Self Report Co-Morbidity and Health Related Quality of Life – A Comparison with Record Based Co-Morbidity Measures

The purpose of this projectwas to compare three hospital-based measures ofco-morbidity to patient self-reportco-morbidity and to determine the relativeproportion of outcome predicted by each of theco-morbidity measures in a population ofindividuals receiving major joint arthroplasty. Baseline measures using the SF-36 generalhealth questionnaire and the Western OntarioMcMaster Osteoarthritis Index (WOMAC) wereobtained from 518 persons undergoing total kneeor hip replacement. A second measure wasobtained six months post-surgery. Co-morbiditywas calculated by summing the self-reportedco-morbidity at baseline, using both thechart-based and administrative data version ofCharlson's Co-morbidity Index, and by summingthe number of International Classification ofDiseases – Version 9 (ICD-9) codes appearingin the electronic health record. Linearregression was used to determine how much ofthe variation in outcome was explained by eachof the co-morbidity measurement methods. Self-report co-morbidity explained as muchvariation in outcome as the hospital-basedindices. Self-report co-morbidity did notperform as well as the other methods inexplaining the variance in health systemutilization. It was concluded that self-reportco-morbidity is minimally as useful asrecord-based systems when measuring the impactof co-morbidity on health related quality oflife (HRQL). This is an important finding, asprivacy legislation, the time until data isavailable and cost are all barriers to usingrecord-based co-morbidity measures.