Rare doesn't always mean mild. I think that's a misconception. Just like in genetics, dominant doesn't mean more frequent. Another misconception. I think treatment, luck, etc, has a lot to do with it. If you look at worldwide mutation prevalence, as according to Cystic Fibrosis: Genotypic and Phenotypic Variations" by J. Zielenski and L.-C. Tsui in Annual Rev. of Genet. (29: 777-807, 1995), all mutations except for DeltaF508 are rare (66%). Ry's was considered fairly common amoung Jewish Cfers, but only 1.2% of the CFers in the world have his mutation.
Anywho, what the classes mean in severity is how much they affect things on a molecular level.
Class I mutations, which affect fewer than 7% of cystic fibrosis patients, are stop mutations that directly interfere with CFTR protein production.
Class II mutations disrupt the trafficking mechanism that takes CFTR from its point of origin on the endoplasmic reticulum to its site of operation on the apical membrane. Approximately 85% of cystic fibrosis patients have at least one copy of the primary class II mutation, delta-F508
Class III mutations, which affect fewer than 3% of patients, cause defective ATP- or phosphorylation-dependent regulation of CFTR. The protein has reached the apical membrane, but it does not become activated and inactivated in the normal fashion
Class IV mutations disrupt conductance through CFTR ion channels, despite normal placement and regulation of the molecule
Finally, class V mutations result in reduced synthesis of functional CFTR.
Prevalence of class IV and V mutations is difficult to assess because the effects tend to be mild and not associated with early lung or gastrointestinal problems.
So you could see how a class one mutation could be more symptomatic than a class 3,4,5. It's not an end all and be all, but it's a molecular level fact. But not anyway of predicting anything, they've proven they don't really know WHY some CFers do well and some not, mutation is just a part of it.