Predictors and Prospective Course of PANS: A Pilot Study Using Electronic Platforms for Data Collection

Abstract

Objectives:

Little is known about the longitudinal course of pediatric acute-onset neuropsychiatric syndrome (PANS) because existing literature is primarily cross-sectional. To begin to address this gap, two digital platforms were used to prospectively monitor neuropsychiatric symptoms in children with PANS. The aim was to identify baseline clinical characteristics that would predict the course of neuropsychiatric symptoms over 12 weeks. We compared relative compliance between two electronic data acquisition platforms and evaluated agreement between parent–child ratings of symptoms.

Methods:

For 12 weeks, 20 children with PANS and their parents completed weekly rating scales of neuropsychiatric symptoms on Research Electronic Data Capture (REDCap) and concurrently parents completed tri-weekly ratings on My Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections (PANDAS) Chart, a symptom monitoring website. Longitudinal data were analyzed by using regression analyses.

Results:

Greater duration of time between onset of PANS and study enrollment was associated with worsening of parent-rated neuropsychiatric symptoms over 12 weeks (p = 0.05). Higher scores on parents' Caregiver Burden Inventory at baseline predicted that children would report more severe symptoms over the 12-week period (p = 0.01). Compliance rates for parents were 86.3% for the weekly REDCap PANS Symptoms Rating Scale compared with 53.8% for the tri-weekly My PANDAS Chart ratings. There was moderate agreement between children and parents on the PANS Symptom Rating Scale (r = 0.55, p < 0.0001).

Conclusion:

Our study highlights the utility of electronic methods for tracking longitudinal symptoms in children with PANS and suggests that particular baseline characteristics (e.g., delay in identification and treatment of PANS, greater caregiver burden) may be indicative of a differential trajectory of PANS course, with more severe symptoms over the short term. clinicaltrials.gov NCT04382716.

Introduction

Pediatric Acute-Onset Neuropsychiatric Syndrome (PANS) is characterized by the sudden onset of obsessive-compulsive disorder (OCD) and/or restrictive eating with comorbid neuropsychiatric symptoms from at least two of seven categories (anxiety, emotional lability/depression, irritability/aggression/severely oppositional behaviors, behavioral/developmental regression, deterioration in school performance, sensory or motor abnormalities, somatic symptoms) (Swedo et al. 2010; Chang et al. 2015). Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections (PANDAS) is a subtype of PANS, classified by the abrupt onset of OCD and/or tics in a prepubertal child, triggered by Group A streptococcal infection (Swedo et al. 1998). The course is described as relapsing and remitting, with relapses or “flares” triggered by new infections (Murphy et al. 2012; Frankovich et al. 2015; Swedo et al. 2017).

Most characterization studies of children with PANS and PANDAS have been cross-sectional (Swedo et al. 1998; Perrin et al. 2004; Bernstein et al. 2010; Frankovich et al. 2015; Murphy et al. 2015) or retrospective chart reviews (Brown et al. 2017a, 2017b). Thus, little is known about the short-term and long-term course of PANS based on prospective studies. These studies are needed to inform both our basic understanding of symptom course and trajectory, as well as clinical decision making. For example, identifying factors that predict the course of a current episode could enhance patient/family psychoeducation, inform treatment selection, and potentially point to novel treatment targets. Many parents ask whether their children will “outgrow” the condition. Parents desire to know whether it is possible to predict the course of the current episode (e.g., are there characteristics that indicate whether children will improve quickly or get worse?) They wonder whether additional episodes or flares will occur in the near or distant future.

Advances in Internet-based and mobile technologies present unique opportunities for tracking symptoms in youths with mental health conditions. A growing number of websites and smartphone applications (apps) are designed to assess and track clinical symptoms, provide therapeutic interventions, and monitor treatment adherence and outcome across a variety of clinical populations, including mood and anxiety disorders, psychosis, eating disorders, and substance-related disorders (Luxton et al. 2011). Apps and other digital platforms allow users to access mental health assessments and interventions in “real time” in natural environments, thereby decreasing patient burden and strengthening the ecological validity of collected data (Wen et al. 2017). Despite the increasing availability and sophistication of these electronic tools, investigations into the feasibility and benefits of using specific platforms are scarce (Shen et al. 2014; Roberts et al. 2018), especially in pediatric populations (Wu et al. 2016; Dubad et al. 2018). It may be particularly informative to track symptoms in PANS longitudinally since this is a relapsing and remitting disease with heterogeneity of symptoms that can change quickly. Use of longitudinal data will help to establish a PANS diagnosis and detect the onset of new flares. To our knowledge, no studies have used these technologies to track PANS symptoms prospectively.

To test the feasibility of digital symptom assessment and explore the short-term course of PANS, we monitored the neuropsychiatric symptoms of children with PANS prospectively for 12 weeks by using two electronic platforms: (1) My PANDAS Chart (unpublished website, Rodney Lusk and 3PointData, Inc.), a symptom monitoring website with text-messaging reminders for parents to rate the severity of their children's PANS symptoms on a smartphone three times per week, and (2) Research Electronic Data Capture (REDCap) (Harris et al. 2009), a customizable survey website with email reminders for both parents and children to rate the severity of PANS symptoms once per week.

The aim of the study was to begin to characterize the longitudinal trajectory of neuropsychiatric symptoms in youths with PANS. For this purpose, we sought to identify baseline demographic and clinical characteristics that would predict the severity of symptoms at baseline and the course of symptoms over 12 weeks.

Another aim was to compare the data collected from the two electronic data acquisition platforms, including relative compliance and agreement between parents and children on neuropsychiatric symptoms by using the PANS Symptom Rating Scale (PANS Scale) (unpublished instrument, Tanya Murphy and Gail Bernstein).

Methods

Participants

Inclusion criteria

English-speaking boys and girls ages 4–16 years who met the following criteria for PANS: (1) abrupt onset of OCD or severely restricted food intake, (2) at least two equally abrupt-onset concurrent neuropsychiatric symptoms from seven categories (anxiety, emotional lability/depression, irritability/aggression/severely oppositional behaviors, behavioral/developmental regression, deterioration in school performance, sensory or motor abnormalities [e.g. tics], somatic symptoms [e.g., urinary, sleep]), and (3) neuropsychiatric symptoms are not better explained by a known neurological or medical disorder. Most of the participants underwent a comprehensive child psychiatric evaluation to establish the PANS diagnosis, including medical record review in the University of Minnesota PANS/PANDAS Clinic before recruitment into this study. Youth deemed to be too unstable psychiatrically or medically to participate safely in the protocol, per clinical judgment of the Principal Investigator or Co-Investigator, were excluded.

Procedures

The study was approved by the University of Minnesota Institutional Review Board (IRB). Twenty-five parents of potential participants completed phone screens to evaluate their eligibility to participate according to study-specific inclusion and exclusion criteria. Two youths were not scheduled for a baseline study visit due to scheduling difficulties, and one because of the low likelihood of meeting diagnostic criteria for PANS. After written informed consent from parents and written or verbal assent from children were obtained, clinical interviews and rating scales were administered by the study coordinator (a master's-level clinician trained in the administration of study instruments) during an in-person baseline assessment. The study coordinator also trained participants in the use of My PANDAS Chart and REDCap. After the initial study visit, parents were asked to complete three My PANDAS Chart surveys and one REDCap survey per week for 12 weeks. Included in the REDCap survey was a 12-item short form of the PANS Scale that children with at least a second grade reading level (per parent/child report) were asked to complete once per week. Twelve weeks after study enrollment, participants returned for a final study visit to repeat clinical interviews and rating scales. At that time, they were compensated for their participation.

Research Electronic Data Capture

REDCap is a web-based program that allows researchers to design and customize online surveys for data collection, storage, and dissemination (Harris et al. 2009). The REDCap database for this study was modified from the one developed by Jennifer Frankovich, MD and Margo Thienemann, MD, at Stanford University (J. Frankovich and M. Thienemann, pers. comm.). At baseline, parents were asked to complete a comprehensive REDCap survey that included the PANS Scale, Modified Overt Aggression Scale (MOAS) (Sorgi et al. 1991), and Caregiver Burden Inventory (CBI) (Novak and Guest 1989). After the baseline survey, parents completed a weekly REDCap survey for 12 weeks that included the PANS Scale, PANS Global Impairment Score (GIS) (Liebold et al. 2019), and free response items to denote changes to the child's health (e.g., a new infection) or treatment plan. Children were asked to complete a 12-item short form of the PANS Scale embedded in the baseline and weekly REDCap surveys. The baseline survey took parents and children 30–60 minutes to complete, whereas weekly surveys took up to 20 minutes. Parents received an email once per week from REDCap prompting them to complete the weekly survey. Parents received one reminder email from REDCap for any survey that was not completed within 48 hours. Access to REDCap survey data was limited to the research team.

My PANDAS Chart

My PANDAS Chart is an unpublished data acquisition website developed by Rodney Lusk, MD and 3PointData, Inc. The site offers several tools for parents and families of children with PANS/PANDAS to digitally record information regarding their children's medical history, medications, and psychosocial interventions. My PANDAS Chart also includes an 11-item short form of the PANS Scale for parents to monitor their child's neuropsychiatric symptoms. A notable feature of My PANDAS Chart is its use of text messages to prompt users to complete the abbreviated PANS Scale and to enter information regarding changes to the child's treatment plan. For this study, parents were sent text-message reminders three times per week on the days and times of their choosing. The short form of the PANS Scale is typically completed within a few minutes, and scores for the total and individual items of each completed scale are visible to users on the My PANDAS Chart homepage.

Recruitment

Participants were recruited primarily from an IRB-approved departmental research recruitment registry. Other participants were recruited via referral from a local outpatient clinic, postings on the department website, and from the local PANS/PANDAS Support Group.

Measures

The PANS Scale is a 31-item instrument that is used to rate the severity of PANS symptoms over the previous week. Each item is rated from 0 to 4 on a Likert scale, with higher scores indicating greater severity. The PANS Scale may be administered by a clinician or used as a parent-report measure. Short forms of the PANS Scale were developed for this study, including an 11-item parent-report measure accessed through My PANDAS Chart and a 12-item child-report measure on REDCap. The psychometric properties of the PANS Scale and the short forms of the PANS Scale have not been studied. The PANS Scale was administered by a trained member of the research team to parents and children conjointly at the baseline and final assessments. Parents were asked to complete the PANS Scale once per week on REDCap, and the short form of the Scale three times per week on My PANDAS Chart. Children with at least a second-grade reading level completed the short form once per week on REDCap, independent of parent assistance. Children with a reading level between kindergarten and second-grade completed the short form once per week on REDCap with parents reading the questions, as needed. Children with less than a kindergarten reading level did not complete the short form of the Scale.

Children's Yale-Brown Obsessive Compulsive Scale (CY-BOCS) is a clinician-administered measure used to rate the presence and severity of obsessions and compulsions in children and adolescents with a total score of 0–40 (Scahill et al. 1997). Psychometric properties are adequate (Storch et al. 2006; Yucelen et al. 2006). The CY-BOCS was administered by a trained member of the research team at baseline and final assessments.

The PANS GIS is a caregiver-rated score of global impairment. It is a single-item that is rated 0–100, with lower scores indicating less impairment. The GIS has been validated in the PANS population (Leibold et al. 2019). Parents completed the GIS once per week on REDCap.

The MOAS is a parent-report measure that is used to rate the severity and frequency of aggressive behavior related to four categories: verbal aggression, aggression against objects, aggression against self, and aggression against others (Sorgi et al. 1991). Parents completed the MOAS on REDCap at baseline.

The CBI is a 24-item self-report measure for rating the level of burden that caregivers experience across five categories: time-dependence burden, developmental burden, physical burden, social burden, and emotional burden (Novak and Guest 1989). Scores range from 0 to 96, with higher scores indicating greater burden. The psychometric properties of the CBI have been studied in the PANS population (Farmer et al. 2018). Parents completed the CBI on REDCap at baseline.

Statistical analysis

Summary statistics were generated for demographic and baseline measures. Compliance rates were calculated for My PANDAS Chart usage by entry (out of three possible entries per week) and by week (defined as at least one entry in a week) and for weekly REDCap surveys. Correlations between different raters and/or different platforms were estimated by Pearson's correlation coefficients.

Longitudinal analysis of ratings over time (weeks in study) were conducted with mixed-effects regression models. Agreement between parent and child answers to equivalent survey questions was examined in a model, with week of survey and age and gender of child as fixed covariates and random intercepts by subject and question. Trends in parental and child overall PANS scores were examined in models with week of survey, age and gender of child, time from initial onset, and baseline measures as fixed covariates and random intercept by child. The effects of baseline measures on longitudinal trends were examined by interaction terms from individual mixed-effects regression models with week of survey, the baseline measure, and their interaction as fixed covariates and random intercept by subject.

All analyses were conducted with R (version 3.6.1). A significance level of 0.05 was used for all hypothesis tests.

Results

Participants

Twenty-two children were enrolled. Twenty participants completed baseline measures. Of these 20 participants, the mean age at study entry was 9.0 years (standard deviation [SD] = 2.1) and mean age at onset of PANS was 6.9 years (SD = 2.2). All participants were White and not Hispanic or Latinx. Primarily mothers completed baseline rating scales. Comorbidities included attention-deficit/hyperactivity disorder, OCD, hoarding, and anxiety disorders (Table 1).

Table 1.

Demographic and Clinical Characteristics of Pediatric Acute-Onset Neuropsychiatric Syndrome Participants

Characteristic	PANS (n = 20)
Age at onset, mean (SD)	6.9 (2.2)
Age at enrollment, mean (SD)	9.0 (2.1)
Male, n (%)	11 (55)
Ethnicity, n (%)
White	20 (100)
Hispanic/Latinx	0 (0)
PANDAS, n (%)	16 (80)
First flare of PANS, n (%)	6 (30)
Medications at enrollment, n (%)
Antibiotic	7 (35)
NSAID	6 (30)
Antibiotic + NSAID	4 (20)
SSRI	5 (25)
Other psychotropic	4 (20)
CY-BOCS, mean (SD)
Total	8.6 (8.1)
Obsessions	3.6 (3.9)
Compulsions	5.1 (5.1)
Baseline REDCap Informant, n (%)
Mother	14 (70)
Father	1 (5)
Both parents	5 (25)
Comorbid diagnoses, n (%)
ADHD	1 (5)
OCD	3 (15)
Hoarding	1 (5)
Anxiety disorder (any)	2 (10)

ADHD, attention-deficit/hyperactivity disorder; CY-BOCS, Children's Yale-Brown Obsessive Compulsive Scale; NSAID, nonsteroidal anti-inflammatory drug; OCD, obsessive-compulsive disorder; PANDAS, Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal Infections; PANS, pediatric acute-onset neuropsychiatric syndrome; SD, standard deviation; SSRI, selective serotonin reuptake inhibitor; REDCap, Research Electronic Data Capture.

Mean baseline CY-BOCS total score was low at 8.6 (SD = 8.1), and mean PANS Scale score was mild at 15.8 (SD = 13.5). These findings suggest that many of the participants were unlikely to be in an acute stage of illness at the time of entering the study. Three of the 20 participants dropped out, and they did not return for the final assessment. Dropouts occurred at weeks 1, 9, and 12. Data from dropouts were included in analyses.

Compliance

Five parents did not use My PANDAS Chart even once to record their children's symptoms. Excluding the five parents who did not use My PANDAS Chart (non-compliers), symptoms were recorded for an average of 11.3 weeks with a mean of 2.3 entries per week. Including the five parents who did not use My PANDAS Chart, parents recorded symptoms for an average of 8.5 weeks with a mean of 1.7 entries per week.

Twenty parents recorded their children's symptoms on REDCap. Parents demonstrated an overall compliance rate of 86.3% for the weekly PANS Scale on REDCap, where compliance is defined as completing the weekly survey. When non-compliers (parents who did not record any data on My PANDAS Chart) were excluded, the compliance rate increased to 92.2%. The overall compliance rate by entry (of three possible entries each week) for My PANDAS Chart was 53.8% and it increased to 75.4% when non-compliers were excluded. If compliance rate was calculated as completing at least one of three entries per week, overall compliance for My PANDAS Chart was 70.8%, which increased to 94.4% if non-compliers were excluded.

Agreement between parents' ratings of neuropsychiatric symptoms in their children on two electronic platforms

Correlation between the PANS Scale scores on My PANDAS Chart and REDCap as completed by parents was r = 0.73 (p < 0.0001), suggesting that parents rate symptoms similarly on both platforms.

Agreement between different raters on PANS Scale

Correlation between the clinician-rated and the parent-rated PANS Scale score at baseline was r = 0.67 (p = 0.0006). Correlation between the clinician-rated and the child-rated PANS Scale total at baseline was r = 0.68 (p = 0.0005). Correlation between the PANS Scale scores as completed by parents and by children on REDCap showed r = 0.55 (p < 0.0001). Correlation between the parent PANS Scale short form (on My PANDAS Chart) and the child PANS Scale short form (on REDCap) showed r = 0.47 (p < 0.0001).

Longitudinal analyses

Parent/child agreement

The regression model examining parent/child agreement on common survey questions found that the child's answer was the strongest predictor of the parent's answer (p < 0.00001). The only other significant predictor was the child's gender, with parents reporting less severe symptoms in females (p = 0.03).

Parent scores

PANS Scale scores decreased significantly over the 12-week study, as indicated by the estimate of −0.27, p = 0.0006 (Table 2). Higher baseline scores on PANS GIS were significantly associated with higher parental PANS Scale scores (p < 0.0001) (Table 2). Although not significant, higher baseline aggression scores on the MOAS appeared to be associated with higher parental PANS scores (p = 0.08). Age of the child, gender, age of onset, CY-BOCS total score, and CBI score did not significantly affect parental PANS scores.

Table 2.

Factors that Affect Parent-Rated Pediatric Acute-Onset Neuropsychiatric Syndrome Symptom Rating Scale Scores Over Time

	Estimate	t-Value	p-Value
(Intercept)	−1.31	−0.26	0.80
Week	−0.27	−3.48	0.0006
Age	0.003	0.07	0.95
Female gender	2.25	1.07	0.31
Age at onset of PANS	0.07	1.36	0.20
Aggression on MOAS	0.75	1.90	0.08
CY-BOCS total	0.13	0.98	0.35
Caregiver Burden Inventory	0.08	1.14	0.27
Global Impairment Score	0.30	10.80	<0.0001
Interactions
Age	−0.003	−0.84	0.40
Female gender	0.11	0.55	0.58
Age at onset of PANS	0.01	1.94	0.05
Aggression on MOAS	0.08	1.90	0.06
CY-BOCS total	0.009	0.68	0.50
Caregiver Burden Inventory	−0.0004	−0.07	0.95
Global Impairment Score	0.006	1.51	0.13

CY-BOCS, Children's Yale-Brown Obsessive Compulsive Scale; MOAS, Modified Overt Aggression Scale; PANS, Pediatric Acute-Onset Neuropsychiatric Syndrome.

Interaction effects demonstrated that a longer duration of time between initial onset of PANS and study enrollment was significantly associated with an increase in parent-rated PANS symptoms over the 12-week period (p = 0.05) (Table 2). Although not significant, higher baseline aggression scores appeared to be associated with an increase in parent-rated PANS symptoms over time (p = 0.06).

Child scores

Older age of the child (p = 0.005), female gender (p = 0.001), higher baseline CY-BOCS total score (p = 0.003), higher baseline CBI score (p = 0.01), and higher baseline GIS (p = 0.0001) were associated with higher child-rated PANS Scale scores (Table 3).

Table 3.

Factors that Affect Child-Rated Pediatric Acute-Onset Neuropsychiatric Syndrome Symptom Rating Scale Scores Over Time

	Estimate	t-Value	p-Value
(Intercept)	−14.69	−4.59	0.0005
Week	0.02	0.43	0.67
Age	0.09	3.40	0.005
Female gender	5.50	4.14	0.001
Age at onset of PANS	0.07	1.96	0.07
Aggression on MOAS	−0.21	−0.86	0.41
CY-BOCS total	0.30	3.68	0.003
Caregiver Burden Inventory	0.13	2.78	0.01
Global Impairment Score	0.07	4.12	0.0001
Interactions
Age	0.0005	0.28	0.78
Female gender	0.05	0.50	0.62
Age at onset of PANS	0.00	0.02	0.99
Aggression on MOAS	0.01	0.57	0.57
CY-BOCS total	0.004	0.71	0.48
Caretaker Burden Inventory	0.007	2.51	0.01
Global Impairment Score	0.002	0.87	0.39

CY-BOCS, Children's Yale-Brown Obsessive Compulsive Scale; MOAS, Modified Overt Aggression Scale; PANS, Pediatric Acute-Onset Neuropsychiatric Syndrome.

There was a significant interaction effect between CBI at baseline and time (p = 0.01) such that higher CBI scores at baseline predicted that children would report greater PANS symptom severity over time (Table 3).

Discussion

Compliance

Twenty-five percent of parents did not use My PANDAS Chart to record their children's neuropsychiatric symptoms even once. This is consistent with research on mental health app utilization indicating that sizeable numbers of individuals install health-related apps but do not sustainably engage with the content of the app (Ng et al. 2019). Specific reasons that parents did not use My PANDAS Chart were not assessed, but one possible explanation is that parents were expected to use the app three times per week, which may have been perceived as too time consuming. Compliance rate for parents' weekly ratings on REDCap was high at 86%. Compliance rate for at least once-weekly ratings on My PANDAS Chart was 71%. However, recording symptoms three times per week on My PANDAS Chart revealed a lower compliance rate (54%). The mean number of tri-weekly ratings ranged from 1.7 to 2.3 entries per week, depending on whether non-compliers were included. It appears that parents were reliable in recording symptoms once a week but less so for three times per week. This suggests that, within the PANS population, parents can be reasonably expected to digitally monitor their child's neuropsychiatric symptoms on a weekly basis. Compliance rates of 71%–86% for weekly digital recording of symptoms are substantially higher than completion rates of paper and pencil rating scales in research studies. For example, Stone et al. (2002) reported a compliance rate of 11% for adults completing daily paper and pencil measures of chronic pain. Electronic platforms may, therefore, be a more reliable method for tracking symptoms longitudinally.

There was strong agreement between parents' PANS Scale scores on the two electronic platforms (r = 0.73, p < 0.0001). Digital platforms appear to be a good method for monitoring neuropsychiatric symptoms in PANS patients and should be considered in future research and clinical endeavors with this population, particularly given the episodic nature of the illness. Future research should explore digital strategies to monitor psychiatric symptoms in children with other diagnoses such as anxiety, depression, mania, and eating disorders.

Parent–child agreement

It is well established in the pediatric literature that agreement between parent- and youth-reports of children's emotional and behavioral functioning is low to moderate (Comer and Kendall 2004; De Los Reyes and Kazdin 2005; Van Roy et al. 2010). In a meta-analysis of 119 studies, Achenbach et al. (1987) reported a mean correlation of r = 0.25 between parent- and child-reports of the child's emotional and behavioral symptoms, with stronger agreement between parents and youths when rating externalizing (e.g., hyperactivity) versus internalizing (e.g., anxiety) symptoms.

We found moderate agreement between parents and children on the PANS Scale with r = 0.55. Items on the PANS Scale include both internalizing and externalizing symptoms. Further, there was moderate agreement between clinician and parent ratings (r = 0.67) and between clinician and child ratings (r = 0.68). Although PANS research has typically relied on parents' report of children's symptoms due to the young age of the affected youth and the impaired functioning of those with a moderate to severe presentation, it appears that children can be accurate reporters of their PANS symptoms. Children with PANS should, therefore, be included as raters in clinical and research settings when possible.

Gender differences

Previous research has demonstrated inconsistent gender differences in the perception of children's mental health symptoms, with some studies reporting greater parent–child disagreement for boys (Salbach-Andrae et al. 2009) and others describing larger parent–child discrepancy for girls (Sourander et al. 1999; Grills and Ollendick 2003). Our study reveals some interesting gender-related findings. When comparing parents' and children's responses to equivalent items on the PANS Scale, parents rated total symptoms as more severe for boys than for girls (p = 0.03). In contrast, results from the regression indicate that girls compared with boys reported higher scores on the child-rated PANS Scale (p = 0.001). Thus, parents rated PANS symptoms as more severe for boys whereas the opposite was true for child ratings (i.e., girls self-reported higher PANS scores than boys). Of note, the PANS Scale includes both internalizing and externalizing symptoms, making interpretation of our gender findings more challenging. However, as proposed by van der Meer et al. (2008), observed differences may reflect gender-related discrepancies or biases in how psychiatric symptoms are experienced, expressed, and observed. For example, the authors noted that across psychiatric disorders, externalizing symptoms are more likely to be endorsed in males, and internalizing symptoms are more likely to be endorsed in females. In our sample, it is possible that parents rated boys as more severely affected by PANS because they more readily recognized externalizing symptoms, whereas internalizing symptoms experienced by girls went under-recognized. This finding underscores the importance of collecting child self-ratings of PANS symptoms in research and clinical settings.

Predictors of course of PANS

Greater duration of time between illness onset and study entry predicted worsening of parent-rated PANS symptoms over the 12-week study. Reducing the time from symptom onset to diagnosis and treatment might mitigate this finding. Supporting this idea, two retrospective studies have demonstrated that early intervention for a PANS flare (within 14 days) using immunomodulatory treatments will reduce the duration of a 12-week flare by 2 weeks with nonsteroidal anti-inflammatory drugs (Brown et al. 2017a) or by 4 weeks with oral steroids (Brown et al. 2017b).

Higher parent-rated baseline CBI scores predicted worsening of child-reported PANS symptoms over the course of 12 weeks. The CBI assesses the impact of caregiving on parents' physical health, emotional well-being, and social relationships and it evaluates burden level due to children's dependency and developmental issues. Children with high levels of neuropsychiatric symptoms are challenging to care for, which likely leads to greater caregiver burden and stress. Due to high caregiver burden, parents of children with PANS may struggle to parent effectively, which may contribute to higher levels of neuropsychiatric symptoms over time. Further, children may sense their parents' stress and frustration and this may further exacerbate the children's emotional and behavioral symptoms. Attention to psychosocial stressors and burden on parents and other family members is important in effectively treating PANS patients (Thienemann et al. 2017; Frankovich et al. 2018).

Consistent with our finding that early identification and treatment of children with PANS may predict reduction of symptom severity over a 12-week period, Frankovich et al. (2018) administered the CBI longitudinally to 94 families in a multidisciplinary PANS clinic. The 94 parents (71% were mothers) completed 892 CBIs. The researchers found that a shorter duration of time between PANS onset and clinical treatment was associated with more rapid reduction of CBI scores over time. During flares of neuropsychiatric symptoms compared with during quiescent periods, CBI scores were higher. Each additional year after initial entry into the clinic predicted a decrease in CBI score. Thus, treatment and support in a multidisciplinary PANS clinic were associated with a reduction in caretaker burden over time.

We also found an interaction, although not significant, between parents' ratings of their children's baseline aggression and time (p = 0.06). This suggests that aggressive behavior may be a risk factor for poor short-term outcome in children with PANS. During flares of neuropsychiatric symptoms, many PANS patients show symptoms of rage, aggression, and unpredictable behaviors such as threatening their parents and siblings or trying to jump out of moving cars (Frankovich et al. 2018). Rage episodes and course of illness are often unpredictable in PANS (Frankovich et al. 2018). Thus, parents may be afraid of their children's aggression and worried about their own safety and the safety of other children in the home. Targeting rage, aggression, and oppositional behaviors is an important treatment goal in many PANS patients.

The primary shortcoming of the study is its small sample size. Thus, we did not have the power to adequately address many questions. In addition, our sample was composed entirely of White non-Hispanic/Latinx participants. Future investigations need to recruit a diverse sample to allow for generalizability of study findings to other populations. Also, since parents were allowed to assist children in kindergarten to second grade with reading items while the children completed their PANS Scale on REDCap, there may have been parental influence on children's selection of answers, thus overinflating correlations between child- and parent-report measures. Further, there was no clinician dashboard on My PANDAS Chart to monitor participant engagement and identify families who were not completing measures. If this feature was added for digital monitoring, it would be possible to remind families to complete measures and improve adherence. Finally, we did not collect repeated measures of certain baseline questionnaires (i.e., MOAS, CBI). In future studies, it would be important to administer these measures longitudinally to view patterns of change over time, particularly to observe how the PANS course may impact the trajectory of psychosocial sequelae and vice versa.

Clinical Significance

Our findings highlight the utility of digital methods in tracking longitudinal neuropsychiatric symptoms and suggest that particular baseline characteristics (e.g., delay in diagnosis and treatment of PANS, high caregiver burden) may be indicative of a differential trajectory of PANS course with more severe symptoms over the short-term. Digital monitoring of neuropsychiatric symptoms can aid in early detection of neuropsychiatric symptoms in children with PANS and facilitate the monitoring of response to treatment.

Clinical Trial Registration Information

Pediatric Acute-Onset Neuropsychiatric Syndrome (PANS): Clinical Characterization and Prospective Course; https://clinicaltrials.gov/; NCT04382716.

Footnotes

Acknowledgments

The authors extend their special thanks to the children and their parents who participated in this study. The authors gratefully acknowledge Madison Warner, BS, University of Minnesota, for data management for this study. The authors report no biomedical financial interests or potential conflicts of interest.

Disclosures

No competing financial interests exist.

References

Achenbach

, McConaughy

, Howell

: Child/adolescent behavioral and emotional problems: Implications of cross-informant correlations for situational specificity. Psychol Bull, 101:213–232, 1987.

Bernstein

, Victor

, Pipal

, Williams

: Comparison of clinical characteristics of pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections and childhood obsessive-compulsive disorder. J Child Adolesc Psychopharmacol, 20:333–340, 2010.

Brown

, Farmer

, Freeman

Jr ., Spartz

, Farhadian

, Thienemann

, Frankovich

: Effect of early and prophylactic nonsteroidal anti-inflammatory drugs on flare duration in pediatric acute-onset neuropsychiatric syndrome: An observational study of patients followed by an academic community-based pediatric acute-onset neuropsychiatric syndrome clinic. J Child Adolesc Psychopharmacol, 27:619–628, 2017a.

Brown

, Farmer

, Freeman

, Spartz

, Farhadian

, Thienemann

, Frankovich

: Pediatric acute-onset neuropsychiatric syndrome response to oral corticosteroid bursts: An observational study of patients in an academic community-based PANS clinic. J Child Adolesc Psychopharmacol, 27:629–639, 2017b.

Chang

, Frankovich

, Cooperstock

, Cunningham

, Latimer

, Murphy

, Pasternack

, Thienemann

, Williams

, Walter

, Swedo

: Consortium PC: Clinical evaluation of youth with pediatric acute-onset neuropsychiatric syndrome (PANS): Recommendations from the 2013 PANS Consensus Conference. J Child Adolesc Psychopharmacol, 25:3–13, 2015.

Comer

, Kendall

: A symptom-level examination of parent-child agreement in the diagnosis of anxious youths. J Am Acad Child Adolesc Psychiatry, 43:878–886, 2004.

De Los Reyes

, Kazdin

: Informant discrepancies in the assessment of childhood psychopathology: A critical review, theoretical framework, and recommendations for further study. Psychol Bull, 131:483–509, 2005.

Dubad

, Winsper

, Meyer

, Livanou

, Marwaha

A systematic review of the psychometric properties, usability and clinical impacts of mobile mood-monitoring applications in young people. Psychol Med, 48:208–228, 2018.

Farmer

, Thienemann

, Leibold

, Kamalani

, Sauls

, Frankovich

: Psychometric evaluation of the caregiver burden inventory in children and adolescents with PANS. J Pediatr Psychol, 43:749–757, 2018.

10.

Frankovich

, Leibold

, Farmer

, Sainani

, Kamalani

, Farhadian

, Willett

, Park

, Sidell

, Ahmed

, Thienemann

: The burden of caring for a child or adolescent with pediatric acute-onset neuropsychiatric syndrome (PANS): An observational longitudinal study. J Clin Psychiatry, 80:17m12091, 2018.

11.

Frankovich

, Thienemann

, Pearlstein

, Crable

, Brown

, Chang

: Multidisciplinary clinic dedicated to treating youth with pediatric acute-onset neuropsychiatric syndrome: Presenting characteristics of the first 47 consecutive patients. J Child Adolesc Psychopharmacol, 25:38–47, 2015.

12.

Grills

, Ollendick

: Multiple informant agreement and the anxiety disorders interview schedule for parent and children. J Am Acad Child Adolesc Psychiatry, 42:30–40, 2003.

13.

Harris

, Taylor

, Thielke

, Payne

, Gonzalez

, Conde

: Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform, 42:377–381, 2009.

14.

Leibold

, Thienemann

, Farhadian

, Willett

, Frankovich

: Psychometric properties of the pediatric acute-onset neuropsychiatric syndrome global impairment score in children and adolescents with pediatric acute-onset neuropsychiatric syndrome. J Child Adolesc Psychopharmacol, 29:41–49, 2019.

15.

Luxton

, McCann

, Bush

, Mishkind

, Reger

: MHealth for mental health: Integrating smartphone technology in behavioral healthcare. Prof Psychol Res Pract, 42:505–512, 2011.

16.

Murphy

, Patel

, McGuire

, Kennel

, Mutch

, Parker-Athill

, Hanks

, Lewin

, Storch

, Toufexis

, Dadlani

, Rodriguez

: Characterization of the pediatric acute-onset neuropsychiatric syndrome phenotype. J Child Adolesc Psychopharmacol, 25:14–25, 2015.

17.

Murphy

, Storch

, Lewin

, Edge

, Goodman

: Clinical factors associated with pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections. J Pediatr, 160:314–319, 2012.

18.

, Firth

, Minen

, Torous

: User engagement in mental health apps: A review of measurement, reporting, and validity. Psychiatr Serv, 70:538–544, 2019.

19.

Novak

, Guest

: Application of a multidimensional caregiver burden inventory. Gerontologist, 29:798–803, 1989.

20.

Perrin

, Murphy

, Casey

, Pichichero

, Runyan

, Miller

, Snider

, Swedo

: Does group A beta-hemolytic streptococcal infection increase risk for behavioral and neuropsychiatric symptoms in children?. Arch Pediatr Adolesc Med, 158:848–856, 2004.

21.

Roberts

, Chan

, Torous

: New tests, new tools: Mobile and connected technologies in advancing psychiatric diagnosis. NPJ Digit Med, 1:20176, 2018.

22.

Salbach-Andrae

, Klinkowski

, Lenz

, Lehmkuhl

: Agreement between youth-reported and parent-reported psychopathology in a referred sample. Eur Child Adolesc Psychiatry, 18:136–143, 2009.

23.

Scahill

, Riddle

, McSwiggin-Hardin

, Ort

, King

, Goodman

, Cicchetti

, Leckman

: Children's Yale-Brown Obsessive Compulsive Scale: Reliability and validity. J Am Acad Child Adolesc Psychiatry, 36:844–852, 1997.

24.

Shen

, Levitan

, Johnson

, Bender

, Hamilton-Page

, Jadad

, Wiljer

: Finding a depression app: A review and content analysis of the depression app marketplace. JMIR Mhealth Uhealth, 3:e16, 2014.

25.

Sorgi

, Ratey

, Knoedler

, Markert

, Reichman

: Rating aggression in the clinical setting: A retrospective adaptation of the Overt Aggression Scale: Preliminary results. J Neuropsychiatry Clin Neurosci, 3:S52–S56, 1991.

26.

Sourander

, Helstelä

, Helenius

: Parent-adolescent agreement on emotional and behavioral problems. Soc Psychiatry Psychiatr Epidemiol, 34:657–663, 1999.

27.

Stone

, Shiffman

, Schwartz

, Broderick

, Hufford

: Patient non-compliance with paper diaries. BMJ, 324:1193–1194, 2002.

28.

Storch

, Murphy

, Adkins

, Lewin

, Geffken

, Johns

, Jann

, Goodman

: The Children's Yale-Brown Obsessive Compulsive scale: Psychometric properties of child- and parent-report formats. J Anxiety Disord, 20:1055–1070, 2006.

29.

Swedo

, Frankovich

, Murphy

: Overview of treatment of pediatric acute-onset neuropsychiatric syndrome?. J Child Adolesc Psychopharmacology, 27:562–565, 2017.

30.

Swedo

, Leckman

, Rose

: From research subgroup to clinical syndrome: Modifying the PANDAS criteria to describe PANS (pediatric acute-onset neuropsychiatric syndrome). Pediatr Ther, 2:113, 2010.

31.

Swedo

, Leonard

, Garvey

, Mittleman

, Allen

, Perlmutter

, Lougee

, Dow

, Zamkoff

, Dubbert

: Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: Clinical description of the first 50 cases. Am J Psychiatry, 155:264–271, 1998.

32.

Thienemann

, Murphy

, Leckman

, Shaw

, Williams

, Kapphahn

, Frankovich

, Geller

, Bernstein

, Chang

, Elia

, Swedo

: Clinical management of pediatric acute-onset neuropsychiatric syndrome: Part I–psychiatric and behavioral interventions. J Child Adolesc Psychopharmacol, 27:566–573, 2017.

33.

Van der Meer

, Dixon

, Rose

: Parent and child agreement on reports of problem behaviour obtained from a screening questionnaire, the SDQ. Eur Child Adolesc Psychiatry, 17:491–497, 2008.

34.

Van Roy

, Groholt

, Heyerdahl

, Clench-Aas

: Understanding discrepancies in parent-child reporting of emotional and behavioural problems: Effects of relational and socio-demographic factors. BMC Psychiatry, 10:56, 2010.

35.

Wen

CKF

, Schneider

, Stone

, Spruijt-Metz

: Compliance with mobile ecological momentary assessment protocols in children and adolescents: A systematic review and meta-analysis. J Med Internet Res, 19:e132, 2017.

36.

, Torous

, Harper

: A gap in the literature: Clinical role for smartphone applications for depression care among adolescents?. J Am Acad Child Adolesc Psychiatry, 55:630–631, 2016.

37.

Yucelen

, Rodopman-Arman

, Topcuoglu

, Yazgan

, Fisek

: Interrater reliability and clinical efficacy of Children's Yale-Brown Obsessive-Compulsive scale in an outpatient setting. Compr Psychiatry, 47:48–53, 2006.