Timothy J Daskivich1, Lorna Kwan2, Atreya Dash3, Sheldon Greenfield4, Mark S Litwin5. 1. Department of Urology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA; Greater Los Angeles Veterans Affairs Medical Center, Los Angeles, CA, USA. Electronic address: Tdaskivich@ucla.edu. 2. Department of Urology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA. 3. Department of Urology, University of Washington, Seattle, WA, USA. 4. Center for Health Policy Research and Department of Medicine, University of California, Irvine, Irvine, CA, USA. 5. Department of Urology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA; Jonsson Comprehensive Cancer Center, University of California, Los Angeles, Los Angeles, CA, USA; Department of Health Policy and Management, Fielding School of Public Health, University of California, Los Angeles, Los Angeles, CA, USA.
Abstract
BACKGROUND: Clinicians need a simple yet accurate method to predict other-cause mortality to inform medical decision making for men with prostate cancer (PCa). OBJECTIVE: To compare weighted and unweighted Charlson Comorbidity Index scores in predicting long-term, other-cause mortality in men with early-stage PCa. DESIGN, SETTING, AND PARTICIPANTS: A retrospective cohort study of 1482 men with early-stage PCa diagnosed in 1998-2004 at two Southern California Veterans Affairs medical centers. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Subhazard ratios and cumulative incidence of other-cause mortality associated with weighted and unweighted Charlson scores, calculated by competing-risks regression accounting for cancer mortality, along with Harrell concordance index (C-index) values. RESULTS AND LIMITATIONS: Weighted and unweighted Charlson scores were identical in 88.6% of subjects (1313 of 1482 men) across all scores and in 91.7% of subjects (1359 of 1482 men) across scores of 0, 1, 2, and ≥3. In competing-risks analysis, hazards of other-cause mortality were similar when comparing weighted and unweighted scores. Men with weighted scores of 1, 2, and ≥3 (vs. 0) had subhazard ratios of 2.3 (95% confidence interval [CI], 1.6-3.2), 4.1 (95% CI, 2.9-5.8), and 8.3 (95% CI, 5.9-11.5), respectively. Men with unweighted scores of 1, 2, and ≥3 (vs. 0) had subhazard ratios of 2.5 (95% CI, 1.8-3.5), 4.5 (95% CI, 3.2-6.3), and 10.3 (95% CI, 7.2-14.7), respectively. The C-indexes for prediction of other-cause mortality were nearly identical for weighted scores (0.759 [95% CI, 0.715-0.780]) and unweighted scores (0.756 [95% CI, 0.717-0.780]). The difference in C-index between the two methods was -0.003 (95% CI, -0.01 to 0.004). CONCLUSIONS: An unweighted Charlson score yields similar strength of association and variance in predicting long-term, other-cause mortality compared with a weighted Charlson score. PATIENT SUMMARY: A simple count of major comorbidities provides similar accuracy to a weighted index in predicting death from other causes in men with early-stage prostate cancer. Published by Elsevier B.V.
BACKGROUND: Clinicians need a simple yet accurate method to predict other-cause mortality to inform medical decision making for men with prostate cancer (PCa). OBJECTIVE: To compare weighted and unweighted Charlson Comorbidity Index scores in predicting long-term, other-cause mortality in men with early-stage PCa. DESIGN, SETTING, AND PARTICIPANTS: A retrospective cohort study of 1482 men with early-stage PCa diagnosed in 1998-2004 at two Southern California Veterans Affairs medical centers. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Subhazard ratios and cumulative incidence of other-cause mortality associated with weighted and unweighted Charlson scores, calculated by competing-risks regression accounting for cancer mortality, along with Harrell concordance index (C-index) values. RESULTS AND LIMITATIONS: Weighted and unweighted Charlson scores were identical in 88.6% of subjects (1313 of 1482 men) across all scores and in 91.7% of subjects (1359 of 1482 men) across scores of 0, 1, 2, and ≥3. In competing-risks analysis, hazards of other-cause mortality were similar when comparing weighted and unweighted scores. Men with weighted scores of 1, 2, and ≥3 (vs. 0) had subhazard ratios of 2.3 (95% confidence interval [CI], 1.6-3.2), 4.1 (95% CI, 2.9-5.8), and 8.3 (95% CI, 5.9-11.5), respectively. Men with unweighted scores of 1, 2, and ≥3 (vs. 0) had subhazard ratios of 2.5 (95% CI, 1.8-3.5), 4.5 (95% CI, 3.2-6.3), and 10.3 (95% CI, 7.2-14.7), respectively. The C-indexes for prediction of other-cause mortality were nearly identical for weighted scores (0.759 [95% CI, 0.715-0.780]) and unweighted scores (0.756 [95% CI, 0.717-0.780]). The difference in C-index between the two methods was -0.003 (95% CI, -0.01 to 0.004). CONCLUSIONS: An unweighted Charlson score yields similar strength of association and variance in predicting long-term, other-cause mortality compared with a weighted Charlson score. PATIENT SUMMARY: A simple count of major comorbidities provides similar accuracy to a weighted index in predicting death from other causes in men with early-stage prostate cancer. Published by Elsevier B.V.
Authors: Joo Yong Lee; Ho Won Kang; Koon Ho Rha; Nam Hoon Cho; Young Deuk Choi; Sung Joon Hong; Kang Su Cho Journal: J Cancer Res Clin Oncol Date: 2015-12-12 Impact factor: 4.553
Authors: Michael Froehner; Rainer Koch; Stefan Propping; Dorothea Liebeheim; Matthias Hübler; Gustavo B Baretton; Oliver W Hakenberg; Manfred P Wirth Journal: Asian J Androl Date: 2017 Mar-Apr Impact factor: 3.285
Authors: Gregory L Snow; Joseph R Bledsoe; Allison Butler; Emily L Wilson; Susan Rea; Sarah Majercik; Jeffrey L Anderson; Benjamin D Horne Journal: PLoS One Date: 2020-05-21 Impact factor: 3.240
Authors: Andrew Swale; Fabio Miyajima; Paul Roberts; Amanda Hall; Margaret Little; Mike B J Beadsworth; Nick J Beeching; Ruwanthi Kolamunnage-Dona; Chris M Parry; Munir Pirmohamed Journal: PLoS One Date: 2014-08-29 Impact factor: 3.240