Rehabilitation Services, Self-Advocacy and Psychosocial Adaptation as Determinants of Employment among Persons with Spinal Cord Injury

Spinal Cord Injury (SCI) often results in devastating impacts on physical, psychological, and social well-being of individuals. In the United States, the prevalence of SCI in 2016 has been estimated at approximately 282,000 persons with incidence at approximately 54 cases per million populations [1]. Although, medical care for individuals with SCI has improved considerably in recent decades, resulting in longer life expectancy; measures of psychosocial adaptation and community participation, such as employment or work have not improved [2,3].

Employment is central and critical for overall wellbeing of individuals. The significance of employment to life satisfaction has been found to be a critical factor to adjustment after a disability [4]. Historically, employment of persons with disabilities in general, and among people with SCI in particular, is considered an important rehabilitation goal, although difficult to achieve. For persons with SCI, employment is fundamental to regaining their lives after the injury. According to research, [5-7] return to work by persons with SCI is regarded a mark of successful rehabilitation and adjustment to the disability. Despite the importance and need for participation in the community such as in employment multiple barriers hinder such essential activity. These barriers to employment of persons with SCI include functioning limitations due to the SCI, negative societal attitudes toward disabilities, psychological and adjustment issues related to the disability [8], and lack of appropriate type of rehabilitation services (such as self-advocacy skills training; Umeasiegbu [7]). According to Craig, Perry, Guest, Tran, and Middleton [9] “restrictions in social participation can become significant barriers to employment or educational activity”.

Despite the importance of employment to overall wellbeing and high unemployment rate among individuals with SCI, little research attention has been committed to understanding the interactions between multiple environmental and personal factors such as rehabilitation services, self-advocacy, and psychosocial adaptation and how these factors facilitate or hinder employment of individuals with SCI. For example, very little is known about the impacts of self-advocacy after SCI. The unsatisfactory participation in employment calls for re-evaluation of the rehabilitation services that individuals with SCI receive after their injury and how rehabilitation outcomes can be improved. The objective of this study was to explore the associations between multiple variables (demographic characteristics, SCI-related variables, and rehabilitation-related variables) and employment among a sample of persons with SCI. The following research questions were investigated in this study:

1. What associations exist between SCI-related variables (level of injury, SCI-complications, use of wheelchair) and rehabilitation-related variables (received rehabilitation services, self-advocacy, and psychosocial adaptation) and employment among persons with SCI?
2. What demographic characteristics of persons with SCI predict employment among a sample of persons with SCI?

A quantitative research method was used in this study; specifically, a cross-sectional design was used. Purposive sampling technique was used to recruit the study participants. Sample size determination was in accordance with Kline [10,11]. There is debate in the Structural Equation Modeling (SEM) literature regarding the sample size determination. Some SEM researchers recommend sample size of 200 and above while others posit that a sample size of 100 and above is an adequate sample size [10-14] and Schumacker and Lomax [15] posit that a ratio of the number of participants to the number of model parameters can be used as a guide for sample size in SEM. A ratio of 20:1 (20 participants to every model parameter) is regarded a large effect while a ratio of 10:1 (10 participants to every model parameter) is considered a medium effect. Path analysis which is a type of SEM is used in the current study. Since path analysis was used for data analysis, a ratio of the number of participants to the number of model parameters can be used as a guide for sample size in path analysis. Hence, a 10 parameters model will need a minimum sample size of 100 for a medium statistical effect. This study involves 7 parameters model, hence a sample size of 101 is deemed adequate for path analysis.

This study was approved by the University of Kentucky Institutional Review Board (IRB protocol #: 12-1016-X4B). Following approval from the IRB, data were collected using two formats; paper-and-pencil (survey package mailed to the participants) and web-page hosting of the survey link. The questionnaire was posted on a web page concurrently with the mailing of the paper-and-pencil questionnaire. The participants were recruited from the patient and member mailing lists of the SCI rehabilitation hospitals and National Spinal Cord Injury chapter of two Southeastern states and one Northeastern state of USA. The researchers provided the rehabilitation hospitals and chapter staff with research packages (including informed consent information, a paper survey with business-reply envelope, and a web address for participants wishing to complete the survey via the web) for distribution to 500 selected mailing-list members. An optional $10 gift-card was provided to participants who completed the survey in its entirety, and who indicated that they wanted the gift-card. This optional gift-card was in order to increase interest and participation in the study. The optional gift-card was included in the research protocol and received IRB approval.

The following instruments were used for data collection (a) Community Integration Questionnaire; (b) Self-advocacy Questionnaire; (c) Disability Centrality Scale; and (d) Received Rehabilitation Services Scale. The Cronbach’s alpha of these instruments can be found on table 1.

 

			Education
			Vocational/Technical School	Some College/Associate Degree	College Graduate
Gender	Male	Count	2	20	12
		% within gender	3.20%	32.30%	19.40%
		% within education	28.60%	66.70%	60.00%
		% of total	2.00%	20.00%	12.00%
	Female	Count	5	10	8
		% within gender	13.20%	26.30%	21.10%
		% within education	71.40%	33.30%	40.00%
		% of total	5.00%	10.00%	8.00%
Total		Count	7	30	20
		% within gender	7.00%	30.00%	20.00%
		% within education	100.00%	100.00%	100.00%
		% of total	7.00%	30.00%	20.00%

Table 1: Gender and education.

Gender * Education cross tabulation

Community Integration Questionnaire (CIQ) [16] was used to measure community participation; in this case, employment (employment is regarded in this study as a measure of community participation). CIQ was initially designed for assessment of community integration in persons with traumatic brain injury, but the instrument has been used in the population of persons with SCI [17] and adults with physical disabilities [18].

Self-Advocacy (SA) was assessed using a modified version of Self-Advocacy Questionnaire (SAQ) [19]. SAQ is a questionnaire initially developed for studying of self-advocacy among college students with disabilities. The SAQ asks participants to rate their perceived self-advocacy level based on a 7-point Likert scale (1 = Little Knowledge/Skill; 7 = Full Knowledge/Expert Skill).

Psychosocial Adaptation (PA) was assessed using Disability Centrality Scale (DCS) [20]. DCS is a multidimensional, quality of life-based model of psychosocial adaptation. The DCS is designed to measure the level of satisfaction, Chronic Illness and Disability (CIAD) related impact, and perceived control within 10 life domain, as well as the level of importance attributed to each of the domains. In two separate studies [21,22], the Cronbach alpha of DCS were 0.73 and 0.82 respectively.

Information on the types of rehabilitation services received by persons with SCI during inpatient and outpatient rehabilitation was collected using Received Rehabilitation Services Scale (RRSS). The RRSS is a 9-item questionnaire designed to measure types of rehabilitation-related services received by persons with SCI [7].

Data analyses were performed using the Statistical Package for the Social Sciences (SPSS) version 21.0. The various statistical analyses performed include descriptive statistics, correlation analysis, regression analysis, and path analysis. We used descriptive statistics to summarize the data. Correlation analysis was used to investigate the associations between selected domains of the study. Regression analysis was used to determine predictive functions of selected demographic information. Path analysis (SPSS’s Analysis of Moment Structures; AMOS) was used to estimate the strengths of the direct and indirect relationships among the select variables of the study. The goodness of fit for the current study was assessed using the chi square test, Bentlers’ Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI) and the Root-Mean-Squared Error of Approximation (RMSEA). The use of multiple measures of goodness of fit was to determine the overall fit from different perspectives [23].

A total of 101 individuals with SCI completed the questionnaire battery for this study. Summarized in table 1 is demographic and SCI-related information of the participants. Approximately 66% of the sample was between the ages of 30 and 59. Mean age of participants was 49 years (SD = 15.0). White (non-Hispanic) participants made up 84% of the sample. Concerning SCI-related characteristics of the sample, 62% of the sample sustained their injury between the ages of 20 and 39, and approximately 47% of the injuries were caused by automobile/vehicle accident. About 96% of the sample had medical complications related to SCI. Mean duration since injury was 13.94 years (S.D = 11.73). In addition, the gender distribution of selected variables (education, employment status before injury, and employment status after injury) is represented in tables 2-4.

 

Characteristic	N	(%)
Current age category
21-29	9	8.9
30-39	20	20
40-49	21	21
50-59	26	26
60-69	16	16
70 or older	9	8.9
Age at injury category
18-20	20	20
21-29	28	28
30-39	14	14
40-49	16	16
50-59	14	14
60-69	4	4
70 or older	4	4
Gender
Male	63	62
Female	38	38
Race/Ethnic group
White	84	84
African American	13	13
Native American/American Indian/Alaskan natives	1	1
Others	2	2
Marital status
Married	48	50
Never married	30	31
Divorced	6	6.2
Widowed	8	8.2
Living with partner/significance other	5	5.2
Education
9th grade or less	1	1
Grades 10-12 (some high school)	5	5
High school graduate/GED	20	20
Vocational/technical school	7	6.9
Some college/associate degree	30	30
College graduate	20	20
Graduate school or professional school	18	18
Residential location
Urban	17	18
Suburban	38	40
Rural	40	42
Employment at injury
Full-time	62	61
Part-time	9	8.9
Student	18	18
Unemployed (not student, not retired)	9	9
Retired	3	3
Employment after injury
Full-time	17	17
Part-time	11	11
Student	3	3
Unemployed (not student, not retired)	48	49
Retired	20	20
Level of Injury
Cervical 1 (C1)-Cervical 3 (C3)	10	9.9
C3-C7	36	36
Thoracic 1 (T1)-Thoracic 5 (T5)	12	12
T6-T12	38	38
Lumbar region	5	5
SCI-related complications
Pressure ulcer	31	31
Low blood pressure	15	15
Depression	41	41
Infections	37	37
Bowel and bladder problems	64	63
Sexual dysfunction	38	38
Pain	65	64
Spasticity	56	55
Respiratory problems	6	5.9
Osteoporosis	17	17
Diabetes	1	1
Years since injury
1-3 years	12	12
4-6 years	22	22
7-10 years	20	20
11-15 years	15	15
16 years and over	31	31

Table 2: Dynamic and static light scattering data for PIC micelles in 0.1 M NaCl.

 

Count
		Gender		Total
		Male	Female
Emp at injury	Employed full-time (more than 30 hours a week)	39	23	62
	Employed part-time (less than 30 hrs a week)	7	2	9
	Volunteer employment	0	1	1
	Unemployed, seeking employment	3	0	3
	Unemployed, not seeking employment	2	2	4
	Student	10	7	17
	Homemaker (manage home/family)	0	1	1
	Retired	1	2	3

Table 3: Gender and employment status before SCI.

 

			Employ Yes 1 no 0		Total
			No	Yes
Gender	Male	Count	44	18	62
		% within gender	71.00%	29.00%	100.00%
		% within employ yes 1 no 0	60.30%	66.70%	62.00%
		% of total	44.00%	18.00%	62.00%
	Female	Count	29	9	38
		% within gender	76.30%	23.70%	100.00%
		% within employ yes 1 no 0	39.70%	33.30%	38.00%
		% of total	29.00%	9.00%	38.00%
Total		Count	73	27	100
		% within gender	73.00%	27.00%	100.00%
		% within employ yes 1 no 0	100.00%	100.00%	100.00%
		% of total	73.00%	27.00%	100.00%

Table 4: Gender and employment status after SCI.

Gender * employ-after-injury: Cross tabulation

Emp-at-Injury * Gender cross tabulation

Table 5 represents the Spearman’s Rank Order inter-correlations of the primary survey data. The correlation analysis was performed to assess the ability of the primary survey items to reproduce the proposed study model. Several of the study variables exhibited high correlation. High correlation existed between the productivity subscale (CIQPD) and being employed full-time (r = 0.61, p = 0.01). Wheelchair use exhibited high correlation with level of injury as well (r = 0.37, p = 0.01).

 

Measure	1	2	3	4	5	6	7	8	9	10
LI
WC	0.372**
Emp-FT	-0.038	0.039
DCS	-0.161	-0.033	0.313**
Comp	0.048	-0.084	-0.201*	-0.504**
SAQ	0.039	0.048	0.171	0.499**	-0.193
RSR	-0.057	0.174	0.217*	0.265*	0.014	0.241**
CIQH	0.049	0.183	0.207*	0.200*	-0.246*	0.134	-0.024
CIQS	-0.007	-0.039	-0.057	-0.224**	0.172	-0.194	-0.212*	-0.19
CIQPD	0.071	0.115	.611**	0.333**	-0.251*	0.161	0.168	0.301**	-0.134
CIQ Total	0.09	0.189	0.396**	0.217*	-0.212*	0.088	-0.037	0.739**	264**	0.659**

Table 5: Spearman rank order coefficients for primary questionnaire data.

Note: LI = Level of Injury, WC = Wheelchair Use, Emp-FT = Employed Full-Time, DCS = Disability Centrality Scale, Comp = SCI-related Complications, SAQ = Self-advocacy Questionnaire, RRSS = Received Rehabilitation Services Scale, CIQH = Community Integration-home integration, CIQS = Community Integration-social integration, CIQPD = Community Integration Questionnaire-productivity, CIQ Total = Community Integration Questionnaire Total Scores

**p = 0.01, *p = 0.05

The researchers explored the reliability of the survey instruments used in the study. Table 6 represents the reliability information of the study instruments.

Variable	No of Items	Cronbach’s Alpha
SAQ	11	0.88
RRSS	9	0.84
CIQ	14	0.7
DCS	10	0.89

Table 6: Cronbach’s alpha scores for instrument scales.

Note: SAQ = Self-Advocacy Questionnaire, RRSS = Received Rehabilitation Services Scale, CIQ = Community Integration Questionnaire, DCS = Disability Centrality Scale (Satisfaction mean)

Research question 1: What associations exist between SCI-related variables (level of injury, SCI-complications, use of wheelchair) and rehabilitation-related variables (received rehabilitation services, self-advocacy, and psychosocial adaptation) and employment among persons with SCI?

Research question 1 was investigated using path analysis. The researchers hypothesized that all the selected variables; level of injury, SCI-complications, use of wheelchair, received rehabilitation services, self-advocacy, and psychosocial adaptation will have direct association with employment after SCI. The results of several re-specifications of the hypothesized associations are represented in figure 1. After the re-specification of this model, the chi square test became nonsignificant (?2 (5) = 4.597, p = 0.467). The CFI = 1.000; TLI = 1.022, and RMSEA = 0.000, indicating that this model represents a good fit to the data. The re-specified model is shown in figure 1.

In order to address research question #2, data were analyzed using multiple regression analysis with stepwise method of entry for predictive variables. The criterion variable was employment and the predictor variable was the demographic information (age, education, gender, ethnicity, marital status, years since injury and residential location). Of the 7 predictor variables only education and age in years predicted employment. The linear combination of predictor variables (age and education) accounted for approximately 9% of the variance in employment; adjusted R2 = 0.094; F (2, 87) = 5.63, p = 0.005. The results show that education significantly predicted employment (β = 0.288, p = 0.005), as did age (β = 0.221, p<0.005).

Research question 1 focused on exploring the associations between multiple factors of SCI-related characteristics, rehabilitation service-related characteristics, and employment. The findings of this study suggest significant positive associations between level of injury and wheelchair use; received rehabilitation services and self-advocacy; self-advocacy and psychosocial adaptation; and psychosocial adaptation and employment (as depicted in figure 1).

A surprising result of this study is the lack of association between level of injury and employment. A review of the literature shows that several past studies had inconsistency in the reported relationships between severity of injury and employment. In some studies, it was reported that employment is not dependent on severity of injury [24,25], while in others contrary results were reported [26,27]. Since there is no consensus in the literature on the relationship between level of injury and employment, it is possible that there are some other unidentified variables that either mediate or moderate this relationship in certain situations. In this current study level of injury had a strong effect on wheelchair use, although this study did not explore details about this relationship such as whether this relationship differs between users of motorized wheelchairs and manual wheelchairs. 

Strong positive associations are noted between received rehabilitation services and self-advocacy; between self-advocacy and psychosocial adaptation; and between psychosocial adaptation and employment. In the past, rehabilitation researchers have suggested that psychosocial adaptation to disability and chronic illness has a relationship with Quality of Life (QOL) [28-30]. In a study Livneh and Martz [31] posit that coping strategies and coping resources increase psychosocial adaptation to SCI. The findings of this current study aligns with previous literature. Particularly, the findings of our study suggest that self-advocacy (a form of coping strategy) has a direct association to psychosocial adaptation. This study has shown that since psychosocial adaptation correlates positively with both self-advocacy and employment; psychosocial adaptation can directly or indirectly increase community participation and hence QOL of persons with SCI. In other words, an individual with increased self-advocacy skills may adapts better to psychosocial impacts of SCI. Increased psychosocial adaptation will more likely lead to employment and employment retention, and hence increase in QOL.

This study suggests a strong positive association between received rehabilitation services and self-advocacy (see Figure 1). Received rehabilitation services also show minimal positive associations with wheelchair use, psychosocial adaptation, and employment. These minimal associations should be interpreted with caution due to small sample size and the self-report nature of the study. These associations underscore the importance of rehabilitation during inpatient and outpatient phases of SCI to employment post SCI. Unfortunately, very little is known in the research literature on comprehensive rehabilitations needs of individuals with SCI or the impacts of rehabilitation services on community participation such as employment of persons with SCI. According to Gutenbrunner, Blumenthal, Geng, and Egen [32] individuals with SCI need long term rehabilitation services that is interdisciplinary in nature. Such rehabilitation services include services that address community re-entry issues such as assistive technology, mental health issues, vocational rehabilitation, education, adaptation to living environment, and others. The results of this current study suggest that rehabilitation professionals who work with persons with SCI need to consider a wide range of variables in both acute and post-acute phases of rehabilitation in order to promote community participation such as employment. Exploring the relationships between this study’s variables (level of injury, SCI-related complications, and wheelchair use, self-advocacy, psychosocial adaptation, employment, received rehabilitation services, and community reintegration) in rehabilitation process can facilitate and increase positive rehabilitation outcomes. The results of this study can be valuable in addressing the psychosocial concerns of persons with SCI by rehabilitation counselors and other rehabilitation professionals. Areas of to focus on can include case management, self-advocacy training, vocational service, and adjustment counseling.

This study explored a second research question that examined the demographic information that may predict employment among persons with SCI. The results of multiple regression analysis (stepwise method) show that demographic variables of age and education were found to be significant predictors of employment. Employment is regarded as a measure of successful rehabilitation and community reintegration after SCI, but employment can be impacted by multiple factors [33], such as level of education. In this study, age and education accounted for 9% of the variance in employment with older age being associated with having a college degree and being employed full-time. This finding supports prior research that has suggested that age at time of injury plays a role in vocational functioning, and that persons who sustained SCI at a younger age have a greater probability of securing employment [4,34,35]. According to Crewe [36] young adults generally have been thought to have more adaptability and flexibility to change. In a 2015 study, Craig, Perry, Guest, Tran, and Middleton [9] posit that the demographic factor of age predicted social participation (employment is a type of social participation), although surprisingly, education did not predict social participation in their study. 

Furthermore, several prior studies have found a wide range of demographic variables to predict employment status. These variables include age, level of education, gender, severity of injury, and marital status [37-39]. With the exception of age and education, these demographic variables were not predictive of employment in this current study. James, DeVivo, and Richards [40] found a relationship between race and gender in employment; their study reported that White women were less likely than White men to be employed, whereas African American women were more likely than African American men to be employed. The inability of other demographic information to predict employment in this study could be the result of the small sample size (n = 101). Education is a key determinant of economic independence and quality of life. According to NSCISC (2016), “over half of persons with SCI are high school graduates (51.5%) at time of their injury” (p. 2). The report of the NSCISC is consistent with the finding in this study; 59.5% of the participants of this study have at least a high school diploma at the time of injury. According to Bryce et al., [41] “education has been found to be the factor most strongly associated with post injury employment. Only 5% of persons in the National SCI Database with less than 12 years of education were found to be employed” (p. 1315) compared to 68.8% among individual with SCI who had doctoral degrees.

The findings of this study should be interpreted with caution for a couple of reasons. First, the descriptive and correlational nature of the study design through the use of self-report (questionnaires) may lead to self-report errors and bias. Due to the descriptive and correlational nature of this study, no definitive causal attribute can be obtained from the relationships between the independent variables and employment. Secondly, the recruitment of participants through convenience (purposive) sampling instead of random sampling further limits the generalization of the results.

SCI is a condition that brings multiple adverse implications that affect the physical, functional and psychosocial aspects of a person’s life; and requires interventions in multiple areas. The results of this study suggest that psychosocial adaptation (strongest predictor), age, and education are strong predictors of employment in a sample of adults with SCI. In addition, received rehabilitation services increase both self-advocacy and psychosocial adaptation. This study has implications for both SCI-related clinical practice and research include the need to examine the impacts of self-advocacy and psychosocial adaptation on rehabilitation outcomes. Issues of self-advocacy among individuals with SCI have not been given adequate attention. Self-advocacy could lead to improved psychosocial adaptation to the impacts of SCI. Understanding the impact of self-advocacy on psychosocial adaptation and community participation such as employment may provide insights into areas where skill development for individuals with SCI may prove beneficial. Self-advocacy skills among individuals with SCI are crucial particularly in areas such as employment, acquiring assistive technology, and closing gaps of health disparities.