Improving identification of familial hypercholesterolaemia in primary care: Derivation and validation of the familial hypercholesterolaemia case ascertainment tool (FAMCAT)

Weng SF, Kai J, Neil HA et al.
Atherosclerosis 238 (2015) 336e343

Familial hypercholesterolaemia (FH) is the commonest autosomal dominant disorder, and about 1/200-1/500 have been estimated to have heterozygous FH (HeFH) [1]. People with HeFH have high serum cholesterol levels, and consequently a strongly increased risk of mortality from coronary heart disease (CHD) as compared to unaffected adults [2]. Treatment with statins can greatly improve life expectancy, thus early identification and treatment is important [3].
Despite established clinical assessment guidelines like the Simon-Broome Register criteria in the United Kingdom [2] and the Dutch Lipid Clinic criteria in the rest of Europe [4], it is estimated that in most European countries only 20% of FH cases are detected [1].
Developing a validated tool based on data routinely available in primary care records may help to improve identification of FH in the general population. This study aimed to develop and validate a predictive modelling tool (FAMCAT) to allow clinicians to identify individuals with the highest likelihood of having FH. Prospectively collected data of 2.971.562 individuals in the Clinical Practice Research Datalink (CPRD) were used. A random sample of 75% of the complete cohort (n=2.228.562) was taken as the derivation cohort, and the remaining 25% (n=742.851) functioned as validation cohort.

Main results

• A multivariate regression analysis revealed that total cholesterol > 75 mmol/L or LDL-c>4.9 mmol/L were the strongest indicators of being diagnosed with FH (Males [M]: Adjusted OR (AOR) 37.97, 95%CI: 30.99-46.52; Females [F]: AOR: 43.08, 95%CI: 35.43-52.40).
• Having a family history of FH was also strongly predictive of diagnosis of FH (M: AOR: 10.99, 95% CI: 9.23-13.08; F: AOR 8.21, 95%: CI 7.01-9.61), and to a lesser extent having family histories of raised cholesterol levels or MI.
• Being prescribed a high potency statin during the cholesterol assessment was also a strong predictor of FH (M: AOR: 10.64, 95%CI 8.95-12.65; F: AOR: 6.31, 95%CI 5.22-7.63).
• The FAMCAT model showed high model performance (AUC 0.860, 95%CI: 0.848-0.871), with improved discrimination as compared to models based on variables from the Dutch Lipid Clinic or Simone-Broome.

Conclusion

A new prediction tool for familial hypercholesterolaemia was developed and validated. This new tool, FAMCAT, can identify individuals in primary care with the highest probability of the condition. These individuals can be further clinically assessed, with referral for diagnosis and preventive care as appropriate. This will allow efficient use of limited resources. FAMCAT uses coded variables that are routinely available in general practice electronic health records, which enhances the discriminatory accuracy because the data are generally well recorded.

Find this article online

References

[1] B.G. Nordestgaard, M.J. Chapman, S.E. Humphries, et al., 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. 34 (45) (2013) 3478e3490.
[2] Simon-Broome Register Group, Mortality in treated heterozygous familial hypercholesterolaemia: implications for clinical management, Atherosclerosis 142 (1) (1999) 105e112.
[3] A. Neil, J. Cooper, J. Betteridge, et al., Reductions in all-cause, cancer, and coronary mortality in statin-treated patients with heterozygous familial hypercholesterolaemia: a prospective registry study, Eur. Heart J. 29 (21) (2008) 2625e2633.
[4] Z. Reiner, A.L. Catapano, G. De Backer, et al., ESC/EAS guidelines for the management of dyslipidaemias: the task force for the management of dyslipidaemias of the European Society of Cardiology (ESC) and the European Atherosclerosis Society (EAS), Eur. Heart J. 32 (14) (2011) 1769e1818.