Skip to content
Buy This GLP
← Research
Safety

Erectile dysfunction: a finding the study's own bias check removed

The raw hazard ratio was 1.26 and statistically significant. The authors then tested their method against impossible outcomes, and the signal did not survive.

Wesley Jenkins6 min read
Erectile dysfunction, vs DPP-4 drugs1.0as measured1.26after bias calibrationno longer significantThe authors built a bias detector in, and it fired.

Sexual function is one of the least studied effects of these drugs, and what little exists is collected in four questions about contraception, one answered. A target trial emulation in a US health system asked directly whether starting a GLP-1 drug was followed by more erectile dysfunction than starting a DPP-4 inhibitor [1].

The raw answer was yes. Incidence was 35.2 per 1,000 person-years among GLP-1 initiators against 28.0 among DPP-4 initiators, a hazard ratio of 1.26 (95% CI 1.08–1.46). The finding held across subgroups, sensitivity analyses and an external validation cohort.

That technique is worth understanding because most observational research does not do it. A study without a negative control reports its 1.26 and stops; the reader has no way to tell how much of it is real. Here the authors quantified their own residual confounding and published the corrected result alongside the raw one, which is the opposite of the practice criticized in the prostate cancer cohort, where a large mortality difference arrived with no such check.

There is also a plainer explanation for part of the gap. Erectile dysfunction here is a diagnostic code, which requires a man to mention the problem and a clinician to write it down. Men who have just started an injectable weight drug are in front of clinicians more often, with more appointments in which something might come up. That is detection, not disease, and it is precisely the sort of bias a calibration step is designed to remove.

The honest summary for a buyer is that this study looked for a signal, found one, tested whether its own method could be trusted, and concluded it could not support the finding. Sexual side effects remain poorly characterized for this drug class either way, which is itself worth knowing before assuming silence means safety — a gap that also runs through what patients said when asked and responding is not one thing.

Frequently asked

Do GLP-1 drugs cause erectile dysfunction?
This study does not establish that. The raw association was a hazard ratio of 1.26, but after the authors calibrated against negative control outcomes it was attenuated and no longer significant.
What is negative control calibration?
Testing the same analysis against outcomes the drug could not plausibly cause. If the method finds effects there, it is producing spurious associations, and the size of those is used to correct the real estimate.
Could the raw difference be explained another way?
Partly by detection. The outcome is a diagnostic code, which requires a patient to raise the issue and a clinician to record it, and men starting a new injectable see clinicians more often.

Sources

  1. [1] Tang H, Lu Y, Zhang B, Zhang D, Asch DA, Chen Y (2026). GLP-1 receptor agonist and risk of erectile dysfunction in men with type 2 diabetes: a target trial emulation EClinicalMedicine. PMID 42005929

More in Safety