Assessing the ability of comorbidity indexes to capture comorbid disease in the inpatient rehabilitation burn injury population

Abstract

Objective: Burn patients exhibit comorbidities that influence outcomes. This study examines whether existing comorbidity measures capture comorbidities in the burn inpatient rehabilitation population.

Design: Data were obtained from the Uniform Data System for Medical Rehabilitation from 2002 to 2011 for adults with burn injury. International Classification of Diseases, 9th Revision, codes were used to assess three comorbidity measures (Charlson Comorbidity Index, Elixhauser Comorbidity Index, Centers for Medicare and Medicaid Services Comorbidity Tiers). The number of subjects and unique comorbidity codes (>1% of frequency) captured by each comorbidity measure was calculated.

Results: The study included 5347 patients with a median total body surface area burn decile of 20%-29%, mean age of 51.6 yrs, and mean number of comorbidities of 7.6. There were 2809 unique International Classification of Diseases, 9th Revision, comorbidity codes. The Charlson Comorbidity Index, Elixhauser Comorbidity Index, and Centers for Medicare and Medicaid Services Comorbidity Tiers did not capture 67%, 27%, and 58% of the subjects, respectively. There were 107 unique comorbidities that occurred with a frequency of greater than 1%. Of these, 67% were not captured in all three comorbidity measures.

Conclusions: Commonly used comorbidity indexes do not reflect the extent of comorbid disease in the burn rehabilitation population. Future work is needed to assess the need for comorbidity indexes specific to the inpatient rehabilitation setting.