Scientists Link Lifelong Caffeine Levels to Lower Body Fat and Diabetes Risk

For many people, caffeine is woven into daily life. It arrives with the morning coffee, the afternoon tea, or the occasional soda that helps push through fatigue. Beyond its familiar role as a stimulant that sharpens alertness, caffeine has long intrigued scientists for another reason: its potential influence on body weight, fat accumulation, and metabolic disease.

Short-term clinical trials have shown that caffeine can slightly increase energy expenditure and fat burning. Observational studies have repeatedly reported that people who drink more coffee tend to have a lower risk of type 2 diabetes. Yet these findings have always come with a question mark. Do these associations reflect a true biological effect of caffeine, or are they simply the result of lifestyle differences between coffee drinkers and non-drinkers?

A new study published in BMJ Medicine approaches this puzzle from a different angle. Instead of asking people what they drink, the researchers looked at their genes. By using genetic variants that influence how quickly caffeine is metabolized in the body, the team was able to estimate the effects of lifelong differences in plasma caffeine levels. Their results suggest that caffeine’s role in metabolism may be more than a short-lived boost, and that it could have meaningful long-term implications for body fat and diabetes risk .

The Scientific Problem: Correlation or Cause?

The central challenge in nutrition science is separating cause from correlation. People who consume more coffee may also exercise more, smoke less, or follow different diets. Even sophisticated observational studies struggle to fully account for these overlapping factors.

Randomized controlled trials are the gold standard for establishing causality, but they are difficult to conduct for long-term dietary exposures. It is not easy, or ethical, to assign thousands of people to consume specific amounts of caffeine for decades and then wait to see who develops diabetes or heart disease.

As a result, despite decades of research, scientists have lacked clear evidence on whether caffeine itself directly reduces body fat or diabetes risk, or whether it merely travels alongside other protective behaviors.

A Genetic Shortcut to Causality

To overcome this limitation, the researchers used a method known as Mendelian randomisation. This approach takes advantage of the fact that genetic variants are assigned at conception and remain fixed throughout life. If certain genetic variants are reliably associated with higher plasma caffeine levels, and those same variants are also associated with specific health outcomes, this strengthens the case for a causal relationship.

In this study, the team focused on two well-characterized genetic variants located near the CYP1A2 and AHR genes. These genes play a key role in caffeine metabolism in the liver. People who carry versions of these genes that slow caffeine breakdown tend to have higher plasma caffeine concentrations over time, even though they may drink less coffee or tea.

By using large genome-wide association datasets involving hundreds of thousands of participants, primarily of European ancestry, the researchers examined how genetically predicted caffeine levels related to body mass index, body fat, type 2 diabetes, and major cardiovascular diseases such as heart attack and stroke.

What the Researchers Found

The results were striking in their consistency. Individuals with genetic variants linked to higher lifelong plasma caffeine levels tended to have lower body mass index and lower total body fat. Importantly, this reduction was specific to fat mass, with no meaningful change in fat-free mass such as muscle.

When the researchers turned to disease outcomes, they found that higher genetically predicted caffeine levels were associated with a significantly lower risk of type 2 diabetes. Across two large independent datasets, the reduction in diabetes risk was robust and statistically strong.

To better understand this relationship, the team conducted a mediation analysis. This allowed them to estimate how much of caffeine’s apparent protective effect against diabetes was explained by lower body weight. The analysis suggested that roughly 40 percent of the reduction in diabetes risk could be attributed to lower body mass index. In other words, caffeine appears to reduce diabetes risk in part by helping people carry less body fat.

By contrast, the study did not find strong evidence that higher plasma caffeine levels directly reduced the risk of major cardiovascular diseases such as ischemic heart disease, atrial fibrillation, heart failure, or stroke. While small effects could not be ruled out, any potential benefit or harm appeared modest at best.

Why Body Fat Matters So Much

Excess body fat, particularly when distributed centrally, is a major driver of insulin resistance. Over time, insulin resistance can progress to type 2 diabetes, a condition that now affects hundreds of millions of people worldwide.

The link between caffeine and lower fat mass fits well with what is already known about caffeine’s biological effects. Caffeine stimulates the central nervous system and increases the release of catecholamines, hormones that promote fat breakdown. It also increases thermogenesis, meaning the body burns slightly more energy at rest. While these effects are small on a daily basis, the new findings suggest that over a lifetime, they may add up to meaningful differences in body composition.

This helps explain why the protective effect against diabetes was only partially mediated by body mass index. Caffeine may also influence glucose metabolism through other pathways, such as improved insulin sensitivity or subtle effects on appetite and energy intake.

Putting Coffee Studies Into Perspective

At first glance, these findings may seem to conflict with previous genetic studies of coffee consumption, some of which did not show a protective effect against diabetes. The key difference lies in what is being measured.

This study focused on plasma caffeine concentrations, not the number of cups of coffee consumed. Genetic variants that slow caffeine metabolism lead to higher caffeine levels in the blood, even though carriers of these variants often drink less coffee. This distinction matters because coffee contains hundreds of biologically active compounds besides caffeine, some of which may have beneficial effects and others that may not.

By isolating caffeine itself, the researchers were able to clarify its specific contribution, separate from the broader dietary and cultural context of coffee drinking.

What About the Heart?

Given the close link between diabetes and cardiovascular disease, some readers may wonder why a reduction in diabetes risk did not translate into a clear reduction in heart disease or stroke.

The scientists note several possible explanations. First, the genetic instruments used in the study explain only a small proportion of variation in caffeine levels, which limits statistical power for detecting modest cardiovascular effects. Second, the relationship between caffeine or coffee and heart disease may be non-linear, with moderate intake being beneficial and very high intake offering no additional advantage. Such patterns are difficult to capture with genetic methods that assume linear effects.

It is also possible that the cardiovascular benefits seen in observational studies of coffee consumption are driven by other compounds in coffee, or by lifestyle factors that tend to cluster with moderate coffee drinking.

Strengths and Limits of the Evidence

One of the major strengths of this research is its design. By using Mendelian randomisation, the study reduces the risk of confounding and reverse causation, two persistent problems in nutritional epidemiology. The genetic variants used are directly involved in caffeine metabolism, which strengthens confidence that the observed effects are biologically meaningful.

However, the study also has limitations. The analysis was largely restricted to individuals of European ancestry, which means the findings may not apply equally to other populations. The use of only two genetic variants limits the ability to detect complex or non-linear effects. Finally, genetic evidence does not directly translate into practical recommendations about how much caffeine people should consume.

What This Means for Everyday Life

The findings do not suggest that people should start consuming large amounts of caffeine as a preventive strategy against diabetes. Excessive caffeine intake can cause anxiety, sleep disturbances, and cardiovascular symptoms in some individuals.

Instead, the study adds to a growing body of evidence that caffeine, when consumed in moderation, is unlikely to be harmful and may offer modest metabolic benefits. It also highlights the importance of body weight control as a central pathway through which many dietary factors influence diabetes risk.

The Road Ahead

The scientists emphasize that long-term clinical trials are still needed to test whether non-caloric caffeine-containing beverages can safely reduce body fat or diabetes risk in real-world settings. Such studies would help determine whether the genetic associations observed here can be translated into practical public health strategies.

For now, the research offers a reassuring message for millions of people who enjoy caffeine as part of their daily routine. Far from being just a temporary pick-me-up, caffeine may subtly shape metabolic health over a lifetime, reminding us that even small biological effects can matter when they persist for decades.

The research was published in BMJ Medicine on March 14, 2023.