Cristina A Shea1, Razvan Turcu1, Bonny S Wong2, Michelle E Brassil1, Chloe S Slocum1, Richard Goldstein1, Ross D Zafonte1, Shirley L Shih1, Jeffrey C Schneider3. 1. Department of Physical Medicine and Rehabilitation, Harvard Medical School, Spaulding Rehabilitation Research Institute, Charlestown, MA, USA. 2. St David's Medical Center/St. David's Rehabilitation Hospital, Austin, TX, USA. 3. Department of Physical Medicine and Rehabilitation, Harvard Medical School, Spaulding Rehabilitation Research Institute, Charlestown, MA, USA. Electronic address: jcschneider@mgh.harvard.edu.
Abstract
OBJECTIVES: To quantify the rate of readmission from inpatient rehabilitation facilities (IRFs) to acute care hospitals (ACHs) during the first 30 days of rehabilitation stay. To measure variation in 30-day readmission rate across IRFs, and the extent that patient and facility characteristics contribute to this variation. DESIGN: Retrospective analysis of an administrative database. SETTING AND PARTICIPANTS: Adult IRF discharges from 944 US IRFs captured in the Uniform Data System for Medical Rehabilitation database between October 1, 2015 and December 31, 2017. METHODS: Multilevel logistic regression was used to calculate adjusted rates of readmission within 30 days of IRF admission and examine variation in IRF readmission rates, using patient and facility-level variables as predictors. RESULTS: There were a total of 104,303 ACH readmissions out of a total of 1,102,785 IRFs discharges. The range of 30-day readmission rates to ACHs was 0.0%‒28.9% (mean = 8.7%, standard deviation = 4.4%). The adjusted readmission rate variation narrowed to 2.8%‒17.5% (mean = 8.7%, standard deviation = 1.8%). Twelve patient-level and 3 facility-level factors were significantly associated with 30-day readmission from IRF to ACH. A total of 82.4% of the variance in 30-day readmission rate was attributable to the model predictors. CONCLUSIONS AND IMPLICATIONS: Fifteen patient and facility factors were significantly associated with 30-day readmission from IRF to ACH and explained the majority of readmission variance. Most of these factors are nonmodifiable from the IRF perspective. These findings highlight that adjusting for these factors is important when comparing readmission rates between IRFs.
OBJECTIVES: To quantify the rate of readmission from inpatient rehabilitation facilities (IRFs) to acute care hospitals (ACHs) during the first 30 days of rehabilitation stay. To measure variation in 30-day readmission rate across IRFs, and the extent that patient and facility characteristics contribute to this variation. DESIGN: Retrospective analysis of an administrative database. SETTING AND PARTICIPANTS: Adult IRF discharges from 944 US IRFs captured in the Uniform Data System for Medical Rehabilitation database between October 1, 2015 and December 31, 2017. METHODS: Multilevel logistic regression was used to calculate adjusted rates of readmission within 30 days of IRF admission and examine variation in IRF readmission rates, using patient and facility-level variables as predictors. RESULTS: There were a total of 104,303 ACH readmissions out of a total of 1,102,785 IRFs discharges. The range of 30-day readmission rates to ACHs was 0.0%‒28.9% (mean = 8.7%, standard deviation = 4.4%). The adjusted readmission rate variation narrowed to 2.8%‒17.5% (mean = 8.7%, standard deviation = 1.8%). Twelve patient-level and 3 facility-level factors were significantly associated with 30-day readmission from IRF to ACH. A total of 82.4% of the variance in 30-day readmission rate was attributable to the model predictors. CONCLUSIONS AND IMPLICATIONS: Fifteen patient and facility factors were significantly associated with 30-day readmission from IRF to ACH and explained the majority of readmission variance. Most of these factors are nonmodifiable from the IRF perspective. These findings highlight that adjusting for these factors is important when comparing readmission rates between IRFs.
Authors: Ogochukwu Azuh; Harriet Gammon; Charlotte Burmeister; Donald Frega; David Nerenz; Bruno DiGiovine; Aamir Siddiqui Journal: Am J Med Date: 2016-04-21 Impact factor: 4.965
Authors: Sun Jung Kim; Joo Hun Lee; Boram Han; Julia Lam; Elizabeth Bukowy; Avinash Rao; Jordan Vulcano; Anelia Andreeva; Heather Bertelson; Hyun Phil Shin; Ji Won Yoo Journal: Aging Dis Date: 2015-06-01 Impact factor: 6.745
Authors: L Halasyamani; S Kripalani; E Coleman; J Schnipper; C van Walraven; J Nagamine; P Torcson; T Bookwalter; T Budnitz; D Manning Journal: J Hosp Med Date: 2006-11 Impact factor: 2.960
Authors: Marina Soley-Bori; Rene Soria-Saucedo; Colleen M Ryan; Jeffrey C Schneider; Alex B Haynes; Paul Gerrard; Howard J Cabral; Keith D Lillemoe; Lewis E Kazis Journal: J Gen Intern Med Date: 2015-02-18 Impact factor: 5.128
Authors: Steve R Fisher; Yong-Fang Kuo; Gulshan Sharma; Mukaila A Raji; Amit Kumar; James S Goodwin; Glenn V Ostir; Kenneth J Ottenbacher Journal: J Gerontol A Biol Sci Med Sci Date: 2012-12-19 Impact factor: 6.053
Authors: Shirley L Shih; Ross Zafonte; David W Bates; Paul Gerrard; Richard Goldstein; Jacqueline Mix; Paulette Niewczyk; S Ryan Greysen; Lewis Kazis; Colleen M Ryan; Jeffrey C Schneider Journal: J Am Med Dir Assoc Date: 2016-07-14 Impact factor: 4.669
Authors: Shirley L Shih; Marisa Flavin; Richard Goldstein; Chloe Slocum; Colleen M Ryan; Aneesh Singhal; Jason Frankel; Ross Zafonte; Jeffrey C Schneider Journal: Am J Phys Med Rehabil Date: 2020-01 Impact factor: 3.412