New research suggests how coffee is made may be just as important as how much people drink, showing that filtered and instant coffee have very different associations with biological aging.
Study: Discrepancies in the biological aging of instant and filter coffee: evidence from the UK Biobank. Image credit: Rabizo Anatolii / Shutterstock
In a recent study published in npj food scienceresearchers evaluated the cross-sectional association between coffee consumption and biological aging.
Aging is a complex process characterized by a gradual decline in function. Given the unprecedented rate of aging worldwide, identifying approaches to slow biological aging remains a critical challenge. Diet is one of the factors that influences biological aging. In particular, coffee, one of the most consumed beverages, is being investigated for its health risks and benefits.
Many studies have evaluated the association between coffee consumption and chronic diseases. However, there is still a lack of research on the association between coffee consumption and biological aging, with only a few studies yielding inconsistent results. Additionally, available studies do not account for different coffee brewing methods or different aging measures.
About research
In this study, researchers investigated the association between types of coffee preparation and measures of biological aging. They analyzed data from the UK Biobank (UKB). UKB subjects with missing data on diet, covariates, aging measures, and subjects with fewer than three dietary recalls were excluded. A meal frequency questionnaire was used to determine total coffee intake. Information on coffee type, including filter, instant, and other types, was collected through repeated 24-hour dietary recalls, and the average intake was used for participants who completed multiple recalls.
We evaluated three measures of biological aging: relative leukocyte telomere length (rLTL), PhenoAge Acceleration (PhenoAge Accel), and Klemera-Doubal Method Biological Age Acceleration (KDM-BA Accel). rLTL, a biomarker of cellular aging, was assessed by quantitative polymerase chain reaction and the ratio of telomere repeat copy number to single-copy gene copy number was estimated.
PhenoAge Accel and KDM-BA Accel quantified biological age based on biochemical and clinical parameters. The following covariates were included: age, race or ethnicity, gender, education, geographic region, Townsend Deprivation Index, alcohol intake, smoking status, body mass index (BMI), physical activity, consumption of sugar, artificial sweeteners or milk added to coffee, decaffeinated coffee, and presence of diabetes, cancer, cardiovascular disease, or hypertension.
Multivariate linear regression was used to examine associations between coffee intake and aging measures while adjusting for covariates. Subgroup analyzes were performed based on age, BMI, gender, drinking status, and smoking status. Additionally, we performed mediation analysis using two-stage linear regression and performed sensitivity analysis to assess the stability of the results.
Survey results
A total of 49,414 UKB participants were included, with a mean age of 56.4 years. Approximately 44.3% of participants were female, 91% were British, and 65% had a history of hypertension. On average, participants consumed 0.7, 1.5, and 0.03 cups of filter coffee, instant coffee, and other types of coffee each day. Additionally, 23% of participants reported drinking decaffeinated coffee, and 33% reported drinking coffee with milk.
Compared to non-consumers, instant coffee intake showed a significant overall trend of increasing PhenoAge Accel and KDM-BA Accel while decreasing rLTL, consistent with accelerated biological aging, although all consumption categories were not significantly different from non-consumers. In contrast, the filter coffee group showed a significant overall trend of decreased PhenoAge Accel and KDM-BA Accel while increased rLTL, consistent with slower aging, although rLTL was not significantly higher in the highest intake group. On the other hand, no significant trends were observed for other types of coffee. The authors noted that most of the estimated effects of instant coffee correspond to an age acceleration of approximately 0.1 to 0.2 years, which is of modest clinical significance at the individual level.
Subgroup analyzes showed that among participants who consumed 3 or more cups of instant coffee per day, men, those over 60 years of age, and current smokers had higher KDM-BA acceleration than non-consumers. Higher PhenoAge Accel was observed in current smokers and individuals with BMI < 30 kg/m2. In contrast, men who consumed three or more cups of filter coffee daily had decreased KDM-BA acceleration.
Sensitivity analyzes showed that the results did not change substantially across multiple analyses, including after excluding participants with cardiovascular disease, cancer, diabetes, or participants who consumed coffee with sugar, artificial sweeteners, milk, or decaffeinated coffee. However, these analyzes could not remove residual confounding. Mediation analysis showed that cystatin C and basal metabolic rate (BMR) statistically mediated 14.2% and 16.7% of the association between filter coffee intake and KDM-BA Accel, respectively. In addition, BMR mediated 2.9% of the association between filter coffee intake and PhenoAge Accel, and cystatin C separately mediated 18.0%.
Meanwhile, glycoprotein acetylation (GlycA) mediated 24.6% and 34.6% of the association between instant coffee intake and PhenoAge Accel and KDM-BA Accel, respectively. It also mediated 23.4% and 28.7% of the association between filter coffee intake and PhenoAge Accel and KDM-BA Accel, respectively. The association between filter coffee or instant coffee intake and rLTL was not mediated by cystatin C, BMR, or GlycA.
conclusion
In summary, filter coffee intake was associated with lower biological pro-senescence scores and generally longer rLTL, whereas instant coffee intake was associated with higher pro-senescence scores and shorter rLTL. Because the analysis was cross-sectional, it was not possible to prove whether either coffee type causes changes in biological aging. Self-reported coffee intake, potential for residual confounding, the predominantly white study population, and lack of long-term aging or clinical outcomes also limit interpretation. GlycA, BMR, and cystatin C were identified as potential statistical mediators of the association with PhenoAge Accel and KDM-BA Accel, but not with rLTL. Overall, the results of this study suggest that filter coffee is associated with a more favorable biological aging biomarker profile than instant coffee, but prospective and interventional studies are needed before the results can inform dietary guidelines.
Reference magazines:
- Wang, Y., Xiang, P., Liu, Z. et al. (2026). Discrepancies in biological aging between instant and filter coffee: Evidence from the UK Biobank. npj food science. Doi: 10.1038/s41538-026-00988-0. https://www.nature.com/articles/s41538-026-00988-0
