Start with the comparison the trial was designed to make
A randomized, double-blind study enrolled 58 community-dwelling participants with long COVID. One group received NR for twenty weeks; the other received placebo for ten weeks before switching to NR. Blood NAD increased during NR exposure, but the randomized comparisons found no significant differences in cognition, fatigue, sleep, anxiety or depression. [1]
Later exploratory analyses pooled participants during NR exposure and found improvements from their own baseline on some measures. Those analyses were post hoc and unadjusted for multiple comparisons. Dropout was also substantial and differed between groups. These details make the exploratory findings less secure than a headline about symptom recovery might suggest. [1]
Why before-and-after improvement is not the same result
A person’s symptoms can change during a study for reasons other than the assigned treatment. The purpose of a contemporaneous comparison group is to help interpret that change. Pooling treatment periods and comparing them with baseline asks a different question from comparing randomized groups over the same period.
It is therefore possible for both statements to be accurate: some participants improved during an NR period, and the controlled comparison did not demonstrate an NR benefit. A fair summary includes both, with the controlled result first. It should not present the exploratory analysis as though it reversed the primary conclusion.
For long-COVID readers, this distinction has practical importance. A promise of recovery is consequential. The relevant evidence should measure the difficulties people actually experience and preserve the study’s comparison, rather than rely on a higher NAD number as a substitute.
What dropout and multiple testing leave uncertain
When many participants leave a study, ask how the analysis handled missing observations and whether reasons for leaving differed between groups. The uncertainty is not solved by looking only at those who remained. Nor is a small P value from one of many exploratory tests a complete answer.
The paper is useful because it directly assessed cognition and symptoms, not just metabolism. Its negative between-group findings deserve visibility in product discussions. They do not prove that no future regimen could help, but they do mean this trial cannot support a reliable symptom-recovery promise.
Compare this with the separate hospitalized COVID-19 and kidney-injury trial, which studied a different compound and clinical situation. Persistent symptoms after infection and an acute hospital admission are not interchangeable research settings. Bring questions about ongoing symptoms and supplement use to a clinician familiar with your care; this study does not supply a self-treatment plan. The biomarker guide offers a second way to evaluate claims that skip the symptom results.