Article Text

Cohort study
Statin potency associated with incident diabetes in a real-world evaluation
  1. Dylan L Steen,
  2. Deepak L Bhatt
  1. VA Boston Healthcare System and Brigham and Women's Hospital, Boston, Massachusetts, USA
  1. Correspondence to: Dr Deepak L Bhatt, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA; dlbhattmd{at}

Statistics from

Commentary on: OpenUrlAbstract/FREE Full Text


For widely prescribed medications, such as statins, even the development of uncommon adverse events may have significant public health ramifications. Recent meta-analyses of randomised controlled trials (RCTs) have demonstrated an increased incidence of diabetes in patients taking statins. These have demonstrated a dose–effect relationship, but whether different statins have differential effects on the risk of diabetes is still unclear. In this study, Carter and colleagues further investigate the dose–effect relationship of statin-induced diabetes and whether different statins carry distinct risks of diabetes in a large, real-world population.


The authors examined healthcare records of more than 1.5 million adults aged 66 years or above in Ontario, Canada from August 1997 to March 2010. The analysis excluded patients with established diabetes and those prescribed a statin within 1 year prior to the start of the study. Linked administrative healthcare databases were used to provide information on demographics, clinical characteristics, diagnostic and procedural information, physician billing, vital status and diagnosis of diabetes. Cox proportional hazards regression was used to estimate the relationship between statins and incident diabetes. For statistical analyses, pravastatin was used as the reference comparator.


The authors identified 471 250 patients without a history of diabetes who were newly treated with a statin. Of the baseline characteristics reported, the year of cohort entry and history of prior coronary disease differed between subgroups prescribed individual statins.

The crude incidence of diabetes per 1000 person-years in those prescribed rosuvastatin, atorvastatin and simvastatin was 34.2, 30.7 and 26.2, respectively. This was significantly higher than for pravastatin, fluvastatin or lovastatin, which had incidences of 22.6, 21.5 and 21.8, respectively. After adjustment for baseline differences, rosuvastatin, atorvastatin and simvastatin were associated with 18%, 22% and 10%, respectively, increased risk of incident diabetes compared with pravastatin. The association of statins with incident diabetes was similar when analysed by indication for primary versus secondary prevention.

High-potency statins (ie, atorvastatin and rosuvastatin) were associated with a 22% increased risk of diabetes compared with pravastatin, while a moderate potency statin (ie, simvastatin) demonstrated an 11% increase. Those defined as low-potency statins (ie, fluvastatin and lovastatin) did not demonstrate an increased risk compared with pravastatin. Defining potency using dose in addition to statin type, high-potency therapy demonstrated a 30% increased risk and moderate-potency therapy a 22% increased risk compared with low-potency therapy. Among individual statins, after adjusting for dose atorvastatin and simvastatin were still associated with increased risk, but rosuvastatin was not.


The authors have validated the finding of a higher risk of statin-induced diabetes with more versus less potent statin therapy. These findings have particular relevance as the utilisation of potent statin therapy appropriately continues to increase. Despite being a real-world population, the authors found a 10–22% increased risk, reassuringly similar to (and not higher than) the degree of excess risk found in randomised comparisons of more versus less intensive statin therapy.1–3 A strength of this current study was the use of sensitivity analyses with varying definitions of potency and diabetes to assess the robustness of the findings. The primary limitations relate to the inability to exclude residual confounding, as certain baseline characteristics were neither reported nor adjusted in the analyses.

Whether individual statins carry different risks of diabetes independent of their low-density lipoprotein cholesterol (LDL-C)-lowering potency still remains unclear. In this study, the risk of diabetes correlated with the LDL-C-lowering efficacy of the statin. For rosuvastatin, an increased risk was not found when adjusted for dose; while interesting, this finding needs validation as these subgroup analyses were exploratory. Studies comparing statins to other medications that can reduce LDL-C will be able to provide insight into whether the diabetes risk is specific to statins or is related to the extent of LDL-C lowering.

The risk of statin-induced diabetes must be balanced against the cardiovascular benefits of more potent statin therapy when needed to achieve lowering of LDL-C to target. One important question is whether this phenomenon is simply a matter of worsened glycaemic control or whether these subjects are developing true diabetes with its systemic implications and worsened prognosis.


View Abstract


  • Competing interests DLB is on the advisory board for Elsevier Practice Update Cardiology, Medscape Cardiology, and Regado Biosciences. He is also on the board of directors for the Boston VA Research Institute, the Society of Cardiovascular Patient Care and is chair of the American Heart Association Get With The Guidelines Steering Committee. He has received honoraria from the American College of Cardiology (Editor, Clinical Trials, Cardiosource), Belvoir Publications (Editor in Chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Population Health Research Institute (clinical trial steering committee), Slack Publications (Chief Medical Editor, Cardiology Today's Intervention), and WebMD (CME steering committees). Additionally, he is on Data Monitoring Committees for Duke Clinical Research Institute, Mayo Clinic and the Population Health Research Institute. He has received research grants from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Sanofi, and The Medicines Company and also has done unfunded research with FlowCo, PLx Pharma, and Takeda. Lastly, he is on the Executive Committee of ODYSSEY (a cardiovascular outcome study of a PCSK9 inhibitor made by Sanofi).

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.