Elizabeth T Loggers1, Hongyuan Gao2, Laura S Gold3, Larry Kessler4,5, Ruth Etzioni4,5, Diana Sm Buist2,4,5. 1. Fred Hutchinson Cancer Research Center, Clinical Research Division, 1100 Fairview Ave, D5-380, Seattle, WA 98109, USA. 2. Group Health Research Institute, Seattle, WA, USA. 3. School of Pharmacy, University of Washington, Seattle, WA, USA. 4. Department of Health Services, School of Public Health, University of Washington, Seattle, WA, USA. 5. Fred Hutchinson Cancer Research Center, Public Health Sciences Division, Seattle, WA, USA.
Abstract
AIM: Investigate how the results of predictive models of preoperative MRI for breast cancer change based on available data. MATERIALS & METHODS: A total of 1919 insured women aged ≥18 with stage 0-III breast cancer diagnosed 2002-2009. Four models were compared using nested multivariable logistic, backwards stepwise regression; model fit was assessed via area under the curve (AUC), R2. RESULTS: MRI recipients (n = 245) were more recently diagnosed, younger, less comorbid, with higher stage disease. Significant variables included: Model 1/Claims (AUC = 0.76, R2 = 0.10): year, age, location, income; Model 2/Cancer Registry (AUC = 0.78, R2 = 0.12): stage, breast density, imaging indication; Model 3/Medical Record (AUC = 0.80, R2 = 0.13): radiologic recommendations; Model 4/Risk Factor Survey (AUC = 0.81, R2 = 0.14): procedure count. CONCLUSION: Clinical variables accounted for little of the observed variability compared with claims data.
AIM: Investigate how the results of predictive models of preoperative MRI for breast cancer change based on available data. MATERIALS & METHODS: A total of 1919 insured women aged ≥18 with stage 0-III breast cancer diagnosed 2002-2009. Four models were compared using nested multivariable logistic, backwards stepwise regression; model fit was assessed via area under the curve (AUC), R2. RESULTS: MRI recipients (n = 245) were more recently diagnosed, younger, less comorbid, with higher stage disease. Significant variables included: Model 1/Claims (AUC = 0.76, R2 = 0.10): year, age, location, income; Model 2/Cancer Registry (AUC = 0.78, R2 = 0.12): stage, breast density, imaging indication; Model 3/Medical Record (AUC = 0.80, R2 = 0.13): radiologic recommendations; Model 4/Risk Factor Survey (AUC = 0.81, R2 = 0.14): procedure count. CONCLUSION: Clinical variables accounted for little of the observed variability compared with claims data.
Entities:
Keywords:
MRI; breast cancer; chart abstraction; claims/utilization data; predictive variables; risk factor data; survey data
Authors: J P Costantino; M H Gail; D Pee; S Anderson; C K Redmond; J Benichou; H S Wieand Journal: J Natl Cancer Inst Date: 1999-09-15 Impact factor: 13.506
Authors: Constance D Lehman; Constantine Gatsonis; Christiane K Kuhl; R Edward Hendrick; Etta D Pisano; Lucy Hanna; Sue Peacock; Stanley F Smazal; Daniel D Maki; Thomas B Julian; Elizabeth R DePeri; David A Bluemke; Mitchell D Schnall Journal: N Engl J Med Date: 2007-03-28 Impact factor: 91.245
Authors: Carrie N Klabunde; Julie M Legler; Joan L Warren; Laura-Mae Baldwin; Deborah Schrag Journal: Ann Epidemiol Date: 2007-05-25 Impact factor: 3.797
Authors: Meagan Elizabeth Brennan; Nehmat Houssami; Sarah Lord; Petra Macaskill; Les Irwig; J Michael Dixon; Ruth M L Warren; Stefano Ciatto Journal: J Clin Oncol Date: 2009-10-05 Impact factor: 44.544