Source: Nutrients (2022)
A randomised, placebo-controlled trial gave older adults 250 mg of NMN a day, morning or evening, for 12 weeks — and tracked sleep, fatigue and physical performance. Here's what it did and didn't show.
Quick answer Sleep quality and fatigue improved over the 12 weeks, but mostly in placebo groups too — meaning NMN itself didn't clearly outperform placebo on its primary sleep measure. A few physical performance and evening-dosing results favoured NMN, but the study's own authors flag a likely placebo effect running through several findings.
At a glance
- Design
- This is a randomised controlled trial, the strongest common design for testing whether a treatment works.
- Blinding & control
- Neither participants nor researchers knew who got the real treatment, and it was compared against a placebo — the strongest common protection against bias.
- Precision
- 108 participants — a reasonably sized sample.
- Directness
- Outcomes measured: functional or performance measure.
- Independence
- Funded by industry, but the authors themselves aren't company employees.
- Reporting
- Reporting quality: pre-registered, full text available.
What did this study actually test?
This was a randomised, double-blind, placebo-controlled trial of nicotinamide mononucleotide (NMN), a compound that sits on the NAD+ salvage pathway — one of the routes the body uses to build the coenzyme NAD+, which cellular energy metabolism depends on. Researchers registered the trial (UMIN000038097) and split participants into four groups, comparing 250 mg of NMN a day against a matching placebo, with each of those two groups further divided by timing: morning dosing or evening dosing.
The rationale for splitting by time of day was to see whether when you take NMN — rather than just whether you take it — makes a difference to sleep and fatigue outcomes. That's a more granular question than most NMN trials ask, and it's the main thing that distinguishes this study's design from a simple two-arm comparison.
Who took part, and for how long?
108 older Japanese adults were randomised, and 105 completed the study. All were community-dwelling residents of Tsukuba City, Japan, aged 65 or over, living independently, mobile, without dementia, and not requiring long-term care. The paper stratified participants into two age bands: 65–74 ("young-old") and over 75 ("old-old").
The intervention ran for 12 weeks. Participants took six tablets at once with water, either any time from waking until midday, or from 6pm until bedtime, depending on which arm they were assigned to. Assessments were taken at baseline, at six weeks (for sleep and fatigue only), and at 13 weeks — one week after the 12-week intake period ended. Physical performance was measured only at baseline and at the end, not at the midpoint.
Why measure sleep, fatigue, and physical performance specifically?
The study's two primary outcomes were sleep quality, measured with the Pittsburgh Sleep Quality Index (PSQI) — a validated self-report questionnaire covering sleep duration, disturbances, latency and overall quality — and fatigue, measured using the Jikaku-sho shirabe questionnaire, a subjective symptom checklist. These are both self-reported measures, not lab tests.
Secondary outcomes covered physical performance: grip strength, the 5-times sit-to-stand test (how quickly someone can rise from a chair five times in a row, a proxy for lower-body strength), the timed up-and-go test (a measure of mobility and balance), and habitual walking speed over 5 metres. Together these outcomes reflect common concerns in ageing — sleep, tiredness, and functional strength — rather than any blood-based marker of NAD+ status itself.
Did the study look at safety?
Yes. The paper reported adverse events, and it noted no reported side effects from NMN ingestion, with none of the events recorded classed as related to the intervention. Whether any serious adverse events occurred at all is not stated in the source material.
This safety data covers 12 weeks of daily use at the studied dose only. The trial did not extend observation beyond the post-intervention assessment at week 13, so nothing about longer-term use can be drawn from it.
What did this study report?
The results below separate out three kinds of finding: changes seen equally across all groups over time (which don't tell you NMN did anything specific), differences that showed up specifically when comparing NMN against placebo, and effect sizes reported for individual groups. Reading past the p-values to which comparison each one actually reflects matters here.
- PSQI total sleep quality (interaction effect) — none: no significant group x time interaction. not reported No statistically significant interaction in sleep quality assessed by PSQI
- PSQI subscales (sleep duration, sleep disturbance, daytime dysfunction, sleep quality, total PSQI global score) - main effect of time — improved over time (all groups): p < 0.01. significant Significant main effect of time, not interaction; NMN_PM group had largest effect sizes: sleep latency d=0.56, daytime dysfunction d=0.72, sleep quality score d=0.80, total PSQI global score d=0.68
- Fatigue - drowsiness (interaction effect) — improved (NMN_PM and Placebo_PM at Mid and Post vs Pre): p = 0.02. significant Post hoc: drowsiness improvement at Mid and Post significantly greater than Pre in both NMN_PM and Placebo_PM groups; NMN_PM largest effect size d=0.64
- Fatigue - instability, dullness, total fatigue score (main effect of time) — improved over time: instability p<0.01; dullness p=0.03; total fatigue score p<0.01. significant Main effect of time only, not interaction (except drowsiness)
- 5-times sit-to-stand (5-STS) (interaction effect) — improved (Post vs Pre, all groups): p = 0.04. significant Post hoc: 5-STS significantly improved Post vs Pre in all groups; NMN_PM largest effect size d=0.72
- 5-STS and TUG (main effect of time) — improved over time: 5-STS p<0.01; TUG p<0.01. significant Main effect of time
- 5-STS (main effect of group) — not specified: p = 0.05. significant Authors describe this as significant main effect of group
- TUG effect size in NMN_PM — improved: d = 0.54. not reported Medium effect size, largest among groups, but interaction/group effect significance not stated for TUG specifically
- Grip strength — none. not reported Not described as significant; discussion mentions 'non-significant change in grip strength'
- 5-m habitual walk — none. not reported No significant results reported for this outcome
What doesn't this study show?
Because this trial relied on questionnaires and timed physical tests rather than lab measurements, there's a fair amount it simply can't speak to. The list below sets out what the study didn't measure or report, so its findings aren't stretched further than the data allows.
- Does not report objective sleep measures (e.g., actigraphy, polysomnography)
- Does not measure blood, urine, or muscle NAD+ or physiological biomarkers
- Does not report effects beyond the 12-week (13th week post) assessment
- Does not control or measure dietary NMN intake from food
- Grip strength and 5-m habitual walk results not shown as significant or detailed
What are this study's limitations?
The authors themselves flagged several constraints on how their findings should be read, from uncontrolled diet to the absence of any biological measurement of NAD+ status. Worth reading in full before drawing conclusions from the results above.
- Daily diet potentially containing NMN was not controlled, and no survey on daily nutrient intake was conducted
- Physiological factors (blood, urine, muscle) that could have been affected by NMN intake were not examined; only physical measurements and questionnaires were used
- Sleep quality was assessed only by subjective questionnaire; objective measures such as actigraphy, polysomnography, and sleep diary were not included
- A placebo effect was observed on 5-STS, drowsiness, and sleep disturbance scores, possibly because participants may have perceived the placebo as NMN despite double-blinding, due to necessary explanation of study design during research briefing under ethics guidelines
Who funded this study, and is there a conflict of interest?
The trial was funded by Mitsubishi Corporation Life Sciences Ltd., which provided research funds, the experimental supplements used, and the article-processing charge. The funder is stated to have had no role in data collection, interpretation, writing, or the decision to submit the study for publication. The authors declare no conflicts of interest.
Funding: Mitsubishi Corporation Life Sciences Ltd. provided research funds, experimental supplements, and APC. Funders had no role in data collection, interpretation, writing, or submission.
Conflicts of interest: The authors declare no conflict of interest.
What else do people ask about this study?
Did this study measure NAD+ levels in the blood?
No. The study used questionnaires and physical performance tests only — it did not examine blood, urine, or muscle for NAD+ or any other physiological marker, so it can't tell you whether NMN raised NAD+ levels in this group of participants.
Does timing of dosing (morning vs evening) matter based on this trial?
The study was designed specifically to compare morning and evening dosing, and several of the larger effect sizes reported turned up in the evening-dosing NMN group. But effect size alone isn't the same as a statistically confirmed group difference for every outcome, and the paper doesn't state that evening dosing was formally proven superior across the board.
Were the placebo tablets convincing enough to keep the trial properly blinded?
The authors raise this as a limitation themselves: a placebo effect turned up on some measures, and they suggest this might be because participants had to be told about the study design during ethics-required briefings, which may have let some guess their group assignment.
Is there any data on what happens after 12 weeks?
No. The study's only post-intervention check was at week 13, one week after the 12-week intake ended. There's no reported follow-up beyond that point.
Where can you read the original study?
- Nutrients (2022)
- DOI: 10.3390/nu14040755
- PMID: 35215405
- PMCID: PMC8877443
Leave a comment
This site is protected by hCaptcha and the hCaptcha Privacy Policy and Terms of Service apply.