Jennifer L Lund1,2, Tzy-Mey Kuo2, M Alan Brookhart1,2, Anne-Marie Meyer1,3, Alexandra F Dalton4, Christine E Kistler2,5, Stephanie B Wheeler2,6, Carmen L Lewis4. 1. Department of Epidemiology, University of North Carolina, Chapel Hill, NC, USA. 2. Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, USA. 3. Epidemiology, IQVIA, St. Prex, Switzerland. 4. Department of Medicine, University of Colorado, Denver, CO, USA. 5. Department of Family Medicine, University of North Carolina, Chapel Hill, NC, USA. 6. Department of Health Policy and Management, University of North Carolina, Chapel Hill, NC, USA.
Abstract
PURPOSE: De-implementation of low-value services among patients with limited life expectancy is challenging. Robust mortality prediction models using routinely collected health care data can enhance health care stakeholders' ability to identify populations with limited life expectancy. We developed and validated a claims-based prediction model for 5-year mortality using regularized regression methods. METHODS: Medicare beneficiaries age 66 or older with an office visit and at least 12 months of pre-visit continuous Medicare A/B enrollment were identified in 2008. Five-year mortality was assessed through 2013. Secondary outcomes included 30-, 90-, and 180-day and 1-year mortality. Claims-based predictors, including comorbidities and indicators of disability, frailty, and functional impairment, were selected using regularized logistic regression, applying the least absolute shrinkage and selection operator (LASSO) in a random 80% training sample. Model performance was assessed and compared with the Gagne comorbidity score in the 20% validation sample. RESULTS: Overall, 183 204 (24%) individuals died. In addition to demographics, 161 indicators of comorbidity and function were included in the final model. In the validation sample, the c-statistic was 0.825 (0.823-0.828). Median-predicted probability of 5-year mortality was 14%; almost 4% of the cohort had a predicted probability greater than 80%. Compared with the Gagne score, the LASSO model led to improved 5-year mortality classification (net reclassification index = 9.9%; integrated discrimination index = 5.2%). CONCLUSIONS: Our claims-based model predicting 5-year mortality showed excellent discrimination and calibration, similar to the Gagne score model, but resulted in improved mortality classification. Regularized regression is a feasible approach for developing prediction tools that could enhance health care research and evaluation of care quality.
PURPOSE: De-implementation of low-value services among patients with limited life expectancy is challenging. Robust mortality prediction models using routinely collected health care data can enhance health care stakeholders' ability to identify populations with limited life expectancy. We developed and validated a claims-based prediction model for 5-year mortality using regularized regression methods. METHODS: Medicare beneficiaries age 66 or older with an office visit and at least 12 months of pre-visit continuous Medicare A/B enrollment were identified in 2008. Five-year mortality was assessed through 2013. Secondary outcomes included 30-, 90-, and 180-day and 1-year mortality. Claims-based predictors, including comorbidities and indicators of disability, frailty, and functional impairment, were selected using regularized logistic regression, applying the least absolute shrinkage and selection operator (LASSO) in a random 80% training sample. Model performance was assessed and compared with the Gagne comorbidity score in the 20% validation sample. RESULTS: Overall, 183 204 (24%) individuals died. In addition to demographics, 161 indicators of comorbidity and function were included in the final model. In the validation sample, the c-statistic was 0.825 (0.823-0.828). Median-predicted probability of 5-year mortality was 14%; almost 4% of the cohort had a predicted probability greater than 80%. Compared with the Gagne score, the LASSO model led to improved 5-year mortality classification (net reclassification index = 9.9%; integrated discrimination index = 5.2%). CONCLUSIONS: Our claims-based model predicting 5-year mortality showed excellent discrimination and calibration, similar to the Gagne score model, but resulted in improved mortality classification. Regularized regression is a feasible approach for developing prediction tools that could enhance health care research and evaluation of care quality.
Authors: Louise C Walter; Karla Lindquist; Sean Nugent; Tammy Schult; Sei J Lee; Michele A Casadei; Melissa R Partin Journal: Ann Intern Med Date: 2009-04-07 Impact factor: 25.391
Authors: Carmen L Lewis; Carol E Golin; Chris DeLeon; Jennifer M Griffith; Jena Ivey; Lyndal Trevena; Michael Pignone Journal: BMC Med Inform Decis Mak Date: 2010-09-17 Impact factor: 2.796
Authors: Rachael K Ross; Tzy-Mey Kuo; Michael Webster-Clark; Carmen L Lewis; Christine E Kistler; Michele Jonsson Funk; Jennifer L Lund Journal: J Am Geriatr Soc Date: 2020-09-05 Impact factor: 5.562
Authors: Jennifer L Lund; Paul R Duberstein; Kah Poh Loh; Nikesha Gilmore; Sandy Plumb; Lianlian Lei; Alexander P Keil; Jessica Y Islam; Laura C Hanson; Jeffrey K Giguere; Victor G Vogel; Brian L Burnette; Supriya G Mohile Journal: J Geriatr Oncol Date: 2021-09-02 Impact factor: 3.599
Authors: Michael P Cary; Farica Zhuang; Rachel Lea Draelos; Wei Pan; Sathya Amarasekara; Brian J Douthit; Yunah Kang; Cathleen S Colón-Emeric Journal: J Am Med Dir Assoc Date: 2020-10-29 Impact factor: 7.802
Authors: Salvatore Tedesco; Martina Andrulli; Markus Åkerlund Larsson; Daniel Kelly; Antti Alamäki; Suzanne Timmons; John Barton; Joan Condell; Brendan O'Flynn; Anna Nordström Journal: Int J Environ Res Public Health Date: 2021-12-04 Impact factor: 3.390