Caregiver burden and inflammation in parents of children with special healthcare needs

Children with special healthcare needs (CSHCN) are a vulnerable population that require specialized services and are often cared for by parents. These parents experience psychological, physiological, and potential inflammatory dysfunction related to amplified caregiving burden which may increase with the complexity of the child’s condition. Due to the potential for inflammatory dysregulation, we aimed to compare caregiver burden and inflammation of parents with CSHCN based on the severity of the child’s condition to parents of typically developing children. A cross-sectional design that included parents of typically developing children (n = 60), non-complex chronic disease (n = 28; one chronic condition that does not progress), and complex chronic disease (n = 32) was used. Parents completed the Caregiver Burden Inventory and blood serum was collected to measure inflammation. Multivariate analyses of variance with post-hoc testing was used to determine between group differences. Parents of children with complex disease experienced greater caregiver burden than parents of typically developing children (p < 0.001) and non-complex chronic disease (p = 0.044). Parents of children with non-complex chronic disease reported greater caregiver burden than parents of typically developing children (p = 0.02). Parents of children with complex chronic disease had lower pro- (p = 0.042) and anti-inflammatory (p = 0.002) composite scores, than parents of typically developing children. Parents of children with greater medical complexity experienced more caregiver burden and potential inflammatory dysregulation. Future research should explore inflammatory processes in this specific population and self-care measures to improve psychological and physical well-being.


Introduction
Parents of children with special healthcare needs (CSHCN), are informal caregivers and provide most of their children's in-home care [1][2][3].These parents are more likely to experience caregiver burden which creates an environment of chronic stress [4,5].Consequently, this chronic stress has the potential to increase inflammation [6][7][8].In addition, headaches, exhaustion, decreased physical function, and negative affect are common among parent caregivers [4,5,9].Moreover, child severity may increase caregiver burden and stress [10,11].Hence, the purpose of this study was to examine the

Caregiver burden
Caregiver burden is the perception of complex stress experienced by persons resulting from caring for someone with whom they have a significant personal relationship [27].Caregiver burden has deleterious effects stemming from financial and time constraints, social isolation, and emotional strain which leads to decreased quality of life, illness, poor wellbeing, and unmet medical and social needs [27][28][29].
Severity of the child's illness or condition increases caregiver burden and decreases psychological well-being in parent caregivers [30,31].Autism spectrum disorder (ASD) symptom severity has been found to increase parent depression and caregiver burden [11,32].Similar findings have been reported in parents caring for children with cerebral palsy, leukemia, and cystic fibrosis [30,31].In addition, the severity of the child's chronic condition has been linked to greater ongoing stress, depression, and anxiety [33,34].
Since repeated daily stressors and caregiver burden are common in parental caregiving for CSHCN, they often experience increased rates of depression, stress, and anxiety, putting them at risk for inflammatory dysregulation.Parent caregivers of children with autism and attention deficit hyperactivity disorder had greater levels of CRP than those of typically developing children, indicating inflammation [35,36].Moreover, parents of children with cancer experienced increased IL-6 and TNF-α in serum blood samples [23,37].
In the United States alone, there has been a 6% increase in the CSHCN population since 2001 due to improvements in medical technology, which has increased the survivability of prematurity, diagnosis of atypical neurodevelopment, and chronic illness [38].This inherently increases the need of parents to care for their child with ongoing and greater health-related needs.These parents are significantly at risk for caregiver burden and chronic stress, which may lead to chronic low-grade inflammation [5,7].Subsequently, chronic low-grade inflammation may facilitate the development of cardiovascular disease, insulin resistance, and cancer in caregivers [7,8,39].Since this caregiver population is growing and they may be at an increased risk for health-related problems related to the stress from caregiving, it is essential to better understand the relationships that exist between caregiving and inflammation.

Study aims
Despite these findings, little research exists that explores the relationship between caregiver burden on inflammation in parents of CSHCN.Moreover, few studies have compared differences in caregiver burden and inflammation in parent caregivers of CSHCN and those without.Even fewer studies have examined differences in caregiver burden and inflammation in relation to the severity of the child being cared for.Therefore, our aims to contribute to current research findings are to first explore relationships between caregiver burden and inflammatory biomarkers in parent caregivers of CSHCN.We hypothesize that inflammation will be positively associated with caregiver burden in parent caregivers of CSHCN.
We then aim to compare differences between parent caregivers of CSHCN and those of typically developing children on caregiver burden and inflammation based on severity of the child's condition.We hypothesize that parents of children with more severe conditions will experience greater caregiver burden and increased inflammation compared to parents of children with non-complex chronic disorders and those of typically developing children.In order to accomplish our aims to examine differences in caregiver burden and inflammation in parents of CSHCN and those of typically developing children and further current research, we asked parents to complete a series of questionnaires and provide a serum sample during a single one-on-one visit in this cross-sectional study using the Caregiving Process and Caregiver Burden Among Pediatric Population framework.This framework posits that caregiving demands (i.e.caregiver burden) and the severity of the child's condition and worse behavior negatively influences the physical health, which we measured inflammatory markers to assess, of the parent caregiver [40].

Study design and setting
To compare the child's condition severity on parent caregivers' burden and inflammatory biomarkers, a cross-sectional design was used with all data collected during a single visit.

Inclusion criteria
Participants met inclusion criteria if they were over 18 and the legal guardian or parent of a child under 18 years.Participants were not excluded based on their relationship (mother, father, grandparent, foster parent, etc.) with the child.Parents of typically developing children and those of CSHCN were included in the study.Participants were excluded if they were not a legal guardian of the child, or the child was 18 years of age or older.

Participant recruitment
Participants were recruited via flyers on community boards in libraries, gyms, businesses, and local social media parenting groups.Flyers were also distributed through day cares, schools, and physician offices.Additional recruitment was aided by parent support organizations via their listservs and social media pages.Initial recruitment resulted in 175 participants contacting with interest to participate in the study.Fifty-five either did not meet inclusion criteria, withdrew interest, were unreachable, or failed to show, resulting in 120 parents participating in this study.
Sixty participants were in the control group and 60 in the comparison group.The comparison group was further stratified based on severity of the child's condition using the pediatric medical complexity algorithm with medical diagnosis of the child reported by the participant through structured interview [41].The severity scoring was 0 for parents of typically developing children (i.e.control group), 1 for parents of children with non-complex chronic disease, and 2 for parents of children with complex chronic disease.Non-complex chronic disease is defined as one chronic condition, and can include physical, developmental, or mental health diagnosis, that does not progress, but persists into adulthood with episodes of good health.Examples of non-complex chronic disease are well controlled type-1 diabetes, attention deficit hyperactivity disorder, and asthma.Complex chronic disease was defined as those with considerable physical, developmental, or mental health diagnosis in two or more body systems for more than a year, they can be progressive, technology dependent, or impact life function [41].Examples of complex chronic disease are malignancy, cystic fibrosis, rare congenital disorders, and paraplegia.

Measures
First, demographic data was collected on all participants and included: gender, age, education level, marital status, ethnicity, and income.Next background data was collected which included: caregiver self-reports of diagnosed chronic and psychiatric conditions, medications, sleep health and routines, exercise, and number of and ages for any children in the household.Finally, height and weight were collected and recorded, caregiver burden was assessed, and blood was drawn to examine biomarkers of inflammation.

Caregiver burden
Caregiver burden was measured using the Caregiver Burden Inventory.This is a 24-item measure containing five subscales: time dependence, developmental, behavior, physical burden, social burden, and emotional burden.Scores for each item are based on a five-point Likert scale ranging from 0 (not at all disruptive) to 4 (very disruptive; [42]).Items were summed to create the subscales scores, and the subscale scores were summed to create an overall caregiver burden score with higher scores indicating greater caregiver burden.Good content, concurrent, predictive, and external validity, and good internal and test-retest reliability in caregivers of CSHCN have been reported [43].The overall internal consistency for the scale was good (α = 0.91), as were the five subscales, time (α = 0.85), developmental (α = 0.85), physical health (α = 0.85), emotional health (α = 0.86), and social relationships (α = 0.85).

Inflammation
Approximately, 7 mL of whole blood samples were collected in CPT tubes via antecubital venous puncture for biomarkers of inflammation.Following the research protocols, serum samples were prepared and stored within 12 h of collection in -80 °C freezer following all research protocols.All collection and transportation methods were in accordance with the bloodborne diseases and pathogen requirements and protocols.
Serum levels of inflammatory markers were measured using the Human Essential Immune Response panel LEGENDplex kit following the manufacturer's instructions.The kit consists of a multiplex analysis that quantifies 13 different pro-and anti-inflammatory cytokines in the same sample including: IL-4, IL-2, CXCL10, IL-1β, TNF-α, MCP-1, IL-17A, IL-6, IL-10, IFN-γ, IL-12p70, CXCL8, and TGF-β1.All molecules lay in the assay detection range of 2.4 to 10,000 pg/mL.At least 300 events were acquired per analyte.Intraassay coefficients of variation ranged from 0 to 3.99% and inter-assay ranged from 1 to 1.9%.Analysis was completed using flow cytometry.Gating strategies used for the flow cytometry analysis included stability gating and forward and side scatter gating.Stability gating was used to ensure that there were no instrument issues during the analysis process.The forward and side scatter gating for the cytokines grouped beads based on size with A being the smaller beads and B the larger beads.This allowed for concurrent recognition of all 13 cytokines.Analyte and bead classification were entered in sequential order based on the bead identifications listed in the panels' manuals.

Data collection
Approval from the institutional review board at Florida State University was obtained in May of 2019 and participant recruitment began immediately thereafter.Recruitment and data collection continued through February 2020.Potential participants were contacted, and the study explained.Upon agreeing to participate, participants were scheduled and given the option of meeting at a laboratory located on the university's campus or a setting of their choice (i.e., home, work).Most participants chose to have the researcher meet them in their home.During the scheduled meeting, written consent was reviewed and signed, then the researcher proceeded with data collection.

Statistical methods
First, descriptive statistics were used to summarize the participant characteristics, then they were assessed for skewness.Due to skewness bivariate analyses (chi-square, χ2; ANOVA, independent-samples Kruskal-Wallis U test) were used to compare between group differences on demographic data and to examine influences of potential confounding variables.All significance levels were set at α = 0.05.
Inflammatory cytokines were assessed for non-detectable cytokines and were set to threshold detection levels.Inflammatory cytokines were then assessed for skewness and were found to be positively skewed; therefore, these measures were transformed according to a log-base 10 transformation [42][43][44].The log-transformed biomarkers were evaluated for extreme outliers which were then recoded using 90% winsorization due to the small sample size.Exploratory factor analysis (EFA) was used to create composite scores of inflammatory cytokines given their high levels of inter-correlation.
To calculate composite scores, z-transformations were applied to standardized values and then the mean of z-scores were calculated.Pro-inflammatory composite score included IL-6, TNF-α, and IL-1β, and anti-inflammatory composite score included IL-10 and IL-4.
Correlations between caregiver burden and pro-and anti-inflammatory factor scores and between caregiver group and caregiver burden and pro-and anti-inflammatory factor scores were assessed using Pearson's Point-biserial correlation coefficients.Multivariate analysis of variance (MANOVA) was used to examine the differences between [severity] groups on burden measure scores and both pro-and anti-inflammatory composite scores.Post-hoc tests were selected depending on the violation of homogeneity of variance assumptions (Tukey-Kramer).All analyses were completed in SAS 9.4.

Demographics
All participants were screened, proved eligible, and were included in the analysis.A total of 120 parent caregivers were included in this study, 60 cared for healthy, typically developing children, 28 were parents of children with non-complex chronic conditions, and 32 had children with complex chronic conditions.No data was missing for the psychological measure and a single participant's data was missing for inflammatory markers, due to the inability to obtain the blood sample.The mean age of participants was 38 years and ranged from 26 to 57.Most participants (91.7%) were women and 87.5% were mothers of the child/ren.The median household income was $70,000 and 86.7% of the participants were employed full time.Participants were more likely to be white (89.2%), married (89.2%), and have at least a bachelor's degree (70.8%).Households averaged 2.2 children with a mean age of 7.29 years.
Groups did not differ significantly across participant characteristics except for education level.Controls were also more likely than parent caregivers of children with not-complex chronic disease to have at least a bachelor's degree (83.3% vs 53.6%; p = 0.007), but there were no differences between parents of children with not-complex chronic disease and those whose children had complex chronic disease or controls and complex chronic disease.Across all demographic characteristics, no differences existed between the parents of children with not-complex chronic disease and complex chronic disease.All demographics are presented in Table 1 and children's medical disorders are presented in Table 2.

Exploratory results
To explore relationships among caregiver burden and inflammation we looked at correlations between the caregiver groups, caregiver burden, and pro-and anti-inflammatory factor scores.Groups based on child severity were not significantly correlated with proinflammatory (r = − 0.14, p = 0.12) or anti-inflammatory factor scores (r = − 0.11, p = 0.22).Caregiver group was positively associated with caregiver burden (r = 0.47, p < 0.001).Anti-inflammatory (r = − 0.05, p = 0.57) factors scores were not associated with caregiver burden or any caregiver burden subscale.Proinflammatory composite scores were not associated with total caregiver burden (r = − 0.11, p = 0.21).A significant negative relationship existed between pro-inflammatory composite scores and caregiver burden subscale development (r = − 0.20, p = 0.03).No other relationships between pro-inflammatory composite scores and the subscales of caregiver burden existed.All results are presented in Table 3.

Inflammatory response
Caregiver groups significantly differed across pro-inflammatory cytokine factor scores (F [2, 116] = 3.24, p < 0.05, η 2 = 0.053).Across these, caregiver group explained 5.3% of the variance in pro-inflammatory biomarkers.Post-hoc comparisons indicated that caregivers of children with complex chronic disease children had significantly lower proinflammatory factor scores than parent caregivers of typically developing children (p < 0.05).Whereas parent caregivers of children with not complex chronic disease did not significantly differ from parents of complex disease or typically developing children.
Caregiver groups also differed significantly on the anti-inflammatory factor scores (F [2,116] = 6.08, p < 0.01, η 2 = 0.095).Caregiver group accounted for approximately 9.5% of the variance in anti-inflammatory composite scores.Post-hoc comparisons revealed significant differences between parents of children with complex chronic conditions and parents of typically developing children on the anti-inflammatory composite scores (p < 0.01).Parents of children with complex chronic conditions also had lower levels of anti-inflammatory composite scores than parents of children with

Discussion
We explored relationships between caregiver burden and the inflammatory composite scores in parent caregivers of CSHCN, hypothesizing that caregiver burden would be correlated with the inflammatory biomarkers.We found that no relationships existed between caregiver burden and the inflammatory composite scores.Significant between groups differences were present on caregiver burden between parents of typically developing children and those with complex chronic disease and not complex chronic disease.Parents of children with chronic disease experienced greater burden on time than both groups.Significant between group differences existed in the inflammatory factor scores, parents of children with complex chronic disease had significantly lower levels than parents of typically developing children.

Interpretation
The results of our study are promising and add novel contributions to the current body of research in its use of serum inflammatory biomarkers as a measure of inflammation from chronic stress with caregiving and parents' perceptions of caregiving burden.Our results indicated that parent caregivers of CSHCN reported greater caregiver burden than those of typically developing children.These results were expected as significant research exists that supports caregiving burden is higher in parents of CSHCN [30,31].Uniquely, we explored group differences based on the medical complexity of the child being cared for by the parent.We found significant differences between parents of typically developing children and parents of children with complex chronic disease on caregiver burden time, development, physical health, social relationships, and total burden.Additional differences between parents of typically developing children and parents of children with non-complex chronic disease on caregiver burden development, physical health, social relationships, and total burden.Only, the emotional health subscale of caregiver burden was not significant between the groups; however, this is most likely due to our small sample size or unequal groups.An additional possible explanation for no difference in burden emotional health between groups is regardless of the child's severity or lack thereof, parents consistently report not progressing emotionally, harboring feelings of guilt and regret, and embarrassment related to their child's behavior [45,46].Parents of children with complex chronic disease scored higher on burden time than parents of children with noncomplex chronic disease and typically developing children, while no differences were present between parents of children with non-complex chronic disease and those of typically developing children.Parents of children with complex chronic disease must often learn medical procedures and these intricate procedures require additional time than more basic caregiving tasks [47].Approximately 22.5% of all parents of children with complex chronic disease spend 11 or more hours per week caring for their child compared to 3.9% of parents of children with not complex chronic disease [3], which may explain the differences between these groups.
We found that parents of CSHCN, with both complex and non-complex conditions, experienced greater caregiver burden on their personal development and social relationships compared to their peers.Parents of CSHCN often must delay or sacrifice personal goals they had set for themselves and social relationships due to caregiving demands, triggering feelings of failure in not meeting their own expectations and social isolation [46,48] Parents of CSHCN have reported the desire for their lives to be different or that their child was able to live on their own as they aged [48].They have also reported sacrificing parts of their lives, such as employment, advancing their education, time spent with friends and family, relocating to be closer to needed services for their child [49,50].The sacrifices parents of CSHCN must make to care for their child likely explains the differences in developmental burden and social relationships between parents of typically developing children and those of CSHCN.
Lastly, parent caregivers of children with complex and non-complex chronic disease reported poorer physical health than parents of children without chronic disease.One potential reason for these findings are that social isolation, greater time spent caregiving, caregiver sacrifices and poor mental health are indicated in the worsening physical health of caregivers [9,27,51,52].Since parents of CSHCN in this study reported greater caregiver time commitment, 1 3 poorer social relationships, and more developmental burden than parents of typically developing children, based on prior research, our outcomes of worse reported physical health are expected.
Previous research indicates that caregiving results in chronic low-grade inflammation [15,53].Contrarily, we found that parents of children with complex disease had significantly lower levels of pro-and anti-inflammatory composite scores than parents of typically developing children.Previous research using similar pro-inflammatory cytokine composites found these were increased in adolescents experiencing chronic stress; yet this result may indicate differences in caregiving stress versus other types of chronic stress [54][55][56].Interestingly, our results support the theory of habituation rather than chronic low-grade inflammation in which persistent repeating daily stressors build an inflammatory tolerance decreasing the inflammatory response [24,26].Moreover, our results may be due to elevated glucocorticoid levels, which inhibit suppression of pro-inflammatory cytokine production; however, we did not examine these [23].
Long term stress has been purported to result in the decreased production of anti-inflammatory cytokines IL-10 and IL-4, indicating dysregulation of the inflammatory processes [15,57].We found that parents of children with complex chronic conditions also had significantly lower levels of anti-inflammatory cytokine composite scores than parents of typically developing children.Our results align with previous research that has indicated parents of children with ongoing, chronic conditions experience decreased production of anti-inflammatory cytokines [23,24].
We found that caregiver group had a significant positive relationship with caregiver burden and all associated subscales, this result was expected in that parents of CSHCN consistently report greater experience of caregiver burden than parents of typically developing children.We found caregiver group was not associated with the pro-or anti-inflammatory composite scores.Anti-inflammatory cytokine composite scores were not significantly correlated with caregiver burden or any of the subscales.Caregiver developmental burden had a small yet significant inverse association with the pro-inflammatory cytokine composite scores, meaning that higher developmental burden was associated with lower inflammatory cytokines.Developmental caregiver burden refers to the caregivers' beliefs that they are not developing at the same rate as their peers who do not have the same caregiving responsibilities [58].While pro-inflammatory cytokines are heavily involved with inducing inflammation and long-term exposure to low levels have been indicated in chronic inflammation [59,60].Yet, a habituation-like occurrence may account for the negative association between caregiver burden development and pro-inflammatory cytokines since parents of chronically ill children experience the same stressors every day.

Future directions
Little evidence exists that has explored the effects of complexity of the child's condition on parent caregivers' inflammatory processes.While our findings lend some credence to the "habituation-like phenomenon", future research should further explore the long-term effects of caregiving for chronically ill children on caregiver inflammation.Moreover, the severity of the child's condition should be further explored to determine if greater complexity of care required increases the incidence of the "habituation-like phenomenon" in parent caregivers.
Additionally, these parents may not have been experiencing the effects of inflammation due to their relatively young age, high overall socioeconomic status, education level, and support systems since these act as protective factors against inflammation [21,55,61].Moreover, research has indicated that resilience, coping, and self-efficacy result in decreased levels of IL-6 and TNF-α and positively impacting physical and psychological health [62,63].While we did not assess these measures, it is possible that parents of children with complex chronic disease have developed protective factors against inflammation.It would be prudent for future research to further explore the effects of potential protective factors on parents with medically complex conditions on long-term inflammation and mental well-being.Lastly, it would be pragmatic for future researchers to examine gender and racial differences regarding caregiver burden and inflammation.

Limitations
There are limitations of note in this study.First, the sample was recruited from a single community in the southeast and the sample size was modest.Moreover, our sample was likely affected by gender and racial bias.Participants were predominantly middle class, white, women, which may impact the generalizability of our findings.Like other parent caregiver studies, our sample was largely mothers since most parent caregivers are women.Yet prior research indicates that significant gender differences exist in caregiving.Women are more likely to report greater stress, depression, and burden than their male peers and are more likely to spend greater amount of time caring for their loved-one [64].Our sample was also highly educated, married, and well above the federal poverty line, which may act as protective factors.

Conclusion
Caregiver burden research is widely recognized in elderly caregivers and the effects of caring on inflammation in those caring for the elderly population is rapidly growing.Yet little research has explored caregiver burden and inflammation in parents of CSHCN.Even less research has explored the effects of medical complexity on parent caregivers and their experienced burden and inflammatory processes.We found significant differences between groups on caregiver burden and inflammatory composite scores suggesting that parents of CSHCN experience greater burden and potential inflammatory dysregulation related to caregiving.While promising, additional research is needed to determine the processes behind the inflammation in parents of CSHCN and if it differs from caregivers of the elderly.Moreover, gender effects on these outcomes should be explored.Furthermore, additional research is needed to determine protective factors, coping, resilience, and self-efficacy against inflammatory dysregulation and caregiver burden.Finally, as the number of parents of CSHCN continues to rise, it is critical to explore interventions to alleviate the negative consequences of caregiving.