# Do ED drugs lower Alzheimer's risk? What the studies can and cannot show > That has not been shown. Large insurance-record studies from 2021 and 2024 linked sildenafil and similar ED drugs with fewer Alzheimer's diagnoses, but those were associations. Later studies using stricter comparisons and genetic methods found no lower risk, and no trial has shown that sildenafil prevents Alzheimer's. Viagra is approved for ED, not for dementia. _Last updated: 2026-10-18_ --- Status: FDA-approved prescription medicine for ED; not approved to prevent or treat Alzheimer's disease. Source: Viagra prescribing information (Pfizer), DailyMed, checked 2026-10-17. **In short:** in 2021, a large study of insurance records found that people who took sildenafil, the drug in Viagra, were diagnosed with Alzheimer's disease less often than people who did not. The finding made headlines, and a UK study in 2024 reported a smaller version of the same pattern in men with ED. But those studies show an association, not a cause, and their own authors said so. Studies that used stricter comparisons, and studies that used genetic data, found no lower risk. No randomized trial has shown that sildenafil prevents Alzheimer's, and the Viagra label lists one use: erectile dysfunction ([Viagra label, section 1](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0b0be196-0c62-461c-94f4-9a35339b4501)). | Study (year) | Data and design | What it reported | | ----------------------------------------- | ------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------ | | Fang et al., Nature Aging (2021) | US insurance claims, 7.23 million people; observational | Sildenafil users had 69% lower risk of an Alzheimer's diagnosis | | Gohel et al., J Alzheimer's Dis (2024) | Two US claims databases; observational | Lower Alzheimer's rates in sildenafil users than in users of some heart and blood pressure drugs | | Adesuyan et al., Neurology (2024) | UK primary-care records, 269,725 men with ED | 18% lower risk; the gap shrank and lost significance with a 3-year delay | | Desai et al., Brain Communications (2022) | US Medicare, people with pulmonary hypertension; active comparator | No lower risk of Alzheimer's or related dementia | | Gronich et al., Neuroepidemiology (2024) | Israeli health records, including 133,336 men with ED | No lower dementia risk | | Alcalde-Herraiz et al., Aging Cell (2025) | Genetic (Mendelian randomization) analysis | No evidence that PDE5 inhibition lowers Alzheimer's risk | Sources are linked in each section below. ## What did the studies report? The studies that started the story reported fewer Alzheimer's diagnoses among people who took sildenafil. The best known, by Fang and colleagues at the Cleveland Clinic, used a computer model to rank more than 1,600 FDA-approved drugs, weighed other factors and picked sildenafil as the top candidate, and then checked insurance claims for 7.23 million people; sildenafil use was "significantly associated with a 69% reduced risk of AD" over about six years ([Fang et al., 2021](https://pubmed.ncbi.nlm.nih.gov/35572351/); [NIA summary](https://www.nia.nih.gov/news/viagra-associated-reduced-risk-alzheimers-disease)). The National Institute on Aging noted that the sildenafil users were mostly men, and that the team also saw changes in lab-grown nerve cells. Several later observational studies pointed the same way: - The same group repeated the analysis in two more US databases and reported lower Alzheimer's rates in sildenafil users than in people starting certain blood pressure and heart drugs, such as spironolactone ([Gohel et al., 2024](https://pubmed.ncbi.nlm.nih.gov/38427489/)). - A UK study of 269,725 men newly diagnosed with ED found an adjusted hazard ratio of 0.82 for Alzheimer's in men who started a PDE5 inhibitor, about 18% lower risk, with lower figures in men issued more than 20 prescriptions ([Adesuyan et al., Neurology, 2024](https://www.neurology.org/doi/10.1212/WNL.0000000000209131)). - A US case-control study of patients aged 65 and older found that patients with Alzheimer's or related dementia were less likely to have been exposed to a PDE5 inhibitor than patients without it ([Henry and Pellegrino, 2023](https://pubmed.ncbi.nlm.nih.gov/37851623/)). - A 2025 meta-analysis pooled 5 observational studies with 885,380 patients and reported a hazard ratio of 0.47 for sildenafil use ([Chua et al., 2025](https://pmc.ncbi.nlm.nih.gov/articles/PMC11984433/)). Pooling studies adds up their numbers, but it also carries over the weaknesses of each one. Every one of these papers called for randomized trials. The Fang paper says the association "does not establish causality or its direction" ([Fang et al., 2021](https://pubmed.ncbi.nlm.nih.gov/35572351/)). ## Why can an observational study not prove a drug prevents dementia? An observational study watches what happens to people who chose, or were prescribed, a drug; it does not assign the drug at random, so the users can differ from the non-users in ways that also affect dementia risk. Our guide to [how to read ED evidence](/how-to-read-ed-evidence/) covers this in general terms. For sildenafil and Alzheimer's, three specific problems come up in the research itself. **Confounding by indication.** The comparison group matters. Two researchers writing in the Journal of Alzheimer's Disease warned that comparing sildenafil users with users of certain heart drugs can create a gap that reflects differences between the groups rather than an effect of sildenafil, if the comparison drug is not a fair match ([Wu and Swardfager, 2024](https://pubmed.ncbi.nlm.nih.gov/38995794/)). **Reverse causation.** There is a gap between when Alzheimer's begins and when it is diagnosed. The UK authors noted that early, unrecorded memory problems could lead a man to stop an ED medicine or be taken off it, which can skew a comparison. When they started counting only after a 3-year delay, the hazard ratio moved to 0.93, with a confidence interval that included no difference (0.80 to 1.08). They wrote that this may reflect bias in the main result, though a longer delay also leaves fewer cases to analyze ([Adesuyan et al., 2024](https://www.neurology.org/doi/10.1212/WNL.0000000000209131)). **Unmeasured differences.** The same authors wrote that they could not account for physical and sexual activity, which may be linked both to getting an ED prescription and to dementia risk. Because they also could not adjust for ethnicity, they wrote that "a level of residual confounding may affect our results." They also noted that prescription records do not show whether a man took the medicine. | Problem | What it means here | Where it is discussed | | ------------------------- | ------------------------------------------------------------------------------ | ------------------------------------ | | Confounding by indication | The comparison drug's users may have a different dementia risk to begin with | Wu and Swardfager, 2024 | | Reverse causation | Early, undiagnosed Alzheimer's may affect who gets or keeps an ED prescription | Adesuyan et al., 2024 (3-year delay) | | Unmeasured differences | Activity, health and habits that records do not capture | Adesuyan et al., 2024 (limitations) | | Exposure uncertainty | A prescription record does not show the pills were taken | Adesuyan et al., 2024 (limitations) | ## What did later studies find? Studies designed to reduce those biases mostly found no lower risk. The National Institutes of Health's DREAM study compared Medicare patients with pulmonary hypertension who started sildenafil or tadalafil against similar patients who started a different class of lung drug, balanced 76 possible confounders, and ran four analyses aimed at specific biases, including reverse causation; none showed a reduced risk of Alzheimer's or related dementia ([Desai et al., 2022](https://pubmed.ncbi.nlm.nih.gov/36330433/)). NIH said in a media advisory that sildenafil and tadalafil "may not be suitable candidates for treating dementia" ([NIH, 2022](https://www.nih.gov/news-events/news-releases/nih-dream-study-finds-viagra-cialis-do-not-reduce-risk-alzheimer-s-related-dementias)). The same study found no evidence that sildenafil corrected Alzheimer's-related changes in most of its cell tests. An Israeli study looked at the question twice. Men starting tadalafil for an enlarged prostate had the same dementia risk as men starting a different prostate medicine (hazard ratio 0.99), and in 133,336 men with ED followed for nearly 8 years on average, PDE5 inhibitor use was not linked with lower dementia risk (hazard ratio 0.95, confidence interval 0.86 to 1.04) ([Gronich et al., 2024](https://pubmed.ncbi.nlm.nih.gov/38952132/)). Genetic studies take a different route. Mendelian randomization uses natural gene variants that mimic a drug's effect, which avoids some of the choices that skew prescription records. A 2025 analysis across four large genetic datasets found "no evidence of a causal association" between PDE5 inhibition and Alzheimer's risk ([Alcalde-Herraiz et al., 2025](https://pubmed.ncbi.nlm.nih.gov/41078087/)). A second genetic study reported mixed signals, including higher odds of Alzheimer's but a trend toward lower odds of vascular dementia, and its authors urged more research before any trials ([Brennan and Tinworth, 2025](https://pubmed.ncbi.nlm.nih.gov/39951190/)). Genetic proxies are not the same as taking a pill, so these results are one more line of evidence, not a final answer. ## Has a clinical trial tested sildenafil for Alzheimer's? We found no completed randomized trial showing that sildenafil prevents or slows Alzheimer's disease. A search of ClinicalTrials.gov on 2026-10-17 for sildenafil with Alzheimer's or dementia returned the DREAM data-analysis project and small studies of blood flow in the brain, not a prevention trial. The closest test is a different PDE5 inhibitor, studied under the code name AR1001: in a 2025 randomized, placebo-controlled trial of 210 people with mild to moderate Alzheimer's, it did not beat placebo on memory and thinking scores or overall clinical change after 26 weeks, though some blood markers moved ([Greeley et al., 2025](https://pubmed.ncbi.nlm.nih.gov/40912996/)). That drug is not sildenafil, and a blood marker is not the same as a better outcome for the person. ## Should anyone take sildenafil for this? No study supports taking sildenafil to prevent Alzheimer's, and its label does not list that use: Viagra "is indicated for the treatment of erectile dysfunction" ([Viagra label, section 1](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0b0be196-0c62-461c-94f4-9a35339b4501)). Sildenafil is a prescription medicine with real risks; it must never be combined with nitrate medicines ([Viagra label, section 4](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0b0be196-0c62-461c-94f4-9a35339b4501)), and it has [side effects](/sildenafil-side-effects/) and [interactions](/sildenafil-drug-interactions/) a prescriber checks before writing it. If you take it for ED, the reasons to take it are the ones on the label, which our page on [what sildenafil is](/what-is-sildenafil/) covers. The fair summary is that the dementia question is still open as research, and the strongest studies so far lean toward no effect. If a headline or an online seller presents sildenafil as brain protection, that goes beyond what the evidence shows. This page sits in our [myths and evidence library](/myths/), next to other claims that sound like findings but are not. ## When to talk to a clinician Talk to a clinician if you are worried about memory or thinking changes, or if you want to know whether an ED medicine is safe with your other health conditions and medicines. Those are separate questions, and neither is answered by an observational study. Our guide to [talking to a doctor about ED](/talking-to-a-doctor-about-ed/) has ways to start the conversation. ## Sources - Fang J, Zhang P, Zhou Y, et al. Endophenotype-based in silico network medicine discovery combined with insurance record data mining identifies sildenafil as a candidate drug for Alzheimer's disease. Nature Aging 2021;1(12):1175-1188. Claims data for 7.23 million people, hazard ratio 0.31; statement that the association does not establish causality: [pubmed.ncbi.nlm.nih.gov/35572351](https://pubmed.ncbi.nlm.nih.gov/35572351/) - National Institute on Aging, "Viagra associated with reduced risk of Alzheimer's disease," December 14, 2021. Plain-language summary; users mostly men; researchers say prevention was not shown: [nia.nih.gov](https://www.nia.nih.gov/news/viagra-associated-reduced-risk-alzheimers-disease) - Gohel D, Zhang P, Gupta AK, et al. Sildenafil as a candidate drug for Alzheimer's disease. J Alzheimers Dis 2024;98(2):643-657. MarketScan and Clinformatics comparisons with four other drugs: [pubmed.ncbi.nlm.nih.gov/38427489](https://pubmed.ncbi.nlm.nih.gov/38427489/) - Adesuyan M, Jani YH, Alsugeir D, et al. Phosphodiesterase type 5 inhibitors in men with erectile dysfunction and the risk of Alzheimer disease: a cohort study. Neurology 2024;102(4):e209131. 269,725 men, hazard ratio 0.82; 3-year lag hazard ratio 0.93 (0.80 to 1.08); limitations: [neurology.org](https://www.neurology.org/doi/10.1212/WNL.0000000000209131) - Henry DS, Pellegrino RG. A case-control study of phosphodiesterase-5 inhibitor use and Alzheimer's disease and related dementias among male and female patients aged 65 years and older supporting the need for a phase III clinical trial. PLoS One 2023;18(10):e0292863: [pubmed.ncbi.nlm.nih.gov/37851623](https://pubmed.ncbi.nlm.nih.gov/37851623/) - Chua WY, Lim LKE, Wang JJD, et al. Sildenafil and risk of Alzheimer disease: a systematic review and meta-analysis. Aging (Albany NY) 2025;17(3):726. Five observational studies, 885,380 patients: [pmc.ncbi.nlm.nih.gov/articles/PMC11984433](https://pmc.ncbi.nlm.nih.gov/articles/PMC11984433/) - Wu CY, Swardfager W. Phosphodiesterase-5 inhibitors and dementia risk: confounding by indication in real-world studies. J Alzheimers Dis 2024;100(4):1161-1163: [pubmed.ncbi.nlm.nih.gov/38995794](https://pubmed.ncbi.nlm.nih.gov/38995794/) - Desai RJ, Mahesri M, Lee SB, et al. No association between initiation of phosphodiesterase-5 inhibitors and risk of incident Alzheimer's disease and related dementia: results from the Drug Repurposing for Effective Alzheimer's Medicines study. Brain Commun 2022;4(5):fcac247: [pubmed.ncbi.nlm.nih.gov/36330433](https://pubmed.ncbi.nlm.nih.gov/36330433/) - National Institutes of Health, "NIH DREAM study finds Viagra and Cialis do not reduce risk of Alzheimer's and related dementias," October 4, 2022: [nih.gov](https://www.nih.gov/news-events/news-releases/nih-dream-study-finds-viagra-cialis-do-not-reduce-risk-alzheimer-s-related-dementias) - Gronich N, Stein N, Saliba W. Phosphodiesterase-5 inhibitors and dementia risk: a real-world study. Neuroepidemiology 2025;59(3):193-202 (online 2024): [pubmed.ncbi.nlm.nih.gov/38952132](https://pubmed.ncbi.nlm.nih.gov/38952132/) - Alcalde-Herraiz M, Woolf B, Xie J, et al. Phosphodiesterase-5 inhibition and Alzheimer's disease risk: a Mendelian randomisation study. Aging Cell 2025;24(12):e70265: [pubmed.ncbi.nlm.nih.gov/41078087](https://pubmed.ncbi.nlm.nih.gov/41078087/) - Brennan SO, Tinworth AC. Genetically proxied phosphodiesterase type 5 (PDE5) inhibition and risk of dementia: a drug target Mendelian randomization study. Mol Neurobiol 2025;62(6):7864-7874: [pubmed.ncbi.nlm.nih.gov/39951190](https://pubmed.ncbi.nlm.nih.gov/39951190/) - Greeley D, Nash M, Herskowitz B, et al. A phase 2 randomized, placebo-controlled study on the efficacy and safety of AR1001, a phosphodiesterase-5 inhibitor, in patients with mild-to-moderate Alzheimer's disease. J Prev Alzheimers Dis 2025;12(9):100337: [pubmed.ncbi.nlm.nih.gov/40912996](https://pubmed.ncbi.nlm.nih.gov/40912996/) - VIAGRA (sildenafil citrate) prescribing information, Pfizer, DailyMed. Section 1, Indications and Usage: [dailymed.nlm.nih.gov, VIAGRA label](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=0b0be196-0c62-461c-94f4-9a35339b4501) - ClinicalTrials.gov, searched 2026-10-17 for sildenafil with Alzheimer's, dementia and cognitive terms: [clinicaltrials.gov](https://clinicaltrials.gov/) --- _Affiliate disclosure: some links on the web version of this site go to Bodybuilding Peptides (trybbpeptides.com), an affiliate destination, and we may earn a commission if you buy there. Our coverage is independent of that relationship. Details: https://theedreport.com/affiliate-disclosure/_