Psychosocial and Health Outcomes of Adults With Violently Acquired Pediatric Spinal Cord Injury

This was a cross-sectional study that involved participants from an ongoing longitudinal investigation evaluating the psychosocial and medical outcomes of adults with pediatric-onset SCI. Each participant was cared for at 1 of 3 pediatric SCI programs at Shriners Hospitals for Children located in Chicago, Philadelphia, or Sacramento. Participants were recruited from around the country. The study included participants who were at least 19 years of age, had sustained an SCI between the ages of 0 and 18 years, had no history of significant brain injury, and were able to speak English. Annual interviews were completed in person or over the telephone; data for the current study were from the most recent interview. Informed consent was attained prior to interviews. This study was approved by the Western Institutional Review Board, and all applicable institutional and governmental regulations concerning the ethical use of human participants were followed.

Injury-related information was extracted from Shriners Hospitals for Children medical charts and SCI database. Neurological level was classified using the American Spinal Injury Association Impairment Scale (AIS) according to the International Standards for the Neurological Classification of Spinal Cord Injury.21 Information about medical and demographic characteristics was obtained by interviewing participants. This included asking questions about employment, marital status, and education level as well the occurrence of secondary health complications within the previous year (eg, pressure injuries, urinary tract infections, and the presence of other chronic medical conditions). Participants were also asked to rate the frequency that pain interfered with their day-to-day activities. Responses to this question were then categorized into either “no activity interfering pain” or “activity interfering pain.”

A number of standardized measures were included in the interview to assess psychosocial outcomes. These measures were chosen due to their reliability and validity in previous SCI studies. Community participation and independence were assessed utilizing the Craig Handicap Assessment and Recording Technique (CHART).22 Responses to the CHART can be used to calculate scores in the domains of Physical Independence, Cognitive Independence, Mobility, Occupation, Social Integration, and Economic Self-Sufficiency. Scoring for the subscales is on a 100-point scale, with higher scores typically indicating greater independence. For the purposes of the current analyses, each CHART subscale was dichotomized into some evidence of impairment (scores <100) versus fully independent (scores of 100).

The Patients Health Questionnaire (PHQ-9)23 was used to measure the presence and severity of depression symptoms as defined in the DSM-IV.24 The PHQ-9 has 9 items, and total scores range from 0 to 27. In the current study, a score of 10 or more was classified as moderate to severe depression.

General life satisfaction was measured using the 5-item Satisfaction with Life Scale (SWLS).25 Each statement on the SWLS is rated on a scale that ranges from 7 (strongly agree) to 1 (strongly disagree); summed aggregate scores range from 5 to 35. A cutoff score of 25 or more was used to categorize respondents into a highly satisfied group.

Descriptive statistics were utilized to summarize participant demographic and injury characteristics. Next, analyses to see whether participant characteristics varied as a function of injury etiology (violent injury vs nonviolent injury) were completed. This included Mann-Whitney U tests to compare groups on continuous variables (such as age) and chi-square analyses to compare groups on categorical variables (such as injury severity). Finally, to examine whether violent etiology was associated with participant outcomes, a series of logistic regressions were completed. In the logistic regression models, the psychosocial and health outcome variable of interest was included as the dependent variable and violent etiology was included as a predictor, while also controlling for injury level (dummy coded as paraplegia vs tetraplegia), age of injury, and race (dummy coded as Caucasian vs other ethnicity). Adjusted odds ratios (OR) along with 95% confidence intervals are reported. Statistical significance was based on a threshold of p < .05. Statistical analyses were conducted with IBM SPSS (IBM SPSS Statistics for Windows [computer program], Version 22.0; IBM Corp., Armonk, NY).

Of the 484 participants in the larger longitudinal study, 1 individual was excluded from the current study because injury etiology could not be classified. Of the remaining 483 participants, 62.7% (n = 303) were male, 84.9% (n = 410) were Caucasian, 53.4% (n = 258) had tetraplegia, and 67.9% (n = 327) had complete injuries. Neurological classification of injury was as follows: C1-4 AIS A, B, or C (n = 66; 13.7%); C5-C8 AIS A, B, or C (n = 166; 34.5%); T1-S5 AIS A, B, or C (n = 203; 42.2%); and AIS D (n = 46; 9.56%). Mean age of participants was 32.89 years (SD = 6.81; range, 19–51), mean age at injury was 14.25 years (SD = 4.40; range, 0–18), and average injury duration was 18.64 years (SD = 7.68; range, 1–45). Nearly half of participants reported being employed (n = 224; 46.4%), and a majority reported that they did not have a college degree or higher (n = 280; 58.0%).

Of the 483 participants, 8.7% (n = 42) sustained their SCI as a result of violence; the vast majority were gunshot wounds (n = 39; 92.9%) and a small minority was due to person-to-person contact (n = 3; 7.1%). Additional causes of injury included vehicular or pedestrian (n = 239; 49.5%), sports (n = 118; 24.4%), medical or surgical complications (n = 50; 10.4%), and nonviolent falls (n = 31; 6.4%). For the purposes of analyses, individuals were classified as having either a violent or nonviolent injury etiology.

Demographic and clinical characteristics of participation as a function of etiology type can be seen in Table 1. Compared to those with nonviolent injuries, individuals with violent injuries sustained their SCI at a significantly earlier age (p = .005). A higher proportion of individuals with violent injuries were non-Caucasian and had paraplegia. Additionally, those in the violent injury etiology group were significantly less likely to have a college or professional degree (OR, 0.40; 95% CI, 0.19–0.84, p = .012) and significantly more likely to have no health insurance or public insurance (OR, 3.46; 95% CI, 1.61–7.42, p = .001).

After adjusting for covariates, logistic regression analyses revealed significant differences between persons with violent and nonviolent injuries (Table 2). Those with violent injury etiology were 3.7 times more likely to have moderate to severe depression symptoms (p = .011) and were significantly less likely to be economically independent as assessed by CHART economic scores (p = .015). Injury etiology was not significantly associated with any other domain of participation assessed by the CHART or with life satisfaction. In terms of secondary health complications, those with violent injuries were 2 times more likely to report a pressure injury (p = .034) within the previous year and 2.3 times more likely to report the occurrence of pain that interfered with activities (p = .019).

The results of the current study reveal greater psychosocial health concerns and more secondary health complications for individuals in the violent injury group than for those in the nonviolent injury group. Specifically, those with violent injury etiology were at greater risk for moderate to severe depression symptoms and were also less likely to be economically independent. In terms of secondary health conditions, these patients experienced more pressure injuries within the past year and higher occurrences of pain that interfered with activities than individuals with nonviolent injury etiologies.

Several of the demographic and injury characteristic differences found in the current study are consistent with previous studies of violent injury in adult-onset SCI7 as well as other neurotrauma populations, like traumatic brain injury (TBI).26 As observed in previous research, individuals with violent injury were more likely to be a member of an ethnic minority group and have limited education.7,8,26 As a majority of the violent injuries in the current study were related to gunshot wounds, the higher rate of paraplegia among those with violent injuries was expected and is consistent with the pattern of neurological injury associated with gunshot wounds.27 

Somewhat surprisingly, the incidence of violent injury was not significantly different between girls and boys. This finding contradicts numerous studies showing that males are typically at significantly greater risk for violence-related neurological injuries than females.7,8,26–28 It is important to note that in the current sample there is still a larger percentage of males with violent injuries than females, but this difference was not large enough to reach statistical significance. This divergence from the existing research is likely due to the developmental timing of the injury. In late adolescence and adulthood, it may be the case that males are at significantly greater risk for violence-induced injuries; however, during childhood, the gender disparity in risk for violence-related injuries may not be as large.

With respect to CHART scores, individuals with violent injuries were not significantly different on most dimensions of participation and independence in the community, with one exception. Those participants with violent-related injuries were significantly less likely to report independence in the area of economic self-sufficiency, which reflects financial stability and the ability to sustain customary socioeconomic activity. This finding suggests that, even after controlling for age of injury, race, and injury level, individuals with violence-related injuries may be at greater risk for obtaining lower levels of economic independence. Although those in the violent injury group were not more likely to be unemployed, they were less likely to obtain a college degree. Taken together, this may suggest that those with violent injuries may be at greater risk for underemployment or work that pays a lower wage.

Of concern, individuals with violent injuries were at greater risk for moderate to severe depression symptom levels. There are several possible explanations for this finding. It may be the case that sustaining an injury due to violence results in greater traumatic stress, which in turn could make some people more vulnerable to mental health symptoms in adulthood. The higher rates of depression symptoms may also reflect the economic issues mentioned previously. This would fit with previous research showing that financial instability is associated with higher perceived stress and rates of depression.29 Despite this finding, however, it is interesting to note that injury etiology was unrelated to satisfaction with life scores, another measure of psychosocial well-being.

The same psychosocial and environmental variables that predispose children and adolescents to the risk of violent injury or gunshot wounds may also be related to later economic and psychosocial functioning. Among adolescents, behaviors such as substance use,30 skipping class,31 and previous arrests or encounters with the criminal justice system30,31 have been linked to increased risk of gunshot injuries. Economic concerns, including living in poverty, have also been associated with pediatric gunshot injuries and severe childhood injuries.32,33 Environmental and psychosocial issues such as living with a nonbiological parent, the experience of parental death, and decreased participation in school and extracurricular activities are all factors shown to increase the risk of gunshot wounds in adolescence and childhood.30 A similar profile is seen in the risk for SCI due to gunshots or violence in adult-onset injury.8,34 These same preexisting behavioral, economic, and environmental stressors may also explain why individuals with pediatric-onset SCI due to violent injury are at greater risk for depression and living below the poverty line as adults. Unfortunately, we do not have information available on preinjury functioning to definitively say that this is the case in our current sample. Regardless of the process that gives rise to the increased risk, the current finding adds to the existing literature suggesting that individuals with SCI secondary to violence or gunshot wounds may warrant greater attention and efforts at social service intervention.

In terms of secondary health conditions, there was a higher incidence of pressure injuries and pain among those with violence-related SCIs. Since it was also the case that those with violent injuries were less likely to be economically self-sufficient and more likely to have either public health insurance or no health insurance, this may reflect an issue regarding access to care. It logically follows that inadequate access to quality health care would make it more difficult to obtain the types of preventative care that would reduce susceptibility to medical complications such as pressure injuries and pain.7 

There are some limitations to the current study worth acknowledging. First, our small sample size of individuals with violent injury etiology and our recruitment of participants who only spoke English and who were all from within the same hospital system may limit our application of these results to the broader population of those with SCI. Future research could improve on the current study by increasing the sample size of those with violent injuries and recruiting participants from a variety of treatment facilities and locations. An additional limitation of the current study was the cross-sectional approach, which restricts our ability to make conclusive statements about the causal relationships between violence, secondary health complications, and psychosocial outcomes. For this reason, prospective longitudinal studies should be done in the future to better understand the predictors and consequences of SCI due to violence within this population, including examining factors that may be associated with violent injuries, such as posttraumatic stress. Future studies may also consider exploring preexisting factors, such as poverty, family support, and community violence, and the effect they have on the prevalence of violent injuries leading to SCI.

This research provides evidence that violent injury etiology in pediatric-onset SCI may be a marker of risk and a potential predictor of less favorable outcomes compared to nonviolent injuries. In analyzing multiple indices of well-being, we found that violent injury was a significant predictor of economic instability and moderate to severe depression symptoms. We also found that violent injury was a significant predictor of secondary health conditions such as pressure injuries and pain. These findings warrant future research to better understand how injury etiology in pediatric-onset SCI might relate to outcomes in adulthood and to determine whether children who sustain violent injuries may warrant additional attention and services to promote stability and long-term resilience.