The first chart details how cycling frequency is related to cycling risk.
You see that where more people bike, biking is less risky. There are three possible explanations. The first one is that some places are just inherently safer for biking (e.g., really flat), and that’s why people bike there more. That can’t be the whole story, though, as seen when you look at the trend temporally rather than spatially:
(This chart shows the trend in the Netherlands: as people bike more, biking becomes safer). There are two remaining possibilities: that biking becomes more popular as it becomes more safe, or that it becomes more safe as it becomes more popular. It’s no doubt a combination of these two factors, but my rather strongly held opinion is that the second is more important. Keep in mind that neither Denmark nor the Netherlands have mandatory helmet laws, nor a high rate of helmet usage.