[Skip to Content]
Access to paid content on this site is currently suspended due to excessive activity being detected from your IP address 54.163.129.96. Please contact the publisher to request reinstatement.
[Skip to Content Landing]
Views 159
Citations 0
Comment & Response
March 2014

Cataracts and Statin Use—Reply

Author Affiliations
  • 1VA North Texas Health Care System, Dallas
  • 2Department of Internal Medicine, University of Texas Southwestern Medical Center, Dallas
  • 3Department of Clinical Sciences, University of Texas Southwestern Medical Center, Dallas
  • 4Wilford Hall Ambulatory Surgery Center, San Antonio, Texas
  • 5San Antonio Military Medical Center, San Antonio, Texas
JAMA Ophthalmol. 2014;132(3):365-366. doi:10.1001/jamaophthalmol.2014.75

In Reply We appreciate the feedback from Drs Strandberg and Tarkkanen. Our study’s conclusion of association between statin use and cataract diagnosis was questioned because our outcome was based on medical records rather than a standardized diagnostic protocol. We agree that randomized clinical trials (RCTs) that use standardized diagnostic protocols are preferable to studies using medical records databases. Despite being the gold standard in determining efficacy of medications, RCTs have limitations in assessing adverse events (AEs).1 Randomized clinical trials are usually only powered to detect efficacy, not AEs. They may exclude those individuals at greatest risk for AEs and have excluded as many as 30% of those screened for the study.2 Withdrawal from trials also may result in an underestimation of AEs; an average of 10% of patients drop out from studies, and these patients are potentially those with the most significant AEs.2 Additionally, RCTs are typically shorter and may be prematurely terminated once benefits are documented.1 Despite their rigorousness, RCTs of statins are not devoid of bias3 and potential conflict of interest for their sponsors.4 Sometimes RCTs use questionnaires to collect information, but patients’ surveys may be biased owing to selective nonresponse.5

First Page Preview View Large
First page PDF preview
First page PDF preview
×