Results of a randomized controlled trial on statin use in dialysis patients had no influence on statin prescription

Abstract

Randomized trials provide high-quality evidence for patient care. The Der Deutsche Diabetes Dialyse Studie (4D), a randomized study which demonstrated no benefit of statins among diabetic patients receiving hemodialysis, was published in July 2005. To determine effects of this study we conducted a retrospective, population-based, time series analysis with change-point regression to see if the rate of statin prescription to dialysis patients had been modified. We linked health administrative data for all diabetic hemodialysis patients living in Ontario, Canada, with similar characteristics to the 4D patient cohort. During the nearly 11-year period prior to study publication, the rate of statin use increased almost 14-fold, from 43 to 597 per 1000 patients. For 2.5 years after study publication, rather than diminish, statin use continued to rise to an absolute rate of 676 per 1000 patients. These temporal patterns in statin use closely mimicked trends in the diabetic population not receiving dialysis. The 4D trial had no impact on statin use when we restricted the analysis to incident statin prescriptions or expanded the characteristics of the dialysis patients considered for study. Thus, we found that publication of a large, expensive, randomized controlled trial in patients receiving hemodialysis had no immediate impact on clinical practice. The use of a common cardiovascular medication in this patient population appears to be influenced by other factors.