Blood NAD+ Levels Track With Frailty in Older Adults, Japanese Study Finds

Deep inside every cell of the human body, a tiny molecule called nicotinamide adenine dinucleotide, or NAD+, is quietly running the machinery of life. It shuttles electrons in the reactions that turn food into usable energy, serves as a cosubstrate for sirtuins and other enzymes that govern cellular repair, and participates in signaling pathways that determine how cells respond to stress. For years, laboratory scientists have suspected that the slow decline of NAD+ with age might be one of the molecular threads connecting aging to the loss of physical resilience. Now, a study of more than 500 older adults living in a Tokyo community has brought that suspicion into the real world, finding that people with lower concentrations of NAD+ in their blood are more likely to be frail.

The research, led by Takashi Shida of the Tokyo Metropolitan Institute for Geriatrics and Gerontology and colleagues, drew on participants from the Itabashi Longitudinal Study on Aging, an ongoing community-based cohort in a residential ward of northern Tokyo. The team analyzed data from 529 adults aged 65 and older who took part in the 2025 survey wave, a group in which 54.3 percent were women. Frailty, the state of heightened vulnerability that leaves older adults at risk of falls, disability, hospitalization, and death, was assessed using the revised Japanese version of the Cardiovascular Health Study criteria, a validated instrument that scores weight loss, exhaustion, low physical activity, weakness, and slowness. Of the 529 participants, 32, or 6.0 percent, met the criteria for frailty.

The central measurement was the concentration of NAD+ in whole blood, quantified in micromoles per liter. Whole-blood NAD+ is a composite signal: it reflects the molecule’s presence inside red blood cells, white blood cells, and platelets, as well as whatever circulates in plasma. Because red blood cells lack mitochondria and rely heavily on NAD+ for glycolysis and antioxidant defense, the blood compartment offers a convenient, if imperfect, window into systemic NAD+ metabolism. The researchers found that the median whole-blood NAD+ concentration was measurably lower among participants with frailty than among those without it, at 24.00 versus 25.00 micromoles per liter, a difference that reached statistical significance with a P value of 0.007.

To move beyond a simple group comparison, the team built multivariable regression models that adjusted for the most obvious confounders: age, sex, and fasting time before the blood draw, the last of which matters because NAD+ levels can fluctuate with recent food intake. After these adjustments, each 1 micromole per liter increase in whole-blood NAD+ concentration was associated with 13 percent lower odds of frailty, with an odds ratio of 0.871 and a 95 percent confidence interval spanning 0.780 to 0.970, and a P value of 0.013. In a cohort where the median sits near 25 micromoles per liter, a one-unit shift is a meaningful fraction of the biological range, which makes the effect size noteworthy rather than trivial.

The investigators then subjected their finding to a battery of robustness checks that reveal how carefully the result was constructed. Restricted cubic spline analysis, a technique that allows the relationship between NAD+ and frailty to bend and curve freely rather than assuming a straight line, showed no clear evidence of nonlinearity, suggesting the association behaves approximately linearly across the observed range. Because frailty was relatively rare in the sample, the team also ran a Firth penalized-likelihood logistic regression, a method designed to produce more stable estimates when outcomes are infrequent and conventional models risk bias. Under this broader covariate adjustment, the association held firm, with an odds ratio of 0.880 and a confidence interval of 0.782 to 0.986.

But the story took a more complicated turn when the researchers added hematocrit, the proportion of blood volume occupied by red blood cells, to the model. With that adjustment, the association weakened and lost statistical significance, with the odds ratio drifting to 0.911 and the confidence interval crossing unity at 0.800 to 1.034. This attenuation is not a failure of the study; it is one of its most informative results. It suggests that part of the apparent link between blood NAD+ and frailty may be explained by blood cell composition itself. Frail older adults often have altered hematological profiles, including tendencies toward anemia and shifts in red cell indices, and since the vast majority of whole-blood NAD+ resides within blood cells, differences in cell counts and packing could masquerade as, or at least substantially contribute to, differences in NAD+ concentration.

This caveat places the new findings within a lively and sometimes contentious scientific debate. NAD+ has become one of the most hyped molecules in the longevity world, with supplements such as nicotinamide riboside and nicotinamide mononucleotide marketed aggressively on the premise that restoring youthful NAD+ levels can slow aging. Human evidence, however, has been mixed. Some studies have reported that whole-blood NAD+ contents decline with age, while others, including a recent analysis published in Nature Metabolism, found that human whole-blood NAD+ levels do not vary with age or with lifestyle interventions. Blood NAD+ has also been reported to be reduced in very old patients hospitalized for heart failure, hinting that the molecule may track with disease severity rather than chronological age per se. The Itabashi study adds a crucial piece: a community-dwelling, non-hospitalized population in which the association with frailty, a clinical syndrome rather than a single disease, was detectable.

The study’s design imposes important limits on interpretation. As a cross-sectional analysis, it captures a single moment in time and cannot determine whether low NAD+ contributes to causing frailty, whether frailty and its metabolic consequences deplete NAD+, or whether both are driven by a shared underlying process such as chronic inflammation, mitochondrial dysfunction, or poor nutrition. The frailty group was also small, with only 32 affected participants, which limits statistical power and explains the team’s recourse to penalized regression. And the whole-blood measurement, while practical and increasingly standardized, cannot distinguish between NAD+ dynamics in muscle, brain, immune cells, and other tissues where the molecule’s age-related decline is thought to matter most. The authors themselves emphasize that further studies using cell- and tissue-specific measurements are needed to clarify the relationship between NAD+ metabolism and frailty.

Those caveats notwithstanding, the significance of the work lies in its grounding. Much of the NAD+ literature rests on cell culture experiments, animal models, or small and highly selected human samples. Here, the association emerges from ordinary older adults living at home in a Tokyo ward, measured with a validated frailty instrument and adjusted for the mundane but essential variables of age, sex, and fasting status. If blood NAD+ does prove to be a genuine marker of physiological resilience, even one partly confounded by blood cell composition, it could eventually help clinicians identify which older patients are silently sliding toward frailty before falls and disability make it obvious. Biomarkers of frailty are a recognized public health priority, and most current candidates, from inflammatory cytokines to metabolomic signatures, remain far from clinical deployment.

The road ahead will require longitudinal follow-up within cohorts like Itabashi, where repeated NAD+ measurements can be tested as predictors of incident frailty, and intervention studies in which NAD+-boosting compounds are evaluated against hard clinical endpoints rather than blood chemistry alone. Recent work comparing different NAD+ precursors has shown that they exert differential effects on circulating NAD and microbial metabolism in humans, underscoring how much remains unknown about how these interventions behave in the body. Exercise training, for its part, has been shown to reverse the age-dependent decline in NAD+ salvage capacity in human skeletal muscle, pointing to lifestyle approaches that may act on the same pathway without a pill. For now, the Itabashi findings should be read as a carefully qualified signal rather than a prescription: a molecule central to energy metabolism and cellular signaling appears to run low in the blood of frail older adults, and untangling whether that deficit is a driver, a passenger, or a mirror of declining resilience is now one of the most compelling questions in the biology of aging.

Subject of Research: Association between whole-blood NAD+ concentration and frailty in community-dwelling older adults

Article Title: Association between whole-blood NAD+ concentration and frailty in community-dwelling older adults: the Itabashi Longitudinal Study on Aging

Article References: Shida, T., Hatanaka, S., Kojima, N., Osuka, Y., & Sasai, H. (2026). Association between whole-blood NAD+ concentration and frailty in community-dwelling older adults: the Itabashi Longitudinal Study on Aging. GeroScience. https://doi.org/10.1007/s11357-026-02545-9

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02545-9

Keywords: NAD+, frailty, aging, older adults, whole blood, biomarkers, energy metabolism, sirtuins, hematocrit, GeroScience, Itabashi Longitudinal Study, geriatrics