| アブストラクト | BACKGROUND: Serious muscle disorders are a concerning adverse event of statins, contributing to low statin uptake and poor adherence in the prevention of cardiovascular disease (CVD). We aimed to derive and validate a model predicting the risk of serious muscle disorders in individuals eligible for statin treatment, to better inform appropriate treatment decision making. METHODS: This was a retrospective cohort study using electronic health record data from the Clinical Practice Research Datalink (CPRD), with its Gold database for model derivation and Aurum database for external validation. Men aged 50 years or older and women aged 60 years or older, who were registered with CPRD between Jan 1, 1998 and Dec 31, 2018 in England, were included. A Fine-Gray model was developed to predict the 1-year, 5-year, and 10-year risks of serious muscle disorders associated with hospitalisation or death, adjusting for competing mortality risk. Model discrimination was assessed using C-index and D-statistic, calibration using the expected to observed (E/O) ratio, calibration slope, and calibration plot, and clinical utility using decision curve analysis. FINDINGS: A total of 1 785 207 individuals were included in the model derivation cohort and 3 889 504 individuals in the external validation cohort. The model consisted of 22 predictors, including statin prescription. The predicted risk of serious muscle disorders in most individuals was low (99.6% of the validation cohort had a 10-year risk below 10%). The model showed overall good discrimination and calibration, with C-index 0.782 (IQR 0.782-0.782), D-statistic 2.176 (95% CI 2.139-2.212), E/O ratio 1.151 (IQR 1.131-1.171), and calibration slope 1.063 (95% CI 1.052-1.073) for the 10-year prediction in the external validation. Decision curve analysis suggested a favourable net benefit of the model-assisted clinical decision making. INTERPRETATION: This model provides reliable prediction of personalised risk of serious muscle adverse events of statins, which, together with existing prediction tools for personalised CVD risk, could help patients and physicians better understand the benefit-to-harm balance of statins for an individual and support well informed shared decision making on statin treatment. FUNDING: British Heart Foundation, Wellcome Trust, and Royal Society. |
|---|
| ジャーナル名 | The Lancet. Digital health |
|---|
| Pubmed追加日 | 2026/6/26 |
|---|
| 投稿者 | Cai, Ting; Hirst, Jennifer A; Nicholson, Brian D; McManus, Richard J; Hobbs, F D Richard; Sheppard, James P; Koshiaris, Constantinos |
|---|
| 組織名 | Nuffield Department of Primary Care Health Sciences, University of Oxford,;Oxford, UK.;Oxford, UK; Brighton and Sussex Medical School, University of Brighton and;University of Sussex, Brighton, UK.;Oxford, UK. Electronic address: james.sheppard@phc.ox.ac.uk.;Oxford, UK; Department of Primary Care and Population Health, University of;Nicosia Medical School, Nicosia, Cyprus. |
|---|
| Pubmed リンク | https://www.ncbi.nlm.nih.gov/pubmed/42350256/ |
|---|
