The cut off for the faecal calprotectin (FC) assay mentioned in the NICE Guidance (50µg/g) was found to increase the number of referrals to secondary care as the specificity at that cut off was not sufficient to prevent a high number of patients with IBS being unnecessarily referred to secondary care.

This new pathway addressed this issue by increasing the cut off to 100µg/g and placing risk assessment figures to support GPs with their clinical decision on when and how to refer a patient. This aims of the pathway were to 1) reduce the pressures in secondary care for endoscopy services and gastroenterology outpatient appointments (by reducing the number of unnecessary referrals), 2) improve the patient experience by supporting quicker diagnoses and enabling those patients who could be treated in primary care to do so 3) provide cost saving benefits for the health economy.

Using a series of supporting documentation and educational support to primary care, there has been an adherence to the pathway of at least 85% which is significantly more than previously (the pathway previously used in Leeds had an adherence of 11%).

This pathway was rolled out initially in York with the aim to spread to the rest of the Yorkshire and Humber region. To date, it has been implemented in 9 CCGs (240 GP practices) and the AHSN has facilitated the scaling up of implementation to other regions such as South Tees, Oxford, Bristol and Exeter.

The AHSN took part in a national task and finish group for the spread of FC diagnostic tests and this pathway formed the basis of the national algorithm which is planned for launch in 2018. This was working with professionals from around the country and included NHS England, NHS Business Services Authority, Trusts, CCGs, GPs and AHSNs.

This example was a runner up finalist in the 2018 NICE Shared Learning Awards.

In 2010, Dr Turvill led a programme to devise and implement a structured care pathway for the use of FC testing in primary care. This was designed to overcome the challenge of the high sensitivity but poor specificity of FC tests and to ensure it focused on delivering an improved patient experience and better use of NHS resources.

In 2014, the pathway was piloted as part of a NICE adoption project in 5 GP practices in York. Evaluation on completion demonstrated its safety and effectiveness and a GP survey confirmed trust and support in the pathway. A patient survey was also completed to confirm the patient experience was improved through the implementation of this pathway.

Following on from the pilot, the AHSN knew there would be benefits to the whole Yorkshire and Humber region and worked with Dr Turvill to create a risk assessment tool (pathway) to support GP clinical decision making when there is diagnostic uncertainty. This built upon the previous research which increased the cut off level to 100µg/g and included placing risk assessment figures to support clinical judgement.

This pathway was rolled out initially in York and has spread to a further 8 CCGs in Yorkshire and Humber as well as other areas of the country. There are a further 8 CCGs who are in the process of implementing the pathway in their area. Some of these areas are more urban than others and the patient demographics are mixed indicating that this pathway works well for any area.

Following on from the Pilot in 2014, the Yorkshire & Humber AHSN supported rollout of the pathway and creating an implementation pack for CCGs and GPs including:

• Template business cases for CCGs and Trusts
• Downloadable pathway templates for EMIS and SystmOne
• An educational video and slide deck to explain the FC pathway (https://www.youtube.com/watch?v=ah-EH-q0lys&list=PLw476CR0-KDe45xaS1CKNxD4mFDYImz3n)
• GP leaflet to provide more detail of the pathway and why and when to test
• Patient information leaflet to explain what the test is and why it is being carried out

These were co-created with York Teaching Hospitals NHS FT and Vale of York CCG to ensure they would be fully utilised and relevant to the other organisations in the region - a copy of these can be viewed at http://www.valeofyorkccg.nhs.uk/rss/index.php?id=faecal-calprotectin. By providing these essential resources for primary care, buy in and acceptance to implementation was approved in many CCGs.

Some barriers encountered were reluctance and hesitance of gastroenterologists in the trusts, how this would fit with other pathways and how FIT testing would affect the future of FC testing. Dr Turvill was able to provide the clinical assurance to these questions and support buy in from secondary care to this pathway.

The AHSN also funded an economic impact analysis by the York Health Economic Consortium (YHEC) to provide demonstrable benefits to the health economy and support further roll out. Most of the costs of the implementation were for the development of the above resources and the financial cost of the pathway implementation to CCGs was minimal. The final report is available here.

The project is meeting its initial aims though we are still working on getting the pathway implemented into all CCGs within Yorkshire and Humber.

The health economic evaluation completed by YHEC demonstrated that the improved pathway saves £100 000 to £160 000 per 1000 patients tested; this equated to a saving of £2.5 million in the Yorkshire and Humber region. The evaluation also found that the pathway saves 1 unnecessary colonoscopy and outpatient appointment per 4-6 patients tested (147-262 colonoscopies per 1000 patients). The sensitivity and specificity of the new pathway was found to be 94% and 92% respectively versus 94% and 61% for the other pathway mentioned in the NICE guidance - by increasing the specificity of the test, it has resulted in the above benefits to the health system in terms of capacity and cost savings. Through the modelling used for the evaluation, it calculated that there would be cost and capacity benefits for the new pathway if only 1% of GPs adhere to it and our evidence suggests an adherence of approximately 85%.

When the test was first being rolled out, a patient survey was conducted to verify the results from the pilot work. The results of this survey mirrored the first one and demonstrated that patients perceived a benefit to having the FC. Some patient quotes include: “the test was enough to confirm that is nothing more than IBS”, “I was happy that the calprotectin is enough” and “the calprotectin is better than a colonoscopy”.

The findings were so impactful and we have successfully implemented this pathway, that the NHS Business Services Authority have written a case study on how the AHSN have spread and shared this pathway across our region. They use this as promotional information when presenting on their PACIFIC programme and is accessible here.

One key piece of learning is that when developing an implementation plan with CCGs, that having Dr Turvill (or a clinical champion) speak at GP education events is key. We found that the GP understanding of the pathway and why the pathway was changed caused adherence to be higher if we had the talk before roll out rather than after.

Another challenge was around the FIT pathways being introduced and how this would affect FC testing - we explained that the pathways were currently separate and FC testing should not be used if cancer is suspected. FIT testing for non-cancer diseases still needs a lot of research and so this pathway shouldn’t be delayed because of those questions.

It is important to understand which assay is being used as this work was based on the Buhlman assay. We have performed some compatibility testing with Hull CCG to understand the cut off differences for the Thermofisher assay. We have since agreed that this pathway can remain the same with the different assay but it important that people know which they are using to ensure the optimal pathways.

The AHSN are more than willing to share the resources mentioned above to support other organisations to roll out this pathway - without this pack of resources to provide CCGs with, implementation would have been much more difficult.

The main objectives of implementing point of care testing in primary care are:

• To improve the patient experience for those with IBS and to reduce the number of patients undergoing invasive procedures within secondary care

• To reduce outpatient referrals into secondary care (32%)

• To increase the numbers accessing rapid diagnosis for IBS/IBD

• To improve pathways for the management of IBS

• To provide a structured programme to aid decision making in primary care

Before introducing the testing in primary care, patients with an uncertain diagnosis of IBS/IBD would be referred into secondary care to see a gastroenterologist. This may have involved an initial consultation, a colonoscopy procedure and a follow up appointment to discuss results several weeks later.

The CCG had access to the total number of consultations and procedures taking place to act as a baseline, but this would have been total numbers, and it was not possible to identify just those attending with IBS symptoms from the data available at the time.

The local gastroenterologist supported the evidence that 32% of the patients that he saw ended up with a diagnosis of IBS and could have been managed in primary care without the need to go to hospital. Cannock Chase CCG has a population size of 132,000 and 26 member GP practices and Stafford & Surrounds CCG has a population of 146,000 and 14 member GP practices.

The CCG took several steps to implement the test in primary care, the key success factors are outlined below:

• Local GP engagement was key to the success and was secured in several ways. A full business case for the implementation was taken to the Membership Board, which has a representative from each practice and was agreed unanimously. This was then followed up by a Protected Learning Time (PLT) delivered by local Gastroenterologist Consultant to explain the benefits of the test to a wider group of GPs.

• The CCG worked with the local GP federation, GP First, to manage the logistics of the accounts and ensuring tests got to each of the 40 GP practices across the 2 CCGs.

• Ferring Pharmaceuticals supported the roll out of the project by providing training to practices and printing out waste disposal information. A clinical pathway was developed with the local Gastroenterologist and the CCG Clinical Leads to outline the circumstances that the test should be carried out. This was  supported by a pathway on how to manage IBS in primary care and also delivered at the GP PLT education events.

The costs incurred by the CCG were the cost of the tests and distribution, £15 each paid to the GP Federation for a bulk order upfront and then an additional £15 enhanced service paid directly to the GP practice once the test had been undertaken.

Each GP practice has to send the CCG, on a monthly basis, a monitoring form which lists the number of tests carried out that month, the batch number of the test, the date it was carried out, the result and whether or not a referral to secondary care was required, as a result of the test being carried out.

The results are recorded on a master file for all practices to log the total number of tests that have been carried out and the number which were negative and the GP has confirmed that a referral was not required to secondary care.

Over the last 13 months period (July 14 – Aug 15) there have been a total of 833 tests carried out across the 2 CCGs and 467 of those were reported as negative and the GP has confirmed the patient was not referred to secondary care, a saving of 56%. This is higher than the anticipated saving of 32%.

The cost savings have been calculated on the basis that the CCG has saved 467 first outpatient attendances, 467 colonoscopy procedures and 467 follow up appointments. Once the investment cost of the tests have been deducted this has a financial saving associated at approx. £280,000.

There has not been a like for like deduction in the activity going through the contracts. This may be due to the fact that the age to screen for bowel cancer increased at a similar time that this project went live and there may have seen an increase in procedures for a separate cohort of patients presenting with other symptoms not relating to IBS. Due to the data the CCG has access to it is not possible to evidence the reasons for this.

The following points have been noted as key learning points of the project:

• The mobilisation phase was underestimated. The CCG’s original approach was to get each practice to set up an individual account with the pharmaceutical company and order the tests directly. Due to practice capacity this did not happen in the timescales we anticipated. A new approach was agreed whereby the GP federation would order the tests on behalf of all the practices and arrange for delivery of the tests to each practice. The training needs for each practice around how to use the test and dispose of the sample afterwards were only identified once the pilot had started and delayed a speedy uptake.

• There is still some further education needs to clarify exactly when the test should be carried out, based on some of the results coming through on the reports. For example some GPs are still referring patients when there is a negative outcome and vice versa. This is being addressed on a case by case basis.

• The CCG needs to have realistic expectations of the impact on the acute contracts – awareness of other influencing factors e.g. Increase of bowel screening age and anticipated growth.

• Further analysis is taking place to look at the diagnosis of those who were referred to secondary care – an audit is currently in progress.

The Department of Gastroenterology at York Teaching Hospital NHS FT and Vale of York CCG developed evidence-based guidelines to support the pragmatic use of FC in primary care. The guidelines incorporated: a higher FC cut off value of 100mcg/g rather than the standard normal range of 50mcg/g, a ‘traffic light’ system for risk assessment, a test repeat when of intermediate risk and a clinical management pathway. The guidelines were designed to maintain the high NPV of FC whilst improving the previously poor PPV. It was hypothesized that this would: 1 identify patients highly likely to have functional disease and provide a structured pathway for their treatment and evaluation; 2 identify those patients at risk of organic disease for referral into secondary care without overwhelming that service with false positives. The guidelines were evaluated by means of an audit.

NICE (Diagnostic Guidance 11) have recently recommended FC testing as an option in adults with lower gastrointestinal symptoms for whom specialist investigations are being considered, if cancer is not suspected and it is used to support a diagnosis of IBD or IBS. Currently there is very little direct evidence from primary care to support the implementation of this guidance. The low incidence of IBD in primary care and the poor PPV of FC means that FC testing may not enhance diagnostic evaluation but instead may overwhelm secondary care services because of the inappropriate referral of false positive patients. An evaluation of a structured guideline for the use of FC in primary care was seen as a possible solution.

The audit of guidelines was registered by the Clinical Effectiveness and Improvement Unit at York Hospital, no: 3232. The audit was supported by NICE HTAP.  Lead-in time was 3 months.

Five primary care practices were invited to enter the evaluation. Educational presentations were given to each practice, a site lead identified and the data sets were agreed. On going support was offered.

Patients with new lower gastrointestinal symptoms, aged 18-60years, were entered into the guidelines where:

• cancer was not suspected
• the likely diagnosis was IBS or IBD but where there was clinical uncertainty in that diagnosis
• there were normal or negative initial investigations as judged appropriate by the GP

Guidelines directed as follows:

Patients with a low FC (<100mcg/g) were treated on the presumption that IBS was likely with positive reassurance, local guidance and review at six weeks with routine referral to the Department of Gastroenterology at York Hospital at that point if they were still symptomatic.

In patients with an intermediate FC result (100-250mcg/g) the test was repeated two weeks later and action thereafter was as directed by that result. Non steroidal anti-inflammatory drugs (NSAID) and aspirin were asked to be avoided if clinically safe or reasonable to do so. A repeat result <100mcg/g prompted expectant, positive, local management as outlined above; a repeat of 100-250mcg/g prompted routine referral to the Department of Gastroenterology at York Hospital. Here the Gastroenterologist would investigate and manage as judged clinically appropriate.

A high risk FC result of >250mcg/g directed to a ‘straight to test’ urgent colonoscopy at York Hospital or an urgent outpatient review if the patient was of a poor performance status.

Patients were entered into the audit for six months from March to August 2014. Clinical outcomes were followed for a further six months during which an evaluation was made.

Comparator secondary care referral data from a neighbouring Trust was obtained.

A GP survey of the guideline was undertaken

Two hundred and sixty two patients were evaluated. 67% were female and the mean age was 38y. Presenting symptoms were diarrhoea 43%, alternating bowel habit 11%, constipation 6%, bleeding 6%, abdominal pain 22%, bloat 4%, none given 6%. 67% of patients with FC<100 were successfully  managed locally. 33% were subsequentlyreferred and 15%  had colonic or cross-sectional investigations.

The guidelines for use of FC  in primary care had a NPV of 97% for IBS ( and non-intestinal disease) and a PPV of 40% for  IBD or other organic intestinal disease.

False negatives were: 

• coeliac disease (2 patients)
• diverticulosis (2 patients)
• microscopic colitis (1 patient)
• pancreatic failure (1 patient)
• 30mm low grade tubulovillous adenoma (1 patient)

This compares with 98% and PPV of 21% using the standard FC cut off of >50mcg/g alone.

The guideline outcomes were better than GP clinical judgment alone and delivered a higher diagnostic yield after secondary care referral (21%) than the conventional (that is current) comparator pathway (5%).

89% of patients with IBD (ulcerative colitis; 4 patients and Crohn’s disease 5) had a FC >250mcg/g and patients were diagnosed by ‘straight to test’ colonoscopy on average within three weeks of referral. The C-reactive protein was normal in 67% of patients with IBD.

The guidelines were considered helpful by GPs who made a number of observations:

“It has been really useful to have a clear set of guidelines for investigating and managing patients with symptoms which would suggest IBS being the most likely diagnosis.”

“I found it a very useful part of feeling confident to exclude borderline cases where uncertain if could be inflammatory bowel disease or irritable bowel disease.” “It also helps a great deal when reassuring and educating patients who are otherwise opposed to a diagnosis of IBS.”

“If this is rolled out across the CCG/ nationally, then education would need to be provided to ensure the test is used appropriately and the results understood.”

Conclusion: This audit outlines the potential role for FC in primary care diagnostics. It incorporates risk assessment to support clinical judgement, a cut off of 100, repeat testing if >100 & straight to test for high-risk investigation. A revised guideline has been developed that we propose to implement at scale across Yorkshire and Humber. Key to its success are:

• early identification of stakeholders,
• communication of the pathway and
• educational support