Susan Searles Nielsen1, Mark N Warden1, Alejandra Camacho-Soto1, Allison W Willis1, Brenton A Wright1, Brad A Racette2. 1. From the Department of Neurology (S.S.N., M.N.W., A.C.-S., B.A.W., B.A.R.), Washington University School of Medicine, St. Louis, MO; Departments of Neurology and Biostatistics and Epidemiology (A.W.W.), University of Pennsylvania School of Medicine, Philadelphia; and School of Public Health, Faculty of Health Sciences (B.A.R.), University of the Witwatersrand, Parktown, South Africa. 2. From the Department of Neurology (S.S.N., M.N.W., A.C.-S., B.A.W., B.A.R.), Washington University School of Medicine, St. Louis, MO; Departments of Neurology and Biostatistics and Epidemiology (A.W.W.), University of Pennsylvania School of Medicine, Philadelphia; and School of Public Health, Faculty of Health Sciences (B.A.R.), University of the Witwatersrand, Parktown, South Africa. racetteb@wustl.edu.
Abstract
OBJECTIVE: To use administrative medical claims data to identify patients with incident Parkinson disease (PD) prior to diagnosis. METHODS: Using a population-based case-control study of incident PD in 2009 among Medicare beneficiaries aged 66-90 years (89,790 cases, 118,095 controls) and the elastic net algorithm, we developed a cross-validated model for predicting PD using only demographic data and 2004-2009 Medicare claims data. We then compared this model to more basic models containing only demographic data and diagnosis codes for constipation, taste/smell disturbance, and REM sleep behavior disorder, using each model's receiver operator characteristic area under the curve (AUC). RESULTS: We observed all established associations between PD and age, sex, race/ethnicity, tobacco smoking, and the above medical conditions. A model with those predictors had an AUC of only 0.670 (95% confidence interval [CI] 0.668-0.673). In contrast, the AUC for a predictive model with 536 diagnosis and procedure codes was 0.857 (95% CI 0.855-0.859). At the optimal cut point, sensitivity was 73.5% and specificity was 83.2%. CONCLUSIONS: Using only demographic data and selected diagnosis and procedure codes readily available in administrative claims data, it is possible to identify individuals with a high probability of eventually being diagnosed with PD.
OBJECTIVE: To use administrative medical claims data to identify patients with incident Parkinson disease (PD) prior to diagnosis. METHODS: Using a population-based case-control study of incident PD in 2009 among Medicare beneficiaries aged 66-90 years (89,790 cases, 118,095 controls) and the elastic net algorithm, we developed a cross-validated model for predicting PD using only demographic data and 2004-2009 Medicare claims data. We then compared this model to more basic models containing only demographic data and diagnosis codes for constipation, taste/smell disturbance, and REM sleep behavior disorder, using each model's receiver operator characteristic area under the curve (AUC). RESULTS: We observed all established associations between PD and age, sex, race/ethnicity, tobacco smoking, and the above medical conditions. A model with those predictors had an AUC of only 0.670 (95% confidence interval [CI] 0.668-0.673). In contrast, the AUC for a predictive model with 536 diagnosis and procedure codes was 0.857 (95% CI 0.855-0.859). At the optimal cut point, sensitivity was 73.5% and specificity was 83.2%. CONCLUSIONS: Using only demographic data and selected diagnosis and procedure codes readily available in administrative claims data, it is possible to identify individuals with a high probability of eventually being diagnosed with PD.
Authors: A G Schuurman; M van den Akker; K T J L Ensinck; J F M Metsemakers; J A Knottnerus; A F G Leentjens; F Buntinx Journal: Neurology Date: 2002-05-28 Impact factor: 9.910
Authors: Ewout W Steyerberg; Andrew J Vickers; Nancy R Cook; Thomas Gerds; Mithat Gonen; Nancy Obuchowski; Michael J Pencina; Michael W Kattan Journal: Epidemiology Date: 2010-01 Impact factor: 4.822
Authors: C Warren Olanow; Christopher G Goetz; Jeffrey H Kordower; A Jon Stoessl; Vesna Sossi; Mitchell F Brin; Kathleen M Shannon; G Michael Nauert; Daniel P Perl; James Godbold; Thomas B Freeman Journal: Ann Neurol Date: 2003-09 Impact factor: 10.422
Authors: J F Gagnon; M A Bédard; M L Fantini; D Petit; M Panisset; S Rompré; J Carrier; J Montplaisir Journal: Neurology Date: 2002-08-27 Impact factor: 9.910
Authors: Cathy A Goldstein; Richard B Berry; David T Kent; David A Kristo; Azizi A Seixas; Susan Redline; M Brandon Westover Journal: J Clin Sleep Med Date: 2020-04-15 Impact factor: 4.062
Authors: Alejandra Camacho-Soto; Mark N Warden; Susan Searles Nielsen; Amber Salter; David L Brody; Heidi Prather; Brad A Racette Journal: Ann Neurol Date: 2017-10-31 Impact factor: 10.422
Authors: Alejandra Camacho-Soto; Anat Gross; Susan Searles Nielsen; Anna N Miller; Mark N Warden; Amber Salter; Brad A Racette Journal: Neurology Date: 2020-04-28 Impact factor: 9.910
Authors: Susan Searles Nielsen; Mark N Warden; Markku Sallmén; Markku Sainio; Sanni Uuksulainen; Harvey Checkoway; Christer Hublin; Brad A Racette Journal: Clin Park Relat Disord Date: 2021-04-20