Abstract
Objective: To examine job stability moderating variables and develop a postinjury work stability prediction model.
Design: Multicenter analysis of individuals with traumatic brain injury (TBI) who returned for follow-up at 1, 2, and 3, or 4 years postinjury, were of working age (between 18 and 62 years of age at injury), and were working preinjury.
Setting: Six National Institute on Disability and Rehabilitation Research TBI Model System centers for coordinated acute and rehabilitation care.
Participants: A total of 186 adults with TBI were included in the study.
Main outcome measures: Job stability was categorized as stably employed (employed at all 3 follow-up intervals); unstably employed (employed at one or two of all three follow-up intervals); and unemployed (unemployed at all three follow-up intervals).
Results: After injury, 34% were stably employed, 27% were unstably employed, and 39% were unemployed at all three follow-up intervals. Minority group members, people who did not complete high school, and unmarried people were more likely to be unemployed. Driving independence was highly influential and significantly related to employment stability. A discriminant function analysis, which included age, length of unconsciousness and Disability Rating Scale scores at 1 year postinjury, accurately predicted job stability groupings.
Conclusion: Data analysis provided evidence that employment stability is predictable with a combination of functional, demographic, and injury severity variables. Identification of people at risk for poor employment outcomes early on can facilitate rehabilitation planning and intervention.
TRAUMATIC BRAIN injury (TBI) typically results in a variety of short- and long-term sequelae including cognitive, behavioral, and physical impairments. Perhaps one of the most discouraging consequences is unemployment. Employment has been associated with higher perceived quality-of-life, 1-3 whereas failure to return to work (RTW) has been associated with poorer psychosocial adjustment and physical ailments. 4-6
Rates of employment after TBI vary widely from study to study. Brooks and colleagues 7 found that unemployment levels rose from 14% preinjury to 71% postinjury. Others have reported similar rates of unemployment after TBI, ranging from 55% to 78%, 8-11 whereas some researchers have documented much lower levels of unemployment ranging from 10% to 34%. 12,13 Differing definitions of employment help explain widely varying RTW rates reported by different researchers. For example, some investigators include sheltered or subsidized and unpaid work, which includes volunteer, homemaker, and student. Broader definitions may more accurately portray "productivity" outcome 14-17 as opposed to employment outcome.
Differences in TBI cohort selection criteria likely lead to further variance in reported unemployment rates. TBI predictor studies have consistently found that early injury severity measures strongly correlate with RTW outcome, 9,13,16-27 implying that comparisons are not valid between cohorts with different injury severity mixes. In addition, other moderating variables are often cited to influence postinjury unemployment including older age, 13,23 premorbid unemployment, 9,28 and lower levels of education. 7,21,29 Impairment on acute rehabilitation-based measures such as the Disability Rating Scale (DRS) 9 and the FIM instrument; 18 longer lengths of inpatient rehabilitation stay 28 and poor performance on early neuropsychological testing 16 have also been found to predict unemployment after TBI.
Less clear is the role of ethnicity and marital status on employment after TBI. With regard to ethnicity, some researchers have found no association, 17,18,30 whereas others 15,29 found that minority status was associated with higher rates of unemployment. Ip and colleagues 21 determined that unmarried subjects were less likely to be employed postinjury than married subjects; however, marital status was found to be unrelated to RTW by Greenspan and colleagues. 29
Recent literature has also suggested that variables measured at 6 months and 1 year may add predictive power to earlier measures. 17,31 Felmingham and colleagues examined data collected at 6 months postdischarge to predict employment at 24 months postdischarge. The addition of postdischarge predictors including psychologic distress (General Health Questionnaire), community integration (Community Integration Questionnaire, CIQ), and cognitive status (Functional Assessment Measure cognitive scale) significantly improved the accuracy of predictions of work status for the sample. In another study, 17 researchers collected DRS ratings, CIQ responses, and information on whether subjects had returned to productive activity (return to preinjury-comparable work, full-time school, or homemaking) at 1 year postinjury. Results indicated that greater impairment on the DRS and CIQ was associated with a higher rate of failure to return to productive activity.
Describing employment for a single time period is a noticeable drawback of nearly all TBI RTW studies. Recent longitudinal data suggest that most individuals' employment status changes with time. 32 The present study sought to expand upon earlier studies of RTW using a longitudinal sample of individuals who were employed before injury. With the benefits of a longitudinal sample, job stability across three annual follow-up visits was determined. Injury severity, functional status at 1 year postinjury, and demographic variables were used to predict job stability after TBI.