Post-Traumatic Stress Symptoms in Caregivers and Children with Hydrocephalus

Abstract

Introduction:

Hydrocephalus is a disorder of cerebrospinal fluid dynamics that has traditionally been treated by placement of a ventricular shunt. Shunts are effective but imperfect as they fail in an unpredictable pattern, and the patient’s well being is dependent upon adequate shunt function. The omnipresent threat of shunt failure along with the potential need for invasive investigations can be stressful for patients and caregivers. This study objective is to measure post-traumatic stress symptoms (PTSS) in children with hydrocephalus and their caregivers.

Methods:

A cross-sectional analysis of children with hydrocephalus and their caregivers was conducted. Caregivers completed a measure of their own PTSS (the Post-Traumatic Stress Disorders Checklist for the DSM-V (PCL-V)) and resilience (the Connor Davidson Resilience Scale (CD-RISC)). Pediatric patients rated their own PTSS and resilience using the Acute Stress Checklist for Kids (ASC-Kids) and CD-RISC.

Results:

Ninety-one caregivers completed the PCL-V. Mean score was 17.0 (SD 15.7; median 13.0). 14% scored 33 and above, suggestive of a preliminary diagnosis of post-traumatic stress disorder. There was a statistically significant association between caregiver post-traumatic stress and marital status, child’s race, and caregiver education. 52% of caregivers reported their child’s hydrocephalus as the most significant source of their PTSS. Children did not have markedly elevated levels of PTSS. 41% of caregivers and 60% of children scored in the lowest resilience quartile compared to the general population.

Conclusion:

Results from this study suggest that post-traumatic stress affects caregivers with hydrocephalus, yet levels of resilience for caregivers and pediatric patients are low.

Introduction

Hydrocephalus is a life-altering chronic condition and is the most common reason for neurological surgery in children.¹ Standard treatment of hydrocephalus is surgical placement of a ventricular shunt. However, shunts are not a cure, and frequently become obstructed and fail. Shunt failure is an unpredictable event that typically results in painful symptoms, requires surgical intervention, and offers a real threat to the patient’s life and well-being. Throughout their lifetime, children with hydrocephalus may undergo multiple hospitalizations, invasive investigations, and repeated surgeries. Consequently, hydrocephalus has an adverse impact on the quality of life for pediatric patients².

Post-traumatic stress disorder is a constellation of symptoms that occurs after exposure to a severe or repeated traumatic event. These symptoms are defined in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) and classified into categories of intrusion, avoidance, negative alterations in cognitions and mood, and marked alterations in arousal and reactivity. Symptoms include recurrent, intrusive memories, dissociative reactions, avoidance of distressing memories or external reminders, irritable behavior, sleep disturbance, and problems with cognition¹¹. Post-traumatic stress symptoms (PTSS) are associated with functional impairment, negative impact on quality of life, increased rehospitalization rates, and disease relapses^8,12. Previous studies have shown that medical complexity places pediatric patients and their caregivers at increased risk for post-traumatic stress symptoms.^3-10 The purpose of this study is to measure the prevalence and severity of post-traumatic stress symptoms in children with hydrocephalus and their caregivers.

Methods

Setting

A cross-sectional analysis of children with hydrocephalus and their caregivers was conducted between July 2017 to February 2018. This is a convenience sample, with the enrollment period determined by the availability of the research assistant. Study participants were recruited during the child’s routine neurosurgical clinic visit at a large hospital in the southeastern United States.

Study Population

The study population consisted of two groups: caregivers of children with hydrocephalus, and children with hydrocephalus. Children were eligible for enrollment if they were ages 7 to 21 years, had surgically treated hydrocephalus, and were able to complete the study instruments independently. Assessment of ability to participate was done by both research staff and parent, including consciousness, ability to read or understand survey questions, and overall maturity. Caregivers of children (child age 0-21) with hydrocephalus were eligible for enrollment. Caregivers were eligible for participation even if their children were not eligible or refused and vice versa. Study data were collected and managed using REDCap electronic data capture tools hosted at the University of Alabama at Birmingham¹³.

Data Collection

Institutional Review Board ethical approval was obtained prior to participant enrollment. Consent was obtained by a research assistant trained in research ethics, survey administration, and informed consent. Caregivers completed five questionnaires: The Post-Traumatic Stress Disorders Checklist for the DSM-V (PCL-V) with Criterion A, The Child Stress Disorders Checklist (CSDC), the Psychosocial Assessment Tool (PAT 2.0), the Connor Davidson Resilience Scale (CD-RISC), and the Zarit Burden Interview (ZBI). The PCL-V is a 20-item self-report of traumatic stress symptoms in response to a person’s most traumatic event¹⁴. The Criterion A question in the PCL-V is a free-response field that asks the responder to identify their most traumatic event. The Child Stress Disorders Checklist (CSDC) is a 36-item observer report screening assessment for traumatic stress symptoms in children ages 2-21 years¹⁵. The CSDC yields a total score on a scale of 0-72, with higher scores indicative of more severe traumatic stress symptoms. The PAT 2.0 is a validated parent report screen of psychosocial risk in pediatric health¹⁶. The Zarit Burden Interview (ZBI) is a self-report of burden in caring for a loved one with a medical condition¹⁷. ZBI was initially designed to study caregivers of dependent adults, but it has been used in populations of children with medical needs as well.¹⁸ The CD-RISC is a 10-item measure of one’s own resilience, defined as the ability to cope in the face of adversity¹⁹.

Children enrolled in the study completed two questionnaires: Acute Stress Checklist for Kids (ASC-Kids) and the Connor Davidson Resilience Scale (CD-RISC). The ASC-Kids is a 29-item self-report measure of traumatic stress for children²⁰. The ASC-Kids yields a severity score on a scale of 0-58, with higher scores indicative of more severe stress reactions. The ASC-Kids was completed by all children ages 7-21 years.

Demographic and social variables were collected from the PAT 2.0. These variables included caregiver age, child race, caregiver education, marital status of caregiver, child health insurance type, and child developmental problems. Etiology of hydrocephalus, number of hydrocephalus related procedures, and history of shunt infection were abstracted from a prospectively collected institutional database. Child’s gestational age at birth, total number of non-temporary hydrocephalus procedures, and total number of external ventricular drain placements was collected retrospectively from the electronic medical record. Distance from residence to hospital was obtained using patient’s most recent zip code, as obtained from the electronic medical record. Recognizing that shunts can fail repeatedly, and that this may increase the risk of post-traumatic stress symptoms, we defined a new variable to capture those patients who have a history of a shunt failure cluster. The variable, shunt cluster, was defined as having three or more hydrocephalus operations within a two-month period.

Statistical Analysis

Categorical demographic characteristics were summarized using counts and percentages, while continuous demographic variables and traumatic stress scores were summarized using means and standard deviations. The Kruskall-Wallis test was used to compare the continuous outcomes and stress scores for different categories of a categorical variables that had small numbers in some categories (for example, age of caregivers, etiology of hydrocephalus, and ETV-alone hydrocephalus treatment). In these cases, we used non-parametric tests, as tests for normality may not perform as well as needed with small numbers. For uniformity, we used Kruskall-Wallis for all other categorical variables. Scatterplots were examined, and Pearson correlation was estimated and tested (null hypothesis: correlation=0) to examine the association among pairs of continuous variables. Parametric tests were used for these comparisons because the issue of small numbers noted above was not a concern. In all tables, unadjusted p-values are presented. Analyses were done using SAS version 9.4.

Results

Study enrollment is shown in Figure 1.

Figure 1.

Enrollment Diagram

Post-Traumatic Stress Symptoms

Ninety-one caregivers completed the PCL-V. Mean score was 17.02 (standard deviation 15.72; median 13.0), and scores ranged from 0 to 71 (Table 1). Thirteen caregivers (14%) scored 33 and above, a score suggestive of a preliminary diagnosis of post-traumatic stress disorder. When asked to identify their most traumatic life event, 47 (52%) caregivers identified their child’s hydrocephalus diagnosis or related event. Twenty-four (26%) stated an “other” event, and 19 (21%) declined to state their most traumatic event. Examples of free text responses to PTSD Criterion A question (“Briefly identify your worst event”) can be seen in Table 1. PCL-V scores showed statistically significant correlation with child race, parental marital and educational status (Table 2). Non-white race, less than college education, and marital status of single or without a partner all correlated with significantly higher parental PTSS. No other clinical or demographic factor showed significant association with PCL-V measure of parental PTSS.

Table 1.

PCL-V Criterion A Free Responses

“Briefly Identify the worst event”
Hydrocephalus	“The worst part was when we found out he had hydrocephalus” “First time my sons shunt quit at home” “The first time the doctors told me my son was very sick. He needed surgery right away and then the following months I was informed that the shunt was not working properly” “Long story short my sons shunt was broken for several months and I had to take him to the doctor numerous times before it was addressed” “Finding out my son has hydrocephalus” “Being told about hydrocephalus/something being wrong” “When she had her shunt put in” “ER with infant sick. Beds no rails. Vomiting. Dr's disagreeing over keeping him. Ended up having shunt repair the next morning.” “When he could not wake up cause of headaches” “Shunt failures and revisions” “Christmas Eve 2014 emergency shunt operation” “Shunt placement at 3 weeks old” “When his shunt broke and his head started to swell” “Them opening my son head two times for the same reason” “Etv cpc surgery “[patient name] having a shunt malfunction” “When her temp shunt failed”
Other	“The death of my father” “Lost a sibling in an accident” “Two miscarriages” “A friend's child was murdered” “Domestic violence” “Watching him have a seizure”

Table 2.

Traumatic stress score associations with clinical and demographic variables

	CDSC Completed by caregivers of children ages 2-21 years			ASC-Kids Completed by children ages 8-21years			PCL-V Completed by caregivers of children of all ages
Variable	N	Total Score, Mean (SD)	P-value	N	Total Score, Mean (SD)	P- value	N	Total Score, Mean (SD)	P-value
Caregiver age
Age 21 or over	64	6.1 (7.3)	0.65	33	9.0 (9.7)	0.50	86	16.8 (15.9)	0.45
Under age 21	5	4.9 (8.3)		5	10.2 (9.7)		5	18.2 (10.9)
Child gender
Female	32	6.5 (7.3)	0.39	20	10.0 (10.5)	0.51	36	15.9 (12.8)	0.91
Male	37	5.6 (7.4)		20	7.8 (8.2)		55	17.8 (17.3)
Child age
0-1	-	-	0.24	-		0.89	25	17.7 (17.4)	0.050
1-5	1	1.0		-			29	22.3 (16.4)
5-10	26	8.7 (8.9)		6	6.4 (5.0)		19	11.5 (13.0)
>15	17	5.1 (7.3)		18	11.0 (12.5)		5	21.0 (19.9)
Child race
White	45	6.0 (7.5)	0.98	25	7.7 (7.1)	0.71	59	14.0 (12.9)	0.044
Non-white	24	6.0 (7.1)		15	10.8 (12.3)		32	22.6 (18.7)
Caregiver education
College or higher	36	5.8 (6.8)	0.86	18	10.5 (8.9)	0.20	43	14.0 (14.2)	0.044
Other	33	6.2 (7.9)		20	7.9 (10.2)		48	19.8 (16.5)
Marital status of caregiver
Married/partner	52	6.4 (7.5)	0.22	20	7.3 (7.4)	0.24	55	12.4 (10.9)	0.003
Other	17	4.7 (6.7)		18	11.2 (11.4)		36	24.1 (19.0)
Child’s health insurance
Private insurance	30	5.6 (5.5)	0.55	15	7.4 (7.6)	0.38	29	15.2 (14.4)	0.52
Other	39	6.3 (8.5)		22	10.5 (10.9)		62	17.9 (16.2)
Child with developmental problems			<0.001			0.29			0.11
No	26	2.6 (3.9)		18	6.7 (6.0)		36	13.7 (13.6)
Yes/getting help/sometimes	35	9.0 (8.4)		18	10.9 (11.56)		50	18.6 (15.4)
ETV alone
No	59	6.4 (7.7)	0.059	32	8.1 (9.2)	0.14	74	17.5 (16.4)	0.87
Yes	5	1.6 (3.1)		4	17.0 (12.6)		16	15.3 (12.2)
Etiology
Aqueductal stenosis	2	0.5 (0.7)	0.22	1	4.0	0.64	10	17.4 (15.2)	0.78
Interventricular hemorrhage	19	4.8 (7.3)		8	5.4 (4.4)		23	17.2 (16.8)
Myelomeningocele	17	7.2 (8.6)		8	13.6 (14.7)		28	14.0 (11.7)
Other	27	6.5 (7.2)		20	9.3 (8.7)		30	19.7 (18.1)
History of shunt cluster
No	54	5.9 (7.4)	0.27	28	9.0 (10.0)	0.82	80	16.8 (15.2)	0.43
Yes	9	7.5 (8.5)		7	10.3 (10.2)		9	21.8 (20.3)
History of shunt infection
No	48	6.7 (8.2)	0.54	29	7.8 (7.4)	0.33	82	17.7 (16.0)	0.26
Yes	17	4.0 (4.6)		8	14.2 (15.1)		9	10.8 (9.7)
History of external ventricular drain			0.62			0.24			0.33
No	41	5.7 (7.2)		24	7.4 (7.4)		73	18.2 (16.7)
Yes	23	6.8 (8.1)		12	12.5 (13.2)		17	12.2 (9.5)
Child age (years)	N=63	Rho = −0.11	0.38	N=37	Rho = 0.15		N=90	Rho = −0.12	0.28
Mean	12.1y			14.6y		0.37	5.4y
Std. Dev.	4.2y			3.7y			5.1y
Gestational age at birth (weeks)	N=61	Rho = 0.16	0.21	N=33	Rho = 0.38	0.03	N=86	Rho = −0.003	0.98
Mean	34.0w			34.8w			34.8w
Std. Dev.	6.0w			5.8w			5.5w
Distance from residence to hospital (miles)	N=64	Rho = 0.047	0.71	N=36	Rho = 0.066	0.70	N=90	Rho = −0.14	0.18
Mean	84mi			92mi			92.9mi
Std. Dev.	55			49			64.0
Number of procedures	N=65	Rho = 0.14	0.27	N=37	Rho = 0.15		N = 91	Rho = 0.047	0.66
Mean	4.6			4.4		0.37	2.7
SD	5.4			6.4			3.8

The CSDC provides a parent report of the child’s PTSS. Sixty-nine parents completed the CSDC. Median score was 3.0 (range 0-30, mean 5.9, SD 7.3). Of all variables tested for association (Table 2), only child history of developmental problems showed a statistically significant association with CSDC scores. Children with developmental delay were scored by their caregivers as having greater PTSS than those without (total CDSC score 9.0 versus 2.6, p<0.001).

The ASC-Kids is a child self-report of the child’s PTSS. Forty children completed the ASC-Kids. Median score was 5.0 (range 0-44, mean 8.9, SD 9.3). Among all variables tested, only gestational age at birth showed a statistically significant correlation with child self-assessment of PTSS (Table 2). Children with greater gestational age identified slightly higher levels of PTSS (rho = 0.38, p= 0.03). We performed two separate measurements of PTSS in children, a self-assessment and a parental assessment. Thirty-two children had complete data on both measures. We observed no correlation between caregiver-reported and child self-reported PTSS (rho=0.0095, p=0.96).

Resilience and Caregiver Burden

One hundred twenty-three parents completed the CD-RISC self-assessment of their own resilience. Scores ranged from 0 to 40, with mean resilience score of 30.0 (SD 8.3). CD-RISC scores are reported in quartiles, with cutoff values to compare respondents to the general population. Fifty-one caregivers (41%) had a score of 29 or less, placing them in the lowest resilience quartile compared to the general population (Table 3). Forty children completed a self-assessment of resilience with the CD-RISC. The mean score was 24.1 (range 0-40, SD 12.6). Twenty-four children (60%) scored in the lowest quartile of resilience compared to the general population. We used the Zarit Burden Interview to quantify the burden experiences by parents in caring for children with hydrocephalus. Median ZBI score was 6.0 (range 0-44, mean 8.1, SD 7.8).

Table 3.

Caregiver and Child Resilience Scores

	Mean (SD)	Range	Participants in Q1 *Least resilient	Participants in Q2	Participants in Q3	Participants in Q4 *Most resilient
CD-RISC Parent N=123	30.1 (7.2)	0-40	51 (41%)	21 (17%)	22 (18%)	29 (24%)
CD-RISC Child N=40	24.1 (12.6)	0-40	24 (60%)	6 (15%)	2 (5%)	8 (20%)

Broadly, we observed a weak negative correlation between resilience and PTSS. Among 74 caregivers who completed both PCL-V and CD-RISC, higher resilience was significantly associated with lower severity of PTSS (rho=−0.41, p<0.001). Fewer observations were available for analysis of child resilience versus PTSS. In 38 children who completed both ASC-Kids and CD-RISC, no significant correlation was observed (rho=−0.24, p=0.15). Similarly, no significant association was seen between parent-assessment of a child’s PTSS and the child’s self-assessment of resilience (32 pairs of caregivers CDSC and child CD-RISC: rho=−0.055, p=0.76). Finally, we analyzed the relationship between burden of care and PTSS, measured by ZBI and PCL-V, respectively. There was a weak positive correlation between caregiver burden and PTSS (rho=0.27, p=0.02).

Discussion

Post-Traumatic Stress Symptoms and Pediatric Hydrocephalus

To the authors knowledge, this is the first study of post-traumatic stress in children with hydrocephalus and their caregivers. Results indicate that post-traumatic stress is a significant problem for caregivers of children with hydrocephalus, as 14% of respondents identified post-traumatic stress symptoms significant enough to warrant a formal diagnosis of post-traumatic stress disorder (PTSD). In addition, more than half of parents identify their child’s hydrocephalus as the most significant traumatic event of their entire lives. Published studies of parents in other complex health-care-related scenarios, such as caring for a child with cancer, bone marrow transplantation, or experiencing pregnancy loss, have shown similar severity of post-traumatic stress symptoms²¹. These results show that there is an important need for psychosocial support for pediatric hydrocephalus caregivers. Single parents, parents of minority children, those without a college degree may be at higher risk for significant PTSS.

Children with hydrocephalus display post-traumatic stress symptoms as well, though they may not be as severely affected as their parents. Unfortunately, few previous studies have used the CSDC or ASC-Kids, so there are few data available for comparison. Two studies of child self-report PTSS after traumatic injury found ASC-Kids scores higher than our sample. ASC-Kids scores of 18-26 were seen in 35 children hospitalized after injury, and mean ASC-Kids scores of 13.2 (SD 7.1) in 254 children after hospitalization or injury, compared to mean scores of 5.0 in our sample ^19,22. Likewise, our sample showed similar ASC-Kids scores to the control group (ie. No history of sexual assault) in a study of sexual assault victims²³. Comparing parent-report of child PTSS using the CSDC, we find similar mean score in our population (5.9) compared to a sample of children with single-ventricle congenital heart defects (mean score 7.0)²⁴. While these results suggest that PTSS may not be as profound of an issue in children with hydrocephalus as in parents, there are still some children who may benefit from attention to the psychological comorbidities of hydrocephalus.

Resilience, Burden, and PTSS

Resilience, in the context of psychological response to trauma, is the process by which individuals utilize available resources to maintain well-being in the face of stress²¹. Studies of children with illness have shown that children and caregivers can display widely variable resilience^25-28. In the present study, we find that both children with hydrocephalus and their parents are more likely than expected to score in the lowest quartile for resilience, compared to the general population. Compared to a sample of children with cancer, children with hydrocephalus have lower CD-RISC scores (mean 24 for hydrocephalus, 28 for cancer)²⁷. This is particularly important because it has been shown that resilience is a teachable skill: Children and parents can undergo training to improve resilience²⁷. Therefore, this may represent a specific target for intervention in children with hydrocephalus. Furthermore, in our previous study of quality of life, we found that psychological comorbidities were more closely associated with health-related quality of life than any clinical or demographic factor. Targeting resilience may be part of an approach to improve quality of life for children with hydrocephalus as well.

Finally, we observe a weak but statistically significant correlation between measures of caregiver burden and post-traumatic stress symptoms in parents. Correlation is not causation, so we cannot say if PTSS causes increased perception of burden or vice versa. However, we hypothesize that interventions targeted to reduce PTSS may help lessen the perception of burden for parents caring for children with hydrocephalus.

Limitations

This study is a cross-sectional analysis of a single clinic population for which the period of observation was limited. The decision to supplement parental metrics for PTSS with the PCL-V was made after the study had already commenced because there was no means of assessing parental PTSS prior to instituting the PCL-V questionnaire. This is the reason that there are more parent responses to the CD-RISC than the PCL-V. While the late addition has some potential to introduce some bias, we feel this is unlikely and is offset by a markedly more complete assessment of parental PTSS. Recall and expectancy bias must be considered. Being asked about stress levels while attending a clinic visit focused on one’s child’s hydrocephalus may enhance the recall and perceived importance of hydrocephalus in a parent’s answers. Researcher expectancy bias is also a consideration. Our hypothesis was that PTSS is a relevant concern. Therefore, subtle non-verbal cues by the research team may influence how caregivers respond.

Children with hydrocephalus were only able to participate if they were deemed able by parents and researchers. Therefore, capable children with mild cognitive delay may have been excluded. However, the parent-report of child PTSS (CSDC) was administered regardless of the child’s intellectual abilities and shows results similar to child self-assessment of PTSS. Therefore, we believe it is reasonable to conclude that post-traumatic stress symptoms are less prevalent in children with hydrocephalus than in their parents. Finally, both the CSDC and ASC-Kids have been used in relatively few previous studies. Therefore, we are limited in our ability to interpret results from these metrics.

Conclusion

This is the first study to examine the prevalence and severity of post-traumatic stress symptoms in children with hydrocephalus and their parents. While children do not display significant PTSS more than control groups from previous studies, parents appear to have significantly elevated levels of PTSS. Importantly, when asked what it the most important traumatic event of their lives, over half of parents reply that it is their child’s hydrocephalus. In addition, we find evidence of low resilience in both parents and children, suggesting a potential future target for intervention.

Abbreviations:

PTSS

post-traumatic stress symptoms