The news about coffee just keeps getting better. In a new analysis of one of the country’s largest and longest-running studies, drinking coffee was linked to a lower risk of heart failure, stroke and coronary heart disease. Every extra cup of coffee consumed per day reduced each of these conditions by 8%, 7% and 5%, respectively, up to at least six cups per day.
The preliminary research was presented at the American Heart Association’s Scientific Sessions in Anaheim, California. It has not yet been published in a peer-reviewed medical journal, but it’s not the first research to suggest that coffee protects the heart and arteries.
Researchers from the University of Colorado medical school analyzed data from the Framingham Heart Study, which has tracked the eating patterns and cardiovascular health of more than 15,000 people since the 1940s. They were looking for previously unidentified risk factors for heart failure and stroke. They used a method known as machine learning, a form of artificial intelligence that looks for patterns in big data sets, similar to the way e-commerce websites might predict products a customer mighty like based on their previous shopping history.
“In an ideal world, we would be able to predict cardiovascular disease and stroke with 100% accuracy long before the occurrence of the event,” said first author Laura Stevens, a doctoral student at the University of Colorado School of Medicine, in an email to TIME. “The challenge here is there are so many potential risk factors, and testing each one using traditional methods would be extremely time consuming, and possibly infeasible,” she added.
Out of all the potential links to heart disease the researchers considered, one stood out after the analysis. Coffee was associated with a reduced risk for heart failure, stroke and coronary heart disease. For coffee drinkers, every 8-ounce cup per day reduced these risks by 7%, 8% and 5%, respectively, compared to people who didn’t drink coffee. Almost all of the coffee drinkers in the study (97%) consumed between one and six cups of coffee a day, says Stevens, so the researchers can’t know for sure if the benefits continue at even higher consumption levels.
The researchers even determined that whether someone drank coffee or not could help predict their eventual risk of heart failure or stroke. When they added coffee-drinking to a risk assessment tool that also included other known risk factors (like age, blood pressure and cholesterol), their prediction accuracy improved by 4%.
The researchers then confirmed their findings with more traditional analyses of two additional large study groups: the Cardiovascular Heart Study and the Atherosclerosis Risk in Communities Study.
Because these studies simply observed people’s health and coffee consumption over time, the analyses were only able to determine a link between the two—not a cause-and-effect relationship. But based on these and other findings, experts suspect that coffee may be protective against conditions like heart failure and stroke. Previous research has suggested that coffee’s caffeine content, along with its antioxidant and anti-inflammatory properties, may be responsible for its presumed health benefits.
The new research also supports the idea that machine learning may help researchers identify other unknown risk factors—or protective factors—for disease. The current risk-assessment tools used to predict whether someone might develop heart disease are very good, the authors noted in their presentation, but they’re not 100% accurate, suggesting that more risk factors could still be identified.
Stevens says her team intends to use similar analyses to determine if compounds like caffeine may explain the association with heart health observed in this study. “Ultimately, our key goals are to determine whether coffee consumption is a clinically useful part of cardiovascular disease risk assessment,” she says, “and whether changing coffee or caffeine consumption may be a way of altering that risk.”