| | Employment After Traumatic Brain Injury: Differences Between Men and WomenAbstract Corrigan JD, Lineberry LA, Komaroff E, Langlois JA, Selassie AW, Wood KD. Employment after traumatic brain injury: differences between men and women. ObjectiveTo determine whether there are sex differences in employment 1 year after traumatic brain injury. DesignProspective cohort. SettingAcute care hospitals in South Carolina and Traumatic Brain Injury Model Systems (TBIMS) rehabilitation centers. ParticipantsSubjects in the TBIMS national dataset and the South Carolina Traumatic Brain Injury Follow-up Registry who were expected to be working before injury and followed at 1 year postinjury. InterventionsNot applicable. Main Outcome MeasureChange in employment from preinjury to 1 year postinjury. ResultsWhen other measured influences on change in hours worked were held constant, there were significant interactions for sex by age and sex by marital status. Compared with men, women were more likely to decrease hours or stop working, except in the oldest age group (55−64y) in which men were more likely to stop working. For women, there was a pattern showing better employment outcomes as age increased. Decreased employment for women was most evident for married women, who were much more likely to reduce hours or stop working. There was also a tendency for divorced women to be more likely to stop working when compared with divorced men. ConclusionsThese findings run counter to the current literature. Although definitive explanations must await future studies, causal factors arising from differential societal behavior toward women as well as discriminatory attitudes about women and employment deserve further study. STUDIES OF THE LONG-TERM effects of traumatic brain injury (TBI) have consistently found that adults have significant problems working after their injuries.1, 2, 3 In a large cohort of adults receiving rehabilitation for TBI, rates of unemployment were very high; only 27% of participants in the Traumatic Brain Injury Model Systems (TBIMS) national dataset reported being competitively employed 1 year after injury and 29% 5 years postinjury.4 There also have been consistent findings regarding the following risk factors for not working after injury: injury severity, preinjury educational and occupational status, and age at injury.1, 2, 3 Sex has not been found to be associated with unemployment. Studies that have investigated the relationship between sex and employment have found mixed results; either no differences were observed or men had worse employment outcomes than women.1 However, this research has been limited by methodologic issues, most notably the relatively low number of women compared with men enrolled in studies of TBI.5, 6 In contrast to previous findings, data from 1 large population-based study found that women fared worse than men when other covariates were held constant (J Langlois, W Rutland-Brown, personal communication, May 2006). The potential importance of understanding sex differences affecting employment was the impetus for the present study. The integration of data from the TBIMS and South Carolina Traumatic Brain Injury Follow-up Registry (SCTBIFR)7 provided a unique opportunity to examine whether employment outcomes after TBI are worse for women than men. Injury severity has consistently been identified as an important risk factor for not working after TBI. Doctor et al8 reported incremental increases in the relative risk of unemployment as a function of initial Glasgow Coma Scale score or 1-month Glasgow Outcome Scale score for patients hospitalized with a TBI who had been working preinjury. In studies limited just to people with more severe injuries who received rehabilitation, the association between initial injury severity and employment was not as strong.9, 10, 11 In a study12 from the TBIMS that examined predictors of employment up to 5 years postinjury, length of stay (LOS) in rehabilitation, which is highly associated with severity of the initial injury, was predictive at 1 year although not thereafter. In rehabilitation cohorts in which injury severity is attenuated, the specific impairments from the brain injury rather than its overall severity may influence employment.1, 2 Wehman et al2 have suggested that poor self-awareness, diminished interpersonal skills, cognitive difficulties, and impulsivity, the latter including the propensity for alcohol and other drug abuse, have the greatest effect on employment after TBI. Preinjury educational and occupational status have been consistently related to working after TBI.2, 8, 12 Doctor8 found those with less than a high school diploma had a higher rate of unemployment at 1 year postinjury than those either with a high school diploma or some college training. Those without a high school diploma also had a higher rate of “excess risk of unemployment,” the extent to which a group’s unemployment rate exceeded the rate for similarly educated people in the general population of working age adults. The comparison by Keyser-Marcus et al12 across 5 years postinjury found education was a significant predictor in the first year but not thereafter. Wehman et al2 have suggested that education is related to preinjury occupational status, which, in turn, affects the likelihood of employment after injury. The higher the occupational status the more likely there are health benefits, intrinsic motivation for work, or greater flexibility and accommodations extended to the injured worker seeking to return. The significant effect of preinjury employment at any level has complicated the research on employment post-TBI. Among rehabilitation patients, very different factors have been found to predict return to work versus unemployment.12, 13 When preinjury employment was not taken into consideration, factors associated with preinjury unemployment (eg, substance abuse, socioeconomic status, race, ethnicity) were much more predictive. Beyond preinjury educational and occupational status, age is the only other sociodemographic characteristic consistently associated with employment after TBI. Specifically, those aged 40 and older had a greater risk for unemployment, which was further increased among those aged 60 and older.11, 14, 15, 16 Keyser-Marcus12 found age to be a significant predictor of employment for the first 4 years after injury. Marital status has received limited attention, and, with mixed results, 1 study17 reported higher unemployment for unmarried persons, whereas another found greater return to work for those not married.10 As noted previously, demographic factors associated with prior employment have shown crude associations with later unemployment when not adjusted for preinjury employment. Despite the extensive research on working after TBI,1, 2, 3 the relationship between sex differences and employment has not been closely examined. In studies that have investigated sex differences, results have been inconsistent. Several studies10, 17, 18, 19, 20, 21, 22, 23, 24 that examined the contribution of multiple variables to employment after TBI found that sex was not significant. Studies8, 11, 25 in which differences were found have reported that men had worse employment rates than women. Doctor et al8 found a 2.95 relative risk of unemployment for women who had a TBI when compared with the general population; however, the risk for men was 5.05. McMordie et al11 reported crude rates showing that a higher proportion of women than men were employed, which they attributed to lower severity of injury for women. In many previous studies,12, 26, 27, 28, 29, 30 sex was included but not tested for statistical significance, including Cifu et al26 who showed that women were 20% more likely to be unemployed at 1 year postinjury than men. Several factors have complicated the examination of sex differences in employment and interpretation of subsequent findings. First, women are often underrepresented in TBI samples, particularly in studies of patients with moderate to severe TBI treated in rehabilitation. In almost all studies of adults with TBI, women represented less than 30% of the study population, which in small samples greatly limits statistical power for examining sex differences. As a result, some relatively large differences that have been observed in previous studies19, 23 were not statistically significant. Small subgroup size may also account for the lack of testing of sex differences, even when there appeared to be an association of sex with employment.26 Injury characteristics may also confound the relationship between sex and employment. Some studies conducted on rehabilitation cohorts reported that women had less severe initial injury, which, in turn, could bias employment outcomes.11, 31 The relationship with injury severity is further complicated by sex differences in injury etiology. For instance, women are less likely than men to be injured in an assault,32 which is associated with poorer premorbid employment as well as a greater likelihood of substance abuse, both risk factors for poorer employment postinjury.13 Finally, sex differences in the nature of the injury to the brain could also affect later employment outcomes. Some studies5, 33 have found greater mortality and morbidity related to sex, with women more vulnerable to greater damage. There has been some evidence of a greater tendency for increased intracranial pressure for premenopausal women, which would result in greater brain damage.34 Davis et al35 reported better outcomes for postmenopausal women when compared with age-matched men, whereas premenopausal women showed no differences from men. This study was conducted on a large, population-based sample of adults with moderate and severe TBI. In contrast to research showing greater injury severity among younger women, there are also studies suggesting younger women have better outcomes,36, 37, 38 which have been attributed to neuroprotective effects of hormones for premenopausal women.37, 38 To fully evaluate the impact of TBI on employment among women, an adequate sample size is needed to examine the wide range of factors that could influence this relationship, especially severity and preinjury employment status. The integrated dataset created from the TBIMS and SCTBIFR provided an opportunity for such analysis.7 Although the extant literature showed mixed findings regarding sex and employment after TBI, regression models computed on data from the Colorado TBI Follow-up Registry suggested that when other factors are taken into account women actually have worse outcomes than men (J. Langlois, W. Rutland-Brown, personal communication, May 2006). Based on this modeling, our a priori hypothesis was that there are differences in employment for women versus men resulting in women working fewer hours at 1 year postinjury. Methods  Participants and Settings The sample for the current study consisted of 3444 adults (2487 men, 957 women) hospitalized with a TBI. This study used a dataset, described elsewhere,7 in which data from the TBIMS and SCTBIFR were integrated. The cohorts represented in these datasets differ because of the objectives of the programs that created them and the methodologies used for data collection. The primary objectives of the TBIMS are (1) to develop and show a model system of care for persons with TBI, stressing continuity and comprehensiveness of care; (2) to track the long-term outcomes of persons who go through these systems; and (3) to establish and maintain a centralized, standardized database designed to achieve the first 2 objectives. Although each local TBIMS center has unique attributes, the common denominator throughout the program is that all research participants experience both acute care and acute inpatient rehabilitation within designated hospitals in the local system of care. In addition, all centers collect a common core dataset. Although the core dataset has varied over the years, it currently consists of 112 variables (form I) collected during initial hospitalization and 88 variables (form II) collected at follow-up years 1, 2, 5, and every 5 years thereafter. The definition of TBI in the TBIMS database is “damage to brain tissue caused by an external mechanical force, as evidenced by loss of consciousness due to brain trauma, posttraumatic amnesia, skull fracture, or objective neurologic findings that can be reasonably attributed to TBI on physical examination or mental status examination.”39(p2) Eligible research participants provide informed consent (or have consent provided by an authorized proxy). On giving consent, a subject’s acute hospital medical record is abstracted for information about injury-related characteristics, timing of milestones in recovery of consciousness, and medical complications encountered. Data are collected contemporaneously during the comprehensive rehabilitation stay and include demographic characteristics, premorbid functioning, functional status at admission and discharge, and cost of services received. Follow-up interviews at years 1, 2, 5, and every 5 years thereafter collect information about sociodemographic status, subsequent injury and hospitalization, functional independence, employment, substance use, and subjective well-being. At the time of the present study, the TBIMS included initial demographic and injury-related information on 5020 participants and data at 1-year follow-up on 3125 participants. The TBI Surveillance System, funded by the Centers for Disease Control and Prevention (CDC), has the overarching purpose of collecting data from acute care hospitals on the incidence, etiology, risk factors, severity, and outcomes of TBI in a statewide population.40 In 1998, CDC funded South Carolina as the second state to implement a statewide TBI follow-up system to acquire additional information about TBI outcomes. The SCTBIFR became a component of South Carolina’s surveillance of persons who are hospitalized in acute care facilities or treated and released from the emergency department after acquiring TBI. The specific objectives of the SCTBIFR are (1) to determine the proportion of persons with TBI who report cognitive, behavioral and emotional, and neurosomatic deficits during the follow-up period; (2) to identify specific functional limitations (disability) by person and injury characteristics; (3) to determine level of life satisfaction; (4) to assess the mortality experience of the cohort during the follow-up period; and (5) to identify unmet needs and barriers to receiving services. The SCTBIFR defines a case of TBI as any discharge with a primary or secondary diagnosis of skull fracture and/or intracranial injury in accordance with the CDC case definition for TBI.40 For case identification purposes, nature of injury codes from the International Classifications of Diseases, 9th Revision, Clinical Modification (ICD-9-CM), included were 800.0 to 801.9, 803.0 to 804.9, 850.0 to 854.1, and 959.01. The SCTBIFR cohort was a random sample of all persons, aged 15 years and older, discharged alive from all nonfederal, acute care hospitals in South Carolina. Potential participants were stratified into high (Abbreviated Injury Score [AIS] head ≥3) or low (AIS head=2) TBI severity. The initial sample had a sampling ratio of 2:1 high to low severity. By using this information, health information specialists from the South Carolina Department of Health and Environmental Control abstracted additional variables from the eligible participants’ hospital medical records to validate the existing data and to gather predischarge history and other medical information. The SCTBIFR conducted the first annual interview approximately 12 months after the participant’s hospital discharge. The survey assessed functional limitations, disability, and other adverse outcomes. The domains assessed included general health status, preexisting health conditions, difficulties performing activities of daily living and instrumental activities of daily living, depressive symptoms, services and unmet needs, alcohol and drug use, pre- and postinjury employment, living situation, marital status, education level, health insurance coverage, personal income, transportation, life satisfaction, cognition, social integration, and social support. The SCTBIFR eligible sample consisted of 3715 of 9688 randomly selected patients discharged from acute care facilities from January 1, 1999, through June 30, 2002. First year follow-up interviews were conducted on 2118 participants from May 2000 through July 2003. Proxy respondents reported for 285 of these participants. The 1-year cohorts for both the SCTBIFR and the TBIMS were affected by the loss to follow-up of participants eligible for follow-up who could not be located, refused to continue, died, became incarcerated, or otherwise were unable to complete the follow-up interview. For the SCTBIFR, 57% of the eligible sample was interviewed on follow-up. Participants who had no health insurance or had Medicaid were less likely to be located than participants who had Medicare or commercial insurance. Characteristics of nonrespondents were accounted for in the weighting of the year 1 cohort.41 For the TBIMS dataset, 75% of eligible participants were interviewed at year 1. Based on the comparison by Corrigan et al42 on dropout rates among 3 longitudinal studies, including the TBIMS, variables most likely to be associated with loss to follow-up were socioeconomic disadvantage, a history of substance abuse, and a violent injury etiology. They also observed that severe motor deficits made it more likely that a person would be found for follow-up at year 1. Procedures To integrate the SCTBIFR and TBIMS datasets, a process was developed to determine which variables in both datasets were comparable. Working from lists of variables for each program, those with apparent similar domains were examined for their actual content and methods used to elicit responses. Corresponding levels for comparable variables in each dataset, including instances in which levels were combined to establish equivalence, were identified for an integrated dataset. The source of data, either the SCTBIFR or TBIMS, was included as a variable. The cohort for the current study was limited to persons who were expected to be working preinjury. A person was defined as expected to be working if they were between the ages of 18 and 64 inclusive and were not a student, homemaker, resident of specialized housing, or retired and not working. Our outcome of interest was the difference between the number of hours that the person worked at 1 year postinjury as compared with hours worked at the time of injury (postinjury hours minus preinjury hours). Pre- and postinjury hours worked were defined as the number of hours worked per week. This variable was analyzed as a categorical variable with 4 levels: increase in hours, same number of hours, decrease in hours but still working, and decrease in hours and no longer working. Injury severity was determined by translating the ICD-9-CM diagnosis codes into AIS head region by using ICDMAP-90 software.a ICDMAP-90 is severity scoring software that computes the severity of the trauma from ICD-9-CM nature of injury codes by using an internal translation scheme that takes the patient’s age and clinical descriptors (the fifth digits of ICD-9-CM codes) into account.43 The AIS is capable of classifying over 2000 injuries according to the body region of injury (eg, head, chest, extremity), type of structure injured (eg, nerve, vessel, bone), location of injury within the body region (eg, femur, tibia, or talus), and nature of injury (eg, abrasion, burn, crush). Consistent with the original AIS, ICDMAP-90–derived AIS scores grade each injury according to its associated threat to life on an ordinal scale from 1 (minor) to 6 (unsurvivable).43, 44, 45 ICDMAP-90 computes for each trauma patient an AIS score for each region of the body, including the head. Age was categorized as 18 to 19, 20 to 24, 25 to 34, 35 to 44, 45 to 54, and 55 to 64 years (those older were excluded because they were not expected to be working). The race groups used were white, black, Hispanic, and other. Education was categorized as 12 or fewer years of school, 12 years or a general educational development diploma, some college or more, and unknown. Marital status included single (never married), married, and divorced or separated. Insurance status was categorized as private, uninsured, worker’s compensation, Medicaid, Medicare, and other. Computed tomography (CT) had 3 categories: normal, abnormal, and unknown. LOS was categorized into 4 levels based on the number of days spent in an acute care hospital: 1 to 5, 6 to 12, 13 to 23, and 24 or more. Last, mechanism of injury was categorized into transport, violence, falls, and other. Interaction terms were considered with sex and all demographic and injury severity variables except LOS and preinjury employment hours. Statistical Analysis We used SASb for all analyses. All variables were assessed for crude association with the outcome by using the chi-square test. Any variable attaining significance at P less than .10 was considered in the multivariable model. Change in hours worked was assessed as a categorical variable by using polychotomous logistic regression. Variables were entered into the model simultaneously, and the least significant variables (P>.05) were removed 1 at a time from the model by using the log likelihood chi-square test. Variables that had P less than .05 and 2 measures of severity (AIS, CT) were retained in the model, although the severity measures were not significant. The model was considered final when the log likelihood test attained significance with the removal of additional variables. Proposed interaction terms involving sex were also entered and ruled out by using the same method as was used with the main-effects variables. The predictive power of the model was assessed by using a max-rescaled chi-square test, and the fit of the model was assessed by a likelihood chi-square test. Results Of the 3444 persons who satisfied the inclusion criteria including that they were expected to be working at the time of injury, 327 (9.5%) worked more hours 1 year postinjury, 1257 (36.5%) worked the same number of hours as they had preinjury, 443 (12.9%) worked fewer hours but were still working, and 1417 (41.4%) had stopped working. Table 1 contains the percentages for each independent variable within each level of the outcome. Although all the people in all 6 of our age groups showed, on average, a decline in the number of hours that they worked postinjury, the decline was greatest in the 35-to-44 age group (−19.3h), followed by the 55-to-64 age group (−18.6h), the 25-to-34 age group (−18.3h), and the 45-to-54 age group (−17.3h). Persons in the 18-to-19 age group had the smallest decline in hours worked (−10.7h), followed by the 20-to-24 age group (−15.2h). Men decreased their hours more than women. Persons with less than 12 years of education showed a greater mean decline in work hours than did the other education groups. Persons who were single were more likely to increase their work hours (12% of single persons increased their work hours vs 7% of divorced persons and 8% of married persons). Also, overall, a more severe injury resulted in a greater loss of work hours postinjury. Table 2 shows the results from the logistic regression model. Results indicated that blacks and other races were significantly more likely than whites to stop working. Also, persons with a high school education or more were more likely than persons with less than a high school degree to increase hours or work the same rather than stop working. Persons who were uninsured, used worker’s compensation, or had Medicaid were significantly less likely than those with private insurance to increase or keep their work hours the same rather than stop working. This association was not seen in persons with Medicare or other types of insurance. An association was also seen between LOS in acute care and change in hours after injury. Persons with a longer LOS (>5d) were significantly more likely to stop working. | | |  | Variable | Increase | Same | Decrease and Still Working | |
|---|
| Age (vs 18−19)⁎ | | | | | |  55−64 | 0.11(0.05−0.28) | 0.73(0.39−1.35) | 0.43(0.20−0.95) | | | 45−54 | 0.26(0.12−0.53) | 1.88(1.09−3.24) | 0.89(0.45−1.69) | | | 35−44 | 0.32(0.17−0.63) | 1.55(0.93−2.60) | 0.80(0.43−1.49) | | | 25−34 | 0.68(0.37−1.23) | 2.15(1.32−3.50) | 0.77(0.43−1.40) | | | 20−24 | 0.50(0.53−1.64) | 1.90(1.16−3.12) | 1.13(0.63−2.02) | | | Sex (vs male) | | | | | | Female | 0.17(0.05−0.58) | 0.17(0.06−0.45) | 2.66(0.95−7.51) | | | Race (vs white)⁎ | | | | | | Other | 0.10(0.02−0.47) | 0.65(0.36−1.20) | 0.29(0.10−0.83) | | | Hispanic | 0.56(0.27−1.16) | 0.68(0.43−1.09) | 0.79(0.44−1.41) | | | Black | 0.32(0.22−0.46) | 0.51(0.40−0.65) | 0.55(0.41−0.73) | | | Education (vs <12y)⁎ | | | | | | Unknown | 1.37(0.50−3.78) | 0.89(0.48−1.64) | 1.10(0.46−2.65) | | | College + | 4.97(3.24−7.65) | 1.94(1.47−2.55) | 1.78(1.28−2.48) | | | 12 or GED | 2.69(1.77−4.07) | 1.43(1.10−1.86) | 1.34(0.97−1.86) | | | Marital status (vs married)⁎ | | | | | | Single | 0.45(0.29−0.71) | 0.47(0.35−0.64) | 1.08(0.74−1.57) | | | Divorced | 0.53(0.30−0.83) | 0.54(0.41−0.78) | 0.89(0.60−1.33) | | | Payer group (vs private)⁎ | | | | | | Uninsured | 0.47(0.29−0.76) | 0.40(0.28−0.58) | 0.81(0.55−1.18) | | | Other | 1.37(0.62−3.02) | 0.65(0.36−1.20) | 0.87(0.39−1.95) | | | Medicare | 0.37(0.12−1.19) | 0.91(0.43−1.92) | 0.97(0.20−4.58) | | | Medicaid | 0.53(0.36−0.78) | 0.67(0.52−0.86) | 0.72(0.53−0.99) | | | Worker’s compensation | 0.52(0.28−0.97) | 0.44(0.29−0.66) | 0.62(0.40−0.96) | | | AIS (vs 2) | | | | | | 5 | 0.53(0.29−0.99) | 0.79(0.54−1.17) | 1.10(0.67−1.78) | | | 4 | 0.73(0.48−1.11) | 0.80(0.59−1.08) | 1.18(0.82−1.69) | | | 3 | 0.98(0.63−1.54) | 0.86(0.61−1.21) | 1.24(0.83−1.84) | | | CT group (vs normal) | | | | | | Unknown | 0.86(0.50−1.50) | 1.07(0.73−1.57) | 0.75(0.47−1.20) | | | Abnormal | 0.93(0.61−1.40) | 0.80(0.59−1.07) | 0.80(0.56−1.13) | | | LOS (vs 1−5d)⁎ | | | | | | 24+ | 0.04(0.02−0.06) | 0.08(0.06−0.11) | 0.16(0.11−0.24) | | | 13−23 | 0.13(0.09−0.20) | 0.16(0.12−0.22) | 0.30(0.22−0.43) | | | 6−12 | 0.38(0.26−0.54) | 0.31(0.23−0.41) | 0.47(0.34−0.64) | | | Preinjury employment hours | 0.91(0.90−0.92) | 0.91(0.90−0.92) | 1.03(1.02−1.04) | | | (age by sex)⁎ | | | | | | 55−64 female | 0.06(0.02−0.20) | 1.06(0.48−2.34) | 1.43(0.56−3.63) | | | 45−54 female | 0.05(0.02−0.17) | 0.77(0.38−1.57) | 1.04(0.44−2.44) | | | 35−44 female | 0.08(0.03−0.25) | 0.56(0.29−1.10) | 1.40(0.64−3.08) | | | 25−34 female | 0.08(0.03−0.21) | 0.57(0.28−1.15) | 1.44(0.64−3.20) | | | 20−24 female | 0.18(0.06−0.52) | 0.39(0.18−0.86) | 1.62(0.66−3.98) | | | Marital status by sex⁎ | | | | | | Single female | 0.19(0.08−0.49) | 0.21(0.09−0.48) | 2.07(0.85−5.04) | | | Married female | 0.19(0.05−0.67) | 0.12(0.05−0.33) | 0.92(0.30−2.79) | | | | |
In general and consistent with our a priori hypotheses, women of all age groups had higher probabilities than men to work fewer hours after injury. There were significant 2-way interactions with sex for age and for marital status; however, the 3-way interaction among sex, age, and marital status was not significant. For age, higher probabilities of stopping work were noted among women than men for all age brackets except the 55-to-64 age group. Conversely, women had lower probabilities than men in all age groups of keeping the same hours worked. However, again men in the 55-to-64 age group differed from other age groups; they had a lower probability than women of keeping the same number of hours worked. Table 3 shows the results of the interactions in terms of predicted probabilities. Figure 1 displays the probabilities of decreasing work hours and of stopping working by sex and age. Inspection of these graphs shows that for women the negative impact on hours worked tended to decline as age increased. Figure 2 displays the probabilities of decreasing work hours and of stopping working by sex and marital status. For sex and marital status, although women overall had a higher probability than men of either decreasing their hours or stopping work, this probability was much higher for married women. The probability for divorced women was comparable to men for decreasing hours but still working; however, divorced women showed a greater likelihood of stopping working than divorced men. | | | | Variable | Increase | Same | Decrease Working | Stopped Working | |
|---|
| Age group | | | | | | | Age 18−19 | | | | | | | Female | 16.00 | 23.81 | 38.05 | 22.14 | | | Male | 34.79 | 51.77 | 5.27 | 8.17 | | | Age 20−24 | | | | | | | Female | 14.49 | 46.67 | 19.87 | 18.97 | | | Male | 22.33 | 67.93 | 4.11 | 5.63 | | | Age 25−34 | | | | | | | Female | 5.58 | 61.71 | 15.74 | 16.98 | | | Male | 15.99 | 75.68 | 2.77 | 5.56 | | | Age 35−44 | | | | | | | Female | 6.00 | 61.14 | 15.65 | 17.21 | | | Male | 10.80 | 77.29 | 4.05 | 7.86 | | | Age 45−54 | | | | | | | Female | 3.36 | 72.02 | 9.88 | 14.74 | | | Male | 7.51 | 81.80 | 3.83 | 6.86 | | | Age 55−64 | | | | | | | Female | 2.72 | 75.60 | 10.40 | 11.29 | | | Male | 7.43 | 72.51 | 4.37 | 15.69 | | | Marital status | | | | | | | Single | | | | | | | Female | 6.14 | 67.65 | 10.86 | 15.35 | | | Male | 9.14 | 67.98 | 8.17 | 14.71 | | | Divorced | | | | | | | Female | 9.05 | 60.04 | 7.39 | 23.53 | | | Male | 8.98 | 72.06 | 5.97 | 12.99 | | | Married | | | | | | | Female | 6.00 | 61.14 | 15.65 | 17.21 | | | Male | 10.80 | 77.29 | 4.05 | 7.86 | | | | |
Discussion Findings from the current study provide important, new insights into the relationship between sex and changes in employment 1 year after TBI. These results run counter to much of the existing literature8, 10, 11, 17, 18, 19, 20, 21, 22, 23, 24, 25 that has suggested either that sex is not associated with employment or that men have worse employment outcomes than women. Instead, we found that, when other measured influences on change in hours worked are held constant, women are more likely to decrease hours of employment than men. There were 2 significant interactions with sex: age and marital status. Compared with men, women were more likely to decrease hours or be unemployed, except in the oldest age group (55–64) in which women were less likely to be unemployed than men. For women, there was a pattern showing better employment outcomes as age increased. Decreased employment for women was most evident for married women, who were much more likely to decrease hours or stop working. There was also a tendency for divorced women to be more likely to stop working when compared with divorced men. Intuitively, societal influences related to sex roles would appear to be the most likely source of explanations for these effects of sex on employment after TBI. However, sex differences arising from biologic sources may also contribute. The interaction between sex and marital status may result from men being more likely to be the primary wage earner in the family,46, 47 which could, in turn, provide a greater opportunity for women to either stop working or work fewer hours after TBI. If a person is experiencing greater fatigue, persistent pain, slower cognitive processing, or a host of other common sequelae of TBI and the symptoms are affecting functioning but not prohibiting employment, then economic demands, such as the centrality of one’s income to sustaining the household, may drive the decision whether or not to reduce hours or stop working. Despite changes in the workforce related to sex, in households with married couples men remain almost 3 times more likely to earn a higher wage than their wives.46 In married families in which only 1 spouse works, it is 3 times more likely that it is the man.47 Thus, to the extent that economic need influences returning to work, it might be expected that married women are more likely to reduce hours or stop working. The interaction between sex and age was largely caused by the sharp increase in the proportion of men aged 55 to 64 who stopped working. These results may reflect a general population trend that is not unique to TBI. Private and public disability benefits have historically been seen as bridges to retirement, especially for persons over age 55 who have less time to wait before becoming eligible for Social Security Administration retirement benefits.48 Higher-paid jobs have better disability and retirement benefits, making the financial viability of “bridging” until retirement eligibility more likely. Men are also more likely to qualify for Social Security disability benefits (and thus be able to bridge to retirement even without private benefits) because they have more stable work histories, less interruptions due to childbearing and child-rearing.48 The same phenomena that create income differentials for men versus women (and that increase with age)46 create more continuous employment, which results in more benefits, including private and public disability insurance. We also observed a trend for women indicating that as age increased the proportion of women who reduced hours decreased. Occupations are more likely to be entry level at younger ages. Entry-level jobs for men are more likely physically labor oriented, whereas for women entry-level jobs may be more likely to involve cognitive and social skills. TBI may differentially affect the ability to return to jobs with greater cognitive and social demands than those with physical demands. Having children in the household, which was not measured in this study, may influence both of the interactions we observed. After experiencing a TBI, women with children in the home may be less able to manage both primary child-rearing responsibilities and employment and, therefore, more likely to decrease their work hours or stop working altogether. Because women who are working and have children in the home are both more likely to be younger and married,46 this association could contribute to both the age by sex and marital status by sex interactions observed in this study. Another social factor that may account for sex differences observed in this study is the greater likelihood of complications because of trauma for younger and/or married women. Intimate partner violence accounted for 20% of nonfatal violence against women in 2001 and 3% against men.49 Younger and married women are disproportionately victims of intimate partner violence.50 Several studies51, 52, 53, 54 have reported that women experiencing injury have a greater likelihood of developing posttraumatic stress disorders (PTSDs), which, in turn, may make return to work more difficult. Holbrook et al51, 52 reported that women were at significantly higher risk of later PTSD onset and, regardless of the timing of onset, had lower subjective well-being. Greenspan et al53 reported that 1 year after TBI, women were more likely than men to report subjective complaints consistent with PTSD. Increased stress and greater depression after trauma, including TBI, would be expected to affect employment. A final social factor that may influence sex differences in employment is the extent to which rehabilitation resources may be allocated in a manner reflecting lower value placed on employment for women. A study of vocational rehabilitation after TBI found that women received fewer services than men.55, 56 Differential service provision, if it exists, might result from discriminatory beliefs that it is less important for younger or married women to work when compared with younger or married men.6 The state-federal vocational rehabilitation system may need to reevaluate policies or practices that unintentionally result in differential rehabilitation services for women. For instance, lower priority given to clients who wish to work part-time would affect more women than men. Biologic explanations for the findings in the present study may have heuristic value even though they do not appear to account for the results as readily as do social explanations. It is possible that differences in the nature of the brain injury for women versus men could have a greater impact on work-related skills. Our findings may be consistent with studies that have found greater mortality and morbidity for women.5, 33 The interaction found between sex and age could be consistent with greater damage incurred by premenopausal women as reported in previous studies.34, 34 However, our findings are not consistent with studies suggesting younger women have better outcomes,36, 37, 38 including that attributed to a neuroprotective effect of hormones for premenopausal women.37, 38 As noted earlier, findings from the current study departed from previously reported results that concluded sex was not a significant factor affecting employment after TBI10, 17, 18, 19, 20, 21, 22, 23, 24, 25 or that men had worse employment rates than women.8, 11 Our data found crude rates of change that also suggested that men had worse outcomes than women. It was only when regression modeling allowed other factors to be held constant that a reversal in the effect of sex was evident. Although several covariates contributed to this reversal, the variable most responsible was prior employment. Examining postemployment data without consideration of preinjury employment could result in the apparently erroneous conclusion that employment outcomes are worse for men than women. In the current study, we used prior employment not only as a predictor but defined the dependent variable in terms of change in hours worked. Another advantage of the current study when compared with the extant literature was the large number of subjects provided by integrating the datasets from the TBIMS and the SCTBIFR. The resulting dataset included a sufficient number of women to allow robust regression modeling and the evaluation of interaction terms. Additionally, integrating the datasets provided a substantial number of subjects with both more and less severe TBI; previous studies have not had the advantage of having both ends of the spectrum well represented. The integrated dataset also included variables that allowed injury severity to be accounted for as a covariate.7 Finally, by limiting our examination to persons who were expected to be working before injury and using actual hours of competitive employment to define the dependent variable, we avoided problems that characterized some of the early studies of employment after TBI.57 Study Limitations There were several potential limitations of the current study. The 1-year datasets for both TBIMS and the SCTBIFR were affected by participants who were not available for the follow-up interview. The loss of participants introduces the potential for systematic bias when the reasons for being lost differentially affect certain subgroups of participants or selected variables.42 When data are used to estimate population rates, weighting can be used to compensate for nonrandom effects of loss to follow-up. However, when relationships are being examined for variables collected at 1 year, bias can occur if the variable of interest is associated with the factor affecting loss to follow-up.42, 58 In the current study, being uninsured and prior education were variables of interest that were potentially associated with both loss to follow-up and employment at 1 year after injury. Although one of the strengths of the study was to limit examination to those expected to be working preinjury, this variable could not be operationalized as discretely as might be desirable. For instance, older adults still in the work force and students coming into the work force in the year after their injury were not included in model building. Future prospective studies may want to consider how these groups are captured in employment cohorts. Another potential limitation of the study may have been the use of self-reported work hours. The relationship between self-report and actual hours is not known but was presumed to be veridical. A limitation in modeling was introduced by incomplete data for some variables, especially in the TBIMS dataset. Greater problems in data collection can be anticipated with datasets drawn from clinical versus administrative datasets. The use of a missing category reduced the loss of subjects for analysis; however, this approach also limits interpretation due to uncertainty about the meaning of the category. Finally, modeling would have been enriched by having a greater number and variety of variables that were comparable in the 2 datasets and, thus, available for analysis. Most notable was the lack of shared variables reflecting premorbid and co-occurring conditions. Study Implications There are several implications of the current findings. Most important, additional research is needed to fully explicate sex differences in employment after TBI. To explore underlying factors, future studies should consider including information about income and occupational type, the primary wage earner in the household, whether children are present in the household, if intimate partner violence was the etiology of the injury, whether PTSD or depression are present, and if rehabilitation services were received. Longitudinal examination of job stability also may provide a different perspective on employment outcomes. The current study underscores the need to have datasets of sufficient size or samples appropriately balanced to allow women to be addressed in analytic strategies. Nosek6 has observed that despite advances in research funding for TBI, there is an inequity in our knowledge of the differential impact of this condition on men and women. Both researchers and policy makers may need to consider concrete steps to promote better understanding of the impact of TBI on women. The study further underscores the relative benefits of both population-based follow-up registries that have been funded by the National Center for Injury Prevention and Control at CDC and the longitudinal dataset for persons with more severe TBI that has been funded by National Institute on Disability and Rehabilitation Research. A benefit of the TBIMS dataset is its focus on persons with more severe injuries who are likely to have longer-term consequences. Although the current study only compared year 1 cohorts, the TBIMS is designed to provide longitudinal data on outcomes spanning many years. Sampling designed to represent the population of persons with TBI (in this case, hospitalized patients in South Carolina) is essential for identifying issues related to demographic distributions and, as suggested by the current results, social disparities. The current study is an example of the complementary use of longitudinal and population-based data collection protocols. Conclusions Integration of the datasets from the TBIMS and SCTBIFR allowed a comprehensive examination of sex differences in employment after TBI. In contrast to the existing literature, we found that, when other measured influences on change in hours worked were held constant, women were more likely to decrease hours of employment than men. Compared with men, women were more likely to decrease hours or be unemployed, except in the oldest age group examined (55 to 64) in which women were less likely to be unemployed than men. Diminished employment for women was most evident for married women, who were much more likely to decrease hours or stop working. We have speculated about the sources of these observed differences; definitive explanations must await future studies. However, causal factors arising from differential societal behavior toward women as well as discriminatory attitudes about women and employment appear to deserve further study. Suppliers Acknowledgments We acknowledge Wesley Rutland-Brown, MPH, for his contributions to the conceptualization of this project. The contents of this study are solely the responsibility of the authors and do not necessarily represent the official views of the funding agencies. References 1. 1Ownsworth T, McKenna K. Investigation of factors related to employment outcome following traumatic brain injury: a critical review and conceptual model. Disabil Rehabil. 2004;26:765–783. MEDLINE |
CrossRef
2. 2Wehman P, Targett P, West M, Kregel J. Productive work and employment for persons with traumatic brain injury: what have we learned after 20 years?. J Head Trauma Rehabil. 2005;20:115–127. MEDLINE |
CrossRef
3. 3Yasuda S, Wehman P, Targett P, Cifu D, West M. Return to work for persons with traumatic brain injury. Am J Phys Med Rehabil. 2001;80:852–864. MEDLINE |
CrossRef
4. 4Traumatic Brain Injury National Data and Statistical Center. TBI Model Systems National Data Center Annual Report for 2004. Available at: http://www.tbindsc.org. Accessed December 13, 2006. 5. 5Kraus JF, Peek-Asa C, McArthur D. The independent effect of gender on outcomes following traumatic brain injury: a preliminary investigation. Neurosurg Focus. 2000;8:e5. MEDLINE 6. 6Nosek MA. Health concerns of women with brain injury. Brain Inj. 2006;3:28–32. 7. 7Corrigan JD, Selassie AW, Lineberry LA, et al. Comparison of the Traumatic Brain Injury (TBI) Model Systems national dataset to a population-based cohort of TBI hospitalizations. Arch Phys Med Rehabil. 2007;88:418–426. Abstract | Full Text |
Full-Text PDF (121 KB)
|
CrossRef
8. 8Doctor JN, Castro J, Temkin NR, Fraser RT, Machamer JE, Dikmen SS. Workers’ risk of unemployment after traumatic brain injury: a normed comparison. J Int Neuropsychol Soc. 2005;11:747–752. MEDLINE 9. 9Gollaher K, High W, Sherer M, et al. Prediction of employment outcome one to three years following traumatic brain injury. Brain Inj. 1998;12:255–263. MEDLINE |
CrossRef
10. 10Ip RY, Dornan J, Schentag C. Traumatic brain injury: factor predicting return to work or school. Brain Inj. 1995;9:517–532. MEDLINE |
CrossRef
11. 11McMordie WR, Barker SL, Paolo TM. Return to work (RTW) after head injury. Brain Inj. 1990;4:57–69. MEDLINE |
CrossRef
12. 12Keyser-Marcus LA, Bricout JC, Wehman P, et al. Acute predictors of return to employment after traumatic brain injury: a longitudinal follow-up. Arch Phys Med Rehabil. 2002;83:635–641. Abstract | Full Text |
Full-Text PDF (62 KB)
|
CrossRef
13. 13Bogner JA, Corrigan JD, Mysiw WJ, Clinchot D, Fugate LP. A comparison of substance abuse and violence in the prediction of long-term rehabilitation outcomes following traumatic brain injury. Arch Phys Med Rehabil. 2001;82:571–577. Abstract | Full Text |
Full-Text PDF (96 KB)
|
CrossRef
14. 14Sander A, Kreutzer J, Rosenthal M, Delmonico R, Young ME. A multicenter longitudinal investigation of return to work and community integration following traumatic brain injury. J Head Trauma Rehabil. 1996;11:70–84.
CrossRef
15. 15Crepeau F, Scherzer P. Predictors of work status after traumatic brain injury: a meta-analysis. Neuropsychol Rehabil. 1993;3:5–35.
CrossRef
16. 16Stambrook M, Moore AD, Peters LC, Deviaene C, Hawryluk GA. Effects of mild, moderate and severe closed head injury on long-term vocational status. Brain Inj. 1990;4:183–190. MEDLINE |
CrossRef
17. 17Kreutzer 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. MEDLINE |
CrossRef
18. 18Brown AW, Malec JF, McClelland RL, Diehl NN, Englander J, Cifu DX. Clinical elements that predict outcome after traumatic brain injury: a prospective multicenter recursive partitioning (decision-tree) analysis [published erratum in: J Neurotrauma 2005;22:1503]. J Neurotrauma. 2005;22:1040–1051. MEDLINE |
CrossRef
19. 19Cattelani R, Tanzi F, Lombardi F, Mazzucchi A. Competitive re-employment after severe traumatic brain injury: clinical, cognitive and behavioural predictive variables. Brain Inj. 2002;16:51–64. MEDLINE |
CrossRef
20. 20Devitt R, Colantonio A, Dawson D, Teare G, Ratcliff G, Chase S. Prediction of long-term occupational performance outcomes for adults after moderate to severe traumatic brain injury. Disabil Rehabil. 2006;28:547–559. MEDLINE |
CrossRef
21. 21Fleming 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. MEDLINE |
CrossRef
22. 22Testa JA, Malec JF, Moessner AM, Brown AW. Outcome after traumatic brain injury: effects of aging on recovery. Arch Phys Med Rehabil. 2005;86:1815–1823. Abstract | Full Text |
Full-Text PDF (135 KB)
|
CrossRef
23. 23O’Connell MJ. Prediction of return to work following traumatic brain injury: intellectual, memory, and demographic variables. Rehabil Psychol. 2000;45:212–217. 24. 24Wagner AK, Hammond FM, Howell CS, 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. Abstract | Full Text |
Full-Text PDF (600 KB)
|
CrossRef
25. 25Groswasser Z, Cohen M, Keren O. Female TBI patients recover better than males. Brain Inj. 1998;12:805–808. MEDLINE |
CrossRef
26. 26Cifu DX, Keyser-Marcus L, Lopez E, et al. Acute predictors of successful return to work 1 year after traumatic brain injury: a multicenter analysis. Arch Phys Med Rehabil. 1997;78:125–131. Abstract |
Full-Text PDF (993 KB)
|
CrossRef
27. 27Dikmen SS, Ross BL, Machamer JE, Temkin NR. One year psychosocial outcome in head injury. J Int Neuropsychol Soc. 1995;1:67–77. MEDLINE |
CrossRef
28. 28Hall KM, Bushnik T, Lakisic-Kazazic B, Wright J, Cantagallo A. Assessing traumatic brain injury outcome measures for long-term follow-up of community-based individuals. Arch Phys Med Rehabil. 2001;82:367–374. Abstract | Full Text |
Full-Text PDF (96 KB)
|
CrossRef
29. 29Kersel DA, Marsh NV, Havill JH, Sleigh JW. Psychosocial functioning during the year following severe traumatic brain injury. Brain Inj. 2001;15:683–696. MEDLINE |
CrossRef
30. 30MacMillan PJ, Hart RP, Martelli MF, Zasler ND. Pre-injury status and adaptation following traumatic brain injury. Brain Inj. 2002;16:41–49. MEDLINE |
CrossRef
31. 31Slewa-Younan S, Green AM, Baguley IJ, Gurka JA, Marosszeky JE. Sex differences in injury severity and outcome measures after traumatic brain injury. Arch Phys Med Rehabil. 2004;85:376–379. Abstract | Full Text |
Full-Text PDF (54 KB)
|
CrossRef
32. 32Langlois J, Rutland-Brown W, Thomas K. Traumatic brain injury in the United States: emergency department visits, hospitalizations and deaths. Atlanta: National Center for Injury Prevention and Control; 2004;. 33. 33Farace E, Alves WM. Do women fare worse: a metaanalysis of gender differences in traumatic brain injury outcome. J Neurosurg. 2000;93:539–545. MEDLINE |
CrossRef
34. 34Farin A, Deutsch R, Biegon A, Marshall LF. Sex-related differences in patients with severe head injury: greater susceptibility to brain swelling in female patients 50 years of age and younger. J Neurosurg. 2003;98:32–36. MEDLINE |
CrossRef
35. 35Davis DP, Douglas DJ, Smith W, et al. Traumatic brain injury outcomes in pre- and post-menopausal females versus age-matched males. J Neurotrauma. 2006;23:140–148. MEDLINE |
CrossRef
36. 36Kirkness CJ, Burr RL, Mitchell PH, Newell DW. Is there a sex difference in the course following traumatic brain injury?. Biol Res Nurs. 2004;5:299–310. MEDLINE |
CrossRef
37. 37Bayir H, Kagan VE, Tyurina YY, et al. Assessment of antioxidant reserves and oxidative stress in cerebrospinal fluid after severe traumatic brain injury in infants and children. Pediatr Res. 2002;51:571–578. MEDLINE 38. 38Stein DG, Hoffman SW. Concepts of CNS plasticity in the context of brain damage and repair. J Head Trauma Rehabil. 2003;18:317–341. MEDLINE |
CrossRef
39. 39Harrison-Felix C, Newton C, Hall K, Kreutzer J. Descriptive findings from the traumatic brain injury model systems national data base. J Head Trauma Rehabil. 1996;11:1–14.
CrossRef
40. 40Centers for Disease Control and PreventionDivision of Acute Care Rehabilitation Research and Disability Prevention. Traumatic brain injury in the United States: a report to Congress. Atlanta: CDC; 1999;. 41. 41Selassie A, Pickelsimer E, Furguson P, Gravelle W, Gu J. South Carolina traumatic brain injury follow-registry manual of operations. In: Weighting of data. Charleston: Medical Univ South Carolina; 2004;p. 118–121. 42. 42Corrigan J, Harrison-Felix C, Bogner J, Dijkers M, Terrill M, Whiteneck G. Systematic bias in traumatic brain injury outcome studies because of loss to follow-up. Arch Phys Med Rehabil. 2003;84:153–160. Abstract |
Full-Text PDF (87 KB)
|
CrossRef
43. 43Committee on Injury Scaling. The Abbreviated Injury Scale 1990 revision. Des Plaines: Association for the Advancement of Automotive Medicine; 1990;. 44. 44Petrucelli E, States J, Hames L. The abbreviated injury scale: evolution, usage and future adaptability. Accid Anal Prev. 1981;13:25–35. 45. 45X ICDMAP-90 software user’s guide. Baltimore: Johns Hopkins University and Tri-Analytics Inc; 1997;. 46. 46U.S. Bureau of Labor Statistics. Women in the labor force: a databook. 2005. Available at: http://www.bls.gov/cps/wlf-databook2006.htm. Accessed December 16, 2006. 47. 47U.S. Bureau of Labor Statistics. Employment characteristics of families in 2006. Available at: http://www.bls.gov/news.release/pdf/famee.pdf. Accessed December 16, 2006. 48. 48McMahon B, Danczyk-Hawley C, Reid C, Habeck R, Kregel J, Owens P. An evaluation of the progression of disability benefits among workers in American industry: impact, outcomes and implications. In: Wehman P editors. Progression of disability benefits. Richmond: Virginia Commonwealth Univ; 2002;. 49. 49Rennison C. Intimate partner violence, 1993-2001. Washington (DC): Bureau of Justice Statistics, Department of Justice; 2003;. 50. 50Tjaden P, Thoennes N. Full report of the prevalence, incidence, and consequences of violence against women: findings from the National Violence Against Women Survey. Washington (DC): Department of Justice; 2000;. 51. 51Holbrook TL, Hoyt DB, Stein MB, Sieber WJ. Gender differences in long-term posttraumatic stress disorder outcomes after major trauma: women are at higher risk of adverse outcomes than men. J Trauma. 2002;53:882–888. MEDLINE 52. 52Holbrook TL, Hoyt DB. The impact of major trauma: quality-of-life outcomes are worse in women than in men, independent of mechanism and injury severity. J Trauma. 2004;56:284–290. MEDLINE 53. 53Greenspan AI, Stringer AY, Phillips VL, Hammond FM, Goldstein FC. Symptoms of post-traumatic stress: intrusion and avoidance 6 and 12 months after TBI. Brain Inj. 2006;20:733–742. MEDLINE |
CrossRef
54. 54Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB. Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry. 1995;52:1048–1060. 55. 55Johnstone B, Vessell R, Bounds T, Hoskins S, Sherman A. Predictors of success for state vocational rehabilitation clients with traumatic brain injury. Arch Phys Med Rehabil. 2003;84:161–167. Abstract |
Full-Text PDF (84 KB)
|
CrossRef
56. 56Bounds TA, Schopp L, Johnstone B, Unger C, Goldman H. Gender differences in a sample of vocational rehabilitation clients with TBI. NeuroRehabilitation. 2003;18:189–196. MEDLINE 57. 57Fraser RT, Wehman P. Traumatic brain injury rehabilitation: issues in vocational outcome. Neurorehabilitation. 1995;5:39–48.
CrossRef
58. 58Corrigan J, Bogner J, Mysiw W, Clinchot D, Fugate L. Systematic bias in outcome studies of persons with traumatic brain injury. Arch Phys Med Rehabil. 1997;78:132–137. Abstract |
Full-Text PDF (884 KB)
|
CrossRef
a Department of Physical Medicine and Rehabilitation, Ohio State University, Columbus, OH b Department of Biostatistics, Bioinformatics, & Epidemiology, Medical University of South Carolina, Charleston, SC c Kessler Medical Rehabilitation Research and Education Center, West Orange, NJ d National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA.  Reprint requests to John D. Corrigan, PhD, Dept of Physical Medicine and Rehabilitation, Ohio State University, 480 Medical Center Dr, Columbus, OH 43210
Supported by the Centers for Disease Control and Prevention and the National Institute on Disability and Rehabilitation Research (grant no. H133A011403), the Ohio Regional Traumatic Brain Injury Model System (grant no. H133A020503), the South Carolina Traumatic Brain Injury Follow-up Registry (award no. U17/CCU421926), and the Henry H. Kessler Foundation. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated. PII: S0003-9993(07)01344-5 doi:10.1016/j.apmr.2007.08.006 © 2007 American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation. Published by Elsevier Inc. All rights reserved. | |
|
FULL TEXT01344-5/journalimage?img=PIIS0003999307013445.gr1.lrg.gif&fig=fig1&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)
01344-5/journalimage?img=PIIS0003999307013445.gr2.lrg.gif&fig=fig2&kwhquery=&issn=0003-9993&ishighres=false&allhighres=false&free=yes','journalimage');)