OBJECTIVE: To define a method of recognizing influence of comorbid conditions when studying outcomes of a specific disease or therapy. Background: Patients are often suffering from multiple disease states. When looking at resource utilization, absenteeism, and other outcomes for a specific disease state or therapy it is important to recognize the influence of other diseases to the total costs and outcomes of the disease of interest. METHODS: Several comorbidity indexes have been developed. Well known and described are the Charlson Index with adaptations, Index of Coexistent Disease, Cumulative Illness Rating Scale, Chronic Disease Score and several other disease specific indexes. The typical use of these indexes is predictive, a forecast of mortality or costs based on the current history of a patient. Many of these indexes require clinical information. The focus of the current index is strictly toward administrative (claims) databases. RESULTS: The methodology presented here is a nonlinear transformation of the number of comorbid conditions, scaled so that the probability of significant comorbidity increases monotonically with the number of conditions reported. This AHRQ Comorbidity Index algorithms described by Elixhauser are used to define a set of thirty-one comorbid conditions as defined by one thousand one hundred-eleven ICD9 diagnoses. Sensitivity analysis of the number of expected conditions resulted in a denominator of ten units. CONCLUSIONS: The denominator to yield a percentage of costs, outcomes, etc attributable to the comorbid conditions divides the number of Elixhauser conditions, other than the disease of interest. This index takes advantage of the longitudinal nature of claims data to be inclusive of all disease states within a patient and not just the current cause for treatment. It provides an alternative approach to assign costs, absenteeism, and other outcomes to a disease state that is a more accurate valuation of the outcomes without undue influence from other comorbidities.