Bjornson, Kristie PT, PhD, PCS; Kobayashi, Ana MPH; Zhou, Chuan PhD; Walker, William MD
INTRODUCTION AND PURPOSE
Transitioning from adolescence to adulthood presents numerous challenges for persons with physical disabilities (ie, cerebral palsy [CP], spinal cord injury, or spina bifida [SB]). Participation in adult life may include appropriate paid employment and/or postsecondary education after high school. US federal mandates require appropriate education for children with physical limitations,1,2 yet it is unclear which specific factors and services are associated with paid employment and/or participation in postsecondary education among young adults with physical disabilities in the United States. As many adolescents with physical disabilities receive physical therapy (PT) and/or occupational therapy (OT), it is important to understand the relationship between the therapy given during secondary education and entry into postsecondary education and/or paid employment.
Adults with physical disabilities are significantly less likely to have paid employment than those without disability,3,4 with educational level reported as an important factor in subsequent employment for persons with physical limitations.5–7 A Japanese study in 2000 noted that among adults with CP, mobility and regular education experience were associated with paid employment.7 Murphy et al6 reported that postsecondary education was the strongest factor predicting competitive paid employment for adults with CP, particularly among those with the most severe motor limitations. In their study, 12 of 38 adults with CP and quadriparesis were competitively employed and had completed some college. Similarly, Fowler and colleagues5 reported that educational level was an important factor in subsequent employment for adults with physical limitations due to progressive neuromuscular disorders (ie, spinal muscle atrophy, muscular dystrophy).
In a recent study of 81 young adults aged 18 to 21 years with mild CP and normal cognition, the duration of transition to paid employment was significantly longer than that for their peers who were able-bodied.8 The authors found that a higher attained level of education was associated with longer transition times between high school and paid employment. The authors suggest that the observed delay may be due to youth attending postsecondary education after high school instead of entering the workforce or due to a higher level of education required to obtain employment.
Historically, persons with CP have had low levels of competitive employment. From 1959 to 1975,9–15 competitive employment rates, including skilled and unskilled types of work, were reported as approximately 17% to 35% among adults with CP. In contrast, in 2000 Murphy and colleagues6 documented employment rates of 53% among 101 adults aged 27 to 74 years with moderate to severe CP. Yet, similar to the employment levels of more than 35 years ago, Michelsen et al16 reported that only 33% of adults with CP were competitively employed versus 82% of peers who were nonimpaired. Persons with SB appear to have similar levels of employment, with only 30% of young adults with SB in a 2008 Dutch study17 having full-time paid employment. Male gender, educational level, and self-care independence were reported to be significant predictors of full-time paid employment in the model of van Mechelen et al.17
Employment and secondary education levels among all youth with disabilities have increased over the past 4 decades.18 These increases have been attributed to enhanced home support, advances in rehabilitation technology, and the federal mandates related to education for children and youth with disabilities.6 Expanded and signed in 2004, PL 108-446 Individuals with Disabilities Education Improvement Act has the specific purpose of ensuring that all children with disabilities have available free appropriate public education with specialized instruction and related services individually designed to “prepare them for further education, employment and independent living.”19
“Transition services” within the school environment under PL 108-446 Individuals with Disabilities Education Improvement Act are activities individually designed to facilitate movement from school to postschool activities such as postsecondary education, vocational training, integrated employment, independent living, and community participation.19 This process may begin as early as age 14 years and a transition plan should be in place by age 16 years. Physical and occupational therapists should be involved in transition planning for children with physical disabilities and/or degenerative disorders of the neuromuscular system.20 The needs of a youth with physical disabilities in the postschool setting can be different from the relatively protected public school environment. Therapists can assist in evaluating and/or providing interventions to optimize participation in life after high school. Such interventions may address independence in activities of daily living (ADL), mobility, and access to technology/adaptive equipment to allow independence in the educational and community environment. Along with completion of high school and/or postsecondary education, these factor have the potential to enhance levels of social interaction and life participation for adolescents with physical disabilities and have been documented to be related to levels of employment.7,17 Thus, a better understanding of how therapy received during the secondary education years relates to participation in postsecondary education and/or paid employment is warranted.
The purpose of this study was to assess the levels of participation in postsecondary education and paid employment among young adults with physical disabilities and to examine the relationship of PT and/or OT services during their secondary education years to levels of participation in postsecondary education and paid employment. For this analysis, physical disability was defined as an educational disability categorization of “orthopedic impairment” including CP and SB. We hypothesized that among young adults with physical disabilities, school- and community-based PT and OT provided between ages 13 and 16 years would be positively associated with both enrollment in postsecondary education and attainment of paid employment. Social skills and activities, independence in ADL, and mobility would also be positively associated with both postsecondary education and paid employment.
This study was a secondary analysis of data from the National Longitudinal Transition Study 2 (NLTS2). Funded by the US Department of Education, NLTS2 was a 10-year longitudinal cohort study that follows a large, nationally representative sample of students who at entry into the study received special education services, were aged 13 to 16 years, and were in seventh grade or higher on December 1, 2000.21,22 Data were collected from a variety of sources, including parents and guardians, teachers, principals, school records, and the students themselves.
Information about youth was collected prospectively in 5 waves as they transitioned from adolescence in secondary school to early adulthood. This current analysis used information collected in wave 1 (2000–2001) and wave 4 (2006–2007). Specifically, we queried the wave 1 parent and wave 4 parent/youth interview data to describe the participants' disability characteristics (ie, orthopedic impairment, CP, SB), individual demographics, school experiences, OT and/or PT services inside and outside school, employment outcomes, participation in secondary education, and school and household characteristics. This analysis includes OT/PT services delivered through school as well as individual school characteristics as measured in the wave 1 School Program Survey (Section D–Educational Services).
At entry, the study sample included 1510 youth and their guardians who had completed wave 1 data collection. Participants had been categorized with “orthopedic impairment” as defined by special education diagnostic categories or whose parents reported a diagnosis of CP or SB. The NLST2 employed a complex survey sampling method, using local educational areas as the primary sampling units, and then stratified these educational areas by geographic region, enrollment, and student wealth. A consequence of complex survey sampling is that some groups of subjects may be oversampled whereas other groups may be undersampled. Sampling weights are then needed to correct for imperfections in the sample that might lead to bias and other departures between the sample and the reference population. Thus, the purposes of sampling weights are to compensate for unequal probabilities of selection and nonresponders and to adjust the weighted sample distribution for key variables of interest (eg, age, race, and sex) to make it conform to a known population distribution.
For our analysis, we used the individual youth as the unit of analysis and applied sampling weights to all analyses that correspond to the complex sampling design, thus ensuring that our findings are generalizable at the population level. Without proper sample weighting, our findings would have been restricted to this subgroup only and would not provide inference about other youth with physical disabilities (ie, orthopedic impairment, CP, SB) in the general population. However, with proper sample weighting, such generalization can be achieved.
For this analysis, we examined the data from the 990 youth and parents/guardians who had completed both waves 1 and 4 data collection and were categorized as having orthopedic impairment, CP or SB. The sample included youth with CP (52%) and SB (9%) (see Table 1, subgroup estimates). At wave 1, the average age was 15.1 years (range, 13–17), and at wave 4, it was 21.1 years (range, 19–23). Thirty percent came from families in which the head of household had completed high school. School-based and non–school-based OT and/or PT were provided to 34% of participants at wave 1. Data were collected through a combination of parent, teacher, and school surveys in wave 1 and with parent, youth surveys, and/or telephone interviews in wave 4.
All analyses were weighted with sampling weights provided by NLTS2 data to reflect the complex survey sampling design. We conducted descriptive analysis and multivariable logistic regression analysis, using SPSS 17.0 Complex Samples (SPSS, Chicago, Illinois). “Any type of postsecondary education” and “Worked for pay in last 2 years or currently has a paying job” at age 19 to 21 years were considered the dependent variables for multivariable logistic regression with “received PT or OT in the last 12 months” at age 13 to 16 years as the main predictor of interest. A multidisciplinary research team that included a physical therapist, developmental pediatrician, public health researcher, and a doctoral-level biostatistician reviewed all potential covariates within the NLST2 data set to develop the initial set of predictors for analysis. Demographics, socioeconomic status (SES), physical and mental characteristics, school, and environment factors were considered potential predictors. The variables of age, gender, ethnicity, free/reduced meal status, education level of the head of household, and school wealth were entered as proxy variables for SES as they all were potential confounders for the receiving PT/OT services as well as the outcomes of interest (ie, paid employment and postsecondary education). Expressive and receptive language variables were included as proxies for cognition. The variables of “any social interaction,” “level of overall social skills,” and “social activities” were proxies for psychosocial characteristics. Activities of daily living, hand use, mobility, wheelchair use, and the presence of a feeding tube were employed to characterize the motor function of the participants. Only covariates that were either found to be significant in univariate analysis or a priori considered scientifically related to the primary research questions, regardless of significance in the univariate analysis, were retained for the multivariate analysis model development. The primary research question was: “What is the association of PT/OT services received to levels of enrollment in postsecondary education and attaining paid employment?”
The multivariate model to test the relationship of therapy services to postsecondary education included the covariates of gender, age, education level of the head of household, health insurance status, ability to use both arms and hands normally to hold a pencil or spoon, self-care skills, overall social skills, expressive language skills, wheelchair mobility, and whether the child received a high school diploma or a certificate. The self-care variable was the sum of a dressing and a feeding variable, both coded on a 10-point Likert scale (1 = total dependence, 10 = independent). Expressive language was an average of a speaking and a communicating variable on a 4-point Likert scale (1 = speaks/communicates as clearly as other children/no trouble communicating to 4 = does not speak or communicate at all). The overall social skills variable was a sum of 3 separate variables querying social assertion, self-control, and cooperation. The multivariate model for paid employment included gender, ethnicity, age, any social interactions, parental education, expressive language, wheelchair mobility, any health insurance, received a high school diploma/certificate, overall social skills, and percent of students attending schools where more than 50% qualified for reduced or free lunch (as a proxy for community SES). The social interactions variable was dichotomous and reported whether the youth had or did not have any social interactions (ie, gets together with friends, invited by friends to social activities, and interacts with others via e-mail/chatrooms or phone).
At the wave 4 follow-up, 48% of participants reported participation in some form of postsecondary education in the past 12 months; only 25% were attending postsecondary education on the day of the survey (see Figure 1). Of this group, 46% were enrolled in a 2-year or community college, 25% a 4-year institution, and 14% in a vocational/technical school. Forty-three percent of participants in wave 4 reported current or prior paid employment in the past 2 years; only 20% were employed on the day of survey completion (see Figure 1). Of the remaining 57% of participants, 26% were presently employed and in postsecondary education while 29% were not engaged in any employment, postsecondary education, and/or job training activities.
The results of the logistic regression models examining the relationship of therapy services to levels of postsecondary education and paid employment are presented in Tables 2 to 4. Youth who received OT and/or PT services during their secondary education years were 3.2 times more likely to attend postsecondary education than those who did not receive any PT/OT services (OR = 3.2, P = .03; Table 2). Among those who received therapy, youth having greater independence in self-care skills and the ability to use both their arms and hands to hold a spoon or pencil or who had greater independence in self-care skills were more likely to go onto post–high school education (OR = 4.8, P = .001, and OR = 1.3, P = .014, respectively). Stronger social skills, acquiring a high school diploma/certificate, and higher parental education were associated with significantly increased likelihood of attending postsecondary education among youth who received therapy (OR = 3.2–1.1, P = .03 to <.001).
Receiving therapy services was not significantly associated with paid employment at follow-up (OR = 0.44, P = .09), while adjusting for other covariates. Among respondents in wave 4, those who received therapy and had social interactions or had received a high school diploma had significantly increased odds (OR = 12.2 and 4.2, respectively) of reporting paid employment at the wave 4 follow-up. Among youth who received therapy, those of nonwhite race (OR = 0.5), with limitation in expressive language (OR = 0.5), or used wheelchairs (OR = 0.06) had significantly decreased odds of reporting paid employment. To explore the paid employment outcomes of participants who no longer qualify for special education services (age 21 years or older), model development was replicated with all wave 4 participants aged 21 years or older (n = 650) (see Table 4). Occupational/physical therapy services did not significantly increase the odds of reported paid employment for participants older than 21 years. In contrast, among youth who received therapy, the use of a wheelchair and strong expressive language skills were significantly associated with an increased likelihood of paid employment in this group (OR = 51.4 and 5.0, respectively). Nonwhite young adults, nonreceipt of a high school diploma/certificate, and lower receptive language skills were significantly associated with lowered odds of paid employment.
The International Classification of Function, Disability and Health (Figure 2) framework facilitates discussion and investigation of health status across the components of body structure/function, activity, and participation as well as environmental and personal factors.23 Activity and participation are equally important during adulthood and childhood. Successful participation for young adults with orthopedic impairments may include the goals of independent living, paid employment and for some postsecondary education as preparation for paid employment.
Physical and occupation therapy services received at age 13 to 16 years were significantly associated with enrollment in postsecondary education at ages 19 to 21 years among young adults with orthopedic impairments, CP or SB controlling for gender, age, upper extremity function, parental education, high school graduation status, and social skills. School- and/or community-based PT and OT services for children with physical impairments appear to be an important factor in attending postsecondary education. These results suggest that therapy services that focus on interventions to improve upper extremity function, self-care skills, and social skills could lead to increased levels of participation in postsecondary education. People who have better upper extremity function, self-care skills, and social skills are more likely to be able to care for themselves and complete academic tasks (take written examinations and notes, open doors, operate computers) than their peers who are less independent.24–26 Such skills are particularly important for success in academic environments, and their improvement may translate to improved ease in navigating academic environments.
In contrast, PT and OT services received at age 13 to 16 years were not significantly associated with paid employment at age 19 to 21 years among young adults with orthopedic impairments (Table 3). Evaluation of and opportunities for individualized instruction to enhance social interaction and expressive language appears indicated for children with orthopedic impairments to promote their participation in future paid employment. Our findings also suggest that specialized instruction within the educational setting that focuses on improving social skills and attaining a high school diploma may be related to an increased chance of obtaining paid employment. For participants no longer eligible for special education services (ie, older than 21 years), PT and OT services were not significantly associated with reported levels of paid employment. Such results may be due to the small sample size at wave 4 follow-up, or it truly does take longer for this population to find paid employment. The association of therapy services to reported levels of paid employment may be better addressed in the wave 5 data after all participants have completed secondary education and/or extended educational support services past age 18 years. Young adults using wheelchairs for mobility and those with higher levels of expressive language were significantly more likely to be engaged in paid employment (Table 4) among participants older than 21 years. The odds ratio for wheelchair use was extremely large with a wide confidence interval (Table 4), which may be due to the small number of subjects older than 21 years using wheelchairs. Wheelchair use may be a marker of increased independence in life. Alternately, it is also possible that individuals who use wheelchairs are more impaired and are more likely to be in supportive work environments. The “any paid employment” variable may also influence the regression outcomes, as the types of “work” activities can be quite varied. For example, clerical/office type work activities may be associated with receiving therapy services but are possibly overshadowed in the analysis by the individuals who are employed in work activities that require lower-level skills (ie, “greeters” at a store or movie theater ticket takers).
The 2009 NLST2 summary report (waves 1 and 3 data) of post–high school outcomes in youth with disabilities up to 4 years after high school noted that in the general population (youth without disabilities), about half (51%) of youth have enrolled in college in the past 2 years as compared with 42% of all youth with disabilities.21 Thus, youth with orthopedic impairments are reporting similar levels of postsecondary education as all youth with disabilities.21 Similar to the 2009 summary report, approximately half of our study sample was participating in some type of postsecondary education; the majority was in 2-year or community colleges. Fewer youth with orthopedic impairments reported paid employment at wave 4 compared to all youth with any disability at wave 3.
As a secondary data analysis, this project was limited in its ability to directly answer research questions as the specific variables, definitions, and/or coding of variables as well as the level of information collected were determined prior to development of our research questions. Thus, many of the important constructs were not directly measured and we had to develop proxy measures using available data (ie, the use of a wheelchair was a proxy variable for walking ability). Yet, the analysis of such a large national representative sample has not been described in the pediatric therapy literature. It is also possible that young adults with greater physical limitations were more likely to participate in this project through parental report (vs self-report of less involved participants). There is a lack of specific information in the NLTS2 data set related to disease/disability severity and/or etiology. The intensity and/or specific type of PT/OT services provided to the participants is not available. In addition, these analyses lack information about services and/or education experiences prior to age 13 years. The outcome of paid employment may require further exploration with wave 5 data when participants are older. Participation in some type of postsecondary education/job training after high school may have also resulted in an underrepresentation of paid employment in this wave 4 analysis.
What implications does this work have for the pediatric therapist? These results suggest that the abilities and skills observed and encouraged during adolescence, such as mobility, social skills, and ADL, may directly and/or indirectly affect the levels of participation in future educational and/or work experience as adults. Clearly, therapists do and should focus on developmentally appropriate issues of mobility and ADL early in a child's life. Literature supports the use of group-based task-related training, prolonged standing, and intensive treadmill training to effect walking and gross motor skills in young children with CP.27–29 In 2008, Blank and colleagues30 documented that intensive practice within the context of a conductive education intervention trial enhanced hand coordination skills with parallel improvement in ADL in 64 elementary school-aged children with CP. Salem and Godwin31 reported significant improvements in standing and walk/run/jump skills as well as speed of walking with task-oriented strength training in ambulatory children with CP. Practice and refinement of mobility, self-care, and social activity skills within the school environment/program may be enhanced through PT and/or OT services, adaptive physical education, and/or extracurricular activities and services.20,32,33 With family-centered intervention services structured as activity-based instruction in a child's natural environment,34 practice and support using skills during early childhood through adolescence may affect how these skills are used as an adult.35
First, it is important to replicate these analyses with wave 5 data, when all participants have finished receiving special education services. Future work should also focus on prospective, longitudinal, student-reported outcome studies across severity of impairment and diagnoses to better understand the factors that support the goals of postsecondary education and/or paid employment for persons with orthopedic impairments in young adulthood. Research should explore specific therapy modalities (forced use, treadmill training, etc), frequency of service, and/or model of intervention (ie, school-based consultation vs direct service). Prospective studies are needed to clarify the specific elements of OT/PT services that improve arm/hand use, improve social skills, and increase support to parents/caregivers for their son's or daughter's participation in postsecondary education. For the paid employment in young adulthood, specific interventions to optimize social interaction, mobility, and expressive language should be tested and compared.
Activities of daily living, expressive language skills, acquisition of a high school diploma/GED, parental education level, social skills, normal hand/arm use, as well as receiving PT/OT services during secondary education years appear to enhance the likelihood of postsecondary education for young adults with orthopedic impairment. Only the individual's experience with social interactions, level of social skills, and/or acquisition of a high school diploma (GED) enhanced the likelihood of paid employment. Several of these factors can be directly addressed by pediatric physical, occupational, and/or speech therapists within or outside the educational environment.20,33 All previous studies in this population were limited by small, local samples of children or youth. This analysis of a large national sample provides the first opportunity to generalize these conclusions and recommendations on a national level. Understanding the factors that enhance postsecondary education and paid employment in young adults with orthopedic impairment has the potential to inform future intervention research as well as the practice of pediatric PT in the educational environment. Such knowledge can provide evidence-based guidance for policy and practice issues such as qualification for services, the most appropriate focus for therapy services provided within the secondary educational environment, and appropriate postsecondary transition planning for youth with physical disability in the United States.