Archives of Physical Medicine and Rehabilitation
Volume 90, Issue 10 , Pages 1716-1722, October 2009

A Prospective Study of Factors Influencing Return to Work After Traumatic Spinal Cord Injury in Taiwan

  • Mau-Roung Lin, PhD

      Affiliations

    • Institute of Injury Prevention and Control, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan, Republic of China
    • Corresponding Author InformationReprint requests to Mau-Roung Lin, PhD, Institute of Injury Prevention and Control, Taipei Medical University, 250 Wu-Hsing Street, Taipei 110, Taiwan, R.O.C.
    • ,
  • Hei-Fen Hwang, MS

      Affiliations

    • Department of Nursing, National Taipei College of Nursing, Taipei, Taiwan, Republic of China
    • ,
  • Wen-Yu Yu, MD, MS

      Affiliations

    • Department of Emergency Medicine, Taipei Medical University Hospital, Taipei, Taiwan, Republic of China
    • ,
  • Chih-Yi Chen, MS

      Affiliations

    • Institute of Injury Prevention and Control, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan, Republic of China

Article Outline

Abstract 

Lin M-R, Hwang H-F, Yu W-Y, Chen C-Y. A prospective study of factors influencing return to work after traumatic spinal cord injury in Taiwan.

Objective

To examine comprehensively the effects of physical, psychologic, and sociologic characteristics on employment among persons after a traumatic spinal cord injury (SCI) in Taiwan.

Design

A prospective study with follow-up telephone interviews over a 3-year period.

Setting

To register people who had sustained an SCI, medical records of 4 hospitals were reviewed using codes of the International Classification of Diseases–9th Revision–Clinical Modifications from 806.0 to 806.9 and from 952.0 to 952.9.

Participants

Subjects (N=219) employed at the time of injury.

Interventions

Not applicable.

Main Outcome Measures

Employment status after an SCI.

Results

The employment rate was 32.9%. After controlling for other variables, education level (relative rate [RR]=4.01∼8.17), autonomy in transportation (RR=5.13), professional licensure (RR=1.86), and thrill and adventure-seeking trait (RR=1.12) were positively and significantly associated with employment, while subjects with more severe overall injury severity (RR=0.95), preinjury chronic conditions (RR=0.20), necessity for aids for daily living (RR=0.31), and depression (RR=0.38) were less likely to have been employed than their counterparts.

Conclusions

In addition to education level and traditional physical factors, overall injury severity and psychologic factors such as thrill and adventure seeking and depression can also influence the return to work after an SCI.

Key Words: Depression, Employment, Rehabilitation, Social support

List of Abbreviations: ADL, activity of daily living, AIS, Abbreviated Injury Scale, ASIA, American Spinal Cord Injury Association, BS, boredom susceptibility, CES-D, Center for Epidemiologic Studies Depression Scale, ISS, Injury Severity Score, SCI, spinal cord injury, TAS, thrill and adventure seeking

 

IMPROVEMENTS IN EARLY postinjury care have minimized or eliminated many of the complications that previously accompanied new traumatic SCI1; however, the complex long-term outcomes for people surviving with SCI still need to be improved.2 Employment after an SCI is an important clinical and public health issue because of its physical, psychologic, and social implications.3 Persons with an SCI are more likely to consider work to be a source of personal fulfillment and social recognition,4 and studies have consistently found that persons who find employment after an SCI have better life satisfaction.5, 6 In addition, employment is an important indicator of social integration for this population.7

Only 21% to 67% of persons with an SCI working at the time of injury return to work after an injury,8 which implies that 33% to 79% of those who would like to work remain unemployed. Therefore, predictors of employment after an SCI are important for identifying target groups and developing effective intervention programs. In many studies, demographic and physical factors such as age at injury, sex, education, preinjury employment, functional ability, cause of injury, time since injury, and neurologic severity (completeness of injury and impairment type) have been reported to be significantly associated with the employment status after an SCI,9, 10, 11, 12, 13, 14, 15 in addition to medical conditions such as pain, fatigue, pressure ulcers, urinary dysfunction, urinary tract infection, spasticity, and respiratory problems.16

However, several disadvantages of previous studies can be improved on. First, most of those findings were based on cross-sectional designs and were derived from retrospective data; thus, the temporal sequence between those identified predictors and employment is not clear, resulting in survival bias of risk factors that may have been misidentified as protective. Second, while the physical factors identified explain only a part of the variation in employment participation, psychologic and sociologic characteristics, such as sensation-seeking traits,17 depressive status, and social support, have rarely been included as potential factors affecting the return to work after an SCI. Because sensation-seeking traits are found to predispose certain persons to an SCI and to distinguish persons with an SCI from the general public,18 they might also have a role in resistance to stress and coping with a return to work. Depression and perceived social support appear to be predictive for some rehabilitative outcomes among persons with SCI19, 20; to our knowledge, their effects on employment participation have not been simultaneously evaluated. Finally, the neurologic severity of an SCI, evaluated by the Impairment Scale of ASIA,21 is used to indicate injury severity for persons with an SCI. However, it is common that patients with a traumatic SCI have also sustained injuries to body regions other than the spinal cord, and therefore, the ASIA might not adequately reflect the overall impact of multiple injuries in persons with an SCI.

To address these issues, this prospective study was conducted to examine comprehensively the effects of physical, psychologic, and sociologic characteristics on returning to work after an SCI in Taiwan.

Back to Article Outline

Methods 

Study Participants 

Potential study subjects were recruited from patients with an SCI at 4 teaching hospitals: Mackay Memorial Hospital in Taipei City and its branch in Tamshui Township, Chang-Gung Memorial Hospital in Linkou Township; Tri-Service General Hospital in Taipei City; and Tzu Chi General Hospital in Hualien County. Following a protocol of the US Centers for Disease Control,22 673 eligible subjects were identified from hospital discharge data during a 5-year period from July 1, 1999, to June 30, 2004. Inclusion criteria were based on International Classification of Diseases–9th Revision–Clinical Modifications codes from 806.0 to 806.9 (fracture of the vertebral column with an SCI) and from 952.0 to 952.9 (SCI without evidence of spinal bone injury). Subjects age 15 years or younger and those whose spinal cord problems were not a result of an acute injury but from lesions from degenerative disk disease, infection, and tumors were excluded.

Of these eligible subjects identified, 238 completed the telephone interview of the baseline assessment, and 219 had been employed before the injury. Compared with the 238 participants, the 435 nonparticipants had significantly lower educational levels, more complete lesions of the spinal cord, and higher levels of overall injury severity, while they had similar distributions of other baseline characteristics. A flow diagram of the study population is provided (fig 1). This research was reviewed and approved by the Institutional Review Board of Taipei Medical University, Taiwan, and written consent was obtained from all participants.

Baseline Assessment and Instruments 

Physical assessment 

Baseline characteristics of participants were collected from medical records and telephone interviews. In addition to demographics (eg, age, sex, education), we collected data on injury characteristics (eg, time and cause of the injury event, neurologic severity, level of SCI, injured anatomic parts, diagnosis) as well as work-related characteristics before the injury, ADLs, necessity for aids for ADLs, transportation, and medical problems (eg, urinary function and pain).

Injury severity assessment 

Rather than using a single measure to assess injury severity, 3 measures of injury severity of an SCI were used: the ASIA Impairment Scale, Barthel Index, and ISS. To assess the neurologic severity of a spinal cord lesion, there are 5 categories in the ASIA Impairment Scale: A (complete, no sensory or motor function), B (incomplete, sensory function only), C (incomplete, muscle grade of <3), D (incomplete, muscle grade of ≥3), and E (normative).21 The Barthel Index assesses disability severity in ADLs.23, 24 The 10-item instrument includes self-feeding, getting in/out of bed, grooming, performing one's toileting, bathing, walking, climbing stairs, self-dressing, and controlling the bowels and bladder. Scores of 0 to 60, 61 to 90, 91 to 99, and 100 on the instrument respectively indicate severe, moderate, mild, and no dependency.25 The ASIA Impairment Scale and Barthel Index may also indicate injury severity in persons with an SCI.

An attending neurosurgeon reviewed the abstracted information and computed the AIS and ISS for each subject. The AIS is an ordinal scale of injury severity ranging from 1 (minor injury) to 6 (unsurvivable) for each of 6 body regions.26 The ISS was developed as a method of combining multiple body injuries on the same person into a single measure of overall injury severity.27 It is an anatomical scoring system that provides an overall score for subjects with multiple injuries. Each injury is assigned an AIS score and is allocated to 1 of 6 body regions (head, face, chest, abdomen, extremities and pelvis, external). Only the highest AIS score in each body region is used. The scores of the 3 most severely injured body regions are squared and added together to produce the ISS score. The ISS score ranges from 0 to 75. Any person with an AIS 6 injury is assigned an ISS of 75.

Psychosocial assessment 

Sensation seeking, depression, and social support were examined in the study. Sensation seeking was defined as a trait that requires varied, novel, and complex sensations and experiences such that a person willingly takes physical and social risks for the sake of such experiences.17 The sensation-seeking trait was characterized by 2 subscales of the Sensation-Seeking Scale (Form V) of Zuckerman of TAS and BS, with each subscale consisting of 10 binary-choice items.17 The TAS measures the self-reported tendency for risky adventures (eg, “I often wish I could be a mountain climber,” and “I sometimes like to do things that are a little frightening”), whereas the BS measures aversion to repetitive or dull experiences (eg, “I can't stand watching a movie that I've seen before,” and “I get bored seeing the same old faces”). Psychometric performance on the subscales of the Sensation-Seeking Scale has been reported28, 29; for example, the internal consistency coefficients were .77 to .82 for the TAS and .65 to .56 for the BS, and the test-retest reliabilities at a 3-week interval were .94 for the TAS and .70 for the BS. We hypothesized that employment is a physically thrilling and adventuresome activity and unemployment is a dull, repetitive activity for persons with an SCI.

The CES-D consists of 20 items, emphasizing 6 affective components of a depressed mood, feelings of guilt and worthlessness, feelings of helplessness and hopelessness, psychomotor retardation, loss of appetite, and sleep disorders.30 All items refer to the frequency of symptoms during the past week and are scored on 4-point scales. CES-D scores range from 0 to 60, with a score of 16 or greater indicating depression.31, 32

The Social Support Survey, including the 6 domains of social network, tangible support, affection, positive social interaction, informational support, and emotional support, was used to evaluate social support.33 The open-ended item for social networks was not included in the calculation of the scale score, while the other 19 items are 5-point scales. Scores of the instrument were rescaled to a 0 to 100 range, with scores of 0 to 80 indicating a lack of strong social support.33, 34

Definition and Ascertainment of Employment after Spinal Cord Injury 

Persons were classified as being employed if they were working in the competitive labor market, self-employed, a homemaker, or a student. At each annual follow-up telephone interview from 2003 to 2006, information on the employment status and date of the return to work was gathered.

Quality Control 

Medical record reviewers reviewed and abstracted each subject's medical records to verify the SCI diagnosis and obtained information on the physical assessment. To improve the quality of the abstracted data, these record reviewers received 6 hours of training to understand key areas of the abstraction form, key variable identification, and site-specific logistical concerns. Additionally, 1 researcher (W.-Y.Y.) checked the first 10 charts abstracted by each reviewer and then discussed the results with the reviewers.

For the telephone interviews of assessing injury severity and psychological variables, a maximum of 6 telephone calls, with the first 3 attempts in the evening and the latter 3 in the daytime, were placed to each potential subject. When a subject agreed to participate in the study during a call, the subject was asked to make a period of approximately 30 minutes available within 1 week for a telephone interview. Interview procedures were also standardized through interviewer participation in a 4-hour training course.

Statistical Analysis 

In a preliminary analysis, the Pearson chi-square test was computed to compare the distributions of depressive status, social support, and other baseline categorical variables between employed and unemployed groups, while Student t test was computed for TAS, BS, ISS, and other baseline continuous variables. The distributions of baseline variables between participants and nonparticipants were also compared; those significantly differing were included in the final proportional hazards model to adjust for a potential selection bias.

The proportional hazards model was applied to investigate whether the likelihood of employment significantly differed among groups with respect to demographic, physical, or psychosocial characteristics.35 The dependent variable in the model was the duration from the onset of the SCI to return to work, and data on participants who did not return to work were censored at the time of death or at the end of the study follow-up (ie, June 30, 2006). In the model, the intensity function at a time point, which is an instantaneous rate of occurrence of employment at a time point for a subject, was assumed to be multiplicatively related to the independent variables. While the level of significance was set to .05, variables with a P value less than .25 on the preliminary analysis were included in the initial multivariable analysis to minimize type II errors in variable selection and biased inferences.36 The assumption of proportional hazards for the model was checked by a plot of the estimated log cumulative hazard function against the logarithm of survival function. SAS version 9.1 was used for all statistical analyses.a

Back to Article Outline

Results 

Among the 219 participants employed before the injury, 72 had returned to work over the 3-year follow-up period—that is, the employment rate was 32.9%. Of the 72 subjects, 49 kept the job they had had before being injured. As shown in table 1, compared with unemployed subjects, employed subjects were younger, had higher levels of education, had an engineering/science background, had incomplete lesions with a muscle grade of 3 or greater, had lower ISS scores, had fewer preinjury chronic conditions, had normative urinary function, lived with ADL independence, required fewer aids for ADLs or ambulation, had autonomy in transportation, had a professional license, and had information on job opportunities.

Table 1. Distribution of Demographic and Injury Characteristics of 72 Employed and 147 Unemployed Subjects After Traumatic SCI
CharacteristicEmployed Before the Injury (N=219)P
Employed (n=72)Unemployed (n=147)
Mean ± SD⁎No. (%)†Mean ± SD⁎No. (%)†
Age at injury (y)35.7±11.8 41.7±15.4 .002
ISS (0–75 points)17.7±7.4 19.9±6.7 .032
Length of stay (d)91±138 131±176 .071
Time since discharge (y)3.0±1.3 2.8±1.5 .382
Time since injury (y)3.2±1.3 3.2±1.4 .716
Male sex 62(86.1) 116(78.9).268
Education level
Elementary school or below 6(8.3) 48(32.7)<.001
Junior high 17(23.6) 32(21.8)
Senior high 33(45.8) 54(36.7)
College or above 16(22.2) 13(8.8)
Specialty
General 37(51.4) 110(74.8)<.001
Sociology/humanities 6(8.3) 13(8.8)
Engineering/science 22(30.6) 22(15.0)
Others 7(9.7) 2(1.4)
Marital status
Single/divorced/widowed 30(41.7) 60(40.8)1.000
Married 42(58.3) 87(59.2)
Neurologic severity (ASIA grade)
A (Complete) 10(13.9) 34(23.1).038
B (Incomplete, sensory function only) 9(12.5) 22(15.0)
C (Incomplete, muscle grade <3) 7(9.7) 27(18.4)
D (Incomplete, muscle grade ≥3) 46(63.9) 64(43.5)
Preinjury chronic conditions 3(4.2) 21(14.3).036
Urinary dysfunction 14(19.4) 69(47.3)<.001
Barthel Index (0-100 points)
Severe disability (≤60) 8(11.1) 69(47.0)<.001
Moderate disability (61-90) 8(11.1) 34(23.1)
Mild disability (91-99) 6(8.3) 10(6.8)
Independence (100) 52(69.5) 34(23.1)
Aids for daily living 5(7.0) 43(29.2)<.001
Aids for ambulation
No aids 56(76.4) 56(38.1)<.001
Wheelchair 13(18.1) 71(48.3)
Crutch/cane/brace 4(5.6) 20(13.6)
Autonomy in transportation 49(70.0) 29(20.4)<.001
Preoccupational assessment 5(7.0) 15(10.0).618
Professional licensure 22(31.0) 16(10.9)<.001
Vocational retraining 5(7.0) 11(7.6)1.000
Information on job opportunities 17(23.9) 8(5.6)<.001

⁎The Student t test was used for continuous variables.

†The Pearson χ2 test was used for categorical variables.

Table 2 shows the distributions of psychologic and sociologic characteristics between the 72 employed and 147 unemployed subjects. Employed subjects were more likely to have higher levels of TAS, no depression, and strong social support than the unemployed. No significant difference in BS levels was detected between the 2 groups.

Table 2. Distribution of Psychologic and Sociologic Characteristics by Employment Status After Traumatic SCI
CharacteristicEmployed Before the Injury (N=219)P
Employed (n=72)Unemployed (n=147)
Mean ± SD*No. (%)†Mean ± SD*No. (%)†
TAS (0-10 points)3.0±3.7 1.4±2.7 .001
BS (0-10 points)3.4±2.6 3.0±2.8 .338
Depression (CES-D≥16 points) 18(28.6) 83(59.7)<.001
Strong social support (SSS>80 points) 17(23.6) 19(12.9).045

Abbreviation: SSS, Social Support Survey.

The Student t test was used for continuous variables.

†Pearson's chi-square test was used for categorical variables.

Results of the proportional hazards model are shown in table 3. After adjusting for other variables, compared with subjects who had an education level of elementary school or no formal education, those with higher education levels were more likely to be employed, from an increase of 301% for those with a level of junior high school to 717% for those with a level of college or higher. An increase in each additional point on the ISS was significantly associated with a 4% decline in returning to work. Subjects who had preinjury chronic conditions and required aids for daily living were 80% and 69% less likely to be employed after an SCI and those with autonomy in transportation and professional licensure were 413% and 86% more likely to be employed after an SCI, compared with their counterparts. For psychologic and sociologic characteristics, an increase in each additional point of the TAS was significantly associated with an increase of 12% of returning to work, while those who had higher levels of BS did not have a significantly increased likelihood of returning to work. Moreover, subjects who had depression were 62% less likely to have returned to work, and those who had strong social support did not have a significantly increased rate of returning to work, compared with their counterparts.

Table 3. Results of the Proportional Hazards Model: Adjusted RRs and 95% CIs for Return to Work After Traumatic SCIs
CharacteristicRR(95% CI)
Education level
Elementary school or below1.00
Junior high4.01(1.53–10.5)
Senior high4.92(2.01–12.0)
College or above8.17(3.03–22.0)
ISS0.95(0.92–0.99)
Preinjury chronic conditions0.20(0.06–0.64)
Aids for daily living0.31(0.12–0.82)
Autonomy in transportation5.13(2.72–9.68)
Professional license1.86(1.09–3.18)
TAS1.12(1.05–1.19)
Depression0.38(0.21–0.69)

Abbreviations: CI, confidence interval; RR, relative rate.

Back to Article Outline

Discussion 

This study reveals that in addition to demographic and physical variables of education level, overall injury severity, the presence of preinjury chronic conditions, necessity for aids for daily living, autonomy in transportation, and professional licensure, psychologic variables of the TAS level of the sensation-seeking trait and depressive status were also significantly associated with a return to work after a traumatic SCI. On the other hand, another dimension of the sensation-seeking trait of the BS and social support did not appear to have a significant effect on the return to work. Most of the findings on education level and physical-related variables are consistent with those found in previous studies3, 9, 10, 11, 12, 13, 14, 15; nonetheless, the magnitudes of the effect estimates seem to be larger, probably because of smaller measurement errors in the prospective study. When only 49 employed subjects who had returned to their jobs at preinjury were included in the data analysis, the significant factors identified for employment were similar.

In contrast with another study in Taiwan that reported a 51% employment participation among persons with SCI who had a job before the injury,9 our employment rate was low. Because the definitions of employment in the 2 studies were similar, the difference in the employment rate may have resulted from a survival bias in which people with an SCI need to survive long enough to participate in a cross-sectional study. A prospective study might not only timely reflect the employment status after an SCI but also avoid a possible selection bias caused by a healthy survival effect. For example, 62.7% of the study sample in the study by Jang et al9 had autonomy in transportation, versus 35.6% in our study. Some studies treating time since injury as an indicator of total adjustment to an SCI37, 38 should take the healthy survival effect into account; otherwise, the influence of injury severity and other characteristics that lead to nonparticipation and losses to follow-up among subjects with SCI may be underestimated. Moreover, age at injury was not an independently significant variable for employment because it was strongly correlated with educational level in this study (r=–.50). Older and less-educated subjects may have had a lower socioeconomic status, had a smaller range of job prospects, and been in a job position that required more physical labor before the injury.3, 10, 12, 13 People with functional limitations are less likely to return to positions that require a lot of physical labor.

Consistent with previous studies,12, 13, 14, 15 injury severity and other physical variables were significantly associated with employment after an SCI. While several measures of injury severity were applied in this study, variability in the return to work could be better explained by the ISS than the ASIA Impairment Scale and Barthel Index, even though the last 2 measures were significantly associated with a return to work when excluding the ISS. Because patients with an SCI often have sustained injuries to other body regions such as the head and/or extremities, an overall injury severity such as ISS scores can better reflect the impact of potential multiple injuries on an individual, and injuries other than the SCI may have affected the employment status of persons with SCI. Whether the interaction of multiple injuries or any specific injury other than an SCI influences the return to work can be further clarified in future research.

Although the sensation-seeking level was found to be higher for persons with an SCI than the general public,18 this is the first study to reveal that their adventurous traits of sensation seeking may have resulted in a better likelihood of returning to work after injury. This finding gains some indirect support from other studies. For instance, sensation-seeking traits were reported to motivate self-actualization of subjects with SCI through facilitating spontaneity in life, interest in current activities, and satisfaction with the meaning and purpose in life.39 Moreover, persons with higher levels of sensation seeking tend to learn by discovering environmental contingencies through actively and repeatedly challenging their environment,40 and patients with an SCI are encouraged to confront their disability while still learning problem-solving and independence skills.19, 41 Nevertheless, not all components of the sensation-seeking trait were assessed in the study, and further research on all components of sensation-seeking traits on the return to work after an SCI is warranted.

There is controversy about the role of depression in persons with an SCI. Several investigators believe that depression is an essential component in the adaptation process and a normative response to an SCI.42 In substance, empirical data do not support depression being experienced by all people with SCIs.43, 44 Depression was found to be negatively associated with more adaptive behavior and an internal locus of control,41 and its occurrence may partly reflect a lack of social support.45, 46 Further subgroup analysis in this study showed that there was a strong, inverse correlation between depression and the amount of social support received (r=–.56); social support became a significant factor in the return to work when the depressive status was not included in the data analysis. Further research is needed to explore the interaction mechanisms between social support and depression concerning the return to work.

It was noted that psychosocial interventions of facilitating adjustment/adaptation and increasing employment participation are rarely used for people with an SCI. The most commonly used intervention is to increase an individual's physical skills to overcome the disability after an injury.47 However, traditional training of physical skills such as vocational rehabilitation programs and driving ability might not be sufficient to achieve the highest possible level of independence for persons with an SCI, because successful reintegration into the community, like employment, may also require social support through workplace accessibility (eg, no architectural barriers), employment incentives (eg, no racial or disability discrimination), and political policy (eg, sufficient budgets for critical programs). While not compromising accommodations from external support, facilitating positive psychologic characteristics like an internal locus of control and work motivation among persons with an SCI may also play important roles in a person's return to work.10, 11, 39, 41

Study Limitations 

There are limitations to this study. First, the employment rate might have been overestimated in the study. Because nonparticipants had significantly lower education levels, more complete lesions of the spinal cord, and higher levels of overall injury severity than participants, they may have had a lower rate of returning to work. Nonetheless, our analyses were conducted to assess the evidence of an association in an unbiased manner in that the education level and overall injury severity were controlled for in the study, with the effect of complete lesions of the spinal cord on employment explained away by the overall injury severity. Second, it is not clear whether the sensation-seeking trait evaluated here remained unchanged after the injury, because our data were collected after the onset of an SCI. However, people generally exhibit lower levels of sensation seeking with the passage of time19; accordingly, those who had higher levels of sensation seeking after the SCI probably had higher levels prior to the SCI. Finally, the 3-year follow-up period might not have been long enough for subjects to have adapted to an SCI, and the resulting disability for a return to work and the follow-up time length may have confounded the relations of employment to psychologic and sociologic characteristics. In addition, workplace environments that may also confound the results were not measured in the study.

Back to Article Outline

Conclusions 

Not only demographic and physical factors but also psychologic characteristics of people with an SCI can affect the possibility of returning to work. Overall injury severity rather than neurologic and disability severity can reflect the effects of multiple injuries on employment after an SCI. For increasing employment participation, rehabilitation interventions should simultaneously include physical skill training and psychological adjustment/adaptation.

Supplier

Back to Article Outline

References 

  1. Johnson RL, Gerhart KA, McCray J, et al. Secondary conditions following spinal cord injury in a population-based sample. Spinal Cord. 1998;36:45–50
  2. Lundqvist C, Siösteen A, Blomstrand C, et al. Spinal cord injuries: clinical, functional, and emotional status. Spine. 1991;16:78–83
  3. Meade MA, Lewis A, Jackson MN, Hess DW. Race, employment, and spinal cord injury. Arch Phys Med Rehabil. 2004;85:1782–1792
  4. Ville I, Ravaud JF. Work values: a comparison of non-disabled persons with paraplegia. Disabil Rehabil. 1998;20:127–137
  5. Lin MR, Hwang HF, Chen CY, Chiu WT. Comparisons of the WHOQOL-BREF and SF-36 for persons with spinal cord injury. Am J Phys Med Rehabil. 2007;86:104–113
  6. Leduc BE, Lepage Y. Health-related quality of life after spinal cord injury. Disabil Rehabil. 2002;24:196–202
  7. Anderson CJ, Krajci KA, Vogel LC. Community integration among adults with spinal cord injuries sustained as children or adolescents. Dev Med Child Neurol. 2003;45:129–134
  8. Lidal IB, Huynh TK, Biering-Sørensen F. Return to work following spinal cord injury: a review. Disabil Rehabil. 2007;29:1341–1375
  9. Jang Y, Hsieh CL, Wang YH, Wu YH. Return to work after spinal cord injury in Taiwan: the contribution of functional independence. Arch Phys Med Rehabil. 2005;86:681–686
  10. Krause JS. Employment after spinal cord injury. Arch Phys Med Rehabil. 1992;73:163–169
  11. Murphy GC, Young AE. Employment participation following spinal cord injury: relation to selected participant demographic, injury and psychological characteristics. Disabil Rehabil. 2005;27:1297–1306
  12. Hess DW, Ripley DL, McKinley WO, Tewksbury M. Predictors for return to work after spinal cord injury: a 3-year multicenter analysis. Arch Phys Med Rehabil. 2000;81:359–363
  13. DeVivo MJ, Rutt RD, Stover SL, Fine PR. Employment after spinal cord injury. Arch Phys Med Rehabil. 1987;68:494–498
  14. Wang RY, Yang YR, Yen LL, Lieu FK. Functional ability, perceived exertion and employment of individuals with spinal cord lesion in Taiwan. Spinal Cord. 2002;40:69–76
  15. Tomassen PC, Post MW, van Asbeck FW. Return to work after spinal cord injury. Spinal Cord. 2000;38:51–55
  16. Krause JS. Aging and self-reported barriers to employment after spinal cord injury. Top Spinal Cord Injury Rehabil. 2001;6:102–115
  17. Zuckerman M. Sensation-seeking: beyond the optimal level of arousal. Hillsdale: L. Erlbaum Associates; 1979;
  18. Mawson AR, Biundo JJ, Clemmer DI, et al. Sensation-seeking, criminality, and spinal cord injury: a case-control study. Am J Epidemiol. 1996;144:463–472
  19. Trieschmann R. Spinal cord injury: psychological, social, and vocational adjustment. New York: Demos Publications; 1988;
  20. Elliott TR, Frank RG. Depression following spinal cord injury. Arch Phys Med Rehabil. 1996;77:816–823
  21. American Spinal Injury Association (ASIA). International standards for neurological and functional classification of spinal cord injury—revised 1992. Chicago: ASIA; 1992;
  22. Harrison CL, Dijkers M. Spinal cord injury surveillance in the Unites States: an overview. Paraplegia. 1991;29:233–246
  23. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index. Maryland State Med J. 1965;14:61–65
  24. Leung SO, Chan CC, Shah S. Development of a Chinese version of the modified Barthel Index—validity and reliability. Clin Rehabil. 2007;21:912–922
  25. Shah RB, Yarkony GM, Ortolano D, et al. Prediction of functional outcome by motor capability after spinal cord injury. Arch Phys Med Rehabil. 1989;70:819–822
  26. Association for the Advancement of Automotive Medicine (AAAM). The Abbreviated Injury Scale—1990 Revision. Des Plain: AAAM; 1990;
  27. Baker SP, O'Neill B, Haddon W, et al. The injury severity score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma. 1974;14:187–196
  28. Zuckerman M, Buchsbaum MS, Murphy DL. Sensation seeking and its biological correlates. Psychol Bull. 1980;88:187–214
  29. Young KF. Sensation seeking and the relation with creativity and juvenile delinquency [master's thesis]. Taipei, Taiwan: National Cheng Chih University; 1987;
  30. Radolff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1:385–401
  31. Weissman MM, Sholomskas D, Pottenger M, et al. Assessing depressive populations: a validation study. Am J Epidemiol. 1977;106:203–214
  32. Chien CP, Cheng TA. Depression in Taiwan: epidemiological survey utilizing CES-D. Bull Jpn Soc Neurol Psychiatry. 1985;87:335–338
  33. Sherbourne CD, Stewart AL. The MOS Social Support Survey. Soc Sci Med. 1991;32:705–714
  34. Yu DS, Lee DT, Woo J. Psychometric testing of the Chinese version of the Medical Outcomes Study Social Support Survey (MOS-SSS-C). Res Nurs Health. 2004;27:135–143
  35. Cox DR. Regression models and life-tables. J Roy Stat Soc B. 1972;34:187–220
  36. Mickey RM, Greenland S. The impact of confounder selection criteria on effect estimation. Am J Epidemiol. 1989;129:125–137
  37. Conroy L, McKenna K. Vocational outcome following spinal cord injury. Spinal Cord. 1999;37:624–633
  38. Engel S, Murphy GS, Athanasou JA, Hickey L. Employment outcomes following spinal cord injury. Int J Rehabil Res. 1998;21:223–229
  39. Hammell KR. Psychological and sociological theories concerning adjustment to traumatic spinal cord injury: the implications for rehabilitation. Paraplegia. 1992;30:317–326
  40. Taylor GP. Moderator variable effect on personality test item endorsements of physically disabled patients. J Consult Clin Psychol. 1970;35:183–188
  41. Mahon-Darby J, Ketchik-Renshaw B, Richmond TS, Gates EM. Powerless in cervical spinal cord injury patients. Dimens Crit Care Nurs. 1988;7:346–355
  42. Krueger DW. Rehabilitation psychology. Rockville: Aspen Publishing; 1984;
  43. Judd F, Stone J, Webber J, Brown D, Burrows G. Depression following spinal cord injury. Br J Psychiatry. 1989;154:668–671
  44. Frank RG, Elliott TR, Corcoran JR, Wonderlich SA. Depression after spinal cord injury: is it necessary?. Clin Psychol Rev. 1987;7:611–630
  45. Elliott TR, Herrick S, Witty T, Godshall F, Spruell M. Social support and depression following spinal cord injury. Rehabil Psychol. 1992;37:37–48
  46. Rintala D, Young J, Hart K, Clearman R, Fuhrer M. Social support and the well-being of persons with spinal cord injury living in the community. Rehabil Psychol. 1992;37:155–163
  47. Targett PS, Wehman PH, McKinley WO, Young CL. Successful work supports for persons with SCI: focus on job retention. J Vocat Rehabil. 2004;21:19–26
  • a SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513-2414.

 This work was funded by the National Health Research Institute (NHRI-EX96-9204PP), Taiwan, Republic of China.

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated.

PII: S0003-9993(09)00358-X

doi:10.1016/j.apmr.2009.04.006

Archives of Physical Medicine and Rehabilitation
Volume 90, Issue 10 , Pages 1716-1722, October 2009

Article Tools

* Image Usage & Resolution