Published June 4th in Clinical Infectious Diseases, one of the most quietly significant clinical findings of the year just landed. And it involves a drug so old and so cheap that most pharmacists dispense it without a second thought.
What the ACTIV-6 Trial Actually Found
New findings from the ACTIV-6 randomized clinical trial, published in Clinical Infectious Diseases and co-led by a University of Minnesota Medical School research team, evaluated nearly 3,000 outpatient adults with mild-to-moderate COVID-19 across 90 sites in the United States. Participants were randomized to receive either metformin or placebo within seven days of symptom onset and were followed for six months.
At six months, participants who received metformin experienced a 50% relative reduction in the risk of clinician-diagnosed long COVID compared with those receiving placebo, indicating that metformin cut the risk of a medical diagnosis of long COVID by approximately half. Among participants followed through 180 days, clinician-diagnosed long COVID occurred in 0.56% of those receiving metformin compared with 1.17% of those receiving placebo.
Investigators observed a high probability of benefit for reducing symptom burden and preventing clinician-diagnosed long COVID. No safety concerns emerged during the study.
Why the Reproduction Story Matters More Than Any Single Trial
“Reproducing research is very important, and both trials have also been replicated in analyses of electronic health record data. Together, these independent studies support that in low- to high-risk adults, metformin is an effective strategy to reduce the risk of long COVID,” said David Boulware, MD, MPH, professor at the University of Minnesota Medical School, infectious disease physician and steering committee co-chair of the trial.
This convergence of evidence deserves emphasis. The ACTIV-6 trial directly replicates the COVID-OUT trial, an earlier phase 3 randomized, quadruple-blind study also led by University of Minnesota researchers. That earlier study reported a similar reduction in long COVID incidence in a different patient population. The COVID-OUT trial only included adults with overweight or obesity, defined as a BMI above 25 kg/m², but the ACTIV-6 RCT provided critical confirmation of those results and expanded them to adults with a normal BMI or greater.
Two independent RCTs, at different sites, with different enrollment criteria, producing concordant results, and confirmed by observational real-world analyses of electronic health record data, that is the most robust evidence structure available outside of a meta-analysis. The ACTIV-6 finding is not a single surprising result. It is the second confirmatory signal from two separate, well-designed clinical programs.
The Honest Nuance That Makes This More Interesting, Not Less
This newsletter’s editorial standard has always been to surface study limitations, and this trial requires precise framing.
In low-risk adults, most with prior immunity, metformin did not exceed the efficacy threshold of 0.975 for the prespecified primary endpoint of preventing self-reported COVID-19 symptoms at day 180. The adjusted risk of symptoms on day 180 was 0.8 percentage points lower with metformin, with a posterior probability of efficacy of 0.83, and a risk ratio of 0.79 with a credible interval that crossed one.
What that means in plain terms: the prespecified primary endpoint, symptom presence at 180 days by self-report, was not statistically crossed. The Bayesian efficacy threshold was not met on that measure. This is the technical ground on which critics of the trial’s conclusions will stand.
What it means for the clinical picture: the secondary endpoint of clinician-diagnosed long COVID, the outcome requiring a physician to make a formal diagnosis based on clinical evaluation, was halved with a 96% posterior probability of efficacy and a risk ratio of 0.495 whose credible interval just reached statistical significance. The difference between these two endpoints reflects the difference between a patient self-reporting that they still have symptoms and a physician making a documented clinical diagnosis of long COVID. The latter is a harder, more clinically meaningful outcome. And it was cut in half.
This is not a failed trial. It is a nuanced result that tells a real clinical story about a drug with a favorable safety profile, essentially zero cost, and now two concordant randomized trials pointing in the same direction.
The Population This Trial Addressed
The median age of participants was 47 years, with an interquartile range of 38 to 57; 63% were female; 47% Hispanic and Latino; and 83% reported at least one prior COVID-19 infection or two or more SARS-CoV-2 vaccinations. There were no deaths.
This is a representative outpatient population, not a hospitalized or severely immunocompromised one. These are the patients who come into your pharmacy with a fresh positive test, mild to moderate symptoms, and no clear path to preventing the post-infectious syndrome that will sideline some of them for months.
There are important population boundaries to know. The clinical trials excluded adults over 85 years of age, so caution is warranted when considering metformin for acute SARS-CoV-2 in adults over age 85. Observational studies in individuals with diabetes have shown similar benefits for protection with metformin in those who were taking it before their infection to treat their diabetes.
For patients with existing renal impairment, the standard metformin contraindication based on eGFR thresholds applies, as it would for any metformin prescription. The acute-use context doesn’t change the renal safety considerations.
The Titration Protocol Pharmacists Need to Know
The dosing used in ACTIV-6 is not a flat dose. It is a 14-day ramp-up protocol designed specifically to minimize the GI side effects that lead patients to discontinue metformin, the same side effects that make standard diabetes dosing initiation challenging.
The specific 14-day titration regimen: 500 mg once daily for day 1, followed by 500 mg twice daily for days 2 through 5, then 1,500 mg daily split into 500 mg in the morning and 1,000 mg in the evening for the remaining nine days.
This titration structure is pharmacologically sound for the same reason it works in diabetes initiation: starting low and ramping up reduces the acute GI burden while maintaining the clinical benefit of full therapeutic exposure by the end of the course. A patient who discontinues metformin on day three because of nausea derives zero benefit. The pharmacist who counsels proactively on this, explaining that mild GI symptoms are expected and typically improve by day five or six, is the difference between a patient who completes the course and one who doesn’t.
The Mechanisms Behind the Signal
Why would a 60-year-old diabetes drug prevent the post-infectious syndrome of a respiratory virus? The mechanisms are more compelling than the coincidence suggests.
Several meta-analyses in individuals with diabetes have shown a high certainty of evidence for preventing mortality from COVID-19 with metformin, which suggests the benefits may extend well beyond glucose lowering.
Metformin’s primary molecular target is AMPK activation, which downregulates the mTOR pathway. Dysregulated mTOR activation has been documented in long COVID pathophysiology, particularly in the context of the mitochondrial dysfunction and cellular energy impairment seen in post-infectious fatigue syndromes. Metformin’s anti-inflammatory properties, including reduction of pro-inflammatory cytokine signaling, are also relevant given the persistent low-grade inflammation that characterizes long COVID. Additionally, metformin carries antiviral properties through disruption of viral replication pathways that depend on host metabolic machinery. None of these mechanisms are speculative; all are documented in existing metformin pharmacology literature. The ACTIV-6 trial didn’t discover a new drug. It revealed a new clinical application of a drug whose mechanism was always capable of this.
The Gap This Fills And Why It Matters
There are currently no FDA-approved treatments specifically for the prevention or treatment of long COVID. The condition affects an estimated 17 to 23 million adults in the United States by various definitions, representing one of the largest untreated post-infectious disease burdens in American medical history.
The patient who tests positive for COVID today and asks their pharmacist what else they can do is asking a question that, until this year, had no randomized controlled trial evidence behind a pharmacological answer. ACTIV-6, combined with COVID-OUT and the electronic health record confirmation data, now provides that evidence. The answer is not yet FDA-labeled. But the evidence structure is stronger than the evidence behind many interventions that have received labeled indications.
The Pharmacist’s Role at This Clinical Moment
The pharmacist’s role here is not to prescribe metformin off-label for long COVID prevention. It is to know the data well enough to have the clinical conversation that a patient with a fresh positive test deserves to have with their prescriber, and to facilitate that conversation proactively when the patient is in front of you.
The specific scenario: a patient comes in with a fresh positive COVID test, mild to moderate symptoms, and no current metformin prescription. The pharmacist who knows the ACTIV-6 and COVID-OUT data can tell that patient: “There is published randomized trial evidence suggesting that a 14-day course of metformin started early during a COVID infection may cut the risk of developing long COVID by about half. This isn’t FDA-approved for this use, but the evidence is real and your doctor may want to discuss it with you today. I’m happy to call the office with you or send a note summarizing the trial.” That conversation takes three minutes and represents exactly the clinical value that no algorithm, no automated refill system, and no mail-order pharmacy can replicate.
Your Clinical Protocol Starting Today
Three steps that require no new infrastructure:
Know the dosing. 500 mg once on day 1, 500 mg twice daily days 2 to 5, then 500 mg morning and 1,000 mg evening for days 6 to 14. Write it down. Put it in your counseling reference.
Know the exclusions. Adults over 85, patients with eGFR below 30, and patients for whom GI side effects of metformin are medically contraindicated are the principal populations where caution is warranted. The trial population was adults with mild to moderate COVID, most with prior immunity.
Build the prescriber conversation. If a COVID-positive patient presents who is not already on metformin and has no contraindications, your role is to surface the evidence and facilitate the discussion. “The ACTIV-6 trial, published in Clinical Infectious Diseases this month, showed a 50% reduction in clinician-diagnosed long COVID with a 14-day metformin course started within seven days of symptom onset. No safety signals. Do you want me to put together a note for the prescriber?” That is clinical pharmacy practice at its most useful.
The label hasn’t caught up to the evidence yet. That gap is exactly where the pharmacist’s value lives.
Sources: Clinical Infectious Diseases (Bramante CT et al. Metformin on the Presence of COVID-19 Symptoms 6 Months after Infection: The ACTIV-6 Randomized Clinical Trial. Clin Infect Dis. 2026. doi: 10.1093/cid/ciag335), University of Minnesota Medical School / EurekAlert (New Study Shows Metformin Given During Acute COVID-19 Infection Reduced Risk of Clinician-Diagnosed Long COVID by 50%, June 11, 2026), Pharmacy Times (Metformin Cuts Long COVID Risk by 50%, New Trial Confirms, June 2026), Drug Topics (Metformin Cuts Long COVID Risk by Half in New ACTIV-6 Trial Results, June 2026), News-Medical.net (Metformin Reduces Long COVID Risk According to ACTIV-6 Clinical Trial, June 2026), MedicalXpress (Metformin Halves Long COVID Risk in Acute Infections, June 2026), Mirage News (Metformin Halves Long COVID Risk in Acute Infections, June 11, 2026)