It appears to me that the larger a potential impactor is, the more likely it is to cause catastrophic damage to the planet itself, and the harder it will be to prevent it, but the longer we'll have to do something about it. It's orbit is more likely to be well known and the hazard can be evaluated years in advance and some kind of defensive action planned and implemented. Fortunately, these objects visit us infrequently.

The smaller fragments hit us more often. Smaller objects may not be planet-wreckers, but they might be able to cause intense local damage and perhaps millions of casualties. These objects are less predictable because they're smaller, and even if detectable, there may not be time enough to do anything about it. There may even be insufficient time to precisely calculate it's orbit and evacuate a target area.

In general, as the size of the object increases, the more dangerous it becomes and the more time we have to develop mitigating technology, but also the more difficult it may be to implement that technology. There may be little we can do about the smallest objects, but their potential damage approaches that of more conventional natural disasters.

I presume someone has carried out a rigorous risk analysis here and recommended an optimized program(s) to consider all these factors and direct resources at the most cost-effective solutions: one that minimizes the ratio of resources committed per lives saved.