No, it's not. That's one injury per population amount, so the lower the second number the less people you "need" to get one injury. Assuming a fixed population

No, it's not. That's one injury per population amount, so the lower the second number the less people you "need" to get one injury. Assuming a fixed population size of 1 million, given the 1980 prevalence of injuries you'd expect about ~689 injuries. With the 2012 prevalence of injuries you'd expect about ~865 injuries. So it's actually an increase, not a drop.