High-Dose Vitamin D Did Not Ease COVID Severity, but a Long COVID Signal Emerged

Vitamin D was one of the most discussed potential COVID-19 interventions of the pandemic era. The logic was plausible: the vitamin plays a role in immune regulation, deficiency is common, and observational studies had linked low vitamin D levels to worse COVID outcomes. What was missing was a large, rigorous, randomized trial. The VIVID trial aimed to provide one.

The results, published in The Journal of Nutrition, are largely negative - but with one intriguing exception.

The trial design: 2,024 participants, two countries

The Vitamin D for COVID-19 (VIVID) Trial enrolled 1,747 adults who had recently tested positive for COVID-19 and 277 of their household contacts across the United States and Mongolia. Participants were randomized to receive either high-dose vitamin D3 (9,600 IU per day for two days, then 3,200 IU per day) or a matching placebo for four weeks.

The U.S. arm ran from December 2020 to September 2022; the Mongolia arm from September 2021 to April 2022. The median time between a positive COVID test and starting the study medication was three days. The research team, led by senior author JoAnn Manson, MD, DrPH, of Mass General Brigham, used stratified randomization and statistical weighting to balance age, sex, body mass index, race/ethnicity, and vaccination status between the two groups.

No effect on acute illness

The primary outcome - healthcare utilization including hospitalizations, clinic visits, emergency visits, and death over four weeks - did not differ between the vitamin D and placebo groups. Symptom severity was similarly unaffected. Among household contacts, vitamin D supplementation did not reduce the rate of COVID-19 infection.

These null results are consistent with the majority of randomized evidence to date. While observational studies have repeatedly found associations between low vitamin D and poor COVID outcomes, those associations likely reflect confounding - people who are sicker, older, less active, or more obese tend to have lower vitamin D levels for reasons that have nothing to do with the vitamin itself.

A hint on long COVID

The more interesting finding came from a secondary analysis. Among participants who adhered to their vitamin D regimen, 21% reported at least one persistent symptom at eight weeks, compared with 25% of those taking placebo. The difference was of borderline statistical significance.

Long COVID - which can include fatigue, shortness of breath, cognitive difficulties, and other lasting symptoms - remains poorly understood and difficult to treat. Any signal that a simple, inexpensive supplement might reduce its incidence warrants attention, even when the statistical evidence is preliminary.

But preliminary is the operative word here. The long COVID analysis was secondary, not the trial's primary endpoint. It was limited to the per-protocol population (those who actually took the medication as directed), which introduces potential selection bias. And a four-percentage-point difference on the edge of statistical significance could easily disappear in a larger, purpose-built study.

What this trial can and cannot tell us

The VIVID trial is one of the largest and most rigorously designed randomized studies of vitamin D and COVID-19. Its null findings on acute severity are important and should help close the door on claims that high-dose vitamin D supplementation started after diagnosis can meaningfully alter the course of COVID-19 infection.

But the trial has limitations. Supplementation began a median of three days after the positive test - potentially too late to affect an immune response already in progress. The four-week treatment period was relatively short. The study could not assess whether long-term supplementation started before infection might have different effects, a question that would require a different trial design entirely.

The Mongolia arm introduced additional variability, as baseline vitamin D levels, healthcare systems, viral variants, and vaccination rates differed from the U.S. population. Whether these differences affected outcomes is unclear.

Regarding the long COVID signal, Manson and her team have called for larger studies specifically designed to test whether vitamin D supplementation reduces the risks and severity of persistent symptoms. Until such studies are completed, the VIVID trial offers a suggestive hint - but not a recommendation.