Predicting institutionalization after traumatic brain injury inpatient rehabilitation

Regina S. Eum, Ronald T. Seel, Richard Goldstein, Allen W Brown, Thomas K. Watanabe, Nathan D. Zasler, Elliot J. Roth, Ross D. Zafonte, Mel B. Glenn

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Risk factors contributing to institutionalization after inpatient rehabilitation for people with traumatic brain injury (TBI) have not been well studied and need to be better understood to guide clinicians during rehabilitation. We aimed to develop a prognostic model that could be used at admission to inpatient rehabilitation facilities to predict discharge disposition. The model could be used to provide the interdisciplinary team with information regarding aspects of patients' functioning and/or their living situation that need particular attention during inpatient rehabilitation if institutionalization is to be avoided. The study population included 7219 patients with moderate-severe TBI in the Traumatic Brain Injury Model Systems (TBIMS) National Database enrolled from 2002-2012 who had not been institutionalized prior to injury. Based on institutionalization predictors in other populations, we hypothesized that among people who had lived at a private residence prior to injury, greater dependence in locomotion, bed-chair-wheelchair transfers, bladder and bowel continence, feeding, and comprehension at admission to inpatient rehabilitation programs would predict institutionalization at discharge. Logistic regression was used, with adjustment for demographic factors, proxy measures for TBI severity, and acute-care length-of-stay. C-statistic and predictiveness curves validated a five-variable model. Higher levels of independence in bladder management (adjusted odds ratio [OR], 0.88; 95% CI 0.83, 0.93), bed-chair-wheelchair transfers (OR, 0.81 [95% CI, 0.83-0.93]), and comprehension (OR, 0.78 [95% CI, 0.68, 0.89]) at admission were associated with lower risks of institutionalization on discharge. For every 10-year increment in age was associated with a 1.38 times higher risk for institutionalization (95% CI, 1.29, 1.48) and living alone was associated with a 2.34 times higher risk (95% CI, 1.86, 2.94). The c-statistic was 0.780. We conclude that this simple model can predict risk of institutionalization after inpatient rehabilitation for patients with TBI.

Original languageEnglish (US)
Pages (from-to)280-286
Number of pages7
JournalJournal of Neurotrauma
Volume32
Issue number4
DOIs
StatePublished - Feb 15 2015
Externally publishedYes

Fingerprint

Institutionalization
Inpatients
Rehabilitation
Wheelchairs
Odds Ratio
Urinary Bladder
Wounds and Injuries
Proxy
Locomotion
Traumatic Brain Injury
Population
Length of Stay
Logistic Models
Demography
Databases

Keywords

  • adult brain injury
  • outcome measures
  • predictive modeling
  • rehabilitation
  • traumatic brain injury

ASJC Scopus subject areas

  • Clinical Neurology

Cite this

Eum, R. S., Seel, R. T., Goldstein, R., Brown, A. W., Watanabe, T. K., Zasler, N. D., ... Glenn, M. B. (2015). Predicting institutionalization after traumatic brain injury inpatient rehabilitation. Journal of Neurotrauma, 32(4), 280-286. https://doi.org/10.1089/neu.2014.3351

Predicting institutionalization after traumatic brain injury inpatient rehabilitation. / Eum, Regina S.; Seel, Ronald T.; Goldstein, Richard; Brown, Allen W; Watanabe, Thomas K.; Zasler, Nathan D.; Roth, Elliot J.; Zafonte, Ross D.; Glenn, Mel B.

In: Journal of Neurotrauma, Vol. 32, No. 4, 15.02.2015, p. 280-286.

Research output: Contribution to journalArticle

Eum, RS, Seel, RT, Goldstein, R, Brown, AW, Watanabe, TK, Zasler, ND, Roth, EJ, Zafonte, RD & Glenn, MB 2015, 'Predicting institutionalization after traumatic brain injury inpatient rehabilitation', Journal of Neurotrauma, vol. 32, no. 4, pp. 280-286. https://doi.org/10.1089/neu.2014.3351
Eum, Regina S. ; Seel, Ronald T. ; Goldstein, Richard ; Brown, Allen W ; Watanabe, Thomas K. ; Zasler, Nathan D. ; Roth, Elliot J. ; Zafonte, Ross D. ; Glenn, Mel B. / Predicting institutionalization after traumatic brain injury inpatient rehabilitation. In: Journal of Neurotrauma. 2015 ; Vol. 32, No. 4. pp. 280-286.
@article{edb62e5bf8bf4279aae463da4277497a,
title = "Predicting institutionalization after traumatic brain injury inpatient rehabilitation",
abstract = "Risk factors contributing to institutionalization after inpatient rehabilitation for people with traumatic brain injury (TBI) have not been well studied and need to be better understood to guide clinicians during rehabilitation. We aimed to develop a prognostic model that could be used at admission to inpatient rehabilitation facilities to predict discharge disposition. The model could be used to provide the interdisciplinary team with information regarding aspects of patients' functioning and/or their living situation that need particular attention during inpatient rehabilitation if institutionalization is to be avoided. The study population included 7219 patients with moderate-severe TBI in the Traumatic Brain Injury Model Systems (TBIMS) National Database enrolled from 2002-2012 who had not been institutionalized prior to injury. Based on institutionalization predictors in other populations, we hypothesized that among people who had lived at a private residence prior to injury, greater dependence in locomotion, bed-chair-wheelchair transfers, bladder and bowel continence, feeding, and comprehension at admission to inpatient rehabilitation programs would predict institutionalization at discharge. Logistic regression was used, with adjustment for demographic factors, proxy measures for TBI severity, and acute-care length-of-stay. C-statistic and predictiveness curves validated a five-variable model. Higher levels of independence in bladder management (adjusted odds ratio [OR], 0.88; 95{\%} CI 0.83, 0.93), bed-chair-wheelchair transfers (OR, 0.81 [95{\%} CI, 0.83-0.93]), and comprehension (OR, 0.78 [95{\%} CI, 0.68, 0.89]) at admission were associated with lower risks of institutionalization on discharge. For every 10-year increment in age was associated with a 1.38 times higher risk for institutionalization (95{\%} CI, 1.29, 1.48) and living alone was associated with a 2.34 times higher risk (95{\%} CI, 1.86, 2.94). The c-statistic was 0.780. We conclude that this simple model can predict risk of institutionalization after inpatient rehabilitation for patients with TBI.",
keywords = "adult brain injury, outcome measures, predictive modeling, rehabilitation, traumatic brain injury",
author = "Eum, {Regina S.} and Seel, {Ronald T.} and Richard Goldstein and Brown, {Allen W} and Watanabe, {Thomas K.} and Zasler, {Nathan D.} and Roth, {Elliot J.} and Zafonte, {Ross D.} and Glenn, {Mel B.}",
year = "2015",
month = "2",
day = "15",
doi = "10.1089/neu.2014.3351",
language = "English (US)",
volume = "32",
pages = "280--286",
journal = "Journal of Neurotrauma",
issn = "0897-7151",
publisher = "Mary Ann Liebert Inc.",
number = "4",

}

TY - JOUR

T1 - Predicting institutionalization after traumatic brain injury inpatient rehabilitation

AU - Eum, Regina S.

AU - Seel, Ronald T.

AU - Goldstein, Richard

AU - Brown, Allen W

AU - Watanabe, Thomas K.

AU - Zasler, Nathan D.

AU - Roth, Elliot J.

AU - Zafonte, Ross D.

AU - Glenn, Mel B.

PY - 2015/2/15

Y1 - 2015/2/15

N2 - Risk factors contributing to institutionalization after inpatient rehabilitation for people with traumatic brain injury (TBI) have not been well studied and need to be better understood to guide clinicians during rehabilitation. We aimed to develop a prognostic model that could be used at admission to inpatient rehabilitation facilities to predict discharge disposition. The model could be used to provide the interdisciplinary team with information regarding aspects of patients' functioning and/or their living situation that need particular attention during inpatient rehabilitation if institutionalization is to be avoided. The study population included 7219 patients with moderate-severe TBI in the Traumatic Brain Injury Model Systems (TBIMS) National Database enrolled from 2002-2012 who had not been institutionalized prior to injury. Based on institutionalization predictors in other populations, we hypothesized that among people who had lived at a private residence prior to injury, greater dependence in locomotion, bed-chair-wheelchair transfers, bladder and bowel continence, feeding, and comprehension at admission to inpatient rehabilitation programs would predict institutionalization at discharge. Logistic regression was used, with adjustment for demographic factors, proxy measures for TBI severity, and acute-care length-of-stay. C-statistic and predictiveness curves validated a five-variable model. Higher levels of independence in bladder management (adjusted odds ratio [OR], 0.88; 95% CI 0.83, 0.93), bed-chair-wheelchair transfers (OR, 0.81 [95% CI, 0.83-0.93]), and comprehension (OR, 0.78 [95% CI, 0.68, 0.89]) at admission were associated with lower risks of institutionalization on discharge. For every 10-year increment in age was associated with a 1.38 times higher risk for institutionalization (95% CI, 1.29, 1.48) and living alone was associated with a 2.34 times higher risk (95% CI, 1.86, 2.94). The c-statistic was 0.780. We conclude that this simple model can predict risk of institutionalization after inpatient rehabilitation for patients with TBI.

AB - Risk factors contributing to institutionalization after inpatient rehabilitation for people with traumatic brain injury (TBI) have not been well studied and need to be better understood to guide clinicians during rehabilitation. We aimed to develop a prognostic model that could be used at admission to inpatient rehabilitation facilities to predict discharge disposition. The model could be used to provide the interdisciplinary team with information regarding aspects of patients' functioning and/or their living situation that need particular attention during inpatient rehabilitation if institutionalization is to be avoided. The study population included 7219 patients with moderate-severe TBI in the Traumatic Brain Injury Model Systems (TBIMS) National Database enrolled from 2002-2012 who had not been institutionalized prior to injury. Based on institutionalization predictors in other populations, we hypothesized that among people who had lived at a private residence prior to injury, greater dependence in locomotion, bed-chair-wheelchair transfers, bladder and bowel continence, feeding, and comprehension at admission to inpatient rehabilitation programs would predict institutionalization at discharge. Logistic regression was used, with adjustment for demographic factors, proxy measures for TBI severity, and acute-care length-of-stay. C-statistic and predictiveness curves validated a five-variable model. Higher levels of independence in bladder management (adjusted odds ratio [OR], 0.88; 95% CI 0.83, 0.93), bed-chair-wheelchair transfers (OR, 0.81 [95% CI, 0.83-0.93]), and comprehension (OR, 0.78 [95% CI, 0.68, 0.89]) at admission were associated with lower risks of institutionalization on discharge. For every 10-year increment in age was associated with a 1.38 times higher risk for institutionalization (95% CI, 1.29, 1.48) and living alone was associated with a 2.34 times higher risk (95% CI, 1.86, 2.94). The c-statistic was 0.780. We conclude that this simple model can predict risk of institutionalization after inpatient rehabilitation for patients with TBI.

KW - adult brain injury

KW - outcome measures

KW - predictive modeling

KW - rehabilitation

KW - traumatic brain injury

UR - http://www.scopus.com/inward/record.url?scp=84923284795&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84923284795&partnerID=8YFLogxK

U2 - 10.1089/neu.2014.3351

DO - 10.1089/neu.2014.3351

M3 - Article

VL - 32

SP - 280

EP - 286

JO - Journal of Neurotrauma

JF - Journal of Neurotrauma

SN - 0897-7151

IS - 4

ER -