Skip to main content

Table 2 Background on the use of the Charlson and Elixhauser measures using HES data

From: Weekend admissions may be associated with poorer recording of long-term comorbidities: a prospective study of emergency admissions using administrative data

Although both the Charlson and Elixhauser measures should ideally identify comorbidities by including information reported at prior admissions, [24, 40, 53] often only data from the ‘index’ hospital admission (i.e. the admission of interest) is used to determine comorbidities at the time of admission [54, 55]. When risk adjusting using the Charlson list or index, data can be limited to diagnoses recorded in the first episode of the index hospital admission, to avoid the risk of including complications that arose during the hospital admission. The Elixhauser measure, however, was designed for use on databases that cannot distinguish between comorbidities and complications and limits its set of conditions to either chronic conditions or acute conditions that are unlikely to be potential complications [21]. When risk adjusting using the Elixhauser list, comorbidity data from all episodes within a hospital stay can be used. Depending on the purpose of the risk adjustment, one either includes only secondary diagnoses from the index admission [16, 58] (on the assumption that these fields contain concurrent conditions that may impact on mortality rates, over and above the primary reason for the admission) or all diagnosis fields.