C A Bellera1,2,3, F Artaud1,2, M Rainfray4,5, P L Soubeyran5,6, S Mathoulin-Pélissier1,2,3. 1. Clinical Research and Clinical Epidemiology Unit, Department of Clinical Research and Medical Information, Institut Bergonié, Comprehensive Cancer Centre, Bordeaux. 2. Clinical Epidemiology Unit, INSERM CIC 14.01, Bordeaux. 3. Team EPICENE, University of Bordeaux, INSERM, Bordeaux Population Health Research Center, UMR 1219, F-33000 Bordeaux. 4. Gerontology Service, Centre Hospitalier Universitaire, Bordeaux. 5. University of Bordeaux, Bordeaux. 6. Department of Medical Oncology, Institut Bergonié, Comprehensive Cancer Centre, Bordeaux, France.
Abstract
BACKGROUND: Classification probabilities reflect to what degree a screening test represents the true disease state and include true positive (TPF) and false positive fractions (FPF). With two tests, one can compare TPF and FPF using relative probabilities which offer advantages in terms of interpretation and statistical modeling. Our objective was to highlight how individual and relative TPF and FPF can be easily estimated and compared within a regression modeling framework. This allows the modeling of tests' accuracy while adjusting for multiple covariates, and thus provides valuable information in addition to the crude TPF and FPF. We illustrate our purpose with the G8 and VES-13 screening tests aimed at identifying elderly cancer patients in need for a comprehensive geriatric assessment (CGA). METHODS: Prospective cohort with a paired design. TPF and FPF of each test, as well as relative TPF and FPF were modeled using log-linear models. RESULTS: G8 detected patients in need for CGA better than VES-13 at the expense of misclassifying a large number of normal patients. Both tests had better TPF with older age and poorer performance status (PS), and for all cancer subtypes compared with prostate cancer. Effect of age and PS on TPF was more pronounced with VES-13. Age affected FPF, but not differentially. CONCLUSIONS: Regression modeling helps provide a thorough assessment of the accuracy of diagnostic tests and should be used more frequently. In the context of screening, we encourage the use of G8 as failing to identify patients in need of a CGA might be more problematic than over-detection. Moreover, although we identified variables associated with the sensitivity of these tests, this association was less pronounced for the G8.
BACKGROUND: Classification probabilities reflect to what degree a screening test represents the true disease state and include true positive (TPF) and false positive fractions (FPF). With two tests, one can compare TPF and FPF using relative probabilities which offer advantages in terms of interpretation and statistical modeling. Our objective was to highlight how individual and relative TPF and FPF can be easily estimated and compared within a regression modeling framework. This allows the modeling of tests' accuracy while adjusting for multiple covariates, and thus provides valuable information in addition to the crude TPF and FPF. We illustrate our purpose with the G8 and VES-13 screening tests aimed at identifying elderly cancer patients in need for a comprehensive geriatric assessment (CGA). METHODS: Prospective cohort with a paired design. TPF and FPF of each test, as well as relative TPF and FPF were modeled using log-linear models. RESULTS: G8 detected patients in need for CGA better than VES-13 at the expense of misclassifying a large number of normal patients. Both tests had better TPF with older age and poorer performance status (PS), and for all cancer subtypes compared with prostate cancer. Effect of age and PS on TPF was more pronounced with VES-13. Age affected FPF, but not differentially. CONCLUSIONS: Regression modeling helps provide a thorough assessment of the accuracy of diagnostic tests and should be used more frequently. In the context of screening, we encourage the use of G8 as failing to identify patients in need of a CGA might be more problematic than over-detection. Moreover, although we identified variables associated with the sensitivity of these tests, this association was less pronounced for the G8.
Authors: Arti Hurria; Kayo Togawa; Supriya G Mohile; Cynthia Owusu; Heidi D Klepin; Cary P Gross; Stuart M Lichtman; Ajeet Gajra; Smita Bhatia; Vani Katheria; Shira Klapper; Kurt Hansen; Rupal Ramani; Mark Lachs; F Lennie Wong; William P Tew Journal: J Clin Oncol Date: 2011-08-01 Impact factor: 44.544
Authors: Supriya G Mohile; Kathryn Bylow; William Dale; James Dignam; Kandis Martin; Daniel P Petrylak; Walter M Stadler; Miriam Rodin Journal: Cancer Date: 2007-02-15 Impact factor: 6.860
Authors: L Biganzoli; L Boni; D Becheri; E Zafarana; C Biagioni; S Cappadona; E Bianchini; C Oakman; S U Magnolfi; A Di Leo; G Mottino Journal: Ann Oncol Date: 2012-10-07 Impact factor: 32.976
Authors: Patrick M Bossuyt; Johannes B Reitsma; David E Bruns; Constantine A Gatsonis; Paul P Glasziou; Les M Irwig; Jeroen G Lijmer; David Moher; Drummond Rennie; Henrica C W de Vet Journal: Ann Intern Med Date: 2003-01-07 Impact factor: 25.391
Authors: Eliane Kellen; Paul Bulens; Laura Deckx; Harry Schouten; Marjan Van Dijk; Ilse Verdonck; Frank Buntinx Journal: Crit Rev Oncol Hematol Date: 2010-01-13 Impact factor: 6.312