Could population-based testing for BRCA-1 and BRCA-2 genes be cost-effective?