Letter: Cystic fibrosis: description and genetics.

It is regrettable that the excellent article by Bowman and Mangos on 
cystic fibrosis should be marred by a minor, yet unfortunately common, 
error. The risks of producing a child with cystic fibrosis in 
Table 1 would have been essentially correct if they had been 
stated in fractional terms. They are, however, stated in odds forms, 
and therefore understatements of the true risks. The risk when both 
parents are known to be heterozygous, for example, is stated as 
1:4. This would mean that the second alternative has four times 
as much likelihood of occurring as the first, and it is 
clearly not true. When both parents are heterozygous the odds are 
three times as likely that the child will be normal as 
affected. This is usually stated as a risk of 1/4, or 
1:3, that the child will be affected. 