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.