From early childhood to adolescence: Lessons about traumatic brain injury (TBI) from the Ohio Head Injury Outcomes Study

Abstract

Objective

The Ohio Head Injury Outcomes (OHIO) study was a 12-year longitudinal study of early childhood TBI. This paper reviewed the findings pertaining to caregiver and family functioning and child cognition, behavior, social competence, emotional functioning, and academics. We further considered individual and social-environmental influences on recovery and interventions.

Setting

Recruitment was completed at three children’s hospitals and one general hospital.

Participants

Children aged 3–7 years at time of injury with complicated mild to moderate and severe TBI or orthopedic injury requiring hospitalization were included.

Design

A concurrent cohort/prospective research design was used. A baseline assessment was completed shortly after the injury. Follow-up assessments were completed at 6, 12, 18 months and at an average of 38 and 82 months post injury.

Main measures

At baseline, parents/guardians completed retrospective ratings of their child’s behavioral, emotional, and social functioning pre-injury. At the subsequent assessments, ratings reflected current functioning. Information about current family and caregiver functioning was collected at each time point and cognitive testing was completed at selected time points.

Results and conclusions

Recovery after TBI is complex, varies over time, and involves injury-related and premorbid influences, cognition, genetics, and caregiver and family functioning. A sizable number of children with TBI have persisting unmet clinical needs.

Introduction

Recent research has provided a welcome increase in knowledge of long-term outcomes following early childhood TBI. Anderson and colleagues completed a study of 3–12 year-old children with mild to severe TBI. This study documented persisting cognitive difficulties, particularly in executive functions, along with behavioral, social, and adaptive deficits. These problems were more pronounced in those with severe TBI.¹ Improving our knowledge of brain plasticity and recovery, these authors also found that younger age at injury was associated with poorer recovery of cognitive functioning, especially after severe injury. Comparatively better outcomes were observed in older children with severe TBI and in children with less severe injuries.^2,3 Other notable research was completed by Ewing-Cobb and colleagues, whose findings also highlighted early vulnerability for children with TBI in the areas of intellectual functioning, ^4,5,6 language,⁷ and executive functioning.⁸ Using a longitudinal birth cohort, McKinlay and colleagues showed that mild TBI was associated with increased risk for psychiatric symptoms, including Attention Deficit/ Hyperactivity Disorder (ADHD), conduct/ behavioral problems, substance abuse, and problems with mood in adolescence.⁹ Extending our understanding of non-injury factors that predispose children to TBI, these authors also highlighted the relevance of sex, adverse life events, and parenting style.¹⁰

The Ohio Head Injury Outcomes (OHIO) study was a 12-year investigation of longitudinal outcomes following early childhood TBI relative to children who sustained orthopedic injuries (OI). The study had many strengths, including the use of comprehensive assessments across multiple domains of functioning, including the child’s neuropsychological, adaptive, emotional, behavioral, and social functioning, as well as caregiver and family coping and adjustment. An additional strength was the multi-method approach, which included standardized parent and teacher rating scales, structured interviews, neuropsychological testing, and classroom and home observations. Statistically, novel methods for the examination of longitudinal data were used, including multiple mediator modeling,¹¹ latent class growth modeling,¹² and time-varying moderators.¹³

Although many papers and conference presentations have come from the OHIO study, its findings and contributions have yet to be synthesized. Therefore, we sought to integrate and synthesize findings of the OHIO study in the areas of caregiver and family functioning, as well as children’s genetics, cognition, behavior and social competence, emotional functioning, academics and education, and psychosocial intervention. A further aim was to consider the findings in the context of the broader pediatric TBI literature, to discuss clinical implications, and to identify directions for future research.

Method

The OHIO study used a concurrent cohort/prospective research design with repeated assessments of children aged 3–7 years at injury with complicated mild to moderate and severe TBI or OI requiring hospitalization. They were recruited from three tertiary care children’s hospitals and one tertiary care general hospital. All children were hospitalized for at least one night related to their injuries. The average length of hospital stay was 2.66 days (SD = 4.19, range = 0–32) for the complicated mild to moderate TBI group, 9.91 days (SD = 19.07, range = 1–91) for the severe TBI group, and 0.82 days (SD = 1.06, range = 0–8) for the OI group. Inclusion criteria included having sustained the injury between 36 and 83 months of age, no evidence of child abuse as the cause of the injury, no history of documented neurological problems or developmental delays preinjury, and English as the primary language in the home. Among those with TBI, additional eligibility criteria included an alteration of neurological functioning as measured by a Glasgow Coma Scale (GCS) score less than 13 or evidence of trauma-related abnormality as seen on magnetic resonance imaging or computerized tomography. Children in the severe TBI group had a GCS score less than or equal to 8 as the lowest post-resuscitation score; those in the complicated mild to moderate TBI group had either a GCS score of 9–12 or a GCS score of 13–15 accompanied by evidence of abnormality on imaging. The average GCS was 3.83 (SD = 1.69, range = 3–8) in the severe group and 13.45 (SD = 1.98, range = 9–15) in the complicated mild to moderate group. The OI group included children who sustained a bone fracture, other than to the skull, and had no alterations in consciousness or other symptoms of head trauma. Of note, there were an additional 14 children with mild TBI/ concussion with overnight hospital stay, GCS scores of 13–14, and normal imaging; this subgroup was excluded in most manuscripts.

Participants were enrolled from January 2003- June 2008. A baseline assessment was completed shortly after the injury (mean number of days = 41.73; SD = 23.48). Follow-up assessments were completed at 6, 12, 18 months, at a minimum of 24 months post injury (average of 38 months), and as children entered middle school (average of 82 months post-injury). Note that some papers referred to these time points differently depending on the subset of the sample that was included. Relevant demographic and injury-related information were reported in multiple papers (see references below). See Figure 1 for information about recruitment and retention.

Figure 1.

Recruitment and retention.

See Table 1 for a list of study measures and Table 2 for a summary of commonly used outcomes. At the baseline assessment, parents/guardians completed retrospective ratings of their child’s pre-injury functioning behavioral, emotional, and social functioning. Baseline measures spanned the age range of 3–7 years and addressed emerging language, social development, and executive function. At the subsequent assessments, ratings reflected current functioning and were selected to enhance our understanding of long-term effects on child functioning across settings. Information about current family and caregiver functioning was also collected at each time point.

Table 1.

Study measures and assessments.

Domain/Construct	Measure/ method	Source	Assessment	Time point collected
Adaptive, behavioral, and social-emotional functioning
Real world functioning	CAFAS 80	Parent interview	Lab	82 months
Child behavior	CBCL/TRF 81,82	Parent, Teacher	Lab/school	All time points
Child social competence	HCSBS/SSBS 83	Parent, Teacher	Lab/school	All time points
Victimization	Schwartz Report of Victimization 84	Child, Parent, Teacher	Lab/school	82 months
Neuropsychological functioning
Executive functions in daily life	BRIEF35	Parent, Teacher, Self-report	Lab/school	All time points
Attentional Control	ANT, TEA-Ch Walk/Don’t walk 85,86	Child	Lab	82 months
Planning/Problem Solving	IGT, Tower of London, D-KEFS Verbal Fluency 87,88,89	Child	Lab	82 months
Working Memory	TEA-Ch Code Transmission86	Child	Lab	82 months
	TEA-Ch Creature Counting86	Child	Lab	82 months
Processing speed	WISC-IV Processing Speed Index 90	Child	Lab	82 months
Language Pragmatics	CASL Pragmatics, Discourse 91	Child	Lab	Baseline and 6, 12, 18, and 82 months
Global cognitive ability	WASI 92	Child	Lab	82 months
Academic Achievement	Woodcock Johnson Tests of Achievement- 3rd Edition 93	Child	Lab	Baseline and 12 and 82 months
Memory	Verbal Paired Associates94	Child	Lab	82 months
Social problem solving	Dodge SIP-SR Stories 95	Child	Lab	82 months
Family/ environment
Parent psychological functioning	SCL-90 96	Parents	Lab	82 months
Injury-related burden	FBII 97	Parent interview	Lab	All time points
Family functioning	FAD 98	Parents	Lab	All time points
Caregiver coping	COPE 99	Parents	Lab	Baseline and 6, 12, 18, and 82 months
Stresses and interpersonal	LISRES-A100	Parent interview	Lab	All time points
Marital/Partner Relations	Relationships Interaction Questionnaire 101	Parents	Lab	Baseline and 6, 12, 18, and 82 months
Parenting Practices	Parenting Practices Questionnaire 102	Parents	Lab	All time points
DNA saliva collection	Oragene Collection Kit	Child	Home	82 months

Key: ANT=Attention Network Test; BRIEF=Behavior Rating Inventory of Executive Function; CAFAS=Child and Adolescent Functional Assessment Scale; CASL=Comprehensive Assessment of Spoken Language; CBCL=Child Behavior Checklist; COPE=Caregiver…; D-KEFS=Delis-Kaplan Executive Function System; Dodge SIP-SR=Dodge Social Information Processing-Self-Report; EA HOME=Early Adolescent version of the Home Observation for Measurement of the Environment; FAD=Family Assessment Device; FBII=Family Burden of Injury Interview; HCSBS=Home and Community Social Behavior Scales; IGT=Iowa Gambling Task; LISRES-A=Life Stressors and Social Resources Inventory-for Adults; MS-CISSARS=Main-Stream Code for Instructional Structure and Student Academic Response; PPQ=Parenting Practices Questionnaire; RIQ=Relationship Improvement Questionnaire; RPQ=Relationship Problems Questionnaire; SCL-90=Symptom Checklist-90; SPPC=Self-Perception Profile for Children; SRoV=Self-Report of Victimization; SSBS=School Social Behavior Scales; TEA-Ch=Test of Everyday Attention for Children; TRF=Teacher’s Report Form; WASI=Wechsler Abbreviated Scale of Intelligence; WISC-IV=Wechsler Intelligence Scale for Children-Fourth Edition; WJ=Woodcock-Johnson Tests of Cognitive Abilities

Table 2.

Studies of commonly used outcome measures following pediatric TBI: Main effects, moderators, and other significant main effects.

Measure	Times assessed	Group differences	Moderators	Other significant main effects	Citations
Preschool and Kindergarten Behavior Scales/ Home and Community Social B6ehavior Scales (HCSBS)	6 Months	S > OI (means) S > OI (% elevated) S > CMM (% elevated)	-	Age at Injury, SES, Effortful Control, Executive Functioning	30, 45
	12 Months	S > OI (% elevated) S > CMM (% elevated)	-	-	45
	18 Months	S > OI (means) S > OI (% elevated) S > CMM (% elevated)	-	Baseline/Preinjury Social Functioning, Race, SES, Parenting Style, Form Version, Time Since Injury, Family Structure	45, 51, 20
	38 Months	-	-	-	48
	82 Months	S > OI (means) S > CMM (means)	Group × Authoritative Parenting	Pre-injury functioning, Later Parenting Style, Discourse Gist	39
Child Behavior Checklist (CBCL) Externalizing/Total Problem Scores	6 Months	S > OI (% elevated)  S > CMM (% elevated)		Age at Injury, SES, Parenting Style, Family Functioning, Sleep Disturbance	45, 47
	12 Months	S > OI (means)  CMM > OI (means)  S > OI (% elevated)  S > CMM (% elevated)	Group × Gene	Age at Injury, Genetics, SES, Parenting Style, Family Functioning, Sleep Disturbance	11, 45, 47
	18 Months	S > OI (means)  CMM > OI (means)  S > OI (% elevated)  S > CMM (% elevated)	Group × Emotion Labeling  Group × Authoritarian Parenting × Time  Group × Permissive Parenting × Time	Age at Injury, Parenting Style, Family Functioning, SES, Scaffolding during Play, Time Since Injury, Parent Distress, Pre-Injury Functioning, Form Version, Family Structure, Sleep Disturbance	45, 13, 20, 47, 44
	38 Months	S > OI (means)  S > CMM (means)	-	Pre-injury Functioning	44, 48
	82 Months	S > OI (means)  CMM > OI (means)	Group × Permissive Parenting  Group × Gene	Pre-injury functioning, Later Parenting Style, Genetics	39, 11
Behavior Rating Inventory of Executive Function (BRIEF)	6 Months	S > OI (means)  CMM > OI (means)  S > OI (% elevated)  S > CMM (% elevated)	Group × Gene	SES, Age at Injury, Social Environment, Neuropsychological Test Performance, Genetics, Sleep Disturbance	30, 38, 11, 45, 47
Global Executive Composite Scores (GEC)	12 Months	S > OI (means)  S > OI (% elevated)  S > CMM (% elevated)	Group × Authoritarian Parenting × Time  Group × Gene × Permissive Parenting	Social Environment; Genetics; Sleep Disturbance	38, 56, 45, 47
	18 Months	S > OI (means)  S > OI (% elevated)  S > CMM (% elevated)	Group × Authoritarian Parenting × Time	Social Environment; SES, Family Functioning, Parenting Style; Sleep Disturbance	45, 47, 38
	38 Months	S > OI (means)  S > CMM (means)  S > OI (% elevated)  S > CMM (% elevated)	-	Pre-injury Functioning, SES, Family Functioning, Parenting Style	36, 48
	82 Months	S > OI (means)  CMM > OI (means)	Group × Authoritarian Parenting  Gene × Authoritarian Parenting  Group × Gene  Gene × Authoritarian Parenting × Time	Pre-injury Functioning, Later Parenting Style; Genetics; Academic Performance	39, 56, 11, 55

S > OI (means) = Severe TBI more problems than OI based on means

S > CMM (means) = Severe TBI more problems than complicated mild/ moderate based on means

CMM > OI (means) = Complicated mild to moderate more problems than OI based on means

S > OI (% elevated) = Severe TBI more problems than OI based on rates of clinical elevation

S > CMM (% elevated) = Severe TBI more problems than complicated mild/ moderate based on rates of clinical elevation

CMM > OI (% elevated) = Complicated mild to moderate more problems than OI based on rates of clinical elevation

Note: Pre-injury baseline ratings were controlled for in all models and differed between the severe and OI groups.

Results

Caregiver and family functioning

A notable contribution was with respect to caregiver and family functioning, an area that was previously largely unexplored in this age range^14,15 In the acute period after the injury, caregivers of children with TBI reported more psychological distress and family injury-related burden compared to caregivers in the OI group. Parents of older children with severe TBI (5- to 6-year-olds) reported significantly higher levels of injury-related burden and distress than did parents of younger children with these injuries (3- to 4-year-olds).¹⁶ At 6, 12, and 18 months post-TBI, injury-related family burden was higher for the moderate and severe TBI groups than for the OI group, although this difference diminished over time. Better family functioning buffered against the acute injury-related burden of severe TBI, although these protective effects diminished at 12 and 18 months.¹⁷ Further from injury, caregiver distress persisted, but diminished. Longer-term caregiver distress was related to multiple factors, including poorer parental coping^18,12 and more problems with the child’s academic functioning,¹⁹ behavior,^13,20 and adaptive living skills.¹⁷

Results also suggested that family relationships were impacted by TBI, although mothers and fathers reported differing perceptions. Mothers of children with severe TBI rated their parental relationship as significantly less loving than did their partners at 18 months post-injury. Mothers of children with moderate and severe TBI also endorsed more parenting disagreements than did their partners. However, fathers reported higher levels of family dysfunction than their partners, regardless of injury type or severity.²¹ At 18 months, fathers endorsed more depressive symptoms and more general distress than mothers, although this pattern did not persist at 82 months.¹² Mothers and fathers also differed in their preferred coping method. At 6, 12, and 18 months post-injury, fathers were more likely to use denial following moderate and severe TBI, but not OI. Mothers were more likely to use acceptance and emotion-focused strategies, regardless of injury type.¹⁸ This pattern changed at the 82 month follow-up, at which time mothers of children with severe TBI reported greater denial than fathers. Denial did not differ between mothers and fathers following OI and moderate TBI.¹²

TBI also had implications for the parent-child relationship. Videotaped interactions of free play in the initial weeks after the injury (mean of 39.99 days post injury [SD = 19.38]) showed that mothers of children with TBI exhibited less warm responsiveness to the child and made more directive statements compared to the OI group. Contrary to expectations, the largest differences in warm responsiveness were between the OI and the complicated mild and moderate TBI groups. Notably, however, three of the most severely injured children were unable to complete the baseline assessment.²² Exploring these relationships at 6 and 12 months post-injury, Fairbanks and colleagues²³ reported that mothers in the complicated mild/moderate TBI group, but not the severe group, showed significantly lower levels of maternal warm responsiveness compared to the OI group. These differences were observed at baseline and 6 months post injury, but not at 12 months. This suggests that the adverse effects on parenting diminished with time in families of children with complicated mild/moderate TBI.

Cognition

Cognitive deficits after pediatric TBI have been well-documented^1–8 and were similarly apparent in the OHIO study. Not surprisingly, deficits were most prominent after severe TBI, although problems were also evident after complicated mild to moderate TBIs.^24,25 Deficits were present across multiple domains, including visual learning and memory, working memory, visual-motor skills,^24,25 and processing speed ^26,11 Consistent with the broader literature, executive functions were the most notable difficulty. ^27,28,29

On performance-based measures, severe TBI was associated with poorer inhibitory control at 3 and 6 months post injury.³⁰ At 18 months, children with severe TBI continued to show weaknesses in higher-order language and pragmatics ^31,24,32 and theory of mind,^31,33 skills also likely associated with executive functions. Fluid reasoning and inhibitory control were additional areas of cognitive weakness after severe TBI, with functioning in these areas associated with parent-reported adaptive living skills at the 82 month assessment.¹¹ Also at that time point, children with severe TBI showed poorer inhibition compared to the OI group. Unexpectedly, poorer family environment and more caregiver psychological distress were associated with better test performance. Although the reason for this was unclear, it may be that high levels of caregiver distress reflected efforts to support their child’s recovery.³⁴

Standardized ratings are commonly used to offer an ecologically-valid assessment of executive functions.³⁵ In the OHIO study, parent ratings were sensitive to the presence of TBI.^36,30 On these ratings, problems with executive functions were more prominent with time, a pattern that was attributed to increasing executive demands with age. However, the emergence and persistence of these problems was influenced by a variety of factors, most notably the family environment.³⁷ Potter and colleagues reported that lower socioeconomic status was associated with more parent-reported executive functioning problems at 12 and 18 months after injury, irrespective of injury group.³⁸ Twelve and 18 months after moderate TBI, authoritarian parenting (i.e., high level of control and low parental responsiveness) was associated with more executive functioning problems. At an average of 18 and 39 months post injury, lower family dysfunction was associated with fewer child executive problems.³⁶At the 82 month assessment, higher levels of permissive parenting (i.e., high parental responsiveness and low level of control) were associated with poorer parent-reported executive functions. Conversely, higher levels of authoritative parenting (i.e., high control and high responsiveness) predicted better executive functions.³⁹ Considering parent-reported executive functions at all time points, the use of latent growth curve modeling pointed to two trajectories: normal (70.8%) or clinically elevated (29.9%). Of the injury-related variables, only pre-morbid executive functions significantly predicted trajectory. Interestingly, injury severity was not associated with class membership. Instead, greater family burden and a more permissive parenting style were associated with more executive problems for both trajectories.³⁷

Academic and educational outcomes

Students with TBI often require academic services and accommodations, although school programming often did not meet the apparent needs of the child in the OHIO cohort. In a recent paper, academic need was defined by (1) an impairment on standardized IQ or achievement testing, (2) clinically significant problems with behavior, emotional, or executive functions per parent and/or teacher report, (3) and/or a diagnosis of a learning disorder, attentional disorder, and/or developmental disability. At 82 months post-injury, students with moderate and severe TBI had more academic needs than students with OI. Compared to the OI and the complicated mild to moderate TBI groups, a larger proportion of children with severe TBI received academic services. However, a sizable proportion of students in all groups had unmet academic needs; this ranged from 46.2%– 63.6% across groups.⁴⁰

Perhaps not surprisingly, cognition was strongly related to academic performance. At baseline and 12 months post injury, better verbal story memory and inhibitory control were related to better performance on measures of kindergarten readiness and early reading, spelling, and math.⁴¹At 82 months, students with TBI had lower teacher ratings of academic engagement compared to children with OI. These classroom difficulties were associated with performance-based and parent- and teacher-reported executive functioning.¹¹

Family factors were again relevant when considering academic outcomes. At 18 months post-injury, spelling and overall school readiness deficiencies in the moderate TBI group were found only at higher levels of caregiver stress.²⁵ At 82 months, children with severe TBI scored lower on measures of academic achievement and had more parent-reported academic concerns and teacher-reported concerns about classroom internalizing behaviors. Interestingly, the early family environment consistently predicted academic skills at the 82 month time point and the later family environment more consistently predicted teacher-reported classroom behavior.¹⁹

Behavioral outcome

Externalizing behavior problems, such as hyperactivity and aggression, are a well-documented sequela of TBI.^42,43This was similarly common in the OHIO cohort, particularly after severe TBI,^44,45although there were other relevant factors. Sleep disturbance, which was more common after severe TBI,⁴⁶was associated with more behavior problems at 6, 12, and 18 months.⁴⁷ Extending the work of Anderson and colleagues,^1–3 younger age at injury was also associated with more behavior problems. Chapman and colleagues reported that children injured at a younger age had more parent-reported externalizing problems at 6, 12, and 18 months post-injury compared to older children.⁴⁵ At 38 months post injury, Karver et al. reported that greater time since injury, severe TBI, and young age at injury were associated with more parent-reported symptoms of ADHD.⁴⁸

Family factors were again relevant with respect to behavior. At 6, 12, and 18 months post injury, higher levels of permissive parenting and family dysfunction and lower socioeconomic status were associated with clinically significant externalizing problems.⁴⁵ Extending these findings with the use of time-varying moderators, Treble-Barna et al. reported that behavior problems were exacerbated by less favorable parent behaviors over the first 12 months after severe TBI. However, the role of parenting factors diminished with time.¹³ At 18 months, children with severe TBI showed more behavior problems than children with moderate TBI and OI, regardless of parent behaviors. At the 82 months assessment, parenting factors were again relevant as a permissive parenting style was associated with more behavioral concerns in children with severe TBI.³⁹ In a separate paper that examined parent-reported ADHD symptoms after complicated mild to severe TBI at all time points, the severe TBI group had more symptoms than the OI group. Higher maternal education was associated with lower ADHD symptoms across groups. Greater family dysfunction was associated with more attention-related symptoms in severe TBI group.⁴⁹

Emotional functioning

TBI was also associated with persisting emotional difficulties. At 18 months post injury, the severe TBI group evidenced more internalizing problems compared to the OI.²⁰ A separate paper documented specific problems with anxiety at 12 and 18 months after severe TBI.⁴⁸ Considering other potential influences, poorer sleep was related to more emotional problems across all groups at 6, 12, and 18 months post injury.⁴⁷ As seen in multiple areas of the child’s recovery, emotional problems were related to family functioning. Using auto regressive cross-lag panel modeling to examine the direction of the relationship between maternal distress and child internalizing problems in families of children with TBI at all time points, there was evidence that maternal psychological distress contributed to persisting child internalizing symptoms. Additionally, child internalizing symptoms exacerbated caregiver distress.⁵⁰

Social competence

Social functioning is a common problem after TBI, as reflected by parent concerns about social competence in the first year after the injury³⁰ and peer victimization at the 82 month time point.²⁶ At 18 months post-injury, Chapman and colleagues found lower than expected rates of clinically significant social competence problems per parent-report, although significantly higher than the severe TBI group relative to the other groups (21% in the severe TBI group, 2% moderately TBI, and 0% OI).⁴⁵ In a separate paper that used group means on the same measure also at 18 months, the combined moderate to severe TBI group had more social competence problems than the OI group. Children with TBI showed difficulty labeling emotions based on photographs, although this was unrelated to parent-reported social competence.⁵¹ Thus, the functional implications of emotional labeling remained unclear. A recent paper documented the relevance of parenting practices with respect to social competence at the 82 month assessment. Specifically, higher permissive parenting was associated with poorer social competence. Higher authoritative parenting was related to better social competence.³⁹

Genetics

In recent years, interest in genetic influences on biological responses to injury and recovery has increased. Although various candidate genes have been identified, the apolipoprotein e4 (APOE) gene has received the most attention in the literature.^52,53 The APOE gene is thought to be associated with neuron protection and synaptic repair and preliminary research suggests presence of the APOE e4 allele may confer risk of poorer outcomes following adult TBI.⁵⁴ Similar results have been found in a pediatric TBI population,⁵⁵ although most genetic studies in a TBI population have not considered the role of environmental factors on recovery. In the OHIO cohort, children with TBI who were carriers of the APOE e4 allele had poorer adaptive functioning compared to non-carriers with TBI and children with OI. However, this was only true at low levels of authoritarian parenting. The poorest adaptive functioning was seen in non-carriers with TBI that had high levels of an authoritarian parenting environment.⁵⁶

Genes associated with dopamine may also impact outcome after TBI and may also be salient targets for research given its association with important cognitive functions affected by injury, including executive functions. The catechol-O-methyltransferase (COMT) enzyme degrades catecholinergic neurotransmitters, such as dopamine, epinephrine, and norepinephrine, which may affect recovery after TBI.⁵⁷ Kurowski and colleagues examined the associations of a functional COMT genotype with executive functions at 18 months post-injury. Low activity COMT enzyme genotype, which indicated higher dopamine levels that are thought to be protective, was associated with better scores on neuropsychological measures of verbal fluency and inhibition in all participants. At that time point, the main effect of COMT with respect to parent-reported executive functions was non-significant.⁵⁸ Extending these findings to consider the interactions of parenting style, COMT, and the child’s executive functioning across all time points, Kurowski and colleagues found that the negative effects of authoritarian parenting on executive functioning were buffered by the presence of the COMT genotype (lower enzyme activity/higher dopamine levels). This did not vary by injury group, suggesting that these findings are not limited to TBI and extend to early childhood traumatic injury more broadly.⁵⁹ In a separate paper, Smith-Paine and colleagues⁶⁰ examined parenting practices in relation to dopamine expression and density (ANKK1 rs1800497). Presence of the T allele of this gene has been associated with fewer dopamine D2 receptors⁶¹ and is thought to confer risk for poorer cognitive outcomes, including in an adult TBI population.⁶² Smith-Paine and colleagues found in a positive parenting environment, the combination of TBI and carrier status, specifically the T allele, was associated with worse executive functioning at 12 months post injury. In a negative parenting environment, the combination of TBI and carrier status was associated with similar executive functioning to non-carriers with TBI and carriers with OI. A similar pattern was found 82 months post injury for carriers of the T-allele regardless of injury status, similar to other work.⁶⁰ Interestingly, the Smith-Paine study suggested that the effects of genetic factors on executive function may be more evident for children from more favorable family environments, such that significant genetic effects of the T allele may either be muted or obscured under conditions of environmental disadvantage.⁶⁰

Intervention

Although psychosocial intervention was not a primary focus of the OHIO study, some of the findings offered insights in this area. Karver et al. examined the associations of mental health service utilization and clinical need (i.e., the presence of significant parent-reported emotional or behavioral problems). As expected, clinical need at 18 and 38 months was associated with greater service utilization at 38 months. However, unmet clinical need (i.e., significant clinical need in the absence of services) was not limited to children with TBI and all groups had higher than expected rates (66% severe TBI, 57% complicated mild to moderate TBI, and 50% OI). Not surprisingly, lower socioeconomic status was associated with higher rates of unmet need across groups.⁴⁴ A subsequent paper explored these relationships at the 82 month assessment. At that time, 77% of children had an unmet clinical need, with no injury group differences identified. Behavioral health services were utilized by 10% of children in the sample and utilization again did not differ by group. Early treatment and white race were associated with less service utilization across groups.⁶³

Conclusions and areas for future research

Clinical implications

Many overarching conclusions can be made based on the results of the OHIO study. Recovery is complex, varies over time, and involves the interplay of injury-related and premorbid influences, cognitive ability, genetics, and caregiver and family functioning. A sizable number of children with TBI displayed persisting cognitive, behavioral, and socio-emotional difficulties and these clinical needs were often unmet. These conclusions are perhaps best illustrated when examining executive functions, an area in which problems often emerge with greater time since injury.³⁷ In addition to time since injury and injury severity, long-term recovery was related to family functioning,³⁶ sociodemographic status,³⁸ and genetics.^58,59 Given the persistence of executive functioning problems, children with TBI would likely benefit from school-based supports and behavioral therapy, although many do not receive such services.^40,44,63 In view of these complex dynamics, prognosis after TBI is variable and dependent on many influences, related both to the child and the family system. As clinical needs often change with time, long-term, multidisciplinary care is crucial.

The family and social environment were associated with recovery in nearly every domain of functioning. Considering the relevance of the family environment, more effective parenting behaviors may support the development and recovery of cognitive skills, behavior, social competence, and emotional functioning after TBI. Parents who are exposed to prolonged distress may withdraw or model maladaptive behaviors and coping strategies, which may contribute to increased child behavior and emotional problems. Parental characteristics may also influence how problems are viewed and reported,^21,49 thus highlighting the need to assess functioning from multiple perspectives. In view of the robust influence of the family environment, family-based psychosocial support is an important component of follow-up care. Indeed, randomized clinical trials of post-discharge family-based interventions showed promising results for young children⁶⁴ and adolescents⁶⁵ with TBI. At present, additional research is needed to understand effectiveness in clinical practice. Additionally, empirically-validated interventions for the inpatient setting are also needed to equip families with skills to cope with the diverse challenges that are frequently experienced after TBI.

The results of the OHIO study have increased our knowledge of individual and genetic influences on recovery after injury. Existing research has begun to demonstrate the important interactions of environment and genetics. The exploration of risk alleles in both positive and negative environments will help to understand these complex interactions and increase our knowledge of possible genetic models (e.g., “vulnerability” versus “plasticity” alleles).^66,67 Future research, however, must also acknowledge that the role of specific genes may change over the course of the lifespan. For instance, the influence of the APOE e4 gene may vary over time and the direction of that influence may be environmentally dependent, which is somewhat unexpected in view of adult research suggesting that the APOE e4 allele has detrimental effects in multiple conditions, such as adult TBI,⁶⁸ Alzheimer’s disease,⁶⁹ cardiovascular disease,⁷⁰ and sleep-disordered breathing.⁷¹ One possibility is that the APOE gene has implications that differ across the life span. Indeed, other authors also have posited that some genotypes are beneficial in youth, but become risk factors later life.⁷²

Directions for future research

Continued research is needed to fully understand the range of psychosocial risk and protective factors that influence long-term cognitive and behavioral recovery after TBI. To this end, longitudinal designs will continue to increase our understanding of the relationships of the child’s recovery with the broader environment. 73 There also remains much to be learned about TBI across the age range. The goal of the OHIO study was to specifically examine long-term outcome after early childhood injury. This resulted in a restricted age range, which limited our ability to fully examine the role of age at injury. Thus, comprehensive examination of psychosocial, cognitive, and family outcome across the pediatric age range is warranted. Continued research that extends beyond childhood and into adulthood is also needed. Other authors have highlighted the challenges of transitioning to adulthood after pediatric TBI and identified factors that are associated with a more successful transition, including older age at injury, receipt of rehabilitative services, higher socioeconomic status, and female gender.⁷⁴ Fewer studies have examined psychosocial and cognitive outcomes in transition-age individuals with TBI with limited research suggesting increased risk for negative outcomes in young adulthood.⁷⁵ Additional research is needed to more fully understand the nature of these difficulties and supportive factors. This will help to identify effective clinical, educational, and vocational supports for young adults with a history of childhood TBI.

From an assessment standpoint, refinements in current methods are needed with an eye to ecological validity. For example, the OHIO study largely relied on parent-reported social competence. This approach has notable limitations, including that parents are often not present during typical social interactions with peers, particularly as their child gets older. Issues also arise in the assessment of executive functions because many commonly used performance-based measures may lack ecological validity.^76,77 Moreover, the results of performance-based and parent-report executive measures are often dissimilar, such that children are more likely to be identified as impaired based on parent report.⁷⁷ This lack of convergence highlights the complexity of executive functioning construct, while also demonstrating a need for improved assessment methods.⁷⁸ As an additional consideration related to assessment, the OHIO study relied on retrospective parent ratings of pre-morbid child functioning at the baseline assessment, which may have resulted in a biased report.

Finally, treatments and interventions that support recovery after TBI also need to be explored. Continued genetics research is crucial, as it may lead to treatments that are personalized to the individual on a neurobiological level. In addition to genetics, other factors may influence responsiveness to post-TBI intervention. For example, Karver and colleagues found that lower verbal intelligence was associated with greater improvements in parent-reported executive functioning after intervention for teens with TBI.⁷⁹ Understanding factors that influence response to intervention will enable clinicians to tailor treatments to individuals more precisely. Furthermore, the robust evidence to suggest that family/ environmental factors may account for more variance in long-term outcome than factors directly related to the injury itself indicates that effective intervention programs target modifiable risk factors and promotes protective factors, particularly within the family system. This may include supporting caregivers to reduce distress and providing tools to promote positive parent-child interactions.⁷³

Conclusions

Recovery after early childhood TBI is complex and related to multiple factors. Many children have unmet behavioral, emotional, cognitive, and educational needs that often change with time, thus highlighting the importance of long-term, multidisciplinary follow-up and care. From a psychosocial perspective, the family environment is crucial and continued work is needed to develop and refine family-based interventions. There are multiple other areas for which further research is needed. Studies that focus on psychosocial factors that influence the transition to adulthood, increase our understanding of protective and risk factors, and lead to the refinement of current assessment methods are particularly imperative. An improved understanding in these areas will further optimize care for children with TBI and their families.