Janis D Harse1, C D'Arcy J Holman. 1. School of Population Health, University of Western Australia, Nedlands. jharse@iinet.net.au
Abstract
OBJECTIVE: We examined the predictive validity of Charlson's Index as a tool to measure and adjust for comorbidity in studies of health-related quality of life(HRQOL) outcomes after joint replacement surgery. STUDY DESIGN AND SETTING: SF-36 physical component summary scores were available for a cohort of patients who underwent primary hip or knee replacement surgery at one hospital over a 12-month period. Baseline comorbidity was assessed for the same group of patients using longitudinal hospital morbidity data from the Western Australia Department of Health. The presence or absence of individual conditions was determined, and Charlson's Index scores were calculated for each patient, using varying look-back periods. RESULTS: In regression analysis, Charlson's Index was a poor predictor of the HRQOL outcome scores, explaining a maximum 1.79% of the variance. In contrast, the presence or absence of a small number of individual conditions together explained between 5% and 7% of the variance. CONCLUSION: The findings suggest that Charlson's Index should not be used to adjust for HRQOL outcomes, particularly in this patient group with low levels of serious comorbidity. Alternative methods are needed for use in this context.
OBJECTIVE: We examined the predictive validity of Charlson's Index as a tool to measure and adjust for comorbidity in studies of health-related quality of life(HRQOL) outcomes after joint replacement surgery. STUDY DESIGN AND SETTING: SF-36 physical component summary scores were available for a cohort of patients who underwent primary hip or knee replacement surgery at one hospital over a 12-month period. Baseline comorbidity was assessed for the same group of patients using longitudinal hospital morbidity data from the Western Australia Department of Health. The presence or absence of individual conditions was determined, and Charlson's Index scores were calculated for each patient, using varying look-back periods. RESULTS: In regression analysis, Charlson's Index was a poor predictor of the HRQOL outcome scores, explaining a maximum 1.79% of the variance. In contrast, the presence or absence of a small number of individual conditions together explained between 5% and 7% of the variance. CONCLUSION: The findings suggest that Charlson's Index should not be used to adjust for HRQOL outcomes, particularly in this patient group with low levels of serious comorbidity. Alternative methods are needed for use in this context.
Authors: Joshua D Meade; Garrett R Jackson; Michael S Schallmo; Bradley Young; Robert L Parisien; David P Trofa; Patrick M Connor; Shadley Schiffern; Nady Hamid; Bryan M Saltzman Journal: Int Orthop Date: 2022-06-06 Impact factor: 3.479
Authors: José M Quintana; Antonio Escobar; Urko Aguirre; Iratxe Lafuente; Juan C Arenaza Journal: Clin Orthop Relat Res Date: 2009-05-02 Impact factor: 4.176
Authors: A Judge; R N Batra; G E Thomas; D Beard; M K Javaid; D W Murray; P A Dieppe; K E Dreinhoefer; K Peter-Guenther; R Field; C Cooper; N K Arden Journal: Osteoarthritis Cartilage Date: 2014-01-11 Impact factor: 6.576
Authors: Paul A Scuffham; Michael F Hilton; Alexander Muspratt; Shu-Kay Ng; Harvey A Whiteford; Libby Holden Journal: Popul Health Metr Date: 2011-06-02
Authors: Andy Judge; Nigel K Arden; Rajbir N Batra; Geraint Thomas; David Beard; M Kassim Javaid; Cyrus Cooper; David Murray Journal: BMJ Open Date: 2013-03-01 Impact factor: 2.692