Sandra M Shi1,2, Ellen P McCarthy1,2, Susan Mitchell1,2, Dae Hyun Kim1,2. 1. Marcus Institute for Aging Research, Hebrew SeniorLife, Boston, Massachusetts, USA. 2. Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, USA.
Abstract
OBJECTIVES: Determine the effects of missing data in frailty identification and risk prediction. DESIGN: Analysis of the National Health in Aging Trends Study. SETTING: Community. PARTICIPANTS: About 6206 older adults. MEASUREMENTS: A 41-variable frailty index (FI) was constructed with the following domains: comorbidities, activities of daily living (ADLs), instrumental activities of daily living, self-reported physical limitations, physical performance, and neuropsychiatric tests. We evaluated discrimination after removing single and multiple domains, comparing C-statistics for predicting 5-year risk of mortality and 1-year risks of disability and falls. RESULTS: The full FI yielded a mean of .18 and C-statistics of .72 (95% confidence interval, .70-.74) for mortality, .80 (.77-.82) for disability, and .66 (.64-.68) for falls. Removal of any single domain shifted the FI distribution, resulting in a mean FI ranging from .13 (removing comorbidities) to .20 (removing ADLs) and frailty prevalence (FI ≥ .25) from 16.0% to 28.7%. Among robust participants models missing ADLs misclassified most often, (19% as pre-frail). Among pre-frail and frail participants missing comorbidities misclassified most often(69.2% from pre-frail to robust, 24% from frail to pre-frail, and 4.9% from frail to robust). Removal of any single domain minimally changed C-statistics: mortality, .71-.73; disability, .79-.80; and falls, .64-.66. Removing neuropsychiatric testing and physical performance yielded comparable C-statistics of .70, .78, and .66 for mortality, ADLs, and falls, respectively. However, removal of three or four domains based on likely availability decreased C-statistics for mortality (.69, .66),disability (.75, .70), and falls (.64, .63), respectively. CONCLUSION: While FI discrimination is robust to missing information in any single domain, risk prediction is affected by absence of multiple domains. This work informs the application of FI as a clinical and research tool. J Am Geriatr Soc 68:1771-1777, 2020.
OBJECTIVES: Determine the effects of missing data in frailty identification and risk prediction. DESIGN: Analysis of the National Health in Aging Trends Study. SETTING: Community. PARTICIPANTS: About 6206 older adults. MEASUREMENTS: A 41-variable frailty index (FI) was constructed with the following domains: comorbidities, activities of daily living (ADLs), instrumental activities of daily living, self-reported physical limitations, physical performance, and neuropsychiatric tests. We evaluated discrimination after removing single and multiple domains, comparing C-statistics for predicting 5-year risk of mortality and 1-year risks of disability and falls. RESULTS: The full FI yielded a mean of .18 and C-statistics of .72 (95% confidence interval, .70-.74) for mortality, .80 (.77-.82) for disability, and .66 (.64-.68) for falls. Removal of any single domain shifted the FI distribution, resulting in a mean FI ranging from .13 (removing comorbidities) to .20 (removing ADLs) and frailty prevalence (FI ≥ .25) from 16.0% to 28.7%. Among robust participants models missing ADLs misclassified most often, (19% as pre-frail). Among pre-frail and frail participants missing comorbidities misclassified most often(69.2% from pre-frail to robust, 24% from frail to pre-frail, and 4.9% from frail to robust). Removal of any single domain minimally changed C-statistics: mortality, .71-.73; disability, .79-.80; and falls, .64-.66. Removing neuropsychiatric testing and physical performance yielded comparable C-statistics of .70, .78, and .66 for mortality, ADLs, and falls, respectively. However, removal of three or four domains based on likely availability decreased C-statistics for mortality (.69, .66),disability (.75, .70), and falls (.64, .63), respectively. CONCLUSION: While FI discrimination is robust to missing information in any single domain, risk prediction is affected by absence of multiple domains. This work informs the application of FI as a clinical and research tool. J Am Geriatr Soc 68:1771-1777, 2020.
Authors: Karen Bandeen-Roche; Christopher L Seplaki; Jin Huang; Brian Buta; Rita R Kalyani; Ravi Varadhan; Qian-Li Xue; Jeremy D Walston; Judith D Kasper Journal: J Gerontol A Biol Sci Med Sci Date: 2015-08-21 Impact factor: 6.053
Authors: Jeremy Walston; Karen Bandeen-Roche; Brian Buta; Howard Bergman; Thomas M Gill; John E Morley; Linda P Fried; Thomas N Robinson; Jonathan Afilalo; Anne B Newman; Carlos López-Otín; Rafa De Cabo; Olga Theou; Stephanie Studenski; Harvey J Cohen; Luigi Ferrucci Journal: J Am Geriatr Soc Date: 2019-05-02 Impact factor: 5.562
Authors: Kenneth Rockwood; Xiaowei Song; Chris MacKnight; Howard Bergman; David B Hogan; Ian McDowell; Arnold Mitnitski Journal: CMAJ Date: 2005-08-30 Impact factor: 8.262
Authors: Emily J Guerard; Allison M Deal; YunKyung Chang; Grant R Williams; Kirsten A Nyrop; Mackenzi Pergolotti; Hyman B Muss; Hanna K Sanoff; Jennifer L Lund Journal: J Natl Compr Canc Netw Date: 2017-07 Impact factor: 11.908
Authors: Harvey Jay Cohen; David Smith; Can-Lan Sun; William Tew; Supriya G Mohile; Cynthia Owusu; Heidi D Klepin; Cary P Gross; Stuart M Lichtman; Ajeet Gajra; Julie Filo; Vani Katheria; Arti Hurria Journal: Cancer Date: 2016-08-16 Impact factor: 6.860
Authors: Brianne L Olivieri-Mui; Sandra M Shi; Ellen P McCarthy; Daniel Habtemariam; Dae H Kim Journal: J Am Geriatr Soc Date: 2020-11-25 Impact factor: 5.562
Authors: Quoc Dinh Nguyen; Erica M Moodie; Mark R Keezer; Christina Wolfson Journal: J Gerontol A Biol Sci Med Sci Date: 2021-10-13 Impact factor: 6.591
Authors: Natalie D Jenkins; Emiel O Hoogendijk; Joshua J Armstrong; Nathan A Lewis; Janice M Ranson; Judith J M Rijnhart; Tamer Ahmed; Ahmed Ghachem; Donncha S Mullin; Eva Ntanasi; Miles Welstead; Mohammad Auais; David A Bennett; Stefania Bandinelli; Matteo Cesari; Luigi Ferrucci; Simon D French; Martijn Huisman; David J Llewellyn; Nikolaos Scarmeas; Andrea M Piccinin; Scott M Hofer; Graciela Muniz-Terrera Journal: Innov Aging Date: 2022-01-15
Authors: David Cheng; Clark DuMontier; Cenk Yildirim; Brian Charest; Chelsea E Hawley; Min Zhuo; Julie M Paik; Enzo Yaksic; J Michael Gaziano; Nhan Do; Mary Brophy; Kelly Cho; Dae H Kim; Jane A Driver; Nathanael R Fillmore; Ariela R Orkaby Journal: J Gerontol A Biol Sci Med Sci Date: 2021-06-14 Impact factor: 6.053