SELECTED ARTICLES FROM THE RECENT LITERATURE 2008 8/15/2008 Cluster analysis used to phenotype asthmatic populations Summary These investigators used a technique known as cluster analysis to analyze asthma phenotypes. Cluster analysis refers to a group of multivariant mathematical algorithms that broadly perform two distinct functions: (1) quantification of similarity between individuals in a population on the basis of the specified variables, and (2) grouping of individuals into clusters such that similarity between members of the same clusters is strong. They looked at two asthma patient groups to make this analysis. One group was followed by primary care physicians, and the other group by asthma specialists. Within these two groups, two phenotypes were common. One phenotype was referred to as early onset atopic asthma, and the other as obese, non-eosinophilic asthma. The cluster more unique to the secondary care group was late onset inflammatory predominant asthma. They used sputum eosinophil counts to guide steroid therapy, and found it generally superior to treatment based on "standard therapy." The differences seen with eosinophilic guided therapy were most marked in the more severe patients. They concluded that symptom-led therapy (non-inflammation assessment) would be effective for mild to moderate asthma in the primary care group, but inflammation guided treatrment (using sputum eosinophils as a guide) was more effective in more severe cases. Reference Haldar P, et al. Cluster analysis and clinical asthma phenotypes. American Journal of Respiratory Critical Care Medicine 2008; 178:218-224.   © Copyright 1997-2008 American Academy of Allergy, Asthma and Immunology All rights reserved