doppler examination or the presence of wounds).

Limitations to implementation—administrative

Eighty-four per cent respondents regarded time constraints to be a major or moderate limitation, with the opinion on financial constraints divided in this survey.

Limitations to implementation—structural

Just over half of the respondents viewed the availability of space within the GP surgery as a major or moderate limitation.

Limitations to implementation—ABPI as a test

Under half (43%) of the 74 respondents felt that the clinical significance of ABPI was not a limitation to its use. Interestingly, the majority of respondents, found interpreting ABPI results as a major or moderate (51%).

Limitations to implementation—patient related

Patient willingness to have ABPI as a test was not a limitation in most respondents (83%).

Limitations to implementation—staff related

The vast majority (89%) of the respondents felt staff availability was a major or moderate limitation to ABPI use. Similarly, 72% respondents regarded staff training to be a major or moderate limitation. Opinion on staff willingness being a limitation to ABPI performance was inconclusive.

Limitations to implementation—equipment related

62% regarded the availability of Doppler and cuff availability to be a major or moderate limitation, respectively.

Limitations to implementation—test performance related

A third of the respondents did not view the application of cuff to be a limitation. However, 66% viewed performance of the walk test and 68% viewed the presence of wounds to be main limitations. Although only 3.4% GPs admitted to performing ABPIs, it was interesting to note that 61% of the respondents regarded performing the Doppler examination to be a major or moderate limitation.


Commissioners and providers are primarily responsible for implementing the National Institute for Health and Care Excellence (NICE) pathway for lower-limb PAD in the UK.16 As these are primarily GPs, our survey focused at GPs to provide a better insight into practical aspects that aid or hinder the implementation and facilitation of ABPI use in primary care practice. All participants in our survey were GPs (100%), compared with the cohort studied in the PARTNERS programme (54% physicians), PARTNERS preceptorship (73% physicians) and in the most recent survey by Davies et al (55% GPs).

Our survey findings reflect those of the ATTEST study findings in which French GPs performed ABPI in only a third of the patients with PAD.17 ABPI was mainly performed by nurses (79.5%) in this survey, compared with the PARTNERS programme (38%) and the survey conducted by Davies et al (67.4%). Respondents of this survey agree with previous survey findings, that ABPI is more useful in symptomatic than asymptomatic PAD.15 As a screening tool for PAD in diabetics, ABPI is generally accepted (78%) by survey respondents. This response is in line with current NICE guidelines.16 It is interesting to note that more than half the respondents (55%) felt ABPI to be a useful screening tool for PAD in patients with chronic renal failure or who were elderly, which reflects an understanding by GPs of factors that pose a risk to PAD. Although, the US preventative services task force (USPSTF) does not recommend ABPI as a screening test,18 the NICE guidelines recommends assessment of PAD in those with suspected PAD, whether symptomatic or not.16

Unlike previous surveys,15 where the majority of clinicians in the PARTNERS programme (90%) and preceptorship programme (88%) believed ABPI to be a feasible test, in our survey only two-thirds (69%) regarded it as being feasible. Using a 70% or more arbitrary cut-off as significant limitation, the primary limitations are time constraints (84%), staff availability (89%) and staff training (72%). Time constraints have also been a limitation in previous surveys by Mohler et al15 (54%) and Davies et al14 (72%). It is intriguing that all surveys conducted to date including this one, have highlighted time constraints and staff training to be primary limitations to ABPI implementation. However, if we used a 50% cut-off for major or moderate limitations, as used by previous surveys,14 ,15 financial constraints (61%), space availability (57%), cuff and Doppler availability (62% each), presence of wounds (68%), performance of walk test (66%) and the interpretation of ABPI (51%) are limiting factors.

The NICE guideline development group consider the incremental resource needs for ABPI measurements to be small compared with the benefits of early PAD diagnosis4 and this test adds between 5 and 15 min to the time needed for clinical examination.16 However, attempts to reduce the time for ABPI measurements, such as using the pulse oximeter instead of the handheld Doppler19 ,20 or an oscillometric automated blood pressure device,21 the reliability of such alternatives have not been widely accepted. Under such circumstances, delegation of ABPI performance to competent and trained staff would appear reasonable. Targeted formal training of students have been shown to enhance the reliability of ABPI.22 Although, the targeted educational initiative was studied in the ABI usage survey,15 the follow-up after the training survey was conducted a few months after the training programme. Clinical audit tools for PAD in primary care and an online education tool are available for healthcare professionals in the UK.16 An ideal test should be reliable, quick to perform, easy to use, interpret and be affordable. Such a test would circumvent many of these limitations. Nevertheless, ABPI is the recommended test for PAD in primary care that GPs would be expected to understand and use.23 Further studies to assess the true impact of targeted ABPI training and the sustainability of practice, should be conducted after an adequate period of time.

Strengths and limitations

The advantage of this study is that the feedback is from GPs who play a pivotal role in the diagnosis of PAD and the provision of this diagnostic service. Unlike recent surveys,14 this survey concentrates less on the actual methods for performing the test, but rather attempts to understand reasons behind why ABPI may or may not be used in primary care.

As the majority of respondents who attended the event were from Leicestershire, it is possible, but unlikely, that these results are not representative of the opinions of GPs nationally.


GPs need to be more proactive in assessing the ABPI as part of their clinical examination for the diagnosis of PAD. Issues such as time constraints, staff availability and staff training can be effectively with better time management of patients and perhaps recording the ABPI as part of the routine observations for patients with these symptoms. Better triage in GP practices and collaboration with community nursing staff could enhance this service. Collaborations between vascular diagnostic laboratories and GP practices could improve the technique and confidence of GPs and healthcare professionals of performing the ABPI. It is clear that a well-defined system is required for the diagnosis of PAD in the community.

Main messages

  • Survey response rate was 77.1%.

  • The use of ankle–brachial pressure index (ABPI) to diagnose peripheral arterial disease is still considered important by most general practitioners.

  • Like previous studies, time constraints and staff training are significant limitations. However, additional limitations reported include staff availability, financial constraints, space availability, cuff and Doppler availability, presence of wounds, performance of walk test and the interpretation of ABPI.

  • General practitioners should use the ABPI as part of their clinical assessment of peripheral arterial disease as per the Royal college of General Practitioners (RCGP) curriculum.

  • The effect of targeted ABPI training of general practitioners and allied health professionals in primary care needs to be studied.

Current research questions

  • Is there a better way to increase survey response rates from general practitioners?

  • Does ankle–brachial pressure index (ABPI) play an important role in the diagnosis of peripheral arterial disease?

  • Is ABPI a feasible diagnostic test in primary care?

  • If not feasible, what are the perceived limitations of its use in primary care?

  • How can this practice be improved?


We would like to thank Spire Hospital Leicester.


View Abstract


  • Contributors RYK: conducted the survey and analysis. ND: analysis. RDS and MJB: peer review.

  • Competing interests None declared.

  • Ethics approval Leicester Regional Ethics Committee 1.

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

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.