Linking Dietary Creatine to DNA Methylation-Based Predictors of Mortality in Individuals Aged 50 and above

Abstract

Introduction

Creatine is a conditionally essential nutrient integral to cellular energy homeostasis, with emerging evidence suggesting its potential role in modulating biological aging. However, associations between dietary creatine intake and epigenetic biomarkers of mortality remain unexplored. This study investigates the relationship between dietary creatine intake and DNA methylation-derived mortality indices in US adults aged 50 years and older.

Methods

Data from the NHANES 1999–2002 cycles were analyzed, including dietary creatine intake estimated from 24-h recall interviews and DNA methylation profiles measured using the Illumina EPIC array. Epigenetic mortality predictors GrimAgeMort and GrimAge2Mort were examined in relation to creatine intake.

Results

Among 4,983 participants (mean age 67.6 ± 10.7 years), a significant inverse correlation was observed between dietary creatine and both GrimAgeMort (r = −0.041, p = 0.045) and GrimAge2Mort (r = −0.047, p = 0.019), indicating that higher creatine consumption was associated with lower epigenetic mortality risk scores. These associations persisted as statistically significant after adjustment for demographic variables and pertinent dietary factors.

Conclusions

Higher dietary creatine intake is linked to reduced biological age acceleration and mortality risk as estimated by epigenetic biomarkers. These findings highlight creatine’s potential as a modifiable dietary factor promoting healthy aging and longevity. Further research is warranted to elucidate underlying mechanisms.

Introduction

Creatine is a conditionally essential nutrient that plays a vital role in maintaining cellular energy homeostasis across multiple organ systems [1]. Although the human body can synthesize creatine endogenously, a considerable proportion is acquired through the diet, primarily from animal-based sources. Advancing age is associated with a decline in both endogenous creatine production and dietary intake – particularly among individuals adhering to plant-based or low-protein diets – which may contribute to declines in physical function and increased susceptibility to age-related health conditions [2–4]. While the ergogenic and neurocognitive benefits of creatine are well documented (for a detailed review, see Kreider and Stout [5]), its potential influence on biological aging remains underexplored. Emerging epidemiological evidence suggests a link between creatine intake and longevity; notably, a recent study reported that consuming at least 1 g of creatine per day was associated with a reduced risk of all-cause mortality in US adults over a nearly two-decade follow-up period [6], indicating potential survival benefits of creatine-rich diets. Concurrently, advances in epigenetic research have led to the development of DNA methylation (DNAm)-based biomarkers – such as GrimAgeMort and GrimAge2Mort – which provide robust estimates of biological age and mortality risk [7, 8], thereby offering a novel approach to assess the impact of nutritional factors on aging trajectories. Building upon prior population-based studies, the present study utilizes data from the National Health and Nutrition Examination Survey (NHANES) to examine associations between dietary creatine intake and DNAm-derived mortality indices. We hypothesize that higher dietary creatine intake is associated with reduced epigenetically predicted mortality risk, shedding new light on the potential of creatine as a modifiable dietary factor for promoting healthy aging and longevity.

Methods

We utilized data from the NHANES cycles that include DNAm profiles and epigenetic biomarker estimates, available under the Laboratory Data Domain. Specifically, DNAm was assessed in whole blood samples collected from participants aged 50 and above during the 1999–2000 and 2001–2002 survey cycles. Genomic DNA was extracted and subsequently analyzed using the Illumina Infinium MethylationEPIC BeadChip array (iScan System, Illumina, San Diego, CA, USA), which interrogates over 850,000 CpG sites across the genome. Raw DNAm data matrices were generated, preprocessed, and normalized following standard quality control and bioinformatics protocols to ensure reliability and comparability across samples. For the present analysis, we focused on validated DNAm-derived epigenetic biomarkers known to predict mortality risk, including GrimAge-based mortality scores (GrimAgeMort and GrimAge2Mort), which integrate methylation signatures of smoking history, plasma proteins, sex, and chronological age [9]. These biomarkers provide robust estimates of biological aging and have been widely utilized in epidemiological aging research. Notably, GrimAge2Mort represents an advancement over GrimAgeMort, offering a more sensitive and comprehensive tool for assessing the effects of dietary patterns, lifestyle factors, and interventions on biological aging and mortality risk [10]. Comprehensive documentation of the NHANES DNAm data – including sample eligibility criteria, laboratory and analytical procedures, bioinformatics pipelines, and accompanying data files – is publicly available through the National Center for Health Statistics [11]. Daily dietary creatine intake was estimated using data from the NHANES Dietary Data component, which is based on structured, interviewer-administered 24-h dietary recall interviews conducted in person with each participant. These recalls provide detailed information on all foods and beverages consumed in the 24 h preceding the interview. To quantify creatine intake, we applied standardized estimates of creatine content for various food categories, drawing upon established compositional values reported in the literature [12]. Specifically, average creatine concentrations were assigned as 0.20 g/kg for milk-based foods and 3.88 g/kg for meat-based sources for creatine-containing foods. These values were applied to reported food intakes to calculate individual-level daily creatine consumption. The analysis focused exclusively on creatine obtained from dietary (i.e., food-based) sources; intake from nutritional supplements, ergogenic aids, or pharmacological preparations was not included in the calculations, to isolate the contribution of habitual dietary patterns to creatine exposure. Associations between daily dietary creatine intake (primary exposure) and GrimAge-based mortality scores (outcome variables) were assessed using linear regression analysis. Multivariable models were sequentially adjusted for key demographic covariates, including sex, race/ethnicity, education level, and household income, as well as relevant dietary factors known to influence epigenetic aging, such as saturated fatty acids, folate, vitamin D, and vitamin A [13]. To address the excess zeros in dietary creatine intake data, a two-part zero-inflated modeling approach was employed, consisting of a logistic regression to model the probability of any creatine consumption and a linear regression to assess the association between creatine intake and DNAm-derived mortality indices among consumers. Statistical significance was defined as a two-tailed p ≤ 0.05. All analyses were conducted using IBM SPSS Statistics for Mac, version 24.0 (IBM Corp., Armonk, NY, USA).

Results

A total of 4,983 participants from the NHANES 1999–2002 cohort were included in the final analyses (51.2% female; mean age: 67.6 ± 10.7 years). DNAm-derived mortality indices were available for 2,532 individuals, and dietary creatine intake data were available for 4,243 individuals. The mean GrimAgeMort and GrimAge2Mort scores were 66.4 ± 8.9 points (95% CI: 66.0–66.7) and 72.2 ± 8.8 points (95% CI: 71.9–72.5), respectively. The average dietary creatine intake was 0.77 ± 0.75 g/day (95% CI: 0.75–0.79); a total of 133 participants (3.1%) reported zero daily creatine intake. A statistically significant negative correlation was observed between dietary creatine intake and both GrimAgeMort (r = −0.041, p = 0.045) and GrimAge2Mort (r = −0.047, p = 0.019), indicating that higher creatine consumption was associated with lower DNAm-derived mortality risk scores. Crude linear regression analysis revealed a significant inverse association between daily dietary creatine intake and both GrimAgeMort and GrimAge2Mort scores (p < 0.01). Specifically, each additional gram of creatine consumed per day was associated with an estimated decrease of 1.00 point in GrimAgeMort and 1.08 points in GrimAge2Mort. These associations remained statistically significant in adjusted models. After controlling for demographic variables, creatine intake was associated with a reduction of 1.29 points in GrimAgeMort and 1.32 points in GrimAge2Mort (both p < 0.001). Further adjustment for dietary covariates yielded similarly significant associations (B = −1.12 and B = −1.17, respectively; both p = 0.001). Zero-inflated modeling revealed no discrepancy from the associations reported between dietary creatine intake and DNAm-derived mortality indices, indicating that excess zeros did not substantially impact the observed relationships.

Discussion

To the best of our knowledge, this is the first population-based study to examine the association between dietary creatine intake and DNAm-derived mortality indices. We identified a significant inverse relationship between creatine intake and both GrimAgeMort and GrimAge2Mort scores in US adults aged 50 years and older. Notably, these associations remained robust after adjusting for a broad range of potential confounders, including demographic characteristics and dietary variables. The findings suggest that higher dietary creatine intake is linked to lower epigenetic estimates of mortality risk, implying a potential role for creatine in attenuating biological age acceleration. These results warrant further investigation into the influence of creatine on aging-related health outcomes and longevity.

Although creatine is among the most widely consumed and extensively researched nutritional compounds, its relationship with mortality risk has been only marginally investigated in both preclinical and clinical contexts. Early experimental studies suggested that creatine deficiency may exacerbate mortality under pathological conditions. For example, a seminal study demonstrated that dietary creatine restriction increased mortality in rats subjected to myocardial infarction [14]. In the domain of reproductive health, creatine supplementation has shown potential in reducing fetal and neonatal mortality, as evidenced by findings from both animal models and human pregnancies, as summarized in a comprehensive review by Dickinson and coworkers [15]. More recently, epidemiological data from a nationally representative US cohort revealed that individuals consuming at least 1 g of creatine per day had a 15% lower risk of all-cause mortality compared to those consuming less than 1 gram daily (hazard ratio = 0.85), suggesting a possible survival advantage associated with higher creatine intake [6]. The present study builds upon these preliminary findings by providing further evidence for a protective association between dietary creatine intake and mortality risk. Notably, our research advances the field by utilizing novel epigenetic biomarkers – GrimAgeMort and GrimAge2Mort – derived from DNAm data, which are validated predictors of biological aging and mortality risk. Through the incorporation of these robust aging indices and adjustment for a comprehensive set of covariates, we offer a more nuanced and mechanistic understanding of the link between creatine intake and aging-related outcomes. Our principal finding, based on a nationally representative sample of US adults aged 50 years and older, indicates that higher dietary creatine intake is significantly associated with lower GrimAgeMort and GrimAge2Mort scores. Since lower values on these biomarkers reflect reduced biological age and diminished risk of premature mortality and age-related diseases, our results suggest that creatine intake may contribute meaningfully to healthier aging trajectories. The observed inverse relationship between creatine intake and epigenetically predicted mortality risk supports the hypothesis that individuals with higher creatine consumption exhibit a younger biological profile. Several biological mechanisms may underlie this association. First, creatine is a critical component of cellular energy metabolism, facilitating ATP regeneration in energy-demanding tissues such as the brain and skeletal muscle. This may confer protection against cellular stress and dysfunction commonly associated with aging [16]. Second, creatine possesses anti-inflammatory and antioxidant properties [17], which may mitigate the chronic inflammation and oxidative stress that contribute to accelerated epigenetic aging. Third, creatine plays a key role in preserving muscle mass, strength, and physical function, thereby reducing the risk of sarcopenia and functional decline – both known predictors of increased mortality [18]. Fourth, by enhancing brain energy metabolism and attenuating neuroinflammatory processes, creatine may provide neuroprotective benefits that help reduce cognitive decline and neurodegeneration [19], which are also reflected in elevated GrimAgeMort scores. Fifth, higher creatine intake is typically associated with greater consumption of animal-based proteins, which are rich in essential micronutrients (e.g., vitamin B12, iron, zinc) known to support healthy aging. Thus, creatine intake may reflect both direct physiological benefits and broader dietary patterns associated with slower biological aging. Finally, by sparing S-adenosylmethionine and promoting methyl group availability, creatine may help stabilize DNAm patterns, contributing to more favorable epigenetic aging profiles [20]. Further studies are warranted to disentangle the relative contribution of each of these mechanisms and to clarify whether the observed associations are mediated by creatine itself, correlated dietary factors, or a combination thereof.

Epigenetically predicted mortality risk, as estimated by DNAm-based clocks such as GrimAge and GrimAge2, serves as both a proxy for actual mortality risk and a biomarker of biological aging and systemic vulnerability [21]. These measures integrate DNAm surrogates for clinically relevant predictors (e.g., smoking history, inflammatory markers, plasma proteins) and have demonstrated strong predictive validity for all-cause mortality, even outperforming chronological age in several cohorts [9, 22]. While NHANES includes observed mortality data, the predictive models employed in our study offer a complementary molecular perspective that may capture preclinical risk trajectories not yet reflected in mortality or morbidity outcomes. The use of predicted mortality risk adds value by enabling the early identification of biological alterations associated with modifiable exposures, such as diet, before clinical outcomes manifest. This is particularly relevant given the long latency period of chronic disease and mortality, and the relatively limited duration of follow-up in some NHANES cycles. Regarding the interpretation of our findings, an association between dietary creatine intake and epigenetically predicted – rather than observed – mortality risk suggests that creatine may influence biological pathways relevant to aging and longevity, without necessarily implying a direct or immediate effect on actual mortality within the available follow-up window. While such associations should be interpreted cautiously, they provide meaningful insights into potential mechanisms and serve as a foundation for future longitudinal and interventional research examining creatine’s role in health span and lifespan.

This study has several limitations that should be considered when interpreting the findings. First, the cross-sectional design of the NHANES dataset precludes any causal inferences between dietary creatine intake and DNAm-derived mortality indices; longitudinal or interventional studies are needed to establish directionality and temporality. Second, dietary creatine intake was estimated from a single 24-h dietary recall, which may not accurately reflect habitual intake and is subject to recall bias and day-to-day variability. Third, the analysis excluded creatine derived from supplements or pharmacological preparations, potentially underestimating total creatine exposure, particularly among individuals who use ergogenic aids. Fourth, although the models adjusted for relevant demographic and nutritional variables, residual confounding from unmeasured lifestyle, dietary, or genetic factors may influence both creatine intake and epigenetic aging markers. Fifth, the GrimAgeMort and GrimAge2Mort indices, while robust predictors of mortality, are indirect measures of biological age and may not fully capture the complex physiological processes involved in aging. Sixth, although our analysis focused on epigenetically predicted mortality risk, we did not include GrimAgeAccel (GrimAge acceleration) as an independent outcome measure. The absence of this specific analysis limits our ability to comprehensively evaluate the relationship between dietary creatine intake and biological age acceleration. Finally, the study population was limited to US adults who participated in NHANES during the 1999–2002 cycles, which may limit the generalizability of the findings to more recent or international populations with different dietary patterns or health profiles. Future research employing longitudinal designs, more precise dietary assessment methods, and mechanistic approaches is warranted to validate and expand upon these findings.

Conclusion

This population-based study is the first to demonstrate a significant inverse association between dietary creatine intake and DNAm-derived mortality indices in US adults aged 50 years and older. Higher creatine intake was linked to lower GrimAgeMort and GrimAge2Mort scores, suggesting reduced biological age acceleration and potentially lower mortality risk. These findings support the hypothesis that creatine, a modifiable dietary factor, may promote healthier aging, warranting further longitudinal and interventional studies to confirm these associations and elucidate the biological mechanisms linking creatine intake to epigenetic aging and longevity.

Acknowledgments

S.M.O. expresses gratitude to late Roger C. Harris for his transformative works.

Statement of Ethics

Ethical approval for the NHANES study was granted by the US National Center for Health Statistics Ethics Review Board (#98-12). Informed consent was obtained from all NHANES participants prior to their inclusion in the study, with the research conducted ethically following the World Medical Association Declaration of Helsinki. Ethical approval and consent were not required as the current study was based on publicly available data.

Conflict of Interest Statement

S.M.O. serves as a member of the Scientific Advisory Board on Creatine in Health and Medicine (AlzChem LLC), co-owns patent “Supplements Based on Liquid Creatine” at the European Patent Office (WO2019150323 A1) and patent application “Composition Comprising Creatine for Use in Telomere Lengthening” at the US Patent and Trademark Office (# 63/608,850); has received research support related to creatine during the past 36 months from the Ministry of Science, Technological Development and Innovation; Provincial Secretariat for Higher Education and Scientific Research; AlzChem GmbH; Kaneka Nutrients; Thermolife International; and Vireo Systems, Inc; and is the co-founder of KRE-ALL, a company which develops food products enriched with creatine. I.K. declares no known competing financial interests or personal relationships that could have appeared to influence the authorship of this paper.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

S.M.O.: conceptualization, data curation, formal analysis, investigation, methodology, and writing – original draft. I.K.: conceptualization, investigation, methodology, supervision, and writing – review and editing.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

The datasets supporting the conclusions of this study are publicly available and can be accessed through the National Health and Nutrition Examination Survey (NHANES) repository maintained by the National Center for Health Statistics at https://www.cdc.gov/nchs/nhanes/. The authors do not own the data. For further information, contact the corresponding author.