Parenting Behaviors after Moderate – Severe Traumatic Injury in Early Childhood

Abstract

Purpose:

Examine the impact of traumatic brain injury (TBI) on parenting behavior over time.

Method:

Included 206 children (3–7 years old) with moderate-severe TBI or orthopedic injury, using a prospective longitudinal cohort study design. Assessments completed at baseline, 6-months, 12-months, 18-months, 3.5 years, and 6.8 years after injury. Dependent variables included authoritative, permissive, and authoritarian parenting.

Results:

Injury characteristics had limited impact on parenting behaviors over time. Levels of authoritative parenting remained stable over time; however, levels of warmth and involvement declined over time for those with TBI. Levels of permissive and authoritarian parenting declined for all participants by 3.5 years post injury. SES and stressors impacted parenting behaviors.

Conclusions:

While there was limited effect of TBI on parenting behavior over time, it remains unclear how individuals respond to these parenting behaviors years after injury. Clinicians should monitor family and parenting behaviors to foster an environment to promote positive recovery.

Pediatric traumatic brain injury (TBI) is one of the leading causes of morbidity and mortality in childhood, and the most common source of acquired disability in children ¹. Children who have sustained a moderate to severe TBI often experience significant cognitive and behavioral sequelae ², even as long as 7–10 years after injury ^3–6. Further, the impact of pediatric TBI on parent and family functioning has been well documented, with the influence of TBI severity less clear ^7–12. Managing new stressors associated with the injury as well as emerging challenging behaviors after injury is stressful for parents ¹³, and families experience elevated family burden and parental distress, compared to families of children with OI, even years after TBI ^11,14. The family environment is a significant predictor of recovery following pediatric TBI, with a positive family environment associated with more favorable outcomes ^15–18. Taylor and colleagues ¹³ described the dynamic/bidirectional relationship between child and family outcomes following TBI in early childhood. In their study, child behavior 6 months post injury predicted changes in family burden from 6 to 12 months after injury, and parental distress 6 months after injury predicted changes in child behavior across the same period. Parenting has been shown to impact recovery, but we need more information on how TBI impacts parenting, which thereby impacts recovery.

Parenting behaviors are an aspect of family environment that has been consistently linked with child development and behavioral functioning among typically developing children ^19–23. Baumrind ²⁴ identified three distinct parenting styles: authoritarian, authoritative, and permissive parenting, and their impact on children. Each style has distinct patterns of parent behaviors and associated child outcomes. Authoritarian parenting is associated with high levels of control and low levels of warmth and nurturing behaviors ^24–26. Authoritarian parents are often described as strict and punitive while also restricting the child’s autonomy. Thus, their children are often discontent, withdrawn, and distrustful ^24,25, and often lack social competence ²⁷. Permissive parenting is described as non-controlling, non-demanding, accepting, and warm/nurturing ^24,25,28. Permissive parents allow their children to regulate their own behaviors and avoid control. In general these children tend to be immature, impulsive, aggressive, self-centered, and rebellious ^28,29. In contrast, authoritative parents are high in both control and positive encouragement/warmth ^24,25. They tend to discipline in a rational manner, are receptive to child communication, and value autonomy and independence ^24,25. Children exposed to this parenting style appear to be the most well-adjusted, as they are cheerful, socially responsible, independent, and cooperative ^24,25. Broadly, authoritarian and permissive parenting styles are considered less effective than authoritative parenting.

We know little about the effects of pediatric TBI on parenting behaviors, and one facet of parenting after TBI that has received limited attention is the impact of the injury itself on the how parents interact with and discipline their children. Wade et al ³⁰ and Fairbanks et al ³¹ reported that parents of children with complicated mild/moderate TBI demonstrated lower levels of warm responsiveness within the first 6 months post injury. However the impact of TBI on warm responsiveness was not observed 12-months post injury, suggesting a diminishing effect of complicated mild/moderate TBI over time on parenting. Parenting behaviors that were effective prior to the injury may not be as effective after injury as the child is less able to anticipate consequences, and parents may need to adjust their behavior in response to the onset of challenging behavior. A few studies have examined the effect of parenting on children with TBI. Micklewright et al ²⁶ suggested that stress precipitated an authoritarian parenting style, which in turn had negative consequences for the child’s adaptive functioning. However, this study was cross-sectional in nature, which precluded assessment of potential changes in parenting in response to TBI or determination of the direction of the effects. Some investigators have noted greater negative effects of TBI for children whose parents reported higher levels of permissive ^15,32 or authoritarian parenting ^15,33. Further, the influences of parenting varied as a function of time since injury with high authoritarian parenting associated with better behavioral adjustment 6 months after injury, but worse behavioral outcomes 18-months after injury. However, while ¹⁵these studies investigated child outcomes longitudinally, the parenting practice variables used as moderators were collected at one time point (baseline ^15,33or 18-months post injury ³²) leaving it unclear whether child outcomes were related to potential changes in parenting across the follow-up period. Finally, Narad et al ⁵ examined parenting behaviors as a time varying moderator of executive functioning over time following pediatric TBI. Greater permissive parenting was associated with increased executive dysfunction for children with TBI; however, change in parenting behavior over time was not directly examined.

The extent to which parenting behaviors change over time after TBI, particularly as the child transitions into adolescents, is unclear. Parents may learn to adapt their parenting styles during their child’s recovery as sequelae of the injury change. Parents who initially utilize more authoritarian practices may become more permissive over time as a result of feeling overwhelmed or worn down by their child’s challenging behaviors ²⁶. The bidirectional relationship between parent and child outcomes, coupled with the potential variability in child behavior following TBI, suggests that parenting behaviors of parents of children with a history of TBI may vary significantly over time. Further, developmental research suggests a change in the parent child relationship during the adolescent period³⁴, as well as an increase in parental distress and family dysfunction during this developmental transition ¹¹, making this a critical time to understand parenting behavior. A better understanding of the pattern of parenting practices over time following pediatric injury would extend our understanding of child and family recovery following pediatric TBI. Given the interest in the effect of the family environment on recovery following TBI, we need a better understanding how parenting changes over time post-injury, rather than relying on reports of pre-injury parenting styles. Moreover, parenting behaviors are modifiable via intervention, and identifying factors (injury factors and developmental period) that may be associated with the development of more maladaptive parenting behaviors would allow clinicians to identify at risk families in need of potential intervention. If we know that TBI, or injury more broadly, results in an increase in a specific type of maladaptive parenting behavior, we can design interventions to prevent/mitigate these changes and promote more optimal parenting post injury.

To address these questions, we examined changes in parent-reported authoritarian, authoritative and permissive parenting behavior over time in a cohort of children who sustained moderate to severe TBI or orthopedic injuries (OI) between the ages of 3–7. Consistent with the developmental literature, we expected that parents of children with TBI would report different parenting practices than parents of children with OI. Because parents of those with TBI tend to be more stressed than those with OI, even years after injury ^35,36, and parents under greater stress are more likely to rely on authoritarian styles ^26,37–39, we predicted that parents of children with TBI would endorse increasing levels of authoritarian parenting practices over time than parents of children with OI.

Method

Participants

Institutional Review Boards of all institutions approved all procedures, and written informed consent was obtained from all participants. Participants were recruited from three children’s hospitals and one general hospital in Ohio as part of a prospective, longitudinal study evaluating outcomes of children who sustained a TBI or OI between age 3 and 7 years ⁴⁰. Families completed assessments at multiple time points, including the immediate post-acute period (0 to 3 months after injury), 6-months, 12-months, 18-months, and 3.5-year post injury, as well as a long-term follow-up when the child was entering middle school, at a mean of 6.8 years post injury. Additional inclusion criteria were hospitalization overnight for traumatic injury (TBI or OI), non-abusive cause of injury, no pre-injury neurological problems or developmental delays, and English as the primary language in the home. The severity of TBI was characterized using the lowest post resuscitation Glasgow Coma Scale (GCS) score ⁴¹. Severe TBI was defined as a GCS score less than or equal to 8. Complicated mild to moderate TBI (henceforth referred to collectively as ‘moderate TBI’) was defined as a GCS score of 9–12 or a higher GCS score with abnormal neuroimaging. The OI group included children who sustained a bone fracture (not including skull fractures), had an overnight stay in the hospital, and did not exhibit alterations in consciousness or other signs or symptoms of head trauma or brain injury. A number of manuscripts have been published utilizing data from this longitudinal cohort. A full list of these publications is available from the authors upon request.

A total of 206 children (23 severe TBI, 64 moderate TBI, 119 OI) were included in the analyses. The OI group had higher socioeconomic status (SES; defined as a z-score that combined parental education and median census track income by zip code) than the severe TBI group; however, no other differences in demographic variables were noted between injury groups (Table 1). A total of 134 children (16 severe TBI, 44 moderate TBI, 74 OI) completed the extended follow-up visit, when participants were transitioning into adolescence, an average of 6.8 years after baseline visit. Those that completed all visits did not differ from those that did not in terms of age at injury, injury severity, sex, race, SES, or baseline parenting behavior scores. Injury groups did not differ on any demographic variables at this extended follow-up assessment (Table 1).

Table 1.

Demographic table for participants at baseline and 6.8 year follow up.

	Severe TBI		Moderate TBI		OI
	Baseline	6.8 years	Baseline	6.8 years	Baseline	6.8 years
N	23	16	64	44	119	74
Sex, % male	16 (69.6%)	10 (62.5%)	37 (57.8%)	25 (56.8%)	69 (58.0%)	39 (52.7%)
Race, % non-white	7 (30.4%)	6 (37.5%)	21 (32.8%)	11 (25.0%)	28 (23.5%)	17 (23.0%)
Age at injury	4.96 (1.00)	5.04 (.98)	5.06 (1.20)	5.16 (1.23)	5.12 (1.07)	5.09 (1.07)
SES	.−.48 (.65)	.−.42 (.68)	.−.12 (1.10)	.−.01 (1.05)	.17 (.95)	.11 (.93)

Note: OI = Orthopedic Injury, TBI = Traumatic Brain Injury. General linear models were used to examine injury group differences for age at injury, age at baseline, and zSES. Chi square tests were used to examine group differences for gender and race.

Group comparisons revealed that the OI group had significantly greater SES z-score than the severe TBI group (p = .01). No other group comparisons were significant.

Measures

Parents completed the Parenting Practices Questionnaire (PPQ) at all six assessments. The PPQ, a 62-item instrument that assesses the extent to which parents rate themselves as engaging in authoritarian, permissive, and authoritative parenting behaviors ⁴² and has been used in studies of adolescents with TBI ²⁶. Parents report the degree to which they engage in each behavior with their child on a 5-point Likert scale (1-Never; 2-Once in a while; 3-Approximately half of the time, 4-Very often, and 5-Always). The authoritative scale is derived from four subscales: warmth and involvement (11 items), reasoning/induction (7 items), democratic participation (5 items), and good natured/easy going (4 items) with scores ranging from 27–155. The permissive scale is comprised of three subscales: lack of follow through (6 items), ignoring misbehavior (4 items), and self-confidence (5 items), with scores ranging from 15–75. Finally, the authoritarian scale is made up of 4 subscales: verbal hostility (4 items), corporal punishment (6 items), non-reasoning punitive strategies (6 items), and directness (4 items) with scores ranging from 20–100. The raw total score for each of these dimensions was used as dependent variables to characterize each parent’s use of each of three parenting styles, originally described by Baumrind ^24,43. Subscale scores were also used as dependent variables in exploratory models. The Life Stressors and Social Resource Inventory (LISRES)⁴⁴ was used to assess life stressors through the study period. The LISRES has established reliability and both concurrent and predictive validity⁴⁴. This interview based measure assesses family stressors in the domains of health, work, spouse, extended family, and friends. A stressor variable was created by averaging the T-scores from all stressor subscales to create an overall index of interpersonal stressors. Interpersonal stressors was used as a covariate in analyses to control for any effects that external psychosocial stressors may have on parenting behaviors.

Statistical Analysis

We performed descriptive analyses to generate means and standard deviations for continuous variables and proportions for discrete variables of baseline demographic variables for all three groups (OI, moderate TBI, and severe TBI). We conducted analyses of variance (ANOVA) for continuous variables and chi-square tests for discrete variables to compare the groups on demographic characteristics. A confirmatory factor analysis was completed to examine the factor structure of the three parenting domains within this clinical population. Γ² <.0001, SRMR = .13, RMSEA = .07. While the fit statistics are less than ideal, internal consistency of the three parenting domains (authoritative, permissive, and authoritarian) calculated at each of the study visits were more acceptable. Cronbach’s alpha values are as follows: authoritative parenting ranged from .88-.92, permissive parenting ranged from .76-.84, and authoritarian parenting ranged from .82-.89.

We employed separate linear mixed models to examine changes in parenting style by injury group over time since injury for each of the three parenting behavior dimensions (authoritative, authoritarian, and permissive). This method allows us to utilize all of the data collected rather than exclude those with missing data at any of the follow-up assessments. We examined the main effects of injury group (OI, Severe TBI, Moderate TBI), time since injury, and their interaction. Because of their documented relationship with parenting, race⁴⁵ (white versus non-white), sex⁴⁶, SES⁴⁷, and interpersonal stressors were statistically controlled for in all models. Time since injury was used as a continuous variable in the models; however, any significant effects of time since injury were further examined by using the discrete visit variable to understand how behavior changed over time.

Exploratory analyses, utilizing the above described models, were also completed to examine the specific aspects of parenting via the subscales of the PPQ. While these exploratory analyses included all demographic factors discussed above, only the effects of injury group, time since injury, and their interaction were interpreted to help elucidate what specific aspects of the broader categories of parenting behavior are influenced by injury type and time since injury. Alpha level for these exploratory models was corrected base on the number of subscales analyzed as part of each composite scale as follows: Authoritative parenting: 4 subscales (warmth and involvement, reasoning/induction, democratic participation, good natured/easy going) p = .0125; Permissive parenting: 3 subscales (lack of follow through, ignoring misbehavior, self-confidence) p = .017; Authoritative parenting: 4 subscales (verbal hostility, corporal punishment, non-reasoning punitive strategies, and directiveness) p = .0125.

Results

Mean raw scores for each of the dependent variables (authoritative, permissive, and authoritarian parenting) for each of the injury groups (severe TBI, moderate TBI, and OI) at each of the assessment points are displayed in Figure 1.

Figure 1.

Mean authoritative parenting, permissive parenting, and authoritative parenting for each injury group over time.

Authoritative Parenting Behaviors

No significant effect of injury group, demographic variables, or time since injury was observed, suggesting that these factors did not influence authoritative parenting behaviors. Interpersonal stressors were significantly related to level of authoritative parenting (F(1,896) = 8.12, p = .005). A median split was used to examine the nature of this difference, and those with greater levels of stress reported lower levels of authoritative parenting (M=113.79, SE = .91) than those with lower levels of stress (M=114.42, SE = .93).

A significant injury group*time since injury interaction was observed for warmth and involvement (F(2,773) = 4.28, p = .01; Figure 2). Parents of children with severe and moderate TBI reported a decline in warmth and involvement over time; however, parents of children with severe injuries reported significantly lower levels at the 18-month visit compared to baseline whereas parents of children with moderate injuries did not report lower levels until 6.8 years post injury. For parents of children with OI, none of the scores for warmth and involvement differed significantly from baseline/pre-injury ratings. As such, parents of children with severe TBI reported lower levels of warmth and involvement than parents of children with moderate TBI at 18-months post injury and parents of children with OI at 6.8 years post injury. No significant effect of injury group, time since injury, or their interaction were noted for the reasoning/induction, democratic participation, or good natured/easy going subscales.

Figure 2.

Mean level of warmth and involvement for each injury group over time, reflecting significant injury group by time since injury interaction.

Permissive Parenting Behaviors

SES (F (1,216) = 5.70, p=.02), time since injury (F (1,756) = 26.59, p<.001), and interpersonal stressors (F (1,916) = 10.39, p=.001) were significantly related to permissive parenting behaviors. Specifically, parents of lower SES (based on median split) reported greater levels of permissive parenting behaviors than those of higher SES (low: M = 32.57, SE = .62; High: M = 31.00, SE = .58), and parents reporting lower levels of stress (based on median split) reported lower levels of permissive parenting (M=31.00, SE = .58) than those with higher levels of stress (M=31.57, SE = .57). Additionally, parents reported decreasing levels of permissive parenting behaviors over time regardless of injury characteristics (Figure 2). Levels of permissive parenting behaviors began declining at 3.5 years post injury, when levels fell significantly below those reported at baseline, t(745) = 2.79, p = .005. A further decline in permissive parenting from 3.5 years to the long term follow up was apparent, with levels reported at the long term follow up significantly lower than those at all previous time points (all ps < .05). Consistent with above results, a significant main effect of time since injury was noted for all subscales that create the permissive parenting scale with a decline in the behaviors noted in all cases. (lack of follow through (F(1,757) = 21.28, p <.0001), ignoring misbehavior (F(1, 790) = 8.99, p = .003), and self-confidence (F(1,763) = 9.62, p = .002). A main effect of injury group was also noted for the ignoring misbehavior subscale (F(2,283) = 4.29, p = .01) such that parents of kids with moderate TBI reported less ignoring misbehavior than parents of children with severe TBI and OI. No injury group by time since injury interaction were noted for any of the subscales.

Authoritarian Parenting Behaviors

Interpersonal stressors were associated with authoritarian parenting (F(1,894) = 18.27, p<.0001) such that those with greater levels of stress reported greater authoritarian parenting (M = 39.55, SE = .70) than those with lower levels of stress (M=38.03, SE = .71). Time since injury was also significantly related to levels of authoritarian parenting behaviors regardless of injury characteristics or demographic variables (F (1,755) = 5.73, p=.02; Figure 2). Specifically, at 3.5 years post injury, parents reported significantly lower levels of authoritarian parenting behaviors than at baseline, t(742) = 2.10, p = .04, and authoritarian behaviors at the long term follow up remained significantly lower than baseline reports, t(746) = 2. 65, p = .008.

Exploratory models examining the subscales of Authoritarian parenting revealed a main effect of time since injury on the corporal punishment scale (F(1,270) = 48.95, p<.0001) with parents of all children, regardless of injury group, reporting a decline in corporal punishment over time. No main effect of injury group, time since injury, or their interaction were noted for the other subscales (verbal hostility, non-reasoning punitive strategies, and directiveness).

Discussion

For the most part, changes in parenting practices over time were not significantly related to injury group. Instead, parenting behaviors changed over time in all injury groups, suggesting that changes in parenting over time may be attributed to the child getting older, or child development more generally, rather than injury characteristics or factors explored in the current analyses. Contrary to hypotheses, as well as Micklewright et al’s ²⁶ prediction, parents reported fewer permissive and authoritarian parenting behaviors over time across all injury groups. One exception to this pattern was the differential effect of TBI on warmth and involvement over time, such that parents of children with moderate and severe TBI reported a decline in warmth and involvement even many years post injury. In combination with the absence of significant increases in authoritative parenting, these results suggest the evolution of a less engaged parenting style as the child progresses through development.

Interestingly, the decrease in permissive and authoritarian behaviors became evident several years (3.5 −6.8 years) post injury. At the final assessment (6.8 years post injury), participants were transitioning into adolescence. While research has demonstrated that the empirical associations found between Baumrind’s parenting styles and child outcomes hold true for adolescents ^48,49, parents may change or adapt their parenting styles as a function of children’s development. Luyckx et al.⁵⁰ examined parenting behaviors over time in a large community sample, and reported that all parents, regardless of parenting style, reported decreased monitoring during adolescence, which the authors interpreted as signaling increasing levels of adolescent independence. Further, all parents reported a decrease in warmth/positive parenting practices during adolescence, again potentially related to adolescents’ developmentally appropriate search for independence and increase in parent-child conflict during this developmental period. Interestingly, parents reported a decrease in both permissive and authoritarian parenting behaviors. One potential explanation for this phenomenon may be that teens often spend more time with their peers outside of the home, thereby limiting parents’ opportunity to be responsive as well as moment to moment interactions with their teens. Because this pattern was noted in all children regardless of injury type and severity, it generates a question regarding the origin/function of this pattern.

The somewhat counter intuitive findings of decreasing permissive parenting and decreasing authoritarian parenting, as well as the limited TBI-related findings, may be also be associated with our use of composite scores of parenting behavior (authoritarian, authoritative, permissive) rather than more specific dimensions of parenting behavior (i.e., warmth and involvement, verbal hostility, directiveness). In fact, when we explored the subscales that made up each of the composite scales, a differential effect of injury over time was noted for warmth and involvement behaviors, such that a greater decline in these behaviors were noted for parents of children with severe TBI compared to the other injury groups. This somewhat contradicts the work of Wade et al³⁰ and Fairbanks et al³¹ that report a diminishing impact of injury on parenting behavior during the first 12-months following injury, suggesting that early childhood TBI may in fact have a persistent and long term effect on aspects of authoritative parenting. Our findings, taken together with findings of Luyckx et al⁵⁰ that report declining levels of warm responsiveness over time for all children, indicate that this decline may be more pronounced for parents of children with TBI, especially severe TBI.

While the current study provides useful information regarding parenting style following pediatric injury, it does have limitations. First, the design did not include a healthy, non-injured control group. While the OI group provided an important point of comparison for understanding the specific effects of TBI relative to injury more generally, children who sustain significant traumatic injuries may demonstrate characteristics, such as a predisposition to disruptive behaviors ⁵¹, that influence parenting style. Therefore, we cannot determine if the reduction in permissive and authoritarian behaviors observed in the present sample is reflective of normative development or associated with having a young child sustain a significant traumatic injury. Additionally, small sample size may have limited out ability to identify significant effects of injury severity over time. Visual inspection of the means suggests that those with severe injuries appear to have less decline in permissive and authoritarian parenting over time resulting in greater permissive and authoritarian parenting scores at the 6.8 year follow up compared to other injury groups. However, the small number of participants with severe injuries may have prevented this pattern from reaching statistical significance. In addition, we know that SES affects parenting, although we controlled for SES when examining group differences over time, this approach is not without limitation. Finally, parenting behaviors are greatly influenced by a number of factors not examined in the current study including cultural background Although we examined SES and stressors and found significant contributions to parenting we were unable to examine other salient factors such as cultural background, and family structure and dynamics. Further research is needed to better understand how these various factors interact to influence parenting behaviors, particularly among parents of children and adolescents with a history of TBI.

In conclusion, while there is an absence of injury group effects on the broader domains of parenting behavior, this does not imply that parenting behaviors after TBI is unimportant or that the association of parenting to outcomes is necessarily the same for children with TBI and other groups. In fact, when looking at specific aspects of each parenting domain, parents of children with severe TBI report declining levels of warmth and involvement than other injury groups. Given the literature supporting the benefits of authoritative parenting throughout development, this subdomain may be a potential intervention target for families of children with severe TBI. While we may expect developmental trends or patterns in parenting behavior similar to those observed in typically developing children, the resulting child and family outcomes may differ for children with TBI. It is important to keep this in mind when working with families of children with a history (even distant history) of TBI. These children may require additional support or supervision at various stages of development than typically developing children. Evaluating parent and family factors continuously following TBI is an important part of fostering positive outcomes throughout recovery.