Caregiver Burden,Stress,and Relationship Cohesion Among Self-Identified Caregivers of Children with Pediatric Acute-Onset Neuropsychiatric Syndrome

Abstract

Objective:

Children with Pediatric Acute-onset Neuropsychiatric Syndrome (PANS) experience sudden onset neuropsychiatric symptoms after infection or other triggers. Symptoms range from mild to severe, potentially lasting days, weeks, months, or longer. Exacerbation-related functional decline presents in many aspects of daily life, generally accompanied by family stress and caregiver burden. We sought to investigate the relationship between severity of PANS symptoms and caregiver burden/stress and the relationship between severity of PANS symptoms and degree of caregiver/child cohesion.

Methods:

This cross-sectional online study surveyed caregivers recruited from PANS-related social media support sites. The Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version (PNSS) measured current severity. Caregiver Burden Inventory (CBI) and Caregiver Self-Assessment Questionnaire (CSAQ) assessed caregiver burden/stress. Inclusion of Other in the Self (IOS) scale determined caregiver-perceived current and desired cohesion with their child(ren) with PANS.

Results:

Of the 216 respondents 79.6% exceeded CBI threshold indicating need for respite in adult care receiver populations. On the CSAQ, 72.9% expressed high distress, 80.5% reported feeling overwhelmed, and 58.1% reported crying spells, meeting cutoffs for support/respite used in adult care receiver populations. Most caregivers reported not having the desired degree of cohesion with their child on the IOS (85.5%). Parents of children with more severe PNSS symptoms fared significantly worse on all measures (CBI: H = 57.83; CSAQ: F = 29.26; IOS: H = 38.04; p < 0.001 for all). Content analysis of comments revealed five themes: (1) severe caregiver and/or family emotional distress and trauma; (2) caregivers wondering what happened to their child; (3) lack of awareness and support among health and education professionals; (4) relationship strain with family, friends, and significant others; and (5) financial and/or legal struggles because of their child's diagnosis.

Conclusion:

There is strong need for support and respite for children with PANS and their families. Long-term effects including posttraumatic stress symptoms among family members should be studied.

Introduction

Pediatric Acute-onset Neuropsychiatric Syndrome (PANS) involves rapid onset of obsessive-compulsive disorder (OCD) symptoms and/or avoidant/restrictive food intake disorder (ARFID) along with other neuropsychiatric symptoms, including sensory defensiveness, dysgraphia, urinary frequency, and/or separation anxiety often after infection or other trigger (Swedo et al., 2012), with a range of symptom severity.

Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections (PANDAS), a subtype of PANS, includes rapid onset OCD and tics, along with other neuropsychiatric symptoms. During exacerbation, many have trouble maintaining typical roles, often with deterioration in daily functioning, including eating, personal hygiene and grooming, bowel/bladder management, bathing, handwriting, math, extracurricular activities, organized activities, community and family social participation, and/or free play (Tona et al., 2017).

Caregiving for a child with illness or disability is stressful, often leading to burnout, difficulty coping, and overly cohesive relationships between caregiver and child (Farifax et al., 2019). Experiences of parents of children with PANS are unique due in part to sudden unexpected onset of symptoms, leaving little time for caregiver or child to develop positive coping strategies. Procuring knowledgeable practitioners for diagnosis and effective treatment often takes years, increasing caregiver stress (Calaprice et al., 2017; O'Dor et al., 2022; Tang et al., 2021).

Financial burdens abound as many seek providers beyond their region and insurance network. One large-scale study reported $4000 out-of-pocket expenses on average, and another reported expenses in the $100,000–$200,000 range (O'Dor et al., 2022; Tang et al., 2021). Calaprice et al. (2017) found more persistent PANS symptoms among those with access to care barriers including procuring a knowledgeable practitioner and ability to afford treatments (Calaprice et al., 2023).

PANS family quality of life is often very poor. Parents report feeling traumatized by the experience, with families reporting (1) the feeling that their lives were suddenly turned upside down; (2) a lack of understanding by the medical community; (3) extreme emotional distress; (4) having to make major life alterations; and (5) challenges working with the child's school personnel to implement effective interventions (Demchick et al., 2019).

The entire family system is impacted during exacerbation, with changes in relationship among family members and extreme burden and stress for the caregivers (Dolce et al., 2022). Frankovich et al. (2018) used the Caregiver Burden Inventory (CBI) with 97 parents/guardians of children from one PANS clinic. Caregiver burden was very high, often higher than the average caregiver burden for those caring for individuals with Alzheimer's disease, with degree of burden associated with lag time from symptom onset to treatment.

Burden exceeded recommended thresholds for respite in adult care receiver populations in 49% of the sample at first flare and was higher with more severe patient global impairment and with longer lag in time from symptom onset to start of treatment. Interestingly, CBI scores improved with visits to a multidisciplinary PANS clinic, although it is not clear if this was due to symptom reduction with medical interventions, or to families receiving emotional support from visits.

Expanding the work of Frankovich et al., we aimed to examine caregiver burden and stress among caregivers of children with a range of ages and PANS symptom severity to determine relationship between current symptom severity and caregiver burden and stress from a larger sample of families located across the United States. Working through a family systems lens (Dolce et al., 2022; Olson, 2000) we also sought to explore the degree of cohesion between caregivers and children and the relationship between current symptom severity and cohesion.

Relationship cohesion is generally viewed on a continuum from separate (disengaged) to overly cohesive (enmeshed), with ranges toward the middle often considered ideal (Olson, 2000). We aimed to discern the current perceived degree of cohesion (caregiver-perceived closeness with the child) as well as the caregiver-reported desired degree of cohesion to determine if the current degree of cohesion is the preferred degree of cohesion.

Specifically, we sought to answer the following questions:

To what extent do caregivers of children with PANS experience caregiver burden and stress?

Is there a relationship between the level of caregiver burden and stress and current severity of PANS symptoms in the child?

Do caregivers of children with PANS experience more or less cohesion with the child than desired?

Is the degree of satisfaction with the cohesion, or lack thereof, in the caregiver–care recipient relationship related to the current severity of PANS symptoms the child is experiencing?

Materials and Methods

This cross-sectional exploratory study used an online anonymous survey of caregivers of children with PANS using the Research Electronic Data Capture (REDCap) platform, which did not collect IP addresses. After approval from the University at Buffalo Institutional Review board, the survey was open from January through March 2022.

We used convenience and snowball sampling, posting announcements on PANS support group websites, newsletters, and Facebook pages. To start the anonymous survey (written in English), participants completed eligibility screening questions asking if they were >18 years and if they were the caregiver to at least one child aged 5–21 years, who had been diagnosed with PANS or PANDAS by a physician or other health care provider. Eligible participants viewed the consent form, providing electronic consent before participating.

The main survey began with demographic questions, including the number of children who were diagnosed with PANS. Conditional logic allowed questions for each child for parents with more than one child with PANS, allowing us to establish that each child reported on lived in the home with the caregiver.

Instruments

We used the Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version (PNSS; Leckman et al., 2012), asking caregivers to rate severity of symptoms within the past week on 30 five-point Likert scale questions, from 0 (no symptoms) to 5 (always present and cause extremely severe difficulty and stress).

Caregiver burden and stress were measured with two standardized assessments, the CBI (CBI; Novak and Guest, 1989) and the Caregiver Self-Assessment Questionnaire (CSAQ; American Medical Association, 2015), asking caregivers to consider their feelings and experiences for the past week. The CBI measures caregiver burden, emphasizing the caregiver without considering the symptomatology of the child. The 24 questions are rated on a five-point Likert scale with 0 indicating no disruption in caregiver's daily living and 5 indicating extreme disruption.

Scores of 36 or higher indicate a need for respite services in adult care receiver populations (no cutoff provided for pediatric population). Farmer et al. (2018) analyzed psychometric properties of the CBI in 104 families of children with PANS and report an overall good fit of the model (root mean square error of approximation −0.061), with significant elevation of CBI scores for parents of children whose illness necessitated a change in schools (Cohen's d = 0.75) or parental work reduction (Cohen's d = 0.65).

The CSAQ is an 18-item survey with the first 16 questions rated as either “yes” or “no” and the final two items scored on a rating scale of 1–10. CSAQ assesses effects of caregiving and overall levels of stress caregivers experience.

Caregivers answering “yes” to 10 or more questions, “yes” to either or both questions 4 (felt completely overwhelmed), and 11 (had crying spells) or scoring 6 or higher on either question 17 (current level of stress) or 18 (current health compared with last year) are considered experiencing a high degree of distress and should consider self-care, support, and respite in adult care receiving populations (no cutoff provided for pediatric population). CSAQ has been shown to have good internal consistency (Cronbach's alpha = 50.82; Epstein et al., 2010).

To measure cohesion between caregivers and children, we used the Buffalo Modification of the Inclusion of Other in the Self (IOS) scale. As Figure 1 shows, the IOS consists of seven sets of overlapping circles representing cohesion between the caregiver and child, with one indicating no cohesion (circles adjacent) and seven indicating extreme cohesion (largely overlapped). The first caregiver prompt mirrored the original IOS measure (Aron et al., 1992) stating: The sets of circles below represent how close your relationship is with your child. One circle represents yourself, and the other circle represents your child. Please select the set of circles that best matches your current relationship with your child.

FIG. 1.

Buffalo Modification-Inclusion of Other in the Self (BM-IOS) scale. PANS, Pediatric Acute-onset Neuropsychiatric Syndrome.

In keeping with other studies at the University at Buffalo, we modified this tool by then asking participants to identify their desired degree of cohesion (Merrill et al., 2023). The desired degree was subtracted from the current degree to identify an IOS difference score. IOS difference scores can range from −6 to +6, with a score of zero indicating no discrepancy (ideal cohesion) and larger absolute values indicating larger discrepancy. Relationships more enmeshed than desired have positive scores; relationships more disengaged than desired have negative scores.

Data analysis

Surveys with missing data were excluded for all analyses. Using PNSS scores for PANS severity, quintiles were established with 20% of the sample in each level (Table 1). Level 1 represents lowest severity and Level 5 highest. For participants with more than one child with PANS, the child with the highest PNSS scores was used. Using SPSS 29.0 (IBM Corp., 2022), we used nonparametric statistics (Kruskal–Wallis with pairwise comparisons and Quade nonparametric ANCOVA) for research questions two and four to account for minor variations in the normal distribution of scores.

Table 1.

Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version Quintile Cutoff Scores

Level	Minimum score	Maximum score	N
1	1	32	39
2	33	49	40
3	50	63	42
4	65	92	41
5	94	150	41
Total			203

We report both original p-values and Benjamini–Hochberg False Discovery Rate (FDR) corrected p-values correcting for 33 analyses (Benjamini and Hochberg, 1995). Respondents were invited (but not required) to provide additional comments. To analyze responses to these open-ended comments, conceptual content analysis was used (Columbia University, 2022).

Results

Demographic characteristics

In all, 216 surveys were completed: 179 U.S. respondents from 25 states, and 37 international respondents (20 Canadian; 7 United Kingdom; 3 Australian; 2 each from Ireland, Israel, and Sweden; and 1 Argentinian), which we decided to keep, although they were not targeted in recruitment. As Table 2 shows, most participants were mothers (94.4%), and most children with PANS were boys (61.0%). In all, the 216 respondents were parents to 427 children, 246 of whom were diagnosed with PANS. Most residences included one child with PANS (85.6%) and two adult caregivers (72.2%).

Table 2.

Demographic Information

	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)
Parent/guardian	Mother	Step mother	Father	Nonbinary step parent	n (%)	n (%)
Relationship (N = 216)	206 (95.4)	1 (0.5)	8 (3.7)	1 (0.5)

Children	One	Two	Three	Four	Five	Six
Children in the home (N = 427)	63 (29.2)	104 (48.1)	34 (15.7)	11 (5.1)	2 (.9)	2 (.9)
Children with PANS diagnosis in home (N = 249)	185 (85.6)	29 (13.4)	2 (0.9)	0 (0)	0 (0)	0 (0)

Children with PANS	Female	Male	Nonbinary/other
Sex of children with PANS (N = 246 reported)	100 (40.7)	146 (59.3)
Gender of children with PANSs (N = 246 reported)	89 (36.1)	150 (61.0)	7 (2.8)

PANS, Pediatric Acute-onset Neuropsychiatric Syndrome.

Caregiver burden and stress

Regarding research question one, To what extent do caregivers of children with PANS experience caregiver burden and stress, 79.6% scored 36 or higher on the CBI, surpassing cutoff to determine respite need in adult care receiver populations. Regarding the CSAQ, we analyzed all thresholds indicating high degree of distress and needing respite. Of the 207 parents completing the entire CSAQ, 72.9% scored in the high distress range with scores ≥10.

Next, of the 215 people who responded to questions 4 and 11, 80.5% answered yes to question 4 (caregiver felt completely overwhelmed), 58.1% responded yes to question 11 (caregiver had crying spells), and 56.3% answered yes to both. Finally, of the 215 who completed questions 17 and 18, 76.3% reported a score of ≥6 for question 17 (caregiver's current level of stress), whereas 45.6% reported a score >6 for question 18 (caregiver's current health compared with last year), all meeting cutoff for distress, support, and need for respite in adult care receiver populations.

For our second research question, Is there a relationship between the level of caregiver burden and stress and current severity of PANS symptoms in the child?, we first used Kruskal–Wallis ANOVAs to analyze difference in CBI and CSAQ scores based on the number of children in the home, the number of children with PANS, the child's sex, and the age of the child with PANS, to determine if any were potential covariates.

Of these, only age of the child with PANS was significant (H = 29.40, df = 17, p = 0.031) for CSAQ, so age was used as a covariate in CSAQ analysis. As Table 3 shows, CBI and CSAQ scores increased with each quintile of PANS severity; parents of children with higher PNSS levels report greater caregiver burden. A Kruskal–Wallis one-way ANOVA exploring effect of PNSS levels on CBI scores (N = 179, H = 57.83, df = 4, p < 0.001, FDR-corrected p < 0.001), and a Quade nonparametric ANCOVA exploring the effect of PNSS levels (with child's age as covariate) on CSAQ scores (N = 191, F = 29.26, Degrees of Freedom, Hypothesis (DFH) = 4, Degrees of Freedom, Error (DFE) = 190, p < 0.001, FDR-corrected p < 0.001) both revealed significant differences based on PNSS levels (Table 4).

Table 3.

Caregiver Burden Inventory, Caregiver Self-Assessment Questionnaire, and Inclusion of Other in the Self Difference Scores for Each Level of Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version

PNSS quartile level	N	Mean score	Standard deviation	Standard error	Minimum	Maximum
CBI
1	34	33.79	14.58	2.50	8	63
2	34	43.29	19.68	3.37	3	82
3	39	52.87	15.78	2.48	22	82
4	36	58.81	15.22	2.54	21	87
5	36	66.64	17.65	2.94	32	94
Total	179	51.39	19.98	1.49	3	94
CSAQ
1	38	7.26	3.25	0.53	2	13
2	38	9.71	3.28	0.53	2	14
3	41	11.61	2.85	0.45	5	16
4	38	12.63	2.05	0.33	7	16
5	41	13.46	1.87	0.29	9	16
Total	196	10.98	3.49	0.25	2	16
IOS difference scores
1	39	1.32	1.45	0.23	0	5
2	40	1.37	1.27	0.20	0	4
3	42	1.80	1.25	0.19	0	4
4	41	2.93	1.25	0.20	0	5
5	40	2.48	1.50	0.24	0	5
Total	202	1.99	1.48	0.10	0	5

CBI, Caregiver Burden Inventory; CSAQ, Caregiver Self-Assessment Questionnaire; IOS, Inclusion of Other in the Self; PNSS, Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version.

Table 4.

ANOVA/ANCOVA Difference in Caregiver Burden Inventory, Caregiver Self-Assessment Questionnaire, and Inclusion of Other in the Self Scores by Pediatric Acute-Onset Neuropsychiatric Syndrome Symptom Severity Level

Measure	N	Analysis	Statistic	Degrees of freedom	Significance	FDR adjusted significance
CBI	179	Kruskal–Wallis ANOVA	H = 57.83	4	<0.001^***	<0.001^***
CSAQ	191	Quade nonparametric ANCOVA	F = 29.26	DFH = 4DFE = 201	0.00^***	0.00^***
IOS	202	Kruskal–Wallis ANOVA	H = 38.04	4	<0.001^***	<0.001^***

***

p ≤ 0.001.

CBI, Caregiver Burden Inventory; CSAQ, Caregiver Self-Assessment Questionnaire; DFE, Degrees of Freedom, Error; DFH, Degrees of Freedom, Hypothesis; FDR, False Discovery Rate; IOS, Inclusion of Other in the Self.

Post hoc pairwise comparisons revealed a direct relationship between symptom severity levels and caregiver burden/stress for both CBI and CSAQ (Table 5). As Table 5 and Figures 2 and 3 show, caregivers bore greater caregiver burden/stress according to the CBI and CSAQ with each quintile of increased symptoms severity. Significant differences were seen between nearly every level with both the original p-value and the FDR-corrected values.

FIG. 2.

CBI scores for each Pandas Neuropsychiatric Symptom Severity - Parent Version (PNSS) level. CBI, Caregiver Burden Inventory.

FIG. 3.

CSAQ scores for each Pandas Neuropsychiatric Symptom Severity (PNSS) level. CSAQ, Caregiver Self-Assessment Questionnaire.

Table 5.

Pairwise Comparison of Caregiver Burden Inventory, Caregiver Self-Assessment Questionnaire, and Inclusion of Other in the Self Scores for each Level of the Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version

CBI scores
PNSS levels	Test statistic	Std. error	Std. test statistic	Sig.	FDR Adj Sig
1.00 vs. 2.00	−25.176	12.564	−2.004	0.045	0.060
1.00 vs. 3.00	−50.239	12.154	−4.133	0.000^***	<0.001^***
1.00 vs. 4.00	−67.815	12.388	−5.474	0.000^***	<0.001^***
1.00 vs. 5.00	−83.509	12.388	−6.741	0.000^***	<0.001^***
2.00 vs. 3.00	−25.062	12.154	−2.062	0.039	0.054
2.00 vs. 4.00	−42.638	12.388	−3.442	0.001^***	0.001^***
2.00 vs. 5.00	−58.333	12.388	−4.709	0.000^***	<0.001^***
3.00 vs. 4.00	−17.576	11.973	−1.468	0.142	0.162
3.00 vs. 5.00	−33.27	11.973	−2.779	0.005^**	0.009^**
4.00 vs. 5.00	−15.694	12.21	−1.285	0.199	0.205

CSAQ scores
PNSS levels	T	DF	Sig.	FDR Adj Sig
1.00 vs. 2.00	−2.735	190	0.007^**	0.010^**
1.00 vs. 3.00	−5.829	190	0.000^***	<0.001^***
1.00 vs. 4.00	−7.63	190	0.000^***	<0.001^***
1.00 vs. 5.00	−9.592	190	0.000^***	0.000^***
2.00 vs. 3.00	−3.042	190	0.003^**	0.004^**
2.00 vs. 4.00	−4.895	190	0.000^***	<0.001^***
2.00 vs. 5.00	−6.822	190	0.000^***	<0.001^***
3.00 vs. 4.00	−1.945	190	0.053	0.068
3.00 vs. 5.00	−3.871	190	0.000^***	<0.001^***
4.00 vs. 5.00	−1.865	190	0.064	0.078

IOS scores
PNSS levels	Test statistic	Std. error	Std. test statistic	Sig.	FDR Adj. Sig.
1.00 vs. 2.00	−2.753	12.752	−0.216	0.829	0.829
1.00 vs. 3.00	−19.723	12.602	−1.565	0.118	0.139
1.00 vs. 4.00	−62.823	12.675	−4.956	0.000^***	<0.001^***
1.00 vs. 5.00	−46.616	12.752	−3.655	0.000^***	<0.001^***
2.00 vs. 3.00	−16.97	12.52	−1.355	0.175	0.193
2.00 vs. 4.00	−60.07	12.594	−4.77	0.000^***	<0.001^***
2.00 vs. 5.00	−43.862	12.671	−3.462	0.001^***	<0.001^***
3.00 vs. 4.00	−43.1	12.441	−3.464	0.001^***	<0.001^***
3.00 vs. 5.00	−26.892	12.52	−2.148	0.032^*	0.046^*
4.00 vs. 5.00	16.208	12.594	1.287	0.198	0.205

p ≤ 0.05; ^** p ≤ 0.01; ^*** p ≤ 0.001

CBI, Caregiver Burden Inventory; CSAQ, Caregiver Self-Assessment Questionnaire; FDR Adj Sig., false discovery rate adjusted significance; IOS, Inclusion of Other in the Self; PNSS, Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version; Sig., significance.

Caregiver–child cohesion

Regarding research question 3, Do caregivers of children with PANS experience more or less cohesion with the child than desired, 241 IOS scores were completed on the 246 children with PANS. Of those, 14.5% showed no difference between the caregivers' desired and current level of cohesion with their child (ideal), whereas 22.0% received a negative IOS difference score, indicating the caregiver wished to have a more cohesive relationship with their child. Most cases (64.3%), received a positive IOS difference score, indicating caregivers are more enmeshed with their child than desired and wish to have a less cohesive (more independent) relationship.

For our fourth research question, Is the degree of satisfaction with the cohesion, or lack thereof, in the caregiver–care recipient relationship related to the current severity of symptoms the child is experiencing, we conducted Kruskal–Wallis ANOVAs to analyze effects of the number of children in the home, the number of children with PANS, the child's sex, and the age of the child with PANS on IOS difference scores to identify possible covariates.

No significant effects were found. As Table 3 shows, the IOS difference scores increased with every quintile of PANS severity except quintile 4 to quintile 5. A Kruskal–Wallis ANOVA using absolute values of IOS difference scores as the dependent variable revealed significant differences among symptom severity levels and IOS (N = 202; H = 38.04, df = 4, p < 0.001, FDR-corrected p < 0.001; Table 4).

Caregivers of children experiencing higher levels of symptoms displayed greater differences between current levels of cohesion and desired levels of cohesion when compared with caregivers of children with PANS whose current symptoms are less severe at nearly every level (Fig. 4), and significant differences were seen between 5 of 10 PNSS levels. Interestingly, children at level 4 had higher absolute IOS difference scores than children at level 5, although the difference was not significant.

FIG. 4.

IOS absolute difference scores for each Pandas Neuropsychiatric Symptom Severity - Parent Version (PNSS) level. IOS, Inclusion of Other in the Self.

Conceptual content analysis

The final survey question, an open-ended optional question asking participants if there was anything else they would like us to know, yielded 122 comments. We are including a synopsis of these in this report as these comments contextualize the challenges parents face, impacting degree of caregiver burden. We reviewed all comments and identified fragments (topics), as some participants provided comments on more than one area of concern.

Two researchers identified five thematic categories from the response fragments: (1) severe caregiver and/or family emotional distress and trauma (72 fragments); (2) caregivers wondering what happened to their child (56 fragments); (3) lack of awareness and support among health and education professionals (47 fragments); (4) relationship strain with family, friends, and significant others (26 fragments); and (5) financial and/or legal struggles because of the child's diagnosis (25 fragments).

Severe caregiver and/or family emotional distress and trauma fragments included the following:

Hard to complain because he's in remission. But the scars and worry are always there for his mother and me. And probably his sister too. Literally every day I watch for anger, anxiety and his finger tapping. We think it's fully gone, but there's no forgetting what it does to your child or family.

Another stated:

I have been a caregiver for my husband through severe health issues, and continue to care for my disabled, unwell mother-in-law. I was involved in the care of two aunts who died quite young of cancer and helped to care for a cousin through a bone marrow transplant. None of these come close, not even the tiniest bit, to the level of caregiving stress that comes with this disease.

Fragments coded as caregivers wondering what happened to their child included the following:

Many times I have been attacked and had serious physical altercations from his rage. Large chunks of hair have been ripped from my head, clothing torn off of me, earrings ripped from my ears, he has tried gouging my eyes and I have been bitten so hard it has broken the skin several times. We are constantly walking on eggshells around him.

Lack of awareness and support among health and education professionals included this fragment:

In addition to the strain this disease puts on the patient and the family unit, it is magnified by the lack of information, lack of care and the general gaslighting by the medical community. It's given us all PTSD and quite frankly is a crime.

Relationship strain with family, friends, and others included the following:

This illness is hell on earth for the child and their family… The years of caring for my son has left me exhausted emotionally, physically, and spiritually. My relationship with my husband has suffered greatly. He shut down and has never been involved in my son's healing. I have no family or support system. This has been 100% on me.

Twenty-five caregivers commented on financial and/or legal struggles due to their child's diagnosis. For example, “I paid over $10,000 out of pocket seeing so many different doctors then having to find ones out of state to try and help me diagnose and treat him. I don't wish this on anyone.”

Discussion

Our research supports the work of Frankovich et al. (2018) who collected data using the CBI but did so on a smaller sample from one clinic, over 3 years on at least two occasions, with at least one during a disease flare. From their data, three themes emerged: (1) caregivers suffer high burden overall, (2) the severity of PANS exacerbations increased caregiver burden, and (3) initiation of care from a multidisciplinary clinic reduced burden.

Our study validated this information with a large cross-sectional sample, establishing a clear relationship between severity of a child's symptoms and degree of caregiver stress/burden. Caregivers of children with more severe symptoms fare significantly worse than those with less severe symptoms on CBI. Consistent with Frankovich et al. (2018), the vast majority of caregivers in our study met cutoffs for both respite care and high risk of burnout used in adult care receiver populations, although our study found 79.6% of caregivers exceeded the threshold of 36 on the CBI, whereas Frankovich et al. found 49%. Individuals in the Frankovich et al. study attended a multidisciplinary PANS clinic and researchers found decreasing caregiver burden with each clinic visit. Our sample may be a more accurate representation of the general public, many of whom do not have access to such care.

We added CSAQ as a measure of distress and results were equally compelling. The vast majority met cutoff for adult care receiver populations for distress, with a clear relationship between degree of PANS symptom severity and degree of caregiver distress, emphasizing the need for caregiver support—especially during exacerbations. The findings from the CBI and CSAQ were similar, which should be considered for future studies as caregiver time may be reduced with only the CBI, which has been shown to have a good fit for PANS families (Farmer et al., 2018).

Our study uniquely utilized the PNSS and Buffalo Modified IOS scale. The PNSS, although not norm-referenced, proved to be a useful measure of symptom severity, noted by the significant differences in caregiver burden, stress, and relationship cohesion based on the level of severity. More psychometrically validated measures of PANS severity are currently under development and should be considered in future studies.

The IOS is a widely used instrument but was used in this study in a unique way to determine satisfaction with the caregiver/recipient relationship. Using the scale in this manner accounts for differences in expected degrees of cohesion based on age (e.g., babies would likely be more enmeshed than teenagers) and on family characteristics and culture (collectivistic cultures would likely be more enmeshed than individualistic).

We found most respondents (66.8%) desired a less enmeshed relationship with (more independence from) their child with PANS. Interestingly Merrill et al. (2023) used a similar design with 54 caregivers of children with Tourette syndrome (TS), 31.4% of whom were satisfied with their relationship with their child, 25.9% desired a closer relationship, and 42.6% desired a relationship that was less enmeshed. That same study queried 20 caregivers of children with cerebral palsy (CP) finding 20% were satisfied with their relationship with their child, 6.7% desired a closer relationship and 73.3% desired a relationship that was less enmeshed.

Differences between caregivers of children with TS and those with CP may be explained by the degree of dependence and physical challenges faced by a child with CP as opposed to the more behavioral and emotional challenges faced by a child with TS. Percent of caregivers from our study desiring less enmeshment fell between that of the parents of children with TS and CP. This may speak to functional debility and dependence on caregiver during PANS exacerbation, reinforcing the deterioration in independence in many areas of daily living, school function, play, and leisure during PANS exacerbation (Tona et al., 2017).

The five themes derived from the open-ended comments in this study closely mirror themes from the Demchick et al. (2019) qualitative study, noting the lack of acknowledgment from the medical community as a large contributor to family stress. Consistent with Demchick et al. (2019) and Dolce (2022), this study highlights the emotional toll on families with sudden onset of PANS exacerbations leaving them to feel as if their lives have been turned upside down with caregivers wondering what happened to their child.

Inadequate access to quality care is a recurring issue in PANS literature, impacting the patient's symptoms as well as caregiver burden (Calaprice et al., 2023; Calaprice et al., 2017; O'Dor et al., 2022; Tang et al., 2021). With adequate access to care, caregiver burden has been found to decrease (Frankovich et al., 2018), whereas lack of access to care, including longer lag times from symptom onset to diagnosis, is associated with more severe PANS symptoms (Calaprice et al., 2017) and greater persistence of symptoms (Calaprice et al., 2023).

These studies, viewed together, emphasize the need for access to quality PANS care. Yet families continue to be faced with barriers to care, as even caregivers who are medical professionals themselves experience decreased access to care, clinician lack of awareness, skepticism, overdependence on diagnostic testing, and high out-of-pocket expenses (Tang et al. 2021).

Limitations

Our study is an anonymous survey of self-identified caregivers of children diagnosed with PANS by a physician or health care provider. Inherent limitations include not verifying the PANS diagnosis and recruiting from PANS social media sites, which are more likely to be frequented by caregivers whose children are experiencing difficulty than those with milder courses. However, our findings are consistent with the controlled study of a smaller sample of patients from one PANS clinic (Frankovich et al., 2018). In addition, CSAQ and CBI cutoffs were created for caregivers of aging adults and should be viewed with caution. Comments were optional and may represent more severe cases. We did not explore the relationship of type of symptoms to caregiver burden, which may be an important future study.

Conclusions

Our cross-sectional study of 216 caregivers from a wide geographic area established a clear and compelling relationship between severity of PANS symptoms, caregiver stress, and caregiver satisfaction with their closeness with their child, emphasizing the need for multidisciplinary care. Open-ended comments highlight needs for health care practitioner education and family support while navigating the sudden-onset turbulence of exacerbations.

Clinical Significance

Clinicians and educators must become aware of the high degree of burden and stress experienced by caregivers as they weather PANS exacerbations. Recognition, diagnosis, effective interventions, and family support are very much needed, and should be the focus of future programs and studies, in addition to studying posttraumatic stress among families and children to determine duration of PANS exacerbation trauma.

Footnotes

Acknowledgments

We would like to thank Saptarshi Chakkraborty, PhD, assistant professor of biostatistics for his assistance with data analysis through the Clinical and Translational Science Award (CTSA) from the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1TR001412 to the University at Buffalo. We would also like to thank Dr. Denise Calaprice and Dr. Angela Tang for article review, and the Rehabilitation Science Department of the University at Buffalo for their support.

Authors' Contributions

J.T. conceptualized the study. M.T. honed the methodology and oversaw initial data analysis. They both worked with J.A, E.B., C.C., K.K., E.L., and A.R. to develop the methodology, collect data, analyze data, write, and edit the article.

Disclosures

No competing financial interests exist.

References

American Medical Association. The Caregiver Self-Assessment Questionnaire. 2015. Available from: https://www.healthinaging.org/sites/default/files/media/pdf/Caregiver-Self-Assessment-Questionnaire.pdf [Last accessed: November 2, 2023 ].

Aron

, Aron

, Smollan

. Inclusion of other in the self scale and the structure of interpersonal closeness. J Pers Soc Psychol, 1992; 63:596–612. Tool available from Stanford SPARQ tools https://sparqtools.org/mobility-measure/inclusion-of-other-in-the-self-ios-scale

Benjamini

, Hochberg

. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc B (Methodological), 1995; 57(1): 289–300; doi: 10.1111/j.2517-6161.1995.tb02031.x

Calaprice

, Tang

, Tona

. Factors Associated with Symptom Persistence in PANS: Part I—Access to Care. J Child Adolesc Psychopharmacol, 2023; 33(9):356–364; doi: 10.1089/cap.2023.0022

Calaprice

, Tona

, Parker-Athill

, et al. A survey of PANS characteristics and course. J Child Adolesc Psychopharmacol, 2017; 27(7):607–618; doi: 10.1089/cap.2016.0105

Columbia University, Mailman School of Public Health. Content Analysis. 2022. Available from: https://www.publichealth.columbia.edu/research/population-health-methods/content-analysis#:∼:text=Content%20analysis%20is%20a%20research,words%2C%20themes%2C%20or%20concepts [Last accessed: November 2, 2023 ].

Demchick

, Ehler

, Marramar

, et al. Family quality of life when raising a child with pediatric autoimmune neuropsychiatric disorder associated with streptococcal infections (PANDAS). J Occup Ther Sch Early Interv, 2019; 12(2):182–199; doi: 10.1080/19411243.2019.1592052

Dolce

, LaRusso

, Abadia-Barrero

. Disruptions and adaptations in family functioning: A study of families' experiences with PANS/PANDAS. J Child Fam Stud, 2022; 31:790–806; doi: 10.1007/s10826-021-02101-3

Epstein

, Gaudiano

, Hinckley

, et al. Evidence for the validity of the American Medical Association Caregiver Self-Assessment Questionnaire as a screening measure for depression. J Am Geriatr Soc, 2010; 58(2):388–389; doi: 10.1111/j.1532-5415.2009.02701.x

10.

Fairfax

, Brehaut

, Colman

, et al. A systematic review of the association between coping strategies and quality of life among caregivers of children with chronic illness and/or disability. BMC Pediatr, 2019; 19(1):215; doi: 10.1186/s12887-019-1587-3

11.

Farmer

, Thienemann

, Leibold

, et al. Psychometric evaluation of the caregiver burden inventory in children and adolescents with PANS. J Pediatr Psychol, 2018; 43(7):749–757; doi: 10.1093/jpepsy/jsy014

12.

Frankovich

, Leibold

, Farmer

, et al. The burden of caring for a child or adolescent with pediatric acute-onset neuropsychiatric syndrome. J Clin Psychiatry, 2018; 80(1):17m12091; doi: 10.4088/jcp.17m12091

13.

IBM Corp. Released 2022. IBM SPSS Statistics for Windows, Version 29.0. 2022. Armonk, NY: IBM Corp.

14.

Leckman

, Thienemann

, Swedo

, et al. Pediatric Acute Neuropsychiatric Symptom Scale - Parent Version. 2012. Available from: from https://pandasnetwork.org/wp-content/uploads/2018/11/pandas_pans_scale.pdf [Last accessed: November 2, 2023].

15.

Merrill

, Celebi

, Engel

, et al. A comparison of burden among caregivers of children with cerebral palsy and Tourette syndrome. In: Poster presentation at the American Occupational Therapy Association Annual Conference, Kansas City, MO, 2023.

16.

Novak

, Guest

. Application of a multidimensional caregiver burden inventory. Gerontologist, 1989; 29(6):798–803; doi: 10.1093/geront/29.6.798

17.

O'Dor

, Homayoun

, Downer

, et al. A survey of demographics, symptom course, family history, and barriers to treatment in children with pediatric acute-onset neuropsychiatric disorders and pediatric autoimmune neuropsychiatric disorder associated with streptococcal infections. J Child Adolesc Psychopharmacol, 2022; 32(9):476–487; doi: 10.1089/cap.2022.0063

18.

Olson

. Circumplex model of marital and family systems. J Fam Ther, 2000; 22:144–167; doi: 10.1111/1467-6427.00144

19.

Swedo

, Leckman

, Rose

. From research subgroup to clinical syndrome: Modifying the PANDAS criteria to describe PANS (pediatric acute-onset neuropsychiatric syndrome). Pediatr Ther, 2012; 2:1–8; doi: 10.4172/2161-0665.1000113

20.

Tang

, Appel

, Bennett

, et al. Treatment barriers in PANS/PANDAS: Observations from eleven health care provider families. Fam Syst Health, 2021; 39(3):477–487; doi: 10.1037/fsh0000602

21.

Tona

, Bhattacharjya

, Calaprice

. Impact of PANS/PANDAS exacerbations on occupational performance: A mixed-methods study. Am J Occup Ther, 2017; 71(3):7103220020P1-7103220020P9; doi: 10.5014/ajot.2017.022285