Written by Keen Meds · Reviewed by Licensed Clinical Providers · March 1, 2026
Key Insight: The Vigen 2013 JAMA study was a retrospective observational study of veterans — not a randomized controlled trial — and was later found to contain a significant data error in which 1,132 women had been incorrectly included in the testosterone-treated group.
In November 2013, a study published in the Journal of the American Medical Association (JAMA) sent a wave of alarm through both the medical community and mainstream media. The headline message was stark: testosterone therapy appeared to raise the risk of heart attack, stroke, and death in older men. Within weeks, the FDA was fielding calls for action, lawsuits were being filed, and men across the country were stopping their prescriptions.
Over a decade later, the full picture looks considerably more complicated. The Vigen study had real methodological problems, contained a verified data error, and was ultimately superseded by a large randomized controlled trial that reached a very different conclusion. Understanding what the Vigen study actually showed, and what it could not show, is essential for any man evaluating testosterone therapy today.
About the Study: Design and Why It Was Done
Vigen et al. published their findings in JAMA on November 6, 2013. The study was a retrospective observational analysis, meaning researchers went back through existing medical records rather than designing a prospective experiment with controlled conditions.
The population: 8,709 male veterans who had undergone coronary angiography (a procedure to examine the arteries of the heart) at Veterans Affairs medical centers between January 2005 and December 2011 and who had low testosterone levels documented. Coronary angiography is typically performed when a physician suspects significant coronary artery disease, so by definition, these men had known or suspected cardiac problems before the study began.
Of the 8,709 men, 1,223 received testosterone therapy after their angiography. The remaining 7,486 did not. Researchers then tracked outcomes over a median follow-up of about 531 days.
The study was motivated by a genuine clinical question: as testosterone prescriptions climbed steeply in the United States during the late 2000s and early 2010s, was there any signal of cardiovascular harm in real-world populations? Because conducting a large randomized trial takes years, researchers turned to existing records to look for preliminary signals.
This design approach is legitimate in many contexts. Retrospective observational studies can identify patterns worth investigating. The problem is that they carry inherent limitations that the media coverage of this study largely ignored.
What the Study Found
The published findings showed that 25.7% of men in the testosterone-treated group experienced a major adverse cardiovascular event (defined as myocardial infarction, stroke, or death) compared to 19.9% of the non-treated group. The risk difference persisted after statistical adjustment for covariates.
On its face, this appeared to support a connection between testosterone therapy and cardiovascular harm. The study received extensive media coverage, was cited in congressional testimony, and contributed directly to an FDA advisory committee review of testosterone products in 2014.
However, within months of publication, a critical problem emerged. In February 2014, JAMA published an erratum acknowledging that the study's data had a significant error: 1,132 women had been incorrectly included in the testosterone-treated group, likely due to a database classification error. Because women were counted among the testosterone users, the testosterone group's reported outcomes were affected in ways that cannot be fully disentangled.
The corrected analysis was published, and while it still showed elevated risk in the testosterone group, the confidence in the underlying data had been substantially undermined. Many researchers who had already been skeptical of the methodology now pointed to the erratum as a reason to discount the study's conclusions.
What the Critics Said: Limitations and Context
Even before the data error was discovered, the Vigen study attracted substantial criticism from researchers in endocrinology, urology, and cardiovascular medicine. The core objection was confounding by indication.
Confounding by indication refers to a well-known bias in observational research where the treatment itself is prescribed to sicker patients. In this case, if physicians were more likely to prescribe testosterone to men who also had more severe cardiovascular disease, fatigue, obesity, or other risk factors, those men would be expected to have worse outcomes regardless of testosterone use. Attempting to statistically adjust for this problem is possible, but it cannot fully account for all the unmeasured reasons one man received testosterone and another did not.
The study population also represented an atypical group. These were men referred for coronary angiography, meaning all of them had significant coronary artery disease or suspected disease at baseline. Applying findings from this high-risk population to the broader population of men considering TRT, many of whom are middle-aged and have no established cardiovascular disease, involves a logical leap the study's design could not support.
Multiple editorials in major journals challenged the study's conclusions. Researchers including Abraham Morgentaler at Harvard Medical School published pointed critiques of the methodology. The journal itself took the unusual step of publishing a formal concern from over 200 researchers calling for the paper to be reviewed or retracted.
None of this invalidated the concern that motivated the study. The scientific community agreed that a properly designed randomized controlled trial was needed to settle the cardiovascular safety question definitively.
What This Means for You Today
The Vigen study served a purpose in the history of testosterone research: it made the cardiovascular safety question urgent and helped justify the funding and design of the TRAVERSE trial, which enrolled 5,246 men with hypogonadism and existing cardiovascular disease or high cardiovascular risk and followed them for a median of 3.2 years in a rigorous randomized controlled design.
TRAVERSE, published in the New England Journal of Medicine in June 2023, found a cardiovascular hazard ratio of 0.96 compared to placebo, confirming that testosterone therapy did not meaningfully increase major cardiovascular event risk. Based substantially on that evidence, the FDA removed the black box cardiovascular warning from testosterone products in March 2025.
The Vigen study is not irrelevant history. It is a useful example of how observational data can generate hypotheses and drive research forward, while also misleading clinicians and patients if treated as definitive proof. For men evaluating testosterone therapy, the important takeaway is that the most rigorous available evidence does not support the conclusion that testosterone therapy meaningfully raises cardiovascular risk in men who are appropriate candidates for treatment.
Frequently Asked Questions
Was the Vigen 2013 JAMA study retracted?
No. The study was not retracted, though an erratum acknowledging the inclusion of 1,132 women in the testosterone group was published in 2014. The corrected analysis was published, but the study's methodology continued to be widely criticized.
Did the Vigen study prove that testosterone causes heart attacks?
No. As a retrospective observational study, it could not establish causation. It identified a statistical association in a specific high-risk population. The TRAVERSE randomized controlled trial, published in 2023, is the highest-quality evidence available and showed no significant increase in major cardiovascular events.
Why did so many doctors stop prescribing testosterone after 2013?
The Vigen study received intense media coverage, and the FDA subsequently required label updates and initiated a safety review. Many physicians responded cautiously to the uncertainty. The landscape has changed significantly since then, particularly following the TRAVERSE results and the FDA removing the cardiovascular black box warning in March 2025.
What is "confounding by indication" and why does it matter for this study?
Confounding by indication means that sicker patients are more likely to receive a treatment, making the treatment appear harmful when the underlying illness is responsible for worse outcomes. In the Vigen study, the men who received testosterone may have been prescribed it precisely because they had more severe fatigue, weight gain, and cardiovascular disease, making their outcomes worse independently of testosterone.
What the Research Means for Your Treatment Options
The Vigen study contributed to years of unnecessary uncertainty around testosterone therapy. The TRAVERSE trial provided the rigorous evidence needed to clarify the cardiovascular safety profile of TRT in men with hypogonadism. For appropriately selected men, the evidence base today is substantially stronger than it was in 2013.
Keen Meds connects you with licensed clinical providers for a telehealth evaluation and lab review. If your testosterone is clinically low, you may qualify for Testosterone Spray Rx, a needle-free transdermal testosterone program.
