Higher diet quality at study start linked to lower heart failure risk over 20 years

In the early 2000s, researchers in Greece asked 4,056 adults living around Athens what they ate. They recorded everything—the olive oil, the bread, the processed snacks, the vegetables. Then they waited. Twenty years later, they checked back to see who had developed heart failure.

The results were striking enough to warrant attention, though not the kind that changes medicine overnight. Among the 2,159 people who completed the full study and remained free of heart disease at the start, 241 had developed heart failure by the 20-year mark. That's an 11.2% cumulative incidence. But the distribution was uneven. Those who ate the highest-quality diets at baseline—measured by a scoring system that tallied healthy foods against less healthy ones—had a 6.6% incidence rate. Those who ate the poorest diets had a 20.7% rate. The difference was real and substantial.

The study, published in the journal Nutrients and drawn from the ATTICA cohort, used a tool called the Global Diet Quality Score, or GDQS, to assess what people were eating. The score ranged from 0 to 49 and was split into two components: one measuring healthy-food consumption and another measuring how much people ate from less healthy food groups. Researchers divided participants into three groups based on their scores. The median score was 37, with healthy-food subscores averaging 31 and unhealthy-food subscores averaging 6. People in the lowest-scoring group tended to be older, heavier, and more likely to have high blood pressure, diabetes, and high cholesterol. They were also less physically active.

When the researchers adjusted their analysis for age, sex, body mass index, education, physical activity, smoking, and existing metabolic conditions, the pattern held. For every one-point increase in the overall diet quality score, the odds of developing heart failure dropped by 3%. For every one-point increase in the healthy-food component specifically, the odds dropped by 4%. The unhealthy-food component, by contrast, showed no independent association with heart failure risk—a finding the researchers attributed partly to limited variation in how much unhealthy food people reported eating.

The strongest protective effect appeared among people who had at least one cardiovascular risk factor at the start of the study. Among those with exactly one risk factor, each point increase in diet quality correlated with 4% lower heart failure odds. The benefit seemed to diminish among people who already carried multiple risk factors, though the association remained significant when researchers pooled all participants with at least one risk factor.

The study also revealed a dose-response relationship between the number of existing risk factors and heart failure incidence. Among people with no cardiovascular risk factors at baseline, 3.3% developed heart failure over 20 years. That rose to 8.8% among those with one risk factor, 15.6% with two, and 24.3% with three or more. Diet quality appeared to offer some protection across these groups, but the effect was most measurable in people who were not yet heavily burdened by disease.

What the researchers could not do—and what they were careful to say they could not do—was prove that changing your diet prevents heart failure. This was an observational study, meaning researchers watched what people ate and what happened to them, but could not randomly assign people to eat differently and measure the result. People who eat higher-quality diets tend to exercise more, weigh less, and have better-controlled blood pressure and blood sugar. Any or all of those factors might explain the lower heart failure rates. The study also assessed diet only at the beginning, not over the full 20 years, so researchers could not account for how eating habits changed over time. Some people were lost to follow-up, and those who were tended to be more educated—a detail that hints at how socioeconomic factors might shape both diet and health outcomes in ways the study could not fully capture.

The findings are strongest for the healthy-food component of the diet score and weakest for the unhealthy-food component. The researchers noted that confidence intervals were wide in some comparisons, and a sensitivity analysis that excluded some participants showed weaker results. The study population was Greek, living in a specific region, and the results may not apply equally to other populations with different food environments and dietary traditions. Still, the magnitude of the difference—a threefold gap in heart failure incidence between the highest and lowest diet-quality groups—suggests that what people eat in their 40s and 50s carries real weight for their cardiovascular health two decades later.