From this
authoritative reference...
In fastening failures we have observed that wheel separations generally occur 175 to 3000 miles, and one to fifteen weeks, after a wheel was taken off and put back on during some service, such as a tire installation. The remainder of this article discusses the causes of wheel separations from fastening failures.
Left vs. right side wheel separations
From our experience and laboratory and on-road testing (Bailey and Bertoch, “Mechanisms of wheel separations,” Society of Automotive Engineers 2009) we have found that there is a different pattern of evidence in left vs. right side wheel separations. Left side wheel separations usually occur after the wheel nuts spin off and right side wheel separations tend to occur after the wheel studs break off.
Figure 5 shows a wheel stud and the hole of the left wheel that it was in. The nut has spun off the stud and was never found.
and...
The wheel nuts and studs basically sandwich the wheel and brake components together. The nuts and studs have to squeeze those components together with enormous force, called the clamping force, in order for the sandwich to stay together. The clamping force is made when the nuts are tightened onto the studs. If the clamping force is lost, then the nuts loosen, leading to the different mechanisms of left-side nut spin-off and right-side stud reversed-bending fatigue.
The mechanism for the left side wheel nuts spinning off as the vehicle travels can be understood from the geometry of the wheel and studs when the wheel is slightly loose. Since the stud holes are larger than the studs, the wheel is not perfectly concentric with the axle when the nuts are loose. When the road pushes up on the tire, the wheel tends to be pushed up relative to the axle centerline. This means the wheel centerline is slightly above the axle centerline. This centerline offset gives rise to a relative velocity vector between each wheel nut and the part of the wheel the nuts touch. This vector is in the loosening direction on the left side when the vehicle is driving forward and is the cause of wheel nuts spinning off the left side.
The right side nuts have that same relative velocity vector, but in the opposite (i.e., tightening) direction. It turns out that this vector is not strong enough to make a loose nut tight again. So, on the right side, a loose nut tends to stay loose rather than spin off. But this invites another mechanism – fatigue. When a nut is tight, the clamping force creates large frictional forces at the wheel/hub interface that transfer the vertical forces that support the weight of the car. However, when a nut is loose, there is no clamping force, and the studs now carry the vertical forces. This bends the studs up and down every time the tire rotates. Just as a paper clip breaks when you bend it back and forth a few times, a wheel stud can break when it is bent up and down a few million times. This is called reversed-bending fatigue, and is the reason that right-side studs eventually break off when the nuts are loose.
The starting point for a wheel nut to spin off the left side, or for a stud to break off the right side, is the same: the nut had to be loose. Therefore, the investigation of fastener-related wheel separations should focus on the clamping force.
I'm home, grounded, sick, so I tried a little more research. From that...
Does this trailer use Aluminum (mag, whatever, not Plain Old Steel) Wheels? They seem to promote for clamping force (torque) issues, and also expand/contract from heat differently than the steel in the hubs and brakes.
And speaking of heat, an article mentioned how a hot brake can promote loosening of lug nuts since there's more expansion/contraction than if the brake isn't dragging.