Who will have Sustainable Employment After a Back Injury? The Development of a Clinical Prediction Model in a Cohort of Injured Workers

Heather M. Shearer, Pierre Côté, Eleanor Boyle, Jill A. Hayden, John Frank, William Johnson

Research output: Contribution to journalArticle

Abstract

Purpose Our objective was to develop a clinical prediction model to identify workers with sustainable employment following an episode of work-related low back pain (LBP). Methods We used data from a cohort study of injured workers with incident LBP claims in the USA to predict employment patterns 1 and 6 months following a workers’ compensation claim. We developed three sequential models to determine the contribution of three domains of variables: (1) basic demographic/clinical variables; (2) health-related variables; and (3) work-related factors. Multivariable logistic regression was used to develop the predictive models. We constructed receiver operator curves and used the c-index to measure predictive accuracy. Results Seventy-nine percent and 77 % of workers had sustainable employment at 1 and 6 months, respectively. Sustainable employment at 1 month was predicted by initial back pain intensity, mental health-related quality of life, claim litigation and employer type (c-index = 0.77). At 6 months, sustainable employment was predicted by physical and mental health-related quality of life, claim litigation and employer type (c-index = 0.77). Adding health-related and work-related variables to models improved predictive accuracy by 8.5 and 10 % at 1 and 6 months respectively. Conclusion We developed clinically-relevant models to predict sustainable employment in injured workers who made a workers’ compensation claim for LBP. Inquiring about back pain intensity, physical and mental health-related quality of life, claim litigation and employer type may be beneficial in developing programs of care. Our models need to be validated in other populations.

Original languageEnglish (US)
Pages (from-to)1-11
Number of pages11
JournalJournal of Occupational Rehabilitation
DOIs
StateAccepted/In press - Nov 2 2016

Fingerprint

Back Injuries
Jurisprudence
Low Back Pain
Workers' Compensation
Mental Health
Quality of Life
Back Pain
Health
Cohort Studies
Logistic Models
Demography
Population

Keywords

  • Back injuries
  • Employment
  • Return to work

ASJC Scopus subject areas

  • Rehabilitation
  • Occupational Therapy

Cite this

Who will have Sustainable Employment After a Back Injury? The Development of a Clinical Prediction Model in a Cohort of Injured Workers. / Shearer, Heather M.; Côté, Pierre; Boyle, Eleanor; Hayden, Jill A.; Frank, John; Johnson, William.

In: Journal of Occupational Rehabilitation, 02.11.2016, p. 1-11.

Research output: Contribution to journalArticle

Shearer, Heather M. ; Côté, Pierre ; Boyle, Eleanor ; Hayden, Jill A. ; Frank, John ; Johnson, William. / Who will have Sustainable Employment After a Back Injury? The Development of a Clinical Prediction Model in a Cohort of Injured Workers. In: Journal of Occupational Rehabilitation. 2016 ; pp. 1-11.
@article{487a1ff5ff8e4dd28b243f56d08decc2,
title = "Who will have Sustainable Employment After a Back Injury? The Development of a Clinical Prediction Model in a Cohort of Injured Workers",
abstract = "Purpose Our objective was to develop a clinical prediction model to identify workers with sustainable employment following an episode of work-related low back pain (LBP). Methods We used data from a cohort study of injured workers with incident LBP claims in the USA to predict employment patterns 1 and 6 months following a workers’ compensation claim. We developed three sequential models to determine the contribution of three domains of variables: (1) basic demographic/clinical variables; (2) health-related variables; and (3) work-related factors. Multivariable logistic regression was used to develop the predictive models. We constructed receiver operator curves and used the c-index to measure predictive accuracy. Results Seventy-nine percent and 77 {\%} of workers had sustainable employment at 1 and 6 months, respectively. Sustainable employment at 1 month was predicted by initial back pain intensity, mental health-related quality of life, claim litigation and employer type (c-index = 0.77). At 6 months, sustainable employment was predicted by physical and mental health-related quality of life, claim litigation and employer type (c-index = 0.77). Adding health-related and work-related variables to models improved predictive accuracy by 8.5 and 10 {\%} at 1 and 6 months respectively. Conclusion We developed clinically-relevant models to predict sustainable employment in injured workers who made a workers’ compensation claim for LBP. Inquiring about back pain intensity, physical and mental health-related quality of life, claim litigation and employer type may be beneficial in developing programs of care. Our models need to be validated in other populations.",
keywords = "Back injuries, Employment, Return to work",
author = "Shearer, {Heather M.} and Pierre C{\^o}t{\'e} and Eleanor Boyle and Hayden, {Jill A.} and John Frank and William Johnson",
year = "2016",
month = "11",
day = "2",
doi = "10.1007/s10926-016-9678-6",
language = "English (US)",
pages = "1--11",
journal = "Journal of Occupational Rehabilitation",
issn = "1053-0487",
publisher = "Springer New York",

}

TY - JOUR

T1 - Who will have Sustainable Employment After a Back Injury? The Development of a Clinical Prediction Model in a Cohort of Injured Workers

AU - Shearer, Heather M.

AU - Côté, Pierre

AU - Boyle, Eleanor

AU - Hayden, Jill A.

AU - Frank, John

AU - Johnson, William

PY - 2016/11/2

Y1 - 2016/11/2

N2 - Purpose Our objective was to develop a clinical prediction model to identify workers with sustainable employment following an episode of work-related low back pain (LBP). Methods We used data from a cohort study of injured workers with incident LBP claims in the USA to predict employment patterns 1 and 6 months following a workers’ compensation claim. We developed three sequential models to determine the contribution of three domains of variables: (1) basic demographic/clinical variables; (2) health-related variables; and (3) work-related factors. Multivariable logistic regression was used to develop the predictive models. We constructed receiver operator curves and used the c-index to measure predictive accuracy. Results Seventy-nine percent and 77 % of workers had sustainable employment at 1 and 6 months, respectively. Sustainable employment at 1 month was predicted by initial back pain intensity, mental health-related quality of life, claim litigation and employer type (c-index = 0.77). At 6 months, sustainable employment was predicted by physical and mental health-related quality of life, claim litigation and employer type (c-index = 0.77). Adding health-related and work-related variables to models improved predictive accuracy by 8.5 and 10 % at 1 and 6 months respectively. Conclusion We developed clinically-relevant models to predict sustainable employment in injured workers who made a workers’ compensation claim for LBP. Inquiring about back pain intensity, physical and mental health-related quality of life, claim litigation and employer type may be beneficial in developing programs of care. Our models need to be validated in other populations.

AB - Purpose Our objective was to develop a clinical prediction model to identify workers with sustainable employment following an episode of work-related low back pain (LBP). Methods We used data from a cohort study of injured workers with incident LBP claims in the USA to predict employment patterns 1 and 6 months following a workers’ compensation claim. We developed three sequential models to determine the contribution of three domains of variables: (1) basic demographic/clinical variables; (2) health-related variables; and (3) work-related factors. Multivariable logistic regression was used to develop the predictive models. We constructed receiver operator curves and used the c-index to measure predictive accuracy. Results Seventy-nine percent and 77 % of workers had sustainable employment at 1 and 6 months, respectively. Sustainable employment at 1 month was predicted by initial back pain intensity, mental health-related quality of life, claim litigation and employer type (c-index = 0.77). At 6 months, sustainable employment was predicted by physical and mental health-related quality of life, claim litigation and employer type (c-index = 0.77). Adding health-related and work-related variables to models improved predictive accuracy by 8.5 and 10 % at 1 and 6 months respectively. Conclusion We developed clinically-relevant models to predict sustainable employment in injured workers who made a workers’ compensation claim for LBP. Inquiring about back pain intensity, physical and mental health-related quality of life, claim litigation and employer type may be beneficial in developing programs of care. Our models need to be validated in other populations.

KW - Back injuries

KW - Employment

KW - Return to work

UR - http://www.scopus.com/inward/record.url?scp=84994357451&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84994357451&partnerID=8YFLogxK

U2 - 10.1007/s10926-016-9678-6

DO - 10.1007/s10926-016-9678-6

M3 - Article

C2 - 27807731

AN - SCOPUS:84994357451

SP - 1

EP - 11

JO - Journal of Occupational Rehabilitation

JF - Journal of Occupational Rehabilitation

SN - 1053-0487

ER -