The differences weren’t giant, but they were statistically significant. This is getting a little into the weeds here (skip down a paragraph if you hate statistics), so bear with for a moment. When doing a large observational study like this, researchers use a hazard ratio to measure how much of an effect a variable like coffee consumption has. For example, in this case a hazard ratio of 1 means there’s no difference in how often people die, whereas a hazard ratio of 0.5 would mean the group that drank coffee died only half as often as the non-coffee drinkers. This study grouped participants into six cohorts: those who drank less than one, one, two to three, four to five, six to seven, and more than eight cups of coffee per day. The researchers found that the hazard ratios generally declined as the number of cups consumed increased, indicating that the more coffee people drank, the less likely they were to die during a set period. In order, the hazard ratios were 0.94, 0.92, 0.88, 0.88, 0.84, and 0.86. All but the first ratio are considered statistically significant—meaning that the difference between the groups was profound enough to (probably) be more than random chance.