Occupational Categories and Return to Work After Traumatic Brain Injury: A Multicenter Study

Article Outline

• Abstract
• Methods
  • Participants
  • Measures
    • Category of productive activity
    • Census occupational category
    • Occupation group
    • FIM instrument
    • Duration of unconsciousness
  • Procedure
  • Data Analysis
• Results
• Discussion
  • Study Limitations
• Conclusions
• Acknowledgment
• References
• Copyright

Abstract 

Walker WC, Marwitz JH, Kreutzer JS, Hart T, Novack T. Occupational categories and return to work after traumatic brain injury: a multicenter study.

Objective

To further evaluate determinants of return to work (RTW) after traumatic brain injury (TBI), with focus on the relation between preinjury occupational category and RTW outcome.

Design

Prospective collaborative cohort study.

Setting

Seventeen National Institute on Disability and Rehabilitation Research–designated Traumatic Brain Injury Model Systems.

Participants

Consecutive sample of 1341 patients (age range, 18–62y) who were hospitalized with a TBI diagnosis, received both acute neurotrauma services and inpatient rehabilitation services, consented to participate, were employed before injury, and completed a 1-year follow-up assessment.

Intervention

An inpatient interdisciplinary brain injury rehabilitation program.

Main Outcome Measure

Competitive employment at 1 year postinjury.

Results

Participants were categorized into 1 of 3 groups depending on preinjury occupational title: professional/managerial (n=192), skilled (n=751), or manual labor (n=398). Chi-square analyses were computed to examine changes across occupation groups between preinjury occupation group and postinjury RTW. The rate of successful RTW was greatest for professional/managerial (56%), lower for skilled (40%), and lowest for manual labor (32%), yielding an odds ratio of 2.959 between the highest and lowest groups. Of those with successful RTW, most did so within the same occupational category grouping. A multiple logistic regression showed that preinjury occupation, education level, discharge FIM score, age, sex, marital status, and hospital length of stay each influenced RTW.

Conclusions

Prior research has shown that preinjury employment status (employed vs unemployed) greatly influences the odds of successful RTW after TBI. A related hypothesis, that occupational classification also influences RTW outcome, has been understudied and has yielded conflicting results. The current study shows convincingly that the type of occupation influences RTW outcome, with the best prospect for RTW among people with professional/managerial jobs. Occupational category should be examined in the future development of predictive models for RTW after TBI.

Key Words: Brain injuries, Employment, Occupations, Rehabilitation, Treatment outcome

RETURN TO WORK (RTW) is a significant outcome milestone for survivors of severe traumatic brain injury (TBI), who often face persisting cognitive and physical impairments. Beyond obvious financial significance, RTW has important psychosocial implications. Within our culture employment contributes to self-esteem and symbolizes full reintegration and membership in the community at large.¹ The ability to accurately predict RTW early after TBI would be valuable. Families typically desire early prognostic information about RTW and other long-term implications of TBI,², ³ and rehabilitation professionals use such prognostic information in program planning and goal setting. However, the present ability to predict RTW for a person after TBI is limited at best.

Prior research has identified multiple variables that correlate with RTW after TBI including demographic variables (age, education level, employment status),⁴, ⁵ injury severity markers,⁵, ⁶ cognitive measures,⁷ and functional measures.⁸, ⁹ However, the relative contribution and interdependency of these variables has only recently been explored in multivariate analyses.⁷, ¹⁰ Two recent reviews, one a meta-analysis¹¹ and the other a critical review,¹² point to the complexity of RTW predictors and the likelihood that multiple factors will need to be considered in any predictive model.

Premorbid occupational factors have emerged as a domain of demographic predictor variables that may warrant special attention. Occupational factors not only directly influence employment opportunities but may also serve as markers for “employable” characteristics. Education level and whether the person was employed or unemployed before the injury have already been shown to have a strong relation to RTW in the TBI population.⁵, ⁷, ¹³, ¹⁴ Less is known about a third occupational factor, preinjury job type. The multitude of possible occupations and the challenge of how to characterize and classify them confounds the study of this variable. Grouping of functionally similar jobs eases statistical analysis, but standards are lacking for classifying the various functional dimensions of each occupation. Thus, most health outcomes researchers have defined occupational groupings on the basis of available data rather than previously defined theoretic constructs, limiting interstudy comparisons.¹⁵

The existing literature has presented conflicting evidence on the possible relation between job categories and RTW after severe TBI. MacKenzie et al¹⁶ have reported that people with semiskilled or unskilled manual jobs have a lower probability of RTW. However, Fraser et al¹⁷ found that people in structural occupations (building and trade) were more likely to RTW. Vogenthaler et al¹⁸ found that those who had occupations with technical skills were more likely to RTW. Brooks et al¹⁹ reported a trend (lacking statistical significance) toward higher RTW rates in managerial-class jobs. Ponsford et al⁴ found no association between RTW and occupational category. Ip et al²⁰ found that skilled were more likely than unskilled workers to RTW. Fleming et al²¹ found that people from a prior occupation with a “higher status” were more likely to RTW. In total, the findings from these prior job category studies are difficult to reconcile considering their conflicting findings, differing methodology, and small sample sizes. The advantage of the current study over prior research is a much larger sample size, uniform follow-up protocol, and more clearly delineated operational definitions of job category groupings.

The hypothesis of the current study is that premorbid job category is a significant and unique predictor of RTW after TBI. A large prospective database within the National Institute on Disability and Rehabilitation Research (NIDRR) Traumatic Brain Injury Model Systems (TBIMS) afforded accrual of enough subjects to perform a robust multivariate analysis. Those unemployed at the time of injury were excluded, because preinjury employment category is not applicable to this subgroup. The study’s specific objectives were (1) to determine if employees in different job categories have different RTW success rates, (2) to analyze the relative contributions of occupational categories and other covariates in an RTW prediction model, and (3) to determine the nature of occupational category change in those with successful RTW.

Back to Article Outline

Methods 

Participants 

All subjects were participants in the NIDRR-funded TBIMS program, a collaborative multicenter project initiated in 1987.²², ²³ At the time of this study, there were 17 participating centers. Each center includes emergency medical services, intensive and acute medical care, inpatient rehabilitation, and a spectrum of community rehabilitation services. All patients were admitted to an acute care hospital within 24 hours of injury. People with a history of prior brain injury, preexisting neurologic condition, or substance abuse are included in the TBIMS program. Informed consent was obtained from each patient or responsible family member, and the research was approved by the local institutional boards at each center.

At the time of this investigation, data from 3703 people with TBI were available in the TBIMS national database. Of this number, 1926 met the additional qualifications of working age (between 18 and 62 years) and employment at the time of the injury. One-year follow-up data were available for 1341 (70%) of the 1926 potential participants. The comparability of the groups with and without (30% [N=585]) 1-year follow-up data was statistically examined, and meaningful differences were not identified. Analysis of variance (ANOVA) showed no between-group differences (P>.05) with regard to age, admission Glasgow Coma Scale (GCS) score, length of unconsciousness, admission FIM score, and acute and rehabilitation lengths of stay (LOSs). Chi-square analyses showed no group differences (P>.05) with regard to sex, race, marital status, education, or preinjury employment category.

The sample was predominantly male (77%) and predominantly white (69%), with 22% African-American, 6% Hispanic, and 3% other. By virtue of their enrollment into the TBIMS project during inpatient rehabilitation for TBI, most participants had sustained brain injuries that would be considered moderate to severe. The median admission GCS score was 8 (range, 3–15), and the mean duration of unconsciousness ± standard deviation (SD) was 9.2±13.7 days (range, 0–98d).

Measures 

Through interviews with patients and caregivers and through records review, each patient’s primary area of activity was categorized for preinjury status and at 1 year follow-up as follows: (1) competitively employed; (2) specially employed (eg, sheltered workshop, supported employment); (3) unemployed; (4) student; (5) retired; (6) homemaker; or (7) volunteer.

The major census occupational category that each participant held in the month before injury and at follow-up (if employed at 1y postinjury) was collected. The categories are defined by the 1990 Occupational Classification System of Industries and Occupations.²⁴

To best test our hypothesis and allow for comparisons with past studies in TBI populations, the census occupational categories were clustered into larger groups using the criterion of skill level (training and/or experience required to master the job). As rationale for this grouping criterion, skill level is used by The International Standard Classification of Occupations²⁵ and is increasingly used in occupation-related health research.²⁶ The current study’s authors judged the best skill level fit for each category and clustered participants into the following 3 groups: professional/managerial (highly skilled), skilled (skilled, semi-skilled), or manual labor (unskilled) (table 1).

Table 1. Preinjury Census Occupational Categories and Occupation Groups

The FIM instrument is an 18-item, 7-point scale on which higher values indicate greater levels of independence. The 18 items describe levels of self-care, continence, mobility, communication, and cognition.²⁷, ²⁸, ²⁹, ³⁰

Calculated as the number of days between the onset of injury and the time a participant was able to consistently follow 1-step commands (GCS motor score, 6).

Procedure 

Data were collected at 17 TBIMS rehabilitation centers. An individualized, comprehensive program of inpatient rehabilitation was provided to each patient. The following services were provided at each center: nursing, occupational therapy, physiatry and related medical services, physical therapy, psychology and neuropsychology, therapeutic recreation, social services, and speech and language therapy. Information about medical aspects of injury was obtained from hospital records. The admission FIM was scored within 72 hours of admission, and the discharge FIM was scored within 72 hours of discharge by the interdisciplinary team members using standard protocols.²²

An annual follow-up interview between 10 and 14 months postinjury is attempted with every person entered in the TBIMS database. An in-person interview with each patient is the first choice of follow-up method. If this is not possible, a telephone interview is attempted, and if this is unsuccessful, data are collected using a mail questionnaire and/or interview with a significant other or family member.

Data Analysis 

Descriptive statistics including means, SDs, and percentages were computed for all relevant variables. Chi-square and ANOVA testing were used to determine differences among the preinjury occupational groups. Simple and multiple logistic regression analyses were performed to assess predictor variables’ relations to RTW at the 1-year follow-up, using a full model. For the regression models, employment outcome categories were grouped dichotomously into RTW (competitively employed) versus no RTW (unemployed, homemaker, volunteer), with exclusion of subjects categorized as student, retired, or specially employed. Predictor variables included preinjury occupation group (professional/managerial, skilled, manual labor), educational level (did, did not complete high school), marital status (married, not married), sex, age, duration of unconsciousness, inpatient LOS (including both acute care and rehabilitation), and discharge FIM score.

Back to Article Outline

Results 

As noted in table 1, most participants held skilled positions preinjury (55%). The least, 15%, were in professional/managerial jobs, and 30% were manual laborers. Demographic and injury severity information for the entire sample and by occupation group is given in Table 2, Table 3. There were differences among the preinjury occupation groups with regard to age, sex, minority status, marital status, and education level. Professional/managerial participants were significantly older than participants in skilled professions and manual laborers (F_2,1338=32.74, P<.001). Members of racial/ethnic minorities were less likely to hold professional/managerial or skilled positions compared with whites (χ22 test=15.9, P<.001), and men were more likely to hold manual labor jobs than women (χ22 test=49.1, P<.001). People in skilled positions and manual labor positions were more likely to be unmarried than married (χ22 test=37.5, P<.001). As expected by the chosen skill level criterion, education and occupation group were related. People with higher levels of education were more likely to hold professional/managerial positions compared with those with less education (χ42 test=279.0, P<.001).

Table 2. Study Sample Characteristics for Preinjury Occupation Groups

NOTE. Values are mean ± SD.

Table 3. Study Sample Characteristics for Preinjury Occupation Groups

Characteristics	Percentages
	Professional/Managerial (n=192)	Skilled (n=760)	Manual Labor (n=389)
Sex⁎
Male	67	73	88
Female	33	27	12
Race⁎
Nonminority	79	70	63
Minority	21	30	37
Marital status⁎
Unmarried	45	67	68
Married	55	33	32
Education⁎
Less than high school	3	24	40
High school graduate/some college	41	65	55
College graduate/graduate degree	56	11	5
Admission GCS score
Mild	32	27	23
Moderate	20	15	20
Severe	48	58	57

NOTE. Values are percentages.

For the entire sample, the overall RTW rate was 39%, defined by competitive employment in any occupation, fulltime or parttime, at 1 year postinjury. Approximately half were unemployed, and a relatively small number (9%) classified themselves as students, retired, homemakers, or volunteers. Regarding occupational groups, people with professional/managerial jobs preinjury showed the highest proportion of competitive employment at follow-up (56%). In comparison, RTW rates were lower for people previously employed in skilled (38%) and manual labor (31%) positions.

Postinjury occupational information for the 483 people employed at follow-up is shown in table 4. Similar to preinjury (see table 1), most jobs were in skilled positions (56%). However, postinjury there was a noticeable change toward proportionally more professional/managerial positions and fewer manual labor positions. Preinjury, twice as many workers were in manual labor (30%) versus professional/managerial (15%); the ratio dropped postinjury, with 24% in manual labor and 20% in professional/managerial positions. Table 4 also shows subgroup changes in the occupational categories within the larger occupational groups. For example, the aforementioned drop in the proportion of the sample working as manual laborers occurs most with machine operators, assemblers, and inspectors (9% preinjury, 5% postinjury); to a lesser degree in transportation and material moving (5% preinjury, 3% postinjury), and not with handlers, equipment cleaners, helpers, or laborers (15% preinjury, 15% postinjury).

Table 4. Postinjury Census Occupational Categories and Occupation Group for People Employed

Cross-tabulation was used to further examine the stability of occupational group status from preinjury to postinjury (table 5). Chi-square analyses were computed to compare distributions across groups, and as expected significant differences were noted (χ62 test=347.7, P<.001). Of those people who successfully returned to work at 1 year postinjury, most in each preinjury occupational group did so in that same group. Relative to the other 2 preinjury occupational groups, manual laborers were most likely to have shifted into a different occupational group: 39% returned to work in a different group. Comparatively, the shift rate to a different group postinjury was 30% for professional/managerial workers and 29% for skilled workers.

Table 5. Preinjury Versus Postinjury Occupation Groupings

NOTE. Values are percentages.

Logistic regression analyses were performed to determine the extent to which selected variables predicted employment status at 1 year postinjury. Both simple and multiple logistic regression analyses were performed to determine the predictors’ influence with and without adjustment for the effects of all other predictors. In the multiple logistic regression analysis, preinjury occupation, education level (higher), discharge FIM score (higher), sex (female), marital status (married), and hospital LOS (shorter) made significant, unique contributions to predicting RTW. Length of unconsciousness, however, did not significantly contribute to this prediction model. When variables were analyzed separately using simple logistic regression, all variables were statistically significant except age, marital status, and sex. In both models, the odds ratios (ORs) provide a measure of the magnitude of each variable’s predictive power (table 6). People who completed high school were 2.34 times more likely to RTW than those without a high school diploma. Those in preinjury professional or managerial positions were 2.96 times more likely to RTW by 1 year postinjury than those in manual labor positions. Those scoring at the 75% level on the total discharge FIM were 3.33 times more likely to RTW than those at the 25% level, whereas those with longer hospital LOSs were 3.02 times less likely (0.331 times more likely) to RTW.

Table 6. Logistic Regression Predicting RTW

NOTE: ORs for occupational categories in this table are both referenced to manual labor. ORs for ordinal variables in this table (age, loss of consciousness, LOS, FIM) are all referenced 75% versus 25%.

Abbreviation: SE, standard error.

We performed secondary multivariate analysis using cross-tabulation to further assess the interaction between occupation type and education level in predicting RTW. Occupation group had a consistent effect on RTW regardless of education level (table 7). For each education level, the RTW rate was highest for the managerial category, intermediate for the skilled category, and lowest for the manual labor category. The employment ratio values indicate a similar proportional influence of occupation type on RTW across each education level.

Table 7. Comparison of Education Level and Preinjury Occupation Classification With Postinjury Employment

Education Level and Preinjury Occupation	Employed (%)	Unemployed (%)	Ratio of Employed to Unemployed
Less than high school
Professional/managerial	40	60	0.67
Skilled	29	71	0.40
Manual labor	24	76	0.32
High school graduate/some college
Professional/managerial	51	49	1.04
Skilled	47	53	0.89
Manual labor	34	66	0.52
College graduate/graduate degree
Professional/managerial	63	37	1.70
Skilled	52	48	1.08
Manual labor	47	53	0.89

Back to Article Outline

Discussion 

A number of researchers have contended that postinjury RTW rates are a function of preinjury job classifications. In particular, Fleming,²¹ Vogenthaler,¹⁸ and Mackenzie¹⁶ and colleagues suggested that people in higher-paying or more technically demanding jobs preinjury were more likely to return to work postinjury. With a substantially larger sample and participants from multiple research sites, the present findings provide support for the findings of earlier researchers. People who were previously employed in professional and managerial positions had the greatest likelihood of working at a 1-year follow-up. Less than one third of people working previously in manual labor positions were working at follow-up, with a slightly better RTW rate for people previously working in skilled positions. Further, multivariate regression analysis showed independent predictive power of job categories in relation to other RTW predictor variables including age, duration of unconsciousness, combined hospital LOS, functional status at rehabilitation admission (FIM score), marital status, sex, and education level. Surprisingly, injury severity as measured by duration of unconsciousness did not achieve predictive significance in the multivariate model.

The ORs showed that several variables within the multivariate model were robust RTW predictors including (1) preinjury job category, (2) education level, (3) hospital LOS, and (4) discharge FIM total. The unadjusted ratios examine 1 variable at a time and offer better guidance for clinical prognostication because multivariate computations are rarely practical. In our sample, professionals were 3 times more likely and skilled workers were 1.5 times more likely to RTW than manual laborers. Similarly, high school graduates were 2.3 times more likely to RTW than nongraduates. Those with higher discharge FIM total scores were 3.33 times more likely to RTW, and those with longer LOSs were 3.02 times less likely (0.331 times more likely) to RTW. In contrast, the unadjusted ORs for age, sex, and marital status ranged from 0.855 to 1.261. Such ratios lack utility for individual clinical prognostication. For example, age is unlikely to sway vocational rehabilitation and disability planning, given that a younger person is only 1.170 times more likely to RTW than an older person. Thus, both occupational domain variables (job category, education level) and 2 of the injury severity variables (LOS, FIM score) proved to be unique and clinically salient predictors of RTW in our model.

Given the importance that both job category and education level seem to hold as RTW predictors in our primary analysis, we undertook a secondary analysis to better assess their interaction. Although our multiple logistic regression analysis had shown independent predictive power for each, a large degree of covariance was still possible, such that the amount of unique contribution from the other might not be clinically meaningful. The secondary cross-tabulation clearly showed that job category adds meaningfully to an RTW prediction beyond using education level alone. Across each education level, managers and professionals were approximately twice as likely to RTW as manual laborers, whereas skilled workers had an intermediate RTW likelihood.

Our finding of better RTW outcome for people seeking to return to more cognitively demanding jobs needs reconciliation with the well-documented pattern of predominantly cognitive rather than physical long-term impairments after TBI.³¹, ³², ³³, ³⁴, ³⁵, ³⁶, ³⁷, ³⁸, ³⁹, ⁴⁰ Possibly, physical impairments after TBI are underrecognized because of their subtle presentation. There is increasing evidence that persisting impairments in balance and fine motor skills are frequent and that standard neurologic examination may be insensitive to these problems.⁴¹, ⁴², ⁴³ Higher-level mobility deficits would most hinder those attempting to return to physically demanding occupations. Among possible extrinsic factors, employers may be more accommodating to their professional versus manual labor employees with TBI because of the perceived value of skill sets and/or greater levels of employer commitment. In addition, stakeholders may perceive greater injury risks (eg, heights, heavy machinery) inherent in manual labor jobs. Different financial incentives for RTW may also play a role, with greater incentive for higher-wage earners, and people from higher-echelon jobs may have more employment options with less need for retraining. Finally, job categories may be a marker for premorbid personality or cognitive traits more conducive to successful vocational outcome after TBI (eg, achievement level, mental flexibility, intelligence), in keeping with Satz’s theory⁴⁴ that larger cognitive reserves are protective in TBI.

Of those in our sample who did RTW, the vast majority worked in a postinjury job category that was similar to their preinjury category (see table 5). This is not surprising, because it represents the path of least resistance. The alternative, a shift to a different occupational grouping, is likely to entail barriers such as demands on new learning, the anxiety that accompanies a change in major life roles, and vocational rehabilitation needs that may be difficult to meet. The rate of shift varied by preinjury job grouping. Of the injured manual laborers who were working again at 1 year, 39% had shifted to a job in a different grouping (all to the skilled sector) for which they may have lacked familiarity, prior training, or education. By comparison, the shift rates for the other 2 preinjury occupational groupings were 29% for skilled workers and 30% for professional workers. In addition to being less likely than others to RTW, manual laborers were more likely to RTW in a different job classification. As a result, these people may have more need for vocational rehabilitation services after brain injury.

Study Limitations 

The overall predictive accuracy of our multivariate RTW model was 72%. Correct categorization of outcome in nearly three quarters of this sample indicates the important influence that the chosen variables had on RTW after TBI for people who were working at the time of injury. Conversely, our inability to correctly categorize over a quarter of the sample indicates the need for further research to refine present variables and/or explore unmeasured determinants of RTW. A specific limitation of the current study is that measures of cognitive impairment were not included in the model. Prior research in TBI has shown that cognitive measures derived from neuropsychologic test scores are associated with RTW.⁶, ⁷, ⁹, ¹⁶, ²⁰, ⁴⁵, ⁴⁶, ⁴⁷ Sherer et al⁷ recently evaluated cognitive testing as a predictor of RTW in a multivariate study. Notably, the predictive power of cognitive variables was substantially less than education level (OR, 1.6 vs 3).

Another potential limitation of the current study was the use of RTW at 1 year as the primary employment outcome measure. A dichotomous RTW variable has the advantages of simplicity, ease of standardization, and track record of use in TBI outcomes research. However, employment after TBI is not completely stable over time, as recently shown by Kreutzer et al.⁴⁸ Better measures of long-term vocational success after TBI may exist such as employment ratio or a similar quantitative measure.⁴⁹ Qualitative or subjective indicators, such as job satisfaction and response to vocational rehabilitation efforts, will also be important to consider in future research.

Subject selection bias may limit the degree to which this study’s findings can be generalized. The NIDDR TBIMS cohort is geographically and socially diverse by design but includes only patients admitted to inpatient rehabilitation who consent to participate.²², ²³ This study excluded those employed subjects lacking 1-year follow-up data, creating an additional source of possible selection bias. Moderating this bias is our finding of no differences in select characteristics between the excluded subjects and the final sample.

Like past studies, the grouping of job categories was constrained by our available data source. Specifically, our data were collected under census occupational categories that sometimes contained a wide diversity of job titles (and skill levels). Therefore, it is possible that not all participants were placed in the most appropriate hierarchic skill-level job grouping.

Back to Article Outline

Conclusions 

The NIDRR TBIMS study confirmed the influence that preinjury occupational factors have on RTW after TBI. Among patients employed at the time of injury, those in manual labor jobs had the lowest rate of RTW, implying that this group may have a greater need for early vocational rehabilitation services. Patients in professional/managerial jobs had the highest rate of RTW at 1 year. Age, sex, marital status, hospital LOS, functional status at rehabilitation discharge, preinjury job category, and education level all showed a statistically significant relation to RTW in the multivariate model, with the latter 4 variables also showing clinically salient (>2 or < 0.5) unadjusted ORs. Premorbid occupation category should be considered in clinical prognostication for RTW and vocational rehabilitation planning and controlled for in clinical trials that use RTW as an outcome measure. Further research is needed to better understand the reasons for differing RTW rates among job categories and to evaluate targeted remediation strategies.

Back to Article Outline

Acknowledgment 

The contributions of the NIDRR-funded TBIMS centers are gratefully acknowledged.

Back to Article Outline

References 

1. Prigatano GP, Fordyce DJ, Zeiner HK, Roueche JR, Pepping M, Wood BC. Neuropsychological rehabilitation after closed head injury in young adults. J Neurol Neurosurg Psychiatry. 1984;47:505–513
   • View In Article
   • MEDLINE
   • CrossRef
2. Evans RL, Bishop DS, Ousley RT. Providing care to persons with physical disability: effect on family caregivers. Am J Phys Med Rehabil. 1992;71:140–144
   • View In Article
   • MEDLINE
   • CrossRef
3. Condeluci A. Brain injury rehabilitation: the need to bridge paradigms. Brain Inj. 1992;6:543–551
   • View In Article
   • MEDLINE
   • CrossRef
4. Ponsford JL, Olver JH, Curran C, Ng K. Prediction of employment status 2 years after traumatic brain injury. Brain Inj. 1995;9:11–20
   • View In Article
   • MEDLINE
   • CrossRef
5. Gollaher K, High WM, Sherer M, et al. Prediction of employment outcome one to three years following traumatic brain injury. Brain Inj. 1998;12:255–263
   • View In Article
   • MEDLINE
   • CrossRef
6. Wagner AK, Hammond FM, Sasser HC, Wiercisiewski D. Return to productive activity after traumatic brain injury: relationship with measures of disability, handicap, and community integration. Arch Phys Med Rehabil. 2002;83:107–114
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (600 KB)
   • CrossRef
7. Sherer M, Sander AM, Nick TG, High WM, Malec JF, Rosenthal M. Early cognitive status and productivity outcome after traumatic brain injury: findings from the TBI Model Systems. Arch Phys Med Rehabil. 2002;83:183–192
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (85 KB)
   • CrossRef
8. Greenspan AI, Wrigley JM, Kresnow M, Branche-Dorsey CM, Fine PR. Factors influencing failure to return to work due to traumatic brain injury. Brain Inj. 1996;10:207–218
   • View In Article
   • MEDLINE
   • CrossRef
9. Cifu DX, Keyser-Marcus L, Lopez E, et al. Acute predictors of successful return to work one year after traumatic brain injury: a multicenter analysis. Arch Phys Med Rehabil. 1997;78:125–131
   • View In Article
   • Abstract
   • Full-Text PDF (993 KB)
   • CrossRef
10. Keyser-Marcus L, Bricout J, Wehman P, et al. Acute predictors of return to employment following traumatic brain injury: a longitudinal follow-up. Arch Phys Med Rehabil. 2002;83:635–641
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (62 KB)
    • CrossRef
11. Crepeau F, Scherzer P. Predictors and indicators of work status after traumatic brain injury: a meta-analysis. Neuropsychol Rehabil. 1993;3:5–35
    • View In Article
    • CrossRef
12. Ownsworth T, McKenna K. Investigation of factors related to employment outcome following traumatic brain injury: a critical review and conceptual model. Disabil Rehabil. 2004;26:765–784
    • View In Article
    • MEDLINE
    • CrossRef
13. Bell KR, Sandel ME. Brain injury rehabilitation (4. Postacute rehabilitation and community integration). Arch Phys Med Rehabil. 1998;79(Suppl 1):S21–S25
    • View In Article
    • Abstract
    • Full-Text PDF (683 KB)
    • CrossRef
14. Crisp R. Return to work after traumatic brain injury. J Rehabil. 1992;58(4):27–33
    • View In Article
15. Berkman LF, Macintyre S. The measurement of social class in health studies: old measures and new formulations. IARC Sci Publ. 1997;138:51–64
    • View In Article
16. MacKenzie EJ, Shapiro S, Smith RT, Siegel JH, Moody M, Pitt A. Factors influencing return to work following hospitalization for traumatic injury. Am J Public Health. 1987;77:329–334
    • View In Article
    • MEDLINE
    • CrossRef
17. Fraser R, Dikmen S, Mclean A, Miller B. Employability of head injury survivors: first year post-injury. Rehabil Couns Bull. 1988;31:276–288
    • View In Article
18. Vogenthaler DR, Smith KR, Goldfader P. Head injury, a multivariate study: predicting long-term productivity and independent living outcome. Brain Inj. 1989;3:369–385
    • View In Article
    • MEDLINE
    • CrossRef
19. Brooks N, McKinlay W, Symington C, Beattie A, Campsie L. Return to work within the first seven years of severe head injury. Brain Inj. 1987;1:5–19
    • View In Article
    • MEDLINE
    • CrossRef
20. Ip RY, Dornan J, Schentag C. Traumatic brain injury: factors predicting return to work or school. Brain Inj. 1995;9:517–532
    • View In Article
    • MEDLINE
    • CrossRef
21. Fleming J, Tooth L, Hassell M, Chan W. Prediction of community integration and vocational outcome 2-5 years after traumatic brain injury rehabilitation in Australia. Brain Inj. 1999;13:417–431
    • View In Article
    • MEDLINE
    • CrossRef
22. Dahmer ER, Shilling MA, Hamilton B, et al. A model systems database for traumatic brain injury. J Head Trauma Rehabil. 1993;8(2):12–25
    • View In Article
    • CrossRef
23. Harrison-Felix C, Newton N, Hall K, Kreutzer J. Descriptive findings from the traumatic brain injury model systems national database. J Head Trauma Rehabil. 1996;11(5):1–14
    • View In Article
    • CrossRef
24. 1990 occupational classification system. Alphabetical index of industries and occupations (1990 census of population and housing). Washington (DC): Bureau of the Census, U.S. Department of Commerce; 1992;
    • View In Article
25. United Nations Statistics Division. International Standard Classification of Occupations. New York: United Nations; 1988;
    • View In Article
26. Regidor E, Ronda E, Martinez D, Calle ME, Navarro P, Dominguez V. Occupational social class and mortality in a population of men economically active: the contribution of education and employment situation. Eur J Epidemiol. 2005;20:501–508
    • View In Article
    • MEDLINE
    • CrossRef
27. Forer S. Functional assessment instruments in medical rehabilitation. J Organization Rehabil Evaluators. 1982;2:29–41
    • View In Article
28. Granger CV, Hamilton BB, Keith RA, Zielesny M, Sherwin FS. Advances in functional assessment for medical rehabilitation. Top Geriatr Rehabil. 1986;1:59–74
    • View In Article
29. Guide for the Uniform Data Set for Medical Rehabilitation, version 5.0. Buffalo: State Univ New York; 1996;
    • View In Article
30. Hall KM, Johnston MV. Outcomes evaluation in traumatic brain injury rehabilitation (Part II: measurement tools for a nationwide data system). Arch Phys Med Rehabil. 1994;75(Suppl):SC10–SC18
    • View In Article
    • MEDLINE
31. Rappaport M, Herrero-Backe C, Rappaport ML, Winterfield KM. Head injury outcome up to ten years later. Arch Phys Med Rehabil. 1989;70:885–892
    • View In Article
    • MEDLINE
32. Dikmen S, Machamer J, Temkin N, McLean A. Neuropsychological recovery in patients with moderate to severe head injury: 2 year follow-up. J Clin Exp Neuropsychol. 1990;12:507–519
    • View In Article
    • MEDLINE
    • CrossRef
33. Conzen M, Ebel H, Swart E, Skreczek W, Dette M, Oppel F. Long-term neuropsychological outcome after severe head injury with good recovery. Brain Inj. 1992;6:45–52
    • View In Article
    • MEDLINE
    • CrossRef
34. Macciocchi SN, Reid DB, Barth JT. Disability following head injury. Curr Opin Neurol. 1993;6:773–777
    • View In Article
    • MEDLINE
    • CrossRef
35. McLean A, Dikmen SS, Temkin NR. Psychosocial recovery after head injury. Arch Phys Med Rehabil. 1993;74:1041–1046
    • View In Article
    • MEDLINE
    • CrossRef
36. Dikmen SS, Ross BL, Machamer JE, Temkin NR. One year psychosocial outcome in head injury. J Int Neuropsychol Soc. 1995;1:67–77
    • View In Article
    • MEDLINE
    • CrossRef
37. Ponsford JL, Olver JH, Curran C. A profile of outcome: 2 years after traumatic brain injury. Brain Inj. 1995;9:1–10
    • View In Article
    • MEDLINE
    • CrossRef
38. Masson F, Maurette P, Salmi LR, et al. Prevalence of impairments 5 years after a head injury, and their relationship with disabilities and outcome. Brain Inj. 1996;10:487–497
    • View In Article
    • MEDLINE
    • CrossRef
39. Dombovy ML, Olek AC. Recovery and rehabilitation following traumatic brain injury. Brain Inj. 1997;11:305–318
    • View In Article
    • MEDLINE
    • CrossRef
40. Hanks RA, Temkin NR, Machamer JE, Dikmen SS. Emotional and behavioral adjustment after traumatic brain injury. Arch Phys Med Rehabil. 1999;80:991–997
    • View In Article
    • Abstract
    • Full-Text PDF (883 KB)
    • CrossRef
41. Guerts AC, Ribbers GM, Knoop JA, van Limbeek J. Identification of static and dynamic postural instability following traumatic brain injury. Arch Phys Med Rehabil. 1996;77:639–644
    • View In Article
    • Abstract
    • Full-Text PDF (743 KB)
    • CrossRef
42. Guskiewicz KM. Assessment of postural stability following sport-related concussion. Curr Sports Med Rep. 2003;2:24–30
    • View In Article
    • MEDLINE
43. Williams G, Robertson V, Greenwood K. Measuring high level mobility after traumatic brain injury. Am J Phys Med Rehabil. 2004;83:910–920
    • View In Article
    • MEDLINE
    • CrossRef
44. Satz P. Brain reserve capacity on symptom onset after brain injury: a formulation and review of evidence for threshold theory. Neuropsychology. 1993;7:273–295
    • View In Article
    • CrossRef
45. Ruff RM, Marshall LF, Crouch J, et al. Predictors of outcome following severe head trauma: follow-up data from the Traumatic Coma Data Bank. Brain Inj. 1993;7:101–111
    • View In Article
    • MEDLINE
    • CrossRef
46. Dikmen SS, Temkin NR, Machamer JE, Holubkov AL, Fraser RT, Winn R. Employment following traumatic head injuries. Arch Neurol. 1994;51:177–186
    • View In Article
    • MEDLINE
47. Boake C, Millis SR, High WM, et al. Using early neuropsychological testing to predict long-term productivity outcome from traumatic brain injury. Arch Phys Med Rehabil. 2001;82:761–768
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (58 KB)
    • CrossRef
48. Kreutzer J, Marwitz J, Walker W, et al. Moderating factors in return to work and job stability after traumatic brain injury. J Head Trauma Rehabil. 2003;18:128–138
    • View In Article
    • MEDLINE
    • CrossRef
49. Yasuda S, Wehman P, Targett P, Cifu DX, West M. Return to work for persons with traumatic brain injury: a review of literature. Am J Phys Med Rehabil. 2001;80:852–864
    • View In Article
    • MEDLINE
    • CrossRef