Improved identification of familial hypercholesterolaemia (FH) in primary care

Prevalence data show that FH is substantially underdiagnosed. National audit data suggest that fewer than 12% of people with FH are identified in the UK. Primary care provides an opportunity to systematically identify index FH cases for screening, diagnosis, and treatment using data held within primary care IT systems. This Medway FH audit set out to support primary care physicians to follow the NICE CG71 recommendation and consider the possibility of FH in adults who have raised total cholesterol concentrations, especially if there was a personal or family history of premature CHD. The aim was to improve the identification of people with FH in Medway CCG.

In 2008, NICE CG71 recommended considering the possibility of FH in adults with raised cholesterol, especially in those with a personal or family history of CHD. However, despite these guidelines, substantial improvements in FH diagnosis were not being made, representing a major gap in prevention of CHD. Medway CCG includes 56 GP practices serving a population of 290,000 patients (2014 figures). Prevalence estimates for FH suggest that between one in 500 and one in 200 people have FH in a population. In Medway, more than 600 patients are likely to have FH. The prevalence of FH within Medway was significantly below that estimate and at baseline, only one in 750 people or 343 patients in total had a formal diagnosis of FH. Primary care is well placed to lead on the identification of FH. Most tests for cholesterol are requested by GPs and GP clinical record systems can contain decades of patient data including cholesterol measurements and family history of CHD. To support Medway GPs to deliver care in accordance with NICE CG71, to improve the detection of FH and hence provide FH patients with appropriate disease management.

Medway CCG collaborated with BMJ, utilising their clinical decision support software ('Audit +') and developed an FH audit tool. The BMJ software supports practices to deliver best practice care through prompts during consultations and audit reports for a variety of diseases and act as a performance-enhancing tool. Audits could be conducted within practices on a daily basis. At the CCG level, audits could be conducted weekly, but progress was monitored monthly. In the second phase of the study Medway collaborated with HEART UK - the cholesterol charity, with support from the pharmaceutical company Sanofi, to conduct nurse-led FH clinics to further improve identification of FH cases. Medway CCG considered that putting NICE guidance into practice and increasing detection of patients with FH in the primary care setting would lead to better management of patients, identify additional 'at risk' relatives for screening and evaluation and ultimately prevent cardiovascular events. The CCGs considered that there was a significant benefit to patients as well as an economic benefit to the CCG to diagnose FH and reduce preventable cardiovascular events.

In collaboration with BMJ, Medway CCG established an audit tool for FH. Technology in the form of 'Audit +' (BMJ) was already in use in Medway GP surgeries, helping practices deliver best practice care through prompts during consultations for a variety of diseases. The 'Audit +' software is compatible with multiple GP clinical platforms and is loaded remotely onto GP clinical systems, requiring no additional work for the practices or clinicians. Practices were familiar with the software so no additional training was required.

NICE CG71 recommends that a diagnosis of FH is based on the 'Simon Broome criteria'. These diagnostic criteria identify FH patients and classify the patient as definite FH or possible FH. There was no Read Code available for patients classified as 'possible FH' and so we sought and obtained a Read Code from the NHS in 2010. After a short single-practice pilot, the study was rolled out across the CCG. The software initially identified all patients with a diagnosis of FH to provide a baseline prevalence, and then the study was conducted in two phases, between 2011 and 2014. Phase 1 was a 2-year audit, and in Phase 2 the software was refined and we introduced nurse-led support clinics. The FH audit tool initially identified all patients with a diagnosis of FH, providing a baseline prevalence within Medway of one in 750. Next, we identified those patients with elevated cholesterol levels without an FH diagnosis and unscreened via the Simon Broome criteria ('at risk and unscreened'). Electronic flags were added to the records prompting assessment when the patient next visited the practice. The prompts contained further information and relevant Read Codes, which the GP added directly into the patient record from the prompt screen, and additional triggers to encourage optimisation of FH management. The audit was completed after two years. In Phase 2, the software was enhanced and we included the Dutch Lipid Clinic Network (DLCN) Score to define the severity of FH to further support clinical management. In response to feedback from participating GPs highlighting a need for additional resource to assess patients at risk and to diagnose FH, we started an 'FH Nurse Advisor' programme. This was a collaboration between NHS Medway CCG and HEART UK, supported by Sanofi. Ashfield Healthcare provided the service and employed the Nurse Advisor. We appointed a single nurse, who visited all the practices during Phase 2, which ran for 9 months.

In total, 53 of 56 GP practices in Medway participated in Phase 1 and 47 practices in Phase 2. The audit loop was closed and comparisons with baseline performed at 2 years and at the end of the Nurse Advisor Programme. The baseline prevalence of FH in Medway at the initiation of the audit (October 2011) was 0.13% or one in 750 of the patient population, well below expected population prevalence. Nearly 1600 patients (0.59%) were considered at risk and unscreened. After two years, the prevalence of FH increased by 0.09% to 0.22% (one in 450) but patients 'at risk and unscreened' remained at a similar level of 0.58%. In Phase 2, over the 9 months the number of patients diagnosed with FH increased to 0.28% (one in 357 people) with a reduction in the number of patients at risk and unscreened (reduced from 0.58% to 0.14%). Over the three year period, the study identified and managed an additional 433 index cases of FH. The Medway FH Audit Tool and the FH Nurse Advisor Programme provides a systematic approach to identify new index cases of FH from data already available within GP systems.

The study improved diagnosis of FH and raised awareness of FH with GPs and patients within the CCG. The 'Audit +' software and FH audit tool is an inexpensive and helps diagnose FH and will be available to all GP IT systems via GPSOC2. We have shown that nurse-led clinics can further improve FH identification. We believe this audit provides a model for other CCGs to improve diagnosis and management of FH. Improving diagnosis and treatment of FH helped Medway CCG fulfil its role in the delivery of quality healthcare in accordance with Government policy and clinical guidance, including the NHS Outcomes Framework (with impact on Domain 1 - preventing people dying prematurely; Domain 2 - enhancing quality of life for people with long-term conditions; and Domain 4 - ensuring people have a positive experience of care); the NICE FH guidelines (CG71); the NICE Quality Standard on FH (QS41); the Cardiovascular Disease Outcomes Strategy (2013), with its aspiration to find and treat at least 50% of cases of FH in England.

The Medway study provides a model that can be implemented within other CCGs to identify people at risk of FH and offer them appropriate care. The study shows that the FH Audit Tool and FH Nurse Advisor Programme can potentially double the number of FH patients identified, and from our experience the additional work burden is manageable. For example, a large primary care practice of more than 10,000 patients would have approximately 60 patients with elevated cholesterol. Following assessment, our data suggest about 30 new FH index cases would be identified and require ongoing management. Some challenges exist. Flagging of 'at risk' patients by the FH audit tool is dependent on the completeness of the electronic clinical records, particularly cholesterol measurements. While patients over 40 may have cholesterol measurements as part of NHS health checks, younger patients may not have these measurements recorded. We do not know what proportion of the Medway population has lipid results available for assessment. Patient participation in the Nurse Advisor clinic was low at 52%. Future programmes should address this issue and seek ways to engage patients and inform them of the importance of making a diagnosis of FH if cholesterol levels are elevated and a family history is apparent. The Nurse Advisor programme ran for a finite period of time. Future programmes need to develop tools to support practice nurses to conduct clinics to ensure assessment and management of FH patients is ongoing.

The audit tool was able to measure improvement in diagnosis of FH but we did not capture pre- and post-diagnosis lipid levels. Therefore, we do not know if a diagnosis of FH lead to improved patient management. We have recently revised the FH audit tool to capture lipid levels pre- and post-diagnosis and we can now monitor these at a practice (per patient) and CCG level (change in lipid levels).