Crack can trigger premature labor, as noted in the study. The hypothesis here is that more crack increases the risk of premature labor. It's a hypothesis that might be worth investigating, and I would be surprised if there was no effect observed. As to the extent obviously I cannot say.
> It's a hypothesis that might be worth investigating,
I don't disagree with that and I wouldn't be surprised to learn that such a study already exists.
>I would be surprised if there was no effect observed.
There are a lot of things that can cause premature labor. My personal opinion is that it is just as likely to be some stress otherwise related to their environment, or even stress related to the process of obtaining illegal drugs.
My guess is that if we controlled for environmental and purchasing stress, more crack use alone would lead to more stress, possibly in a dramatic way depending on the levels.
Yes, a related hypothesis is that more crack results in more health problems for a baby. If both are true, then by excluding pre-term babies you are excluding the ones with the most serious health problems.