A 2026 GeroScience study found lower whole-blood NAD+ is associated with greater frailty in older adults, though the link weakened after adjusting for hematocrit.
In 2026, researchers published an analysis in GeroScience using data from the Itabashi Longitudinal Study on Aging. The Japanese study examined 529 community-dwelling adults aged 65 or older to evaluate the relationship between circulating NAD+ and physical frailty.
The primary conclusion of the study is that lower whole-blood NAD+ concentrations are associated with greater frailty in older populations. Researchers observed a statistically significant difference in metabolite levels between frail individuals and their robust peers. After adjusting for basic demographic factors, higher blood concentrations correlated with lower odds of being frail. However, this observational data does not establish that raising NAD+ levels would actually reduce or prevent frailty.
This study was an observational, cross-sectional analysis conducted entirely in humans.
To evaluate any longevity finding, we must first look at the specific population involved. The researchers analyzed a cohort of 529 older adults living independently in the community. Women made up 54.3 percent of this total sample. The research team utilized the revised Japanese Cardiovascular Health Study Criteria to evaluate physical decline.
This specialized diagnostic scale helps researchers objectively quantify the physical vulnerabilities that accompany advanced age. It moves the assessment away from subjective feelings of weakness toward standardized clinical measurements. By applying this rigorous metric, the team ensured their frailty classifications met established gerontological standards. Using this specific diagnostic scale, the study classified exactly 32 participants as frail.
This classification represents just 6.0 percent of the total group. This precise number of frail individuals is a vital piece of context for interpreting the final results. It means the primary statistical estimates rely on a relatively small subset of the overall population. The wide margin between healthy and frail participants heavily influences the resulting mathematical models.
Understanding these baseline characteristics helps ground our broader expectations. These participants were community-dwelling adults rather than hospital patients. This distinction matters deeply when we try to apply the findings to general aging populations. It provides a snapshot of natural variations in cellular metabolism among older adults managing daily life.
The researchers measured whole-blood NAD+ levels to look for defining biological patterns. They reported median whole-blood NAD+ concentrations of 24.00 µmol/L among participants with frailty. In contrast, they found a median concentration of 25.00 µmol/L among those without frailty. This absolute difference of 1 µmol/L was statistically significant in the reported analysis, presenting a P value of 0.007.
Finding a difference in median values is only the first step in this type of research. The subsequent step requires applying statistical models to account for normal human variations. The authors adjusted their initial analytical model for age, sex, and fasting time. Fasting time is an important variable because recent meals can temporarily alter core blood chemistry.
By controlling for these baseline factors, the researchers attempted to isolate the specific impact of the metabolite. In this adjusted model, each 1 µmol/L higher whole-blood NAD+ concentration was associated with lower odds of frailty. The specific odds ratio in this initial model was 0.871. The 95 percent confidence interval ranged from 0.780 to 0.970.
The result remained statistically significant with a P value of 0.013. The mathematical association also held up in a broader Firth regression model. That subsequent Firth regression model demonstrated an odds ratio of 0.880. Its confidence interval spanned from 0.782 to 0.986.
These numbers initially suggest a persistent relationship between the metabolite and the physical condition. However, the statistical picture shifted dramatically when researchers accounted for another common biological variable. Whole-blood measurements can be heavily influenced by a person's core blood-cell composition. To account for this reality, the researchers applied an additional adjustment for hematocrit.
Hematocrit measures the total proportion of red blood cells located in the overall blood volume. After the researchers additionally adjusted for hematocrit, the previous association weakened considerably. The revised odds ratio shifted to 0.911. Crucially, the confidence interval widened to range from 0.800 to 1.034.
Because this new confidence interval included the baseline of 1.0, the association was no longer statistically significant. This mathematical change highlights a crucial nuance in how we must interpret blood tests. According to the authors, the hematocrit result suggests that blood-cell composition may partly explain the initial association. The observed difference in the metabolite might reflect shifts in red blood cell density rather than an independent deficit.
This is a persistent challenge when evaluating biological age testing in human cohorts. It underscores why a single whole-blood reading does not always capture the full biological picture. Different cellular structures hold different concentrations of various molecules. When researchers test whole blood, they capture everything circulating at that exact moment.
It is remarkably easy to misinterpret observational cohort data as a definitive roadmap for clinical intervention. A finding that higher natural levels correlate with robust health often leads to premature assumptions about supplementation. However, a related NutraIngredients report explicitly notes that the data do not establish that increasing NAD+ reduces frailty. The study simply asks if the metabolite is associated with physical decline in this specific cohort.
It does not answer whether artificially raising levels improves health or extends life spans. Readers must strictly keep the precise clinical outcome in view when reviewing these preliminary findings. The paper assessed frailty using a specified frailty scale, and the finding does not establish benefits for lifespan. It also provides absolutely no insight into other complex clinical outcomes associated with systemic aging.
This distinction is critical for anyone trying to navigate longevity interventions and therapeutics accurately. Readers must treat the Japanese cohort result as an early observational signal. It is strictly not a clinical reason to use specialized supplements to prevent physical decline. Observing a correlation in a population simply provides a starting point for further scientific investigation.
The context surrounding whole-blood measurements grew significantly more complex with a separate 2026 report. This secondary report summarized a Nature Metabolism study that analyzed data across multiple distinct cohorts. That expansive study systematically measured whole-blood NAD+ in a total of more than 300 people. It ultimately reported no meaningful association between whole-blood NAD+ and chronological age.
Furthermore, the Nature Metabolism study found no major shifts in the metabolite after standardized lifestyle interventions. The report did note an expected rise in the group taking nicotinamide riboside supplements. However, the general lack of a universal correlation with age raises a highly useful interpretive question. It suggests that whole-blood NAD+ may behave differently as a marker of frailty than as a general marker of age.
Finding no correlation with age challenges the popular narrative surrounding this specific longevity metric. These two findings concern completely different clinical questions within modern longevity science. They do not, by themselves, resolve how this popular molecule relates to fundamental aging biology. A single whole-blood measurement should absolutely not be treated as a universal aging biomarker.
Evaluating longevity science news requires analyzing each specific outcome scale entirely independently. Different studies target different endpoints, meaning their results cannot be seamlessly merged. Conflating chronological age metrics with physical decline scales leads to profound misunderstandings about what a biomarker actually represents.
Every observational analysis carries inherent boundaries that dictate exactly how its conclusions can be utilized. This study contains several important weaknesses:
The authors said further studies using cell- or tissue-specific measurements are needed to clarify this biological relationship.
Determining if a specific longevity metabolite actually reflects human frailty requires rigorously isolating cross-sectional cohort patterns from validated clinical interventions. AgeAmaze directly addresses the confusion over conflicting longevity studies by delivering well-sourced explanations of new biomarker research, ensuring skeptical adults can understand emerging mechanisms clearly. Read the research
Stay current with research on aging biology, biomarkers, nutrition, therapeutics, peptides and longevity technology. AgeAmaze reports what the evidence shows, where uncertainty remains and which claims still need stronger data.
Follow AgeAmaze for careful reporting on what longevity science can show today and what still needs stronger evidence.
read the Blog
