Start with the study, then the claim
- Population: who was studied? Record the health context and numbers enrolled and completing.
- Design: laboratory, animal, observational or human intervention? Was there a comparator and random allocation?
- Intervention: record the exact ingredient, formulation and route.
- Outcome: distinguish a laboratory marker from an outcome people experience.
- Time: a short trial cannot establish effects over years.
- Limitations: note missing data, uncertainty, exclusions, funding and conflicts.
A worked reading example
Martens and colleagues studied oral NR in healthy adults aged 55–79. Thirty people were randomised and 24 completed a double-blind, placebo-controlled crossover trial with two six-week periods. There was no formal washout; the authors checked for carryover and reported none. The physiological observations required larger trials. NIH grants supported the work; ChromaDex supplied study pills, analytical standards and partial funding. The authors declared no competing interests.
This study does not establish longer life, benefits for everyone or the effect of a different marketed product. It is a reading example, not a recommendation to take NR.
Compare without combining unlike studies
A small NMN trial in postmenopausal women with prediabetes and an IV NAD+ metabolism pilot in men cannot be pooled informally into a promise about all NAD products. See the focused study ledger for their different designs and limits.
Useful terms
- Randomised
- Participants are allocated by chance.
- Placebo
- A comparison designed to resemble the intervention.
- Crossover
- Participants receive different conditions in successive periods.
- Washout and carryover
- A washout is a gap between periods; carryover is a remaining effect from an earlier condition.
- Biomarker
- A measured biological feature, not automatic evidence of a meaningful health benefit.
Make a claim-to-source note
Write the claim in one column, the exact paper in another, and what that paper can and cannot establish in a third. A mismatch in route, ingredient, population or outcome is a reason to narrow the statement. A citation does not remove that mismatch.
Sources and scope
- Martens et al., Nature Communications, 2018
- Yoshino et al., Science, 2021
- Grant et al., Frontiers in Aging Neuroscience, 2019
Sources checked 15 October 2026. This is a focused educational resource, not a systematic review or evidence that an Elixir product has a health benefit. No clinician review is claimed.

