Predicting non-adherence in patients with familial hypercholesterolemia

Research output: Contribution to journalArticleAcademicpeer-review

20 Citations (Scopus)


Familial hypercholesterolemia (FH) is an autosomal dominant disorder associated with a high risk of premature coronary heart disease (CHD). CHD prevention consists of lifestyle changes combined with lifelong statin treatment. Good adherence to statins reduces the risk of events substantially. This study was designed to identify determinants of non-adherence and to develop a model predicting non-adherence. A single centre survey included all consecutive heterozygous FH patients above age 18 years, who were treated by a specialized team in the outpatient clinic of a university hospital in The Netherlands between 2008 and 2009. In addition to clinical data, patients completed a questionnaire concerning medication adherence. We analyzed 321 patients (169 women) with a statin prescription whose mean age was 46 +/- 14 years (+/- S.D.), and 13 % of the patients had CHD. The untreated mean total cholesterol was 10 +/- 2.3 mmol/l. On average, patients were ten years on cholesterol-lowering therapy (range 1-29 years). Adherence was reported by 89 % of the patients (> 90 % adherence). Non-adherence was associated with younger age (OR = 10.64, 95 % CI 2.86-39.68), high total cholesterol level during prescription (OR = 4.29, 95 % CI 1.86-9.89) and a relatively low untreated total cholesterol level (OR = 3.94 95 % CI 1.39-11.14). A prediction model based on these three determinants had a c-index of 0.78 and a calibration with P = 0.88. Based on three independent determinants, a prediction model is developed to identify non-adherent FH patients. This model needs to be tested in future prospective research. It might be a first step in improving statin adherence in this extremely high risk group.
Original languageUndefined/Unknown
Pages (from-to)391-397
Number of pages7
JournalEuropean Journal of Clinical Pharmacology
Issue number4
Publication statusPublished - 2014

Cite this