We have evaluated three genotyping techniques to detect allelic association between single nucleotide polymorphisms (SNPs) and disease loci by using pooled DNA samples. The use of DNA pools instead of individual samples saves considerably time, DNA quantity, and reagent costs. The methods that were compared were PCR-restriction fragment length polymorphism (PCR-RFLP), Single nucleotide primer extension, and chip-based mass spectrometry. As a model disease we studied an autosomal recessive disorder, congenital chloride diarrhea (CLD), caused by mutations in the human gene SLC26A3 on chromosome 7q31. Patient, carrier, and control pools were genotyped with eight SNP markers located within 10 cM around the target gene. Measured allele frequencies were consistent between the three platforms and we detected disease association with a cluster of several SNPs, the most significant being the marker located closest to the target gene (p=1.56 x 10(-16)). The largest source of error in pooled genotyping was shown to be the volumetric error in preparing the pools. The PCR-RFLP method was further evaluated by applying the pooling strategy to an ongoing multicenter-study. Measured allele frequencies of six polymorphisms agreed well with those measured individually and confirmed that multiple platforms are suitable for pooled genotyping.