Suboptimal dietary patterns are associated with accelerated biological aging in young adulthood: a twin study

Suvi Ravi, Anna Kankaanpää, Leonie H. Bogl, Aino Heikkinen, Kirsi H. Pietiläinen, Jaakko Kaprio, Miina Ollikainen, Elina SillanpäämedRxiv 2024 · DOI

Abstract

Suboptimal diets are well-known drivers of morbidity and mortality, but their impact on biological aging in early adulthood—before the onset of clinical disease—is less understood. This study investigated the association between dietary patterns and biological aging in a cohort of young adult twins (ages 21–25) from the FinnTwin12 study. Using epigenetic clocks (GrimAge and DunedinPACE) and latent class analysis (LCA), researchers identified six distinct dietary patterns. Results indicated that diets high in fast food and sugar-sweetened beverages, and low in fruits and vegetables, are associated with accelerated biological aging. Within-twin pair analyses suggested that while genetics partially explain these associations, dietary habits remain a significant factor in the biological aging process.

Introduction

Most research linking nutrition to aging focuses on middle-aged or elderly populations, where chronic conditions like type 2 diabetes or cardiovascular disease are already prevalent. However, biological aging is a cumulative, lifelong process. This study sought to determine if dietary choices in young adulthood (ages 21–25) leave a measurable "epigenetic signature" of accelerated aging. By utilizing a twin-study design, the researchers addressed a critical challenge in nutritional epidemiology: confounding factors. This approach allowed the team to control for shared genetics and environmental influences, such as socioeconomic status, to isolate the independent effect of diet on the biological clock.

Methodology

Epigenetic Clocks

To measure biological age, the study employed "second and third-generation" epigenetic clocks:

  • GrimAge: An algorithm designed to predict lifespan and mortality risk by analyzing DNA methylation—chemical "tags" on the genome that shift in response to age and environmental stressors.
  • DunedinPACE: A tool that measures the current pace of aging, effectively acting as a speedometer for biological decline.

Dietary Pattern Identification

Rather than focusing on isolated nutrients, the researchers used Latent Class Analysis (LCA) to categorize participants into six dietary patterns based on a 52-item food frequency questionnaire:

  1. High fast food, low fruits and vegetables (F&V)
  2. Plant-based
  3. Health-conscious
  4. Western with infrequent fish
  5. Western with regular fish
  6. Balanced average

The Twin Model

The study utilized data from the FinnTwin12 study, involving 826 individuals and 363 complete twin pairs (157 monozygotic and 206 dizygotic). By comparing twins within a pair, the researchers could account for shared upbringing and, in the case of monozygotic (identical) twins, 100% of their genetic sequence.

Key Results

Dietary Impact on GrimAge

At the individual level, biological aging (GrimAge acceleration) was significantly slower in the Plant-based, Health-conscious, and Balanced average groups compared to the High fast food/low F&V group. Conversely, the Western with infrequent fish pattern was associated with faster aging compared to the Balanced average. Even after adjusting for sex, energy intake, smoking, and alcohol consumption, the difference between the Balanced average and High fast food patterns remained statistically significant.

The Pace of Aging (DunedinPACE)

The Plant-based group exhibited a significantly slower pace of aging compared to both the High fast food and Western (infrequent fish) groups. However, these effect sizes were attenuated and became non-significant when BMI and physical activity were added to the model, suggesting that diet is part of a broader cluster of lifestyle behaviors.

Genetic Confounding

A pivotal finding was that the association between diet and aging was weaker in identical (monozygotic) twins than in fraternal (dizygotic) twins. This suggests that genetics may partially influence both dietary preferences (e.g., cravings for highly processed foods) and the underlying rate of biological aging.

Discussion

Lifestyle Clustering and BMI

The study highlighted that "suboptimal" diets do not exist in a vacuum; they strongly cluster with smoking, higher BMI, and lower physical activity. Interestingly, a "curvilinear" relationship was observed with BMI: in individuals with a low BMI, a high-fast-food diet was even more strongly associated with accelerated aging. This may be due to a critical lack of protective nutrients, such as polyphenols found in fruits and vegetables, which are necessary to mitigate oxidative stress.

Limitations

While the twin design is robust, the study is cross-sectional, meaning it cannot definitively prove causality. Additionally, dietary data relied on self-reported questionnaires, which are subject to recall bias. The cohort was also limited to individuals of European ancestry, which may affect the generalizability of the results to other populations with different dietary staples.

Conclusion

This research demonstrates that biological aging is already detectable and influenced by lifestyle in the early 20s, decades before the typical onset of age-related diseases. Diets characterized by high intake of fast food, processed meats, and sugar-sweetened beverages are linked to an accelerated biological clock. While genetics play a role in these associations, the findings underscore the importance of establishing healthy dietary patterns early in life to slow the pace of biological aging and potentially delay the onset of age-related morbidity.

Summary updated

Abstract

ABSTRACT Background & aims Suboptimal diets increase morbidity and mortality risk. Epigenetic clocks are algorithms that can assess health and lifespan, even at a young age, before clinical manifestations of diseases. We investigated the association between dietary patterns and biological aging in young adult twins. Methods The data were drawn from the population-based FinnTwin12 study and consisted of twins aged 21–25 years (n=826). Food and beverage intakes were assessed using a food frequency questionnaire. Biological aging was estimated using the epigenetic clocks GrimAge and DunedinPACE. Latent class analysis was used to identify dietary patterns. The association between dietary patterns and biological aging was assessed using linear regression modeling at the individual level, followed by within–twin pair analyses to account for genetic liabilities and shared familial confounders. Results Six dietary patterns were identified: 1) High fast food, low fruits and vegetables (F&V), 2) Plant-based, 3) Health-conscious, 4) Western with infrequent fish, 5) Western with regular fish, and 6) Balanced average. At the individual level, GrimAge acceleration was slower in the Plant-based, Health-conscious, and Balanced-average patterns compared to the High fast food, low F&V, and faster in the Western with infrequent fish pattern compared to the Balanced average, regardless of sex, nonalcoholic energy intake, smoking, and alcohol consumption. After further adjustment for BMI and sports participation, the strengths of the associations modestly decreased; however, the difference between the Balanced-average and High fast food, low F&V patterns remained significant. The pace of aging (DunedinPACE) was slower in the Plant-based pattern compared to the High fast food, low F&V and the Western with infrequent fish patterns after adjustment for sex, nonalcoholic energy intake, smoking, and alcohol. The effect sizes were attenuated and reached a non-significant level when BMI and sports participation were added to the model. Most of the associations were replicated in the within-pair analyses among all twin pairs and among dizygotic twin pairs, but the effect sizes tended to be smaller among monozygotic twin pairs. This suggests that genetics, but not a shared environment, may partially explain the observed associations between diet and biological aging. Conclusion Diets high in fast food, processed red meat, and sugar-sweetened beverages and low in fruits and vegetables are associated with accelerated biological aging in young adulthood. The clustering effect of lifestyle factors and genetic confounders should be considered when interpreting the findings.
BibTeX
@misc{ravi2024suboptimal,
  author = {Suvi Ravi and Anna Kankaanpää and Leonie H. Bogl and Aino Heikkinen and Kirsi H. Pietiläinen and Jaakko Kaprio and Miina Ollikainen and Elina Sillanpää},
  title = {{Suboptimal dietary patterns are associated with accelerated biological aging in young adulthood: a twin study}},
  year = {2024},
  doi = {10.1101/2024.06.25.24309391},
  howpublished = {medRxiv},
}
Figure