Original ArticleIncreasing the Detection of Familial Hypercholesterolaemia Using General Practice Electronic Databases
Introduction
Primary care is central to the increased detection and management of Familial Hypercholesterolaemia (FH) to reduce the corresponding burden of premature coronary artery disease (CAD). [1] Familial hypercholesterolaemia is an autosomal co-dominant disorder associated with elevated low density lipoprotein (LDL-c) cholesterol.[2] Heterozygote FH is thought now to occur in approximately 1 in 250 of the population.[3] There are estimated to be 80,000 people with FH in Australasia with only 10–15% of the affected population being identified and the average general practice likely to encounter 50–100 patients with FH, every year.[4] In Australia, the majority of people with FH are undiagnosed, often not treated to adequate targets, and their families are not tested for the possible 50% of affected first-degree relatives.[5], [6], [7] If FH is left untreated, 50% of men will have CHD by age 50 years and 30% of women by age 60 years.[3]
General practitioners (GP) request over 90% of LDL-c measurements in the community,[8] and are crucial in the management of hypercholesterolaemia and other cardiovascular risk factors for the prevention of cardiovascular disease. Commonly used cardiovascular risk calculators do not accurately assess risk in those with FH. Awareness of FH however is low amongst both GPs and specialist physicians. [9], [10].
We have previously reviewed the role of FH detection in the community by suggesting a multidisciplinary approach including: a broader primary care awareness and education program; auditing of electronic patient information on general practice databases; and, via pathology providers, to highlight people with very elevated LDL cholesterol at risk of FH.[1]
Over the past 25 years, computer use by Australian GPs has increased such that nearly all general practices in Australia use a computer at their practice.[11] Routinely collected health care data in GP electronic health records (GP EHR) can be mined, to allow identification of disease states and population patterns. [12] General practice EHR systems vary across general practices and extraction systems are required to be able to coordinate with different systems. Some GP EHR systems are capable of effectively extracting clinical data although the quality of the data has been found to be of varying consistency. [13] A number of GP EHR approaches have been used in other countries with varying success [14], [15]
We present a simple method to improve detection and to enhance awareness of FH in Australian primary care by using an electronic extraction tool suitably designed for GP EHRs.
Section snippets
Data Source
De-identified data were obtained using electronic extraction tools in five Perth metropolitan practices over two years using the Canning Tool [16]. This extraction tool is able to extract data directly from more than seven types of GP software which is applicable to over 95% of available Australian general practices. We had access to the databases of five general practices with an average of 5.5 full time equivalent GPs (FTE). Full time equivalent is defined as 37 hours worked per week in
Results
We were able to extract data from 157,290 active patients (Table 2).
Of the total number of active patients, 4.84% (n=7605) of extracted patients had been prescribed statins, 1.39% (n=2191) had a diagnosis of vascular disease, 3.73% (n= 5869) had a diagnosis of lipid disorders of which 20 patients (approximately 1 in 7900 active patients) had been coded with a diagnosis of FH. Of those extracted patients prescribed statins 16.8% (n=1276 of 7605) had a persistent measured LDL-c recorded between
Discussion
Familial hypercholesterolaemia is poorly identified in general practice [10]. Applying a simple data extraction tool to general practice software, we were able to detect a cohort of general practice patients at high risk of FH. These patients will require a targeted clinical assessment to assess the likelihood of FH. It is expected that between 30 and 50% of such patients would be found to have phenotypic FH. Clinical assessment of these at risk patients is required using additional
Conclusion
A simple electronic data extraction tool with defined criteria from GP EHRs can successfully identify cases at high risk of FH. Further, it is a useful tool for helping GPs to identify patients at risk of cardiovascular disease to better achieve total cholesterol and LDL-c targets. Identified cases will require further clinical review and the DLCNS assessment to target patients at risk of FH. Such cases on review, can be provided with more intensive treatment and management, appropriate
Conflict of Interest
There is no identifiable conflict of interest.
References (29)
- et al.
Optimising the detection and management of familial hypercholesterolaemia: The central role of primary care and its integration with specialist services
Heart, Lung Circ
(2014) - et al.
A review on the diagnosis, natural history, and treatment of familial hypercholesterolaemia
Atherosclerosis
(2003) - et al.
Prevalence and treatment of familial hypercholesterolaemia in Australian communities
Inter J Cardiology
(2015) - et al.
Evaluation of cholesterol lowering treatment of patients with familial hypercholesterolemia: a large cross-sectional study in The Netherlands
Atherosclerosis
(2010) - et al.
Significant gaps in awareness of familial hypercholesterolemia among physicians in selected Asia-Pacific countries: a pilot study
J Clin Lipidol.
(2015 Jan-Feb) - et al.
Familial hypercholesterolaemia in primary care: knowledge and practices among general practitioners in Western Australia
Heart Lung Circ
(2014 Apr) - et al.
Improving identification of familial hypercholesterolaemia in primary care: derivation and validation of the familial hypercholesterolaemia case ascertainment tool (FAMCAT)
Atherosclerosis
(2015 Feb) Can patients be assessed for FH in Primary care?
Heart Lung Circ.
(2014)- et al.
Prospective analysis of LDL-C goal achievement and self-reported medication adherence among statin users in primary care
Clin Ther.
(2011 Sep) - et al.
Improving identification of familial hypercholesterolaemia in primary care: derivation and validation of the familial hypercholesterolaemia case ascertainment tool (FAMCAT)
Atherosclerosis.
(2015 Feb)
Detecting familial hypercholesterolaemia in general practice
Australian Family Physician
Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease Consensus Statement of the European Atherosclerosis Society
Eur Heart J.
Familial hypercholesterolaemia: a model of care for Australasia
Atherosclerosis
Opportunistic screening for familial hypercholesterolaemia via a community laboratory
Annals of Clinical Biochemistry
Cited by (22)
Clinical decision support for familial hypercholesterolemia (CDS-FH): Rationale and design of a cluster randomized trial in primary care
2022, American Heart JournalCitation Excerpt :However, some studies have found CDS interventions to be cost-effective and even cost-saving.14,20 CDS systems integrated within the electronic health records (EHR), machine learning algorithms and other forms of computer-based screening tools have previously been used to aid in the identification and management of patients with FH.21-51 However, to date, no randomized controlled trial has been conducted investigating the effects of an EHR-based CDS to identify patients with FH.
Electronic health records to facilitate continuous detection of familial hypercholesterolemia
2020, AtherosclerosisCitation Excerpt :In an outpatient setting, failure to recognize FH is often due to the ongoing use of lipid lowering therapy (LLT), which leads to misclassification due to reduced LDL-C levels [11]. In fact, this may well be one of the major reasons that previous efforts to improve FH-detection using electronic screening algorithms in electronic health records (EHR) have only provided modest results [8,12–15]. Since LDL-C is one of the strongest discriminatory factors in the clinical diagnosis of FH using the DLCN criteria, it is essential to use untreated LDL-C values in patients already using LLT [16].
The evaluation and management of patients with LDL-C ≥ 190 mg/dL in a large health care system
2020, American Journal of Preventive CardiologyHeterozygous familial hypercholesterolaemia in specialist centres in South Africa, Australia and Brazil: Importance of early detection and lifestyle advice
2018, AtherosclerosisCitation Excerpt :The generalisability of our findings are constrained by the sampling population being derived from specialist referral centres. FH patients in the primary care setting may be less treated and have a different spectrum of CVD and CVD risk factors [35,36]; we also did not report on homozygous FH and children. A gene founder effect for FH is well-recognised in South Africa [14]; the outcomes of treatment have been conjointly reported with the UK [37].
Detection of familial hypercholesterolemia in patients from a general practice database
2017, Atherosclerosis SupplementsCitation Excerpt :In the current practice of general medicine, the identification of family forms can be only opportunistic (that is, as a result of contacts required by the patient for various reasons), since the active screening procedures are still not compatible with the available resources. This means that GPs have a great diagnostic potential [14,15]: the high number of contacts with most of the population assisted by the GP in a few years allows a systematic and effective approach to this problem, simply applying what is already recommended by the guidelines as normal good clinical practice. The cut-off of 190 mg/dL for LDL-C would be a first screening.