 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
a Department of Pediatrics and Intermountain Injury Control Research Center, University of Utah, Salt Lake City, Utah
b Departments of Social Medicine
c Pediatrics
d University of North Carolina Injury Prevention Research Center, University of North Carolina, Chapel Hill, North Carolina
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
METHODS. All North Carolina-resident children who were hospitalized between January 2000 and December 2001 in any of the state's 9 PICUs and survived a TBI that occurred on or before their second birthday were eligible to participate in the prospective cohort study. Child health status, use of ancillary medical resources, and family characteristics were assessed through maternal caregiver interviews 
2 years after injury. Comparisons were made among injury types (inflicted versus noninflicted).
RESULTS. There were 112 children who survived a TBI during the 2-year study period. Fifty-seven (79%) of the 72 maternal caregivers who had completed an interview 1 year after the child's injury participated in the year 2 interview. Most children (67%) had an outcome of mild disability or better at year 2, with 45% functioning at an age-appropriate level. Children's outcomes did not differ significantly at year 2 according to the mechanism of injury. The majority (67%) of children retained their Pediatric Overall Performance Category scores from year 1 to year 2. Children who changed were as likely to show improvement as deterioration. Children differed very little across time, as measured with the Stein-Jessup Functional Status II (Revised) scale. Families tended to have multiple environmental factors that could put their children at risk for poor developmental outcomes, including living below the poverty level (22%) and low social capital (39%).
CONCLUSIONS. The children in this cohort had relatively stable functional outcomes from year 1 to year 2 after injury. This population of children remains very vulnerable to poor developmental outcomes secondary to the effects of their TBI and environmental factors.
Key Words: traumatic brain injury • outcomes • child abuse • shaken baby syndrome
Abbreviations: CI—confidence interval • FSII(R)—Functional Status II (Revised) • GCS—Glasgow Coma Scale • IQR—interquartile range • POPC—Pediatric Overall Performance Category • RR—relative risk • TBI—traumatic brain injury
The period from infancy to the toddler stage is a time of rapid social, cognitive, and behavioral development. As a result, parents of young children look forward to watching the developmental stages of their infants as they become toddlers. Children who sustain a traumatic brain injury (TBI) may not follow the normal developmental trajectory, although they may make developmental gains.1 How children develop over time after early brain injury is not well studied, although it is thought that rapid gains in development plateau 6 months after injury.2
We monitored, for 2 years after the injury event, a cohort of children who sustained either inflicted or noninflicted TBI. In this report, we examine the functional outcomes of children at 2 years after discharge from the hospital and assess any changes in the children's functional outcomes from 1 to 2 years after discharge from the hospital. We hypothesized that deficits might become more apparent to maternal caregivers at 2 years after injury than at 1 year after injury if children failed to follow a normal developmental trajectory. We characterized the ongoing need for services that children demonstrated between the first and second years after hospital discharge, as well as family characteristics that might affect the meeting of those needs.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
45 minutes.
Children's Outcomes
Children's outcomes were assessed with 3 separate instruments. The first was the Pediatric Overall Performance Category (POPC) scale, which was used at hospital discharge, posttrauma year 1, and posttrauma year 2. This scale is a composite measure of functional morbidity and cognitive outcome.4 The POPC scale has been related significantly to the Bayley Psychomotor Development Index and the Vineland Adaptive Behavior Scales.5 Children's outcomes are ranked from 1 to 6. A score of 1 indicates that the child is healthy, alert, and capable of age-appropriate activities. A score of 6 indicates death. Because of the difficulty in detecting mild delays among very young children, for some analyses we categorized the POPC results into good and poor outcomes. Children were considered to have good outcomes if they scored 1 or 2 on the POPC scale, consistent with mild disability at most. Children were considered to have poor outcomes if they scored 3 or 4 on the POPC scale, consistent with moderate to severe disability. There were no survivors in a vegetative state (POPC score of 5). The POPC score was obtained through direct questions about the child's ability to perform age-appropriate activities, use of rehabilitative services, and physical disabilities.
The second outcome measure used was the short-form (14-item) Stein-Jessup Functional Status II (Revised) [FSII(R)] scale.6 This scale is designed to measure the health status of children who have ongoing health conditions. It measures behavioral inventories associated with the child's condition that interfere with the child's age-appropriate activities. This scale was compared previously with population normative values for ages 1 through 17 years. The short form has validity equal to that of the long form. Among well children, the mean score is 96.1 ± 8.2; among children who are ill, the mean score is 86.8 ± 15.7.6
The third measure was the Global Health Index.7 This measure is used to assess the respondent's perception of 5 areas of their child's development, including the child's general health, physical well-being, role functioning, psychological distress, and social functioning. Each area is measured with a 4-point Likert scale, with higher numbers indicating better performance.
Family Characteristics
Family characteristics, including whether the child was in the home of origin, the age, educational status, and marital status of the maternal caregiver, and whether the maternal caregiver was employed, were assessed. Data on paternal caregivers' educational status and employment status were collected during the maternal interview. A paternal caregiver was defined as a male partner identified by the mother, regardless of marital status. The maternal caregiver's social capital was also assessed. Social capital is a measure of a person's social relationships in his or her community and family.8 The social capital index used for this study was composed of questions about maternal social support, neighborhood support, church attendance, the number of children in the home, and whether the maternal caregiver had a partner. This instrument was used previously in a set of longitudinal studies of preschool-aged children at high risk for poor developmental outcomes because of adverse social or economic conditions. The social capital index was associated strongly with the child's well-being; scores of 
4 were associated with children faring well, and scores of <4 were associated with children faring poorly.9
In addition, in the year 2 interviews, information was collected on the families' socioeconomic status with the Hollingshead 4-Factor Index of Social Status.10 The Hollingshead Index has a range from 8 to 66, with higher numbers indicating higher social status. Social status is determined by factors such as gender, marital status, education, and occupation. Each family's financial capital was assessed through direct questions about all sources of the family's pretax income. The maternal caregiver was asked in which income bracket (in $5000 increments up to $50000) her family resided. Family income was compared with the North Carolina poverty level for the year 2000.11 This information was not collected in the year 1 interviews.
Statistical Analyses
Child and family characteristics were examined with frequencies and percentages. Comparisons were made between children and families with inflicted versus noninflicted injuries with 
2 analysis. Means with SDs were calculated for normally distributed data, and medians with interquartile ranges (IQRs) were calculated for nonnormally distributed data. The change in POPC scores was examined from the hospital discharge to posttrauma year 1 and from posttrauma year 1 to posttrauma year 2. Continuous scales were compared from year 1 to year 2 with the t test if normally distributed and the Mann-Whitney U test if nonnormally distributed. A cumulative logarithmic regression model using the generalized estimating equation to account for the correlation between measures obtained for a patient over time was used to assess whether there were changes in POPC scores over all 3 time periods and to assess interactions between time and mechanism of injury.
A logistic regression analysis was used to examine hospital predictors of year 2 outcomes, dichotomized as good (POPC scores of 1 or 2) or poor (POPC scores of 3 or 4). Covariates that were associated significantly with outcomes at hospital discharge or year 1 or were significant at the .05 level in bivariate analyses were placed in the model. Covariates that did not change the estimate by 10% were then removed in a stepwise manner.
Poverty level was calculated by using the highest number of the self-identified income bracket, divided by the number of people that income supported. This value was then compared with the North Carolina state poverty level.
Medical Resources
Children's use of ancillary medical resources was evaluated through maternal report. Items included whether the child had a primary pediatrician, medication use, use of home health services, and use of occupational, physical, and speech therapies. Children who used 2 therapies >1 time per week were categorized as "high users" of resources. Children were compared between assessments performed at year 1 and year 2.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
POPC outcomes were examined for 3 times, ie, discharge from the hospital, year 1 interview, and year 2 interview (Table 3). The hospital discharge POPC score was generally reflective of the year 2 POPC score. Of the 19 children with normal outcomes (POPC score of 1) at hospital discharge, the majority (73.6%) seemed to have no deficits at the time of the year 2 interview. However, 4 children had mild deficits recognized, and 1 child had a moderate deficit recognized. Sixteen children had mild deficits at hospital discharge (POPC score of 2). Of these 16, 11 were thought to have no deficits by year 2. However, 3 children were recognized to have moderate deficits by 2 years after discharge from the hospital. Therefore, of the 35 children with good outcomes (POPC scores of 1 or 2) at discharge, 31 (89%) still had good outcomes 2 years later. However, 8 children (22.8%) had additional deficits recognized.
|
The majority of children (66.7%) did not change their POPC scores between the year 1 and year 2 evaluations. Of the 31% who exhibited changes, children were as likely to show improvement (17.5%) as deterioration (14.0%). No child changed by >1 POPC point between year 1 and 2, with only 2 children changing from a poor to good outcome. When POPC scores for assessments at all 3 time points were placed in a cumulative logarithmic regression model, there was no significant change in POPC scores overall (P = .1); however, when mechanism of injury was added to the model, there was a significant interaction between time of assessment and injury type (P = .02). Children with noninflicted injuries tended to have increases in their POPC scores over time, compared with children with inflicted injuries.
Overall, children showed no substantive differences between the year 1 and year 2 interviews, as evaluated with the FSII(R) (P = .3). In addition, the FSII(R) showed no appreciable differences between the inflicted-TBI and noninflicted-TBI groups at the year 2 evaluation (P = .4) and no longitudinal differences within the groups across time (inflicted TBI, year 1 to year 2 interviews: P = .2; noninflicted TBI, year 1 to year 2 interviews: P = .8). There was also no appreciable change in the scores on the Global Health Index between year 1 and year 2 evaluations.
Logistic regression modeling was performed to analyze whether there were hospital covariates associated with outcomes (good versus poor on the POPC scale) at year 2. The only significant predictor was a Glasgow Coma Scale (GCS)12 score of 
12. Children with GCS scores of 12 had adjusted odds of 4.2 (95% CI: 1.1–16.3) of having poor outcomes, compared with children with GCS scores of 13, after adjustment for mechanism of injury (inflicted versus noninflicted), presence of seizures, and gender. Whether the child had received cardiopulmonary resuscitation was not included in the model. However, precision was poor because of small numbers in some cells.
A group of children continued to need intensive rehabilitative therapy 2 years after injury (Table 3). More than one fourth (26.3%) of the children remained in formal rehabilitation programs. Sixteen children (28%) had weekly or more physical therapy, 13 (22.8%) required weekly or more occupational therapy, and 17 (29.8%) received weekly or more speech therapy. More children were receiving speech therapy at year 2 than at year 1 (12.3% vs 29.8%). When resource use was categorized into high use (2 of 3 service types at least once weekly) versus low use, one third (29.8%) of children were high resource users. However, this did not vary according to inflicted versus noninflicted injury (RR: 1.3; 95% CI: 0.6–2.6) or race (RR for minority patient having low resource use: 0.8; 95% CI: 0.6–1.1). No child with a good outcome (POPC score) was a high resource user, with the majority of services being focused on children with poor outcomes. Families living below 2 times the poverty level tended to have lower use of ancillary resources (RR of low resource use if below 2 times the poverty level: 0.6; 95% CI: 0.4–1.0), compared with families living above 2 times the poverty level. There were 17 children receiving regular medications. Of the children receiving medications, most (70.5%) were taking anticonvulsants because of ongoing seizure disorders. More children with inflicted injuries required anticonvulsant therapy, compared with children with noninflicted injuries (RR: 1.5; 95% CI: 1.0–2.3). Almost all children (94.7%) had a primary care doctor.
Income and socioeconomic data were collected from the families at the year 2 interview (Table 4). Families tended to be poor, with 50% of families at less than 2 times the poverty level and 20% of families at less than the poverty level. This reflects a higher prevalence of poverty than the 12.3% of North Carolinians reported to be at or below the poverty level throughout the state in 2000.11 Dual employment was common, with 63.4% of 2-parent households and 26 of all 57 families (45.6%) reporting 2 working adults. The Hollingshead Index reflects a wide range of parental social status, which did not differ according to mechanism of injury (median index: 29 and 27 for inflicted and noninflicted TBI groups, respectively; P = .8). When the Hollingshead Index results were divided into groups above and below the median, children's outcomes on the POPC were not associated statistically with the Hollingshead Index results (RR: 0.8; 95% CI: 0.5–1.2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Our study showed that functional outcomes between year 1 and year 2 evaluations did not change substantively, as measured with either the POPC or the FSII(R). Deficits noted in hospital discharge POPC results were maintained in the majority of cases, and POPC scores did not change dramatically between year 1 and year 2. Children with inflicted injuries were less likely to have stability in their POPC scores when examined at 3 time points, including hospital discharge, year 1, and year 2, compared with children with noninflicted injuries; however, year 1 to year 2 POPC scores tended to remain stable in both the inflicted-TBI and noninflicted-TBI groups. This is consistent with results of a previous study of young children with TBI, which showed variable recovery during the first 6 months after recovery from posttraumatic amnesia and then stable persistent deficits.2 The need for an increase in speech therapy as the children grow older is also supported by deficits in expressive language noted among children with mild or moderate TBI.2 These deficits are more likely to be noticed as children grow older. However, it should be noted that the children in the study of TBI by Ewing-Cobbs et al2 had a median age of injury of 41 months and the study excluded children who were victims of abuse; therefore, the cohorts are not entirely comparable. Longitudinal studies of very young children suffering TBI are sparse; however, studies of older children (5–15 years of age) suggest that the full consequences of brain injury may not be seen in the first several years after injury. Importantly, growth curves of academic achievement among older children declined over time in the youngest cohort of children.1
Children with inflicted versus noninflicted injuries had similar functional outcomes at the year 1 and year 2 evaluations. No substantive differences were seen across time within groups. The most important predictor of poor outcome (POPC scores of 3 or 4) at 2 years after discharge was a GCS score of 12, although precision was poor. This highlights the difficulty in predicting longer-term outcomes for children after TBI. Whether these 2 groups of children will diverge in outcomes eventually or might differ importantly with more precise testing is unknown.
Many more families in this injured cohort were living below the poverty level then would be expected on the basis of state normative values. This is despite the fact that many families had 2 working adults. This cohort also had a low median Hollingshead Index, compared with a brain-injured cohort from Houston13 (Hollingshead Index of 46 for the severely injured group), which indicates that many families had a combination of occupations with low social prestige (such as unskilled labor), low educational achievement, and unmarried status. Social capital was low for 40% of families, which puts children at additional disadvantage.9 Children from households with multiple risk factors, including low socioeconomic status, have been shown to have worse developmental and behavioral outcomes.14 A combination of multiple risk factors puts children at higher risk, compared with a single risk factor, showing a dose-response effect. Sameroff et al15 found that environmental factors were more predictive of preschool intelligence than were child factors for a group of uninjured children. Environmental factors that were considered included minority group status, maternal education, family social support, family size, and stressful life events, among others. Caring for a brain-injured child has been shown to be more stressful to families than caring for a child with orthopedic injuries.16 Therefore, this group of children should be considered at high risk for poor development and preschool performance secondary to environmental risk factors, as well as their brain injury.
This study has limitations. There is selection bias, because we were unable to retain all families in the study. It is possible that families with higher levels of coping and organizational skills were more likely to be able to make and keep appointments for telephone interviews, thus biasing our findings toward families with better coping abilities. In addition, it is possible that families whose children were faring well might think that the developmental evaluations would be of less benefit to their children than families of children who were faring poorly. However, when families that were eligible to participate but chose not to have their children evaluated were compared with those whose children were evaluated at year 2, the groups were similar except for the loss of nonwhite and nonblack subjects. The 2 groups could differ in other ways not measured in this study. It is also possible that our instruments were not fine enough to measure more-subtle distinctions in children's outcomes. Cognitive disabilities may not be reflected adequately by functional testing in this group of children and may be more important predictors of future school performance. It is also possible that parents thought that the child was progressing well and did not report difficulties because the child had made progress since the injury event. Finally, it is unclear whether foster or adoptive parents might rate children's deficits differently than biological parents who knew the child before the injury.
This study also has strengths. It represents 2 years of follow-up monitoring for a cohort of children who were injured at similar ages, from a statewide population. Unlike other studies, it does not exclude children with inflicted injuries, who represent approximately one half of all children <2 years of age with brain injuries. It also allows for comparisons of child outcomes and family characteristics between the inflicted-TBI and noninflicted-TBI groups.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Address correspondence to Heather T. Keenan, MDCM, PhD, Intermountain Injury Control Research Center, 295 Chipeta Way, PO Box 581289, Salt Lake City, UT 84158-0289. E-mail: heather.keenan{at}hsc.utah.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Jayawant and J. Parr Outcome following subdural haemorrhages in infancy Arch. Dis. Child., April 1, 2007; 92(4): 343 - 347. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. T. Keenan, S. R. Hooper, C. E. Wetherington, M. Nocera, and D. K. Runyan Neurodevelopmental Consequences of Early Traumatic Brain Injury in 3-Year-Old Children Pediatrics, March 1, 2007; 119(3): e616 - e623. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L Acerini, R. C Tasker, S. Bellone, G. Bona, C. J Thompson, and M. O Savage Hypopituitarism in childhood and adolescence following traumatic brain injury: the case for prospective endocrine investigation. Eur. J. Endocrinol., November 1, 2006; 155(5): 663 - 669. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
