Identification of the molecular basis of the beta-thalassaemias and sickle cell disease (SCD) has made it clear that patients with the same beta-globin genotypes can have very variable patterns of clinical expression. Extensive biochemical and pathophysiological studies over the last 50 years have derived two major modifiers--innate ability to produce fetal haemoglobin and co-inheritance of alpha-thalassaemia, subsequently validated by family and population studies. However, these two modifiers do not explain the full clinical spectrum. Genetic studies have been successful in identifying modifiers if the loci have a major clinical effect and if the genetic variants are common. It is possible that additional modifiers could be uncovered using genetic approaches but success will depend on large sample sizes of well-characterised patients with well-defined phenotypes. Since some of the complications, such as overt stroke in SCD, are relatively rare events, intermediate end-points that contribute to the phenotype, such as Transcranial Doppler velocity (a major predictor of stroke in SCD), could be integrated within the genetic analysis. Integrating multiplex genetic testing with clinical and laboratory data to generate predictive models shows potential, but such genetic approaches also require large datasets.