A lot of people have complaints about their digestive systems and they're looking for answers. From my own experience, it's often not as easy to divine the problem as you'd think. I had just lower than normal lactase levels and I didn't always have severe symptoms. I think it's short-sighted for the insurance company to not cover the testing.
I would also be interested in that 57%. However, given they weren't lactose intolerant, that is a separate question from what they were asking. Your second point is unfounded. This is something you'd only learn in a college statistics or experimental design class, so bear with me. Small sample sizes are a problem only for Type II Errors: the likelihood of failing to reject the null hypothesis when the alternate hypothesis is true. Think of it as a false negative. This can happen when an effect of some variable is very small, but consistent. It can take larger sample sizes to detect a difference. That is not what happened in this study, they found large statistically significant differences in symptoms over time in both raw and pasteurized over both baseline and soy. With effects that large, funding agencies would not support more expansive and expensive study. Having said that, there was a slightly significant difference between raw and pasteurized on day 1 of testing (p=0.04). That would be worth exploring more.
Every study can't answer every question about a subject. The study was supported by the raw milk lobby and that is what they paid for. We should be grateful for it, it's the only one out there. And that's the great thing about science, anyone can go out and collect data that will either agree or disagree with this first study. But, I am in complete agreement that long term studies will be much more valuable. Chronic digestive issues generally don't clear themselves up over night.
Agreed.
ETA: trying to make the stats explanation clearer.