Lisa M Lix1, Marina S Yogendran2, William D Leslie3, Souradet Y Shaw4, Richard Baumgartner5, Christopher Bowman6, Colleen Metge7, Abba Gumel8, Janet Hux9, Robert C James10. 1. Manitoba Centre for Health Policy, University of Manitoba, Canada; Department of Community Health Sciences, University of Manitoba, Canada. Electronic address: lisa_lix@cpe.umanitoba.ca. 2. Manitoba Centre for Health Policy, University of Manitoba, Canada. 3. Department of Medicine, University of Manitoba, Canada. 4. Department of Community Health Sciences, University of Manitoba, Canada. 5. Institute for Biodiagnostics, National Research Council, Winnipeg, Canada. 6. Department of Electrical and Computer Engineering, University of Manitoba, Canada; Institute for Biodiagnostics, National Research Council, Winnipeg, Canada. 7. Manitoba Centre for Health Policy, University of Manitoba, Canada; Faculty of Pharmacy, University of Manitoba, Canada. 8. Department of Mathematics, University of Manitoba, Canada. 9. Institute for Clinical Evaluative Sciences, Toronto, Canada. 10. Private Scholar, Salt Spring Island, British Columbia, Canada.
Abstract
OBJECTIVES: The aim was to construct and validate algorithms for osteoporosis case ascertainment from administrative databases and to estimate the population prevalence of osteoporosis for these algorithms. STUDY DESIGN AND SETTING: Artificial neural networks, classification trees, and logistic regression were applied to hospital, physician, and pharmacy data from Manitoba, Canada. Discriminative performance and calibration (i.e., error) were compared for algorithms defined from different sets of diagnosis, prescription drug, comorbidity, and demographic variables. Algorithms were validated against a regional bone mineral density testing program. RESULTS: Discriminative performance and calibration were poorer and sensitivity was generally lower for algorithms based on diagnosis codes alone than for algorithms based on an expanded set of data features that included osteoporosis prescriptions and age. Validation measures were similar for neural networks and classification trees, but prevalence estimates were lower for the former model. CONCLUSION: Multiple features of administrative data generally resulted in improved sensitivity of osteoporosis case-detection algorithm without loss of specificity. However, prevalence estimates using an expanded set of features were still slightly lower than estimates from a population-based study with primary data collection. The classification methods developed in this study can be extended to other chronic diseases for which there may be multiple markers in administrative data.
OBJECTIVES: The aim was to construct and validate algorithms for osteoporosis case ascertainment from administrative databases and to estimate the population prevalence of osteoporosis for these algorithms. STUDY DESIGN AND SETTING: Artificial neural networks, classification trees, and logistic regression were applied to hospital, physician, and pharmacy data from Manitoba, Canada. Discriminative performance and calibration (i.e., error) were compared for algorithms defined from different sets of diagnosis, prescription drug, comorbidity, and demographic variables. Algorithms were validated against a regional bone mineral density testing program. RESULTS: Discriminative performance and calibration were poorer and sensitivity was generally lower for algorithms based on diagnosis codes alone than for algorithms based on an expanded set of data features that included osteoporosis prescriptions and age. Validation measures were similar for neural networks and classification trees, but prevalence estimates were lower for the former model. CONCLUSION: Multiple features of administrative data generally resulted in improved sensitivity of osteoporosis case-detection algorithm without loss of specificity. However, prevalence estimates using an expanded set of features were still slightly lower than estimates from a population-based study with primary data collection. The classification methods developed in this study can be extended to other chronic diseases for which there may be multiple markers in administrative data.
Authors: Nam-Kyong Choi; Daniel H Solomon; Theodore N Tsacogianis; Joan E Landon; Hong Ji Song; Seoyoung C Kim Journal: J Bone Miner Res Date: 2017-02-07 Impact factor: 6.741
Authors: J M Pritchard; L M Giangregorio; G Ioannidis; A Papaioannou; J D Adachi; W D Leslie Journal: Osteoporos Int Date: 2011-05-12 Impact factor: 4.507
Authors: L M Lix; J P Kuwornu; K Kroeker; G Kephart; K C Sikdar; M Smith; H Quan Journal: Health Promot Chronic Dis Prev Can Date: 2016-03 Impact factor: 3.240
Authors: Shreyasee Amin; Sara J Achenbach; Elizabeth J Atkinson; Sundeep Khosla; L Joseph Melton Journal: J Bone Miner Res Date: 2014-03 Impact factor: 6.741
Authors: Lisa M Lix; Marina S Yogendran; Souradet Y Shaw; Laura E Targownick; Jennifer Jones; Osama Bataineh Journal: BMC Health Serv Res Date: 2010-02-01 Impact factor: 2.655
Authors: Stewart B Harris; Richard H Glazier; Jordan W Tompkins; Andrew S Wilton; Vijaya Chevendra; Moira A Stewart; Amardeep Thind Journal: BMC Health Serv Res Date: 2010-12-23 Impact factor: 2.655