But it does raise a serious issue - they're studying changes that don't necessarily reflect the selective pressures of present-day life.
Think about it: what are the leading causes of death for people in the prime breeding age (15-34)? Car accidents - by a good margin. So isn't this significant selective pressure to beef up the neck against whiplash, the skull against forehead impact, survival during significant blood loss, etc?
#2 is suicide. I don't know how this rate has changed over time or whether the methods modern humans choose for attempts are more effective than would have been chosen in the past. For example, while men commonly turn to firearms, which are a very effective way to commit suicide, women more often turn to prescription medication overdoses as a method, which overwhelmingly fails.
#3 is poisoning. While humans have always been around poisons, the sheer number that we keep in our houses, most of types that we didn't evolve to, suggests that this may be a stronger selective factor now than it was during our agrarian days, perhaps comparable to that when we were hunter-gatherers or worse.
#4 is homicide. We've definitely gotten a lot better at that, a person is far more likely to die from an intentional gunshot wound than a beating or stabbing. Selective pressures: surviving blood loss, mainly. Stronger, thicker bones may help in against low velocity penetrations.
#5 is other injuries. Again, we're not as likely to suffer from, say "crushed by a mastodon" as an injury, but we've developed plenty of new ways to get killed or maimed in our modern lives.
Then it gets more complicated on the basis that the issue isn't just about survival of the individual, but their social group as a whole, so even nonbreeding members can have a major impact...