The effects of racial and socioeconomic disparities on time to diagnosis and treatment of pediatric functional seizures in the United States

Abstract

Purpose.

The present study sought to assess the effects of racial and socioeconomic status in the United States on time to treatment and diagnosis of pediatric functional seizures (FS).

Methods.

Eighty adolescents and their parent/guardian completed a demographics questionnaire and reported date of FS onset, diagnosis, and treatment. Paired samples t-tests compared time between FS onset and diagnosis, onset and treatment, and diagnosis and treatment based on race (White vs racial minority), annual household income (≤$79,999 vs ≥$80,000), maternal and paternal education (≤Associate’s Degree vs Bachelor’s Degree), and combined parental education (≤Post-graduate training vs Graduate degree).

Results.

Adolescents with lower annual household income began treatment >6 months later than adolescents with greater annual household income (p=0.049). Adolescents with lower maternal and paternal education (≤Associate’s Degree vs Bachelor’s Degree) began treatment >4 and ~8.5 months later than adolescents with greater maternal and paternal education (p=0.04; p=0.03), respectively. Adolescents with lower maternal education also received a diagnosis >5 months later (p=0.03). Adolescents without a mother or father with a graduate degree received a diagnosis and began treatment~3 and >11 months later (p=0.03; p=0.01) than adolescents whose mother or father received a graduate degree, respectively. No racial differences were found.

Conclusions.

Adolescents with lower annual household income and/or parental education experienced increased duration between FS onset and treatment and diagnosis. Research is needed to clarify the mechanisms underlying this relationship, and action is needed to reduce these disparities given FS duration is associated with poorer prognosis and greater effects on the brain.

Introduction

Functional seizures (FS), also known as psychogenic non-epileptic seizures or PNES, are a type of functional neurological disorder characterized by seizure-like symptoms without associated epileptiform activity.¹ Pediatric FS account for approximately 10%−20% of patients presenting to outpatient neurology clinics in the United States,^{2, 3} making it one of the most common diagnoses among those referred.⁴ Pediatric FS are severely debilitating for patients and their families and are associated with significantly decreased quality of life.^{5, 6} For example, children and adolescents with FS often experience bullying and are at increased risk of academic difficulties due to frequent absences and negative school responses to FS.⁷ Children and adolescents with FS also report decreased peer relationships and engagement in extracurricular and social activities following the onset of FS.^{8, 9} Caregivers of children and adolescents with FS are also impacted due to frequently taking off work to care for their child and the associated financial burden.⁸ A recent study assessing the mean annual cost of FS for families found estimates ranging from nearly $5,000 to $87,000.¹⁰

Patients in the United States with FS also report high levels of stigmatization stemming from multiple sources, including healthcare providers, family members, and school personnel.⁶ Stigma related to FS is associated with decreased self-esteem, increased stress, and self-isolation.^{6, 11} Additionally, patients with FS who experience stigma are less likely to seek treatment.¹¹ This is concerning, given that previous literature has correlated FS prognosis with duration of symptoms, such that children and adolescents with chronic FS may experience poorer treatment outcomes as compared with patients with an acute history.^12–15 Given this information, as well as the debilitating effects of FS,⁶ assessing other factors that may contribute to or interfere with the timely diagnosis and treatment of pediatric FS is critical.

Race and socioeconomic status have been found to contribute to delayed diagnosis and treatment of various chronic health conditions in the United States.¹⁶ In a study assessing the effects of race on delays in autism spectrum disorder diagnoses, Black and Latin American-origin children were evaluated by upwards of 6–8 providers before receiving a diagnosis.¹⁷ Studies examining delayed diagnosis of pediatric appendicitsis and development delays have found similar results based on race.^{18, 19} Regarding socioeconomic status, children of low-income families were found to be diagnosed with autism spectrum disorder 8 months later than higher-income peers. Proposed mechanisms underlying these relationships include medical mistrust based on prior discrimination²⁰ and lack of access to health care and/or health insurance²¹ often experienced by racial minority youth and families with lower household income, respectively. Racial minority youth and families with lower household income have also been shown to experience poorer primary care quality and decreased access to primary care centers.¹⁸

As earlier intervention for health conditions is often associated with better prognosis,^{22, 23} these findings suggest that racially and socioeconomically minoritized individuals may be at risk of poorer prognosis due to disparities in time to diagnosis and treatment. In a study assessing the effects of social identity on COVID-19 mortality rates, Black individuals were more than twice as likely to die from COVID-19 as compared with their White counterparts.²⁴ Similarly, Black women experience the highest maternal mortality rate in the United States. In 2021, 69.9 per 100,000 Black women experienced maternal mortality during live births, which is nearly three times the rate of White women.²⁵ With regard to educational attainment and cancer mortality rates, individuals with ≤12 years of education experienced 1.43–2.24 increased risk of cancer mortality as compared with individuals with ≥12 years of education.²⁶ More recent research assessing the relationship between neighborhood socioeconomic status and survival rates among patients with breast, prostate, lung, and colorectal cancers found that individuals of disadvantaged neighborhoods experienced 34%, 51%, 21% and 24% increased risk of mortality, respectively.²⁷ These results emphasize the need to identify and address disparities in time to diagnosis and treatment in other health conditions.

Despite a growing body of literature among other health conditions, research has yet to assess the relationships among delays in diagnosis and treatment and race and socioeconomic status in pediatric FS. Understanding this relationship is critical, as patients with FS already experience delays in diagnosis and treatment due to misdiagnosis as another medical condition,²⁸ and prompt treatment is critical for better prognosis.¹² Therefore, the current study sought to assess the effects of race and socioeconomic status, including household income and parental education, on time to diagnosis and treatment of pediatric FS. We hypothesize that racial minority youth, youth with lower household income, and youth with lower parental education will experience increased duration between FS onset and diagnosis, FS onset and treatment, and FS diagnosis and treatment.

Methods

Design Overview

In order to evaluate the relationship between race and socioeconomic status and time to treatment and diagnosis pediatric FS, analyses compared time between 1) FS onset and diagnosis, 2) FS onset and treatment, and 3) FS diagnosis and treatment based on race, annual household income, maternal education, paternal education, and combined parental education (i.e., highest degree obtained by either mother or father). Data were collected as a part of two larger randomized controlled trials conducted in the United States which investigated the treatment effects of Retraining and Control Therapy (ReACT), a cognitive behaviorally based treatment for FS.

Participants

Both studies were approved by the Institutional Review Board at the University of Alabama at Birmingham and are registered on ClinicalTrials.gov (#NCT02801136, #NCT05096273). All participants and their parent/guardian provided informed consent and/or assent prior to enrollment.

Eligibility for the first trial, which began in November 2016, included ages 9–18 years-old and video-electroencephalogram (EEG)-confirmed FS diagnosis. Participants with comorbid epilepsy were permitted if they had no epileptic seizures six or more months before enrollment. Exclusion criteria included substance abuse, psychosis, and severe intellectual disability. Eligibility for the second trial, which began in October 2021, included ages 11–18 years-old and a video-EEG-confirmed FS diagnosis. Exclusion criteria included comorbid epilepsy, less than four FS per month, other paroxysmal nonepileptic events, participation in other mental health therapy, severe intellectual disability, and severe mental illness.

Measurements

Demographics.

Parents/guardians completed a demographics questionnaire assessing the participants’ date of birth, sex, race, annual household income, and maternal/paternal education. Racial categories were chosen in accordance with funding agency categories (i.e., American Indian or Alaska Native, Asian, Black or African American, Hispanic or Latino, Native Hawaiian or Other Pacific Islander).

Time Variables.

Parents/guardians reported date of the participants’ FS onset and first treatment. A subsample of parents/guardians self-reported date of FS diagnosis. For participants who had not received previous treatment, date of first treatment was recorded as the date of their baseline study visit. Time to FS diagnosis was measured by calculating the number of weeks between FS onset and FS diagnosis. Time to first FS treatment was measured by calculating the number of weeks between FS onset and first treatment. Time from FS diagnosis to first treatment was measured by calculating the number of weeks between FS diagnosis and first treatment.

Data Analyses

A power analysis utilizing an alpha level of 0.05 and an effect size of d=0.93, which was determined based on prior literature,²³ indicated a minimum sample size of N=13 to achieve 80% power. All variables were assessed for normality and homogeneity of variances. The assumption of normality was violated for all three time variables. Log transformations were subsequently applied, and the assumption of normality was met. One outlier that was >3 standard deviations over the mean (i.e., 1,057 weeks between FS onset and treatment) was also removed due to its significant effects on primary outcomes. The data underlying this article cannot be shared publicly for the privacy of the individuals that participated in the study. The data may be shared on reasonable request to the corresponding author.

Paired samples t-tests compared time between 1) FS onset and diagnosis, 2) FS onset and treatment, and 3) FS diagnosis and treatment based on race, annual household income, maternal education, paternal education, and combined parental education. Bivariate correlations assessed other factors (i.e., age, sex, seizure frequency) that may be associated with delays in diagnosis and treatment. ANCOVAs controlling for significant correlations were used to determine if these had a significant effect on the models.

A cut point of ≤$79,999 vs. ≥$80,000 was utilized for annual household income comparisons, as this is about the median annual household income in the United States in 2023 (i.e., 74,580).²⁹ For maternal education, paternal education, and combined parental education, a cut point of ≤Associate’s Degree (i.e., two-year undergraduate degree) vs. ≥Bachelor’s Degree (i.e., undergraduate degree typically completed in four years) was utilized for comparisons, as recent census data reports 53% of individuals in the United States reported educational attainment ≤Associate’s degree.³⁰ After there was no difference in combined parental education, an exploratory analysis was completed, which suggested using a cut point of ≤Post-Graduate Training vs. ≥Graduate Degree for comparison. Of note, cut points for annual household income and educational attainment were based on nationally representative data, not geographical region, as the study included participants from 19 states across the United States and one participant from the West Indies. Race analyses compared White adolescents vs. racial minority adolescents, including Black (N=14, 51.85%), Asian (N=4, 14.81%), Hispanic or Latino (N=2, 7.41%), Native Hawaiian or Pacific Islander (N=1, 3.70%), and Other (N=6, 22.22%; i.e., selected more than one Race) adolescents. A sensitivity analysis also compared White adolescents vs. Black adolescents. As there were no differences in outcomes, our final analyses compared White adolescents vs. racial minority adolescents to retain maximum sample size.

Results

Participants included 80 adolescents. Of the 80 participants, 68 provided date of FS onset and date of first treatment, 51 provided date of FS onset and date of diagnosis, and 37 provided date of diagnosis and date of first treatment. See Table 1 for demographics by time variables.

Table 1.

Demographics by Time

		FS Onset to Treatment	FS Onset to Diagnosis	FS Diagnosis to Treatment
		% (N)
Race	Racial Minority	35.3 (23)	29.4 (15)	29.7 (11)
	White	64.7 (44)	70.60 (36)	70.30 (26)
Income	<$79,000	44.78 (30)	31.37 (16)	32.43 (12)
	>$80,000	55.22 (37)	68.63 (35)	67.57 (25)
Father Education	<Associate degree	59.37 (38)	58.82 (30)	62.16 (23)
	>Bachelor’s Degree	40.63 (26)	41.18 (21)	37.84 (14)
Mother Education	<Associate degree	51.52 (34)	39.22 (20)	43.24 (16)
	>Bachelor’s Degree	48.48 (32)	60.78 (31)	56.76 (21)
Parental Education	<Post-Grad Training	72.30 (34)	66.67 (34)	62.16 (23)
	>Graduate Degree	27.70 (13)	33.33 (17)	37.84 (14)
Sex	Male	41.1 (33)	15.70 (8)	18.9 (7)
	Female	58.9 (35)	84.3 (43)	81.1 (30)
		Mean (SD)
Age		15.15 (2.54)	14.98 (2.03)	15.18 (2.06)

Average duration between FS onset and treatment, FS onset and diagnosis, and FS diagnosis and treatment was 47.60, 20.75, and 20.76 weeks, respectively (Table 2).

Table 2.

Number of Weeks by Demographics

		FS Onset to Treatment Mean(SD)	FS Onset to Diagnosis Mean(SD)	FS Diagnosis to Treatment Mean(SD)
Race	Racial Minority	52.94 (90.85)	14.58 (23.05)	16.42 (24.55)
	White	44.81 (48.48)	23.32 (36.58)	22.60 (31.92)
Income	<$79,000	60.88 (83.39)	18.12 (27.91)	18.68 (29.66)
	>$80,000	36.83 (44.83)	21.96 (35.66)	21.77 (30.31)
Father Education	<Associate Degree	59.95 (80.77)	26.39 (39.75)	21.01 (31.20)
	>Bachelor’s Degree	26.47 (29.43)	12.69 (18.64)	20.35 (28.27)
Mother Education	<Associate Degree	57.30 (48.71)	35.42 (37.52)	18.14 (30.01)
	>Bachelor’s Degree	42.05 (74.17)	14.64 (29.67)	22.02 (30.12)
Parental Education	<Post-Grad Training	73.25 (81.83)	23.98 (29.14)	19.34 (29.18)
	>Graduate Degree	29.14 (32.51)	12.29 (40.31)	23.11 (31.55)

Adolescents with annual household income ≤$79,999 experienced significantly longer duration between FS onset and first treatment as compared with adolescents with annual household income ≥$80,000 (t(65)=2.01, p<0.05). Adolescents of mothers with lower education (≤Associate’s Degree) experienced significantly longer duration between FS onset and first treatment (t(64)=1.70, p<0.05) and FS onset and diagnosis (t(49)=1.93, p<0.05) compared to adolescents of mothers with ≥Bachelor’s degree. Adolescents of fathers with lower education (≤Associate’s degree) experienced significantly longer duration between FS onset and first treatment (t(62)=2.24, p<0.05) compared to those with fathers with ≥Bachelor’s degree. No significant differences were found between onset and treatment (t(45)=1.70, p=0.35), diagnosis and treatment (t(35)=−0.52, p=0.91), or onset and diagnosis (t(49)=1.93, p=0.10) based on mother or father (combined parental) education utilizing a cut point of ≤Associate’s Degree vs. ≥Bachelor’s Degree. However, exploratory analyses revealed a significant difference between FS onset and treatment and FS onset and diagnosis when utilizing a cut point of ≤Post-Graduate Training vs. ≥Graduate Degree. Specifically, adolescents with a mother or father with a Graduate Degree experienced decreased time between onset and treatment (t(45)=2.56, p<0.05) and FS onset and diagnosis (t(49)=2.19, p<0.05) compared to those with a mother or father with ≤post-graduate training. No significant differences were found between onset and treatment (t(65)=0.09, p=0.93), onset to diagnosis (t(49)=−0.68, p=0.46), or diagnosis and treatment (t(35)=−0.38, p=0.69) based on race. Sex was significantly correlated with onset to diagnosis but did not have a significant effect on any models when assessed using an ANCOVA. See Table 3 for a summary of the results.

Table 3.

Independent Samples T-tests

		t	df	p-value
FS Onset to Treatment	Race	0.09	65	0.93
	Income	2.01	65	0.049*
	Father Education	2.24	62	0.03*
	Mother Education	1.70	64	0.04*
	Mother or Father Education	2.56	45	0.01*
FS Onset to Diagnosis	Race	−0.68	49	0.46
	Income	−0.22	49	0.83
	Father Education	1.32	49	0.19
	Mother Education	1.93	49	0.03*
	Mother or Father Education	2.19	49	0.03*
FS Diagnosis to Treatment	Race	−0.38	35	0.69
	Income	−0.28	35	0.78
	Father Education	−0.24	35	0.81
	Mother Education	−0.57	35	0.68
	Mother or Father Education	−0.28	35	0.78

^*Indicates p<0.05. Cut points for Father Education and Mother Education are ≤Associate’s Degree vs. ≥Bachelor’s Degree; Cut points for Mother or Father Education are ≤Post-Graduate Training vs. ≥Graduate Degree.

Discussion

The present study highlights the significant effects of socioeconomic status on time to treatment and diagnosis of pediatric FS in the United States, including household income and parental education.

First, adolescents with annual household income ≤$79,999 experienced longer duration between FS onset and first treatment as compared with adolescents with annual household income ≥$80,000. Specifically, adolescents with annual household income ≤$79,999 began treatment >6 months later. These findings are consistent with prior literature that has identified nearly two times greater odds of delayed healthcare delivery and difficulty accessing healthcare among lower income individuals.³¹ While the precise factors underlying these disparities remains unclear, they may be due to lack of access to health insurance often experienced by families with lower household income.²¹ Economically disadvantaged patients also report greater medical mistrust based on prior class discrimination²⁰ and fear of underlying financial motivations,³¹ both of which may contribute to delayed treatment. No significant differences were found between FS onset and diagnosis and FS diagnosis and treatment based on annual household income. These results are likely due to the smaller, less representative, and uneven samples.

Adolescents with lower paternal and maternal education also experienced increased time between FS onset and treatment. Specifically, adolescents whose mothers and fathers received ≤Associate’s Degree began treatment around 4 and 8.5 months later than adolescents of mothers and fathers with greater educational attainment, respectively. Adolescents with lower maternal education also experienced more than 5 months longer duration between FS onset and diagnosis. Further, results also showed that adolescents who have at least one parent with a graduate degree were diagnosed about 3 months earlier and began treatment more than 11 months earlier than adolescents whose mother or father did not receive a graduate degree. These results further emphasize health disparities based on socioeconomic status and suggest a tiered effect of education on increased time to treatment. Prior literature in other patient populations has found similar results, including one study that found a negative relationship between educational attainment and time to diagnosis and treatment of cervical cancer.³² Educational attainment is highly correlated with income, as it contributes to more stable, higher paying jobs.²¹ Thus, these results may also be due to lack of access to health insurance, lack of access to adequate healthcare, and increased medical mistrust.^{21, 31} Educational attainment is also associated with health literacy, which refers to the capacity to obtain and process health-related information and services needed to make health-related decisions.³³ Individuals with lower educational attainment may be less prepared to decipher reliable information pertaining to FS. They may also experience increased difficulty identifying appropriate treatment options which are already scarce.³⁴ No significant differences were found between time from FS diagnosis and treatment based on maternal, paternal, and combined parental education. Further, no significant differences were found between FS onset and diagnosis based on paternal education.

Contrary to our hypothesis, no significant differences were found between time between FS onset and treatment, FS onset and diagnosis, and FS diagnosis and treatment based on race. These findings contradict previous literature,^{35, 36} including one study which found longer time from diagnosis to treatment for lung cancer among Black patients.³⁷ However, this may be explained by the effects of other, more salient factors. One study investigating racial differences in time to treatment for melanoma found that racial differences in time from diagnosis to surgery were correlated with insurance. That is, individuals with Medicaid experienced significantly longer duration between diagnosis and surgery as compared with individuals with private insurance.³⁶ Another explanation may be the small sample of racial minority individuals with FS in this study (N=27, 33.8%).

Finally, it is important to note that mean durations between FS onset and treatment, FS onset and diagnosis, and FS diagnosis and treatment were 47.60, 20.75, and 20.76 weeks, respectively, for all adolescents. These significant delays ranging from more than 5 months to over a year may be due to the misdiagnosis of FS as epilepsy, which has been found to occur in up to 75% of adults who present with FS.¹ Further, this could be due to the lack of confidence experienced by healthcare providers in evaluating and diagnosing FS.³⁸ Related, stigma experiences from healthcare providers, friends and family and/or cultural beliefs about the relationship between disorders that affect both mind and body may result in patient uncertainty and non-acceptance of a FS diagnosis, which may drive patients to seek additional diagnostic testing that can further derail treatment.^{11, 39}These results may also be due to difficulty identifying an evidence-based treatment upon receiving a diagnosis,³⁴ as availability of competitive psychiatric services and treatment options for FS are severely limited.⁴⁰

These results are concerning given that the delayed treatment of FS is associated with poorer prognosis¹³ and adverse long-term psychosocial health outcomes⁴⁰, which may be exacerbated among adolescents of lower socioeconomic status. Delayed treatment of FS may also exacerbate impairments in quality of life among adolescents of lower socioeconomic status, including increased delays in returning to school and extracurricular and social activities.⁹ Families of lower socioeconomic status may also be adversely affected by increased financial burden due to exacerbations in absenteeism from work and repeated healthcare visits.⁸ Research aimed at assessing the mechanisms by which these relationships exist is critical given the significant consequences. Further, providers must consider tangible next steps for decreasing delayed diagnosis among adolescents of lower socioeconomic status. For example, clinicians may consider utilizing technology-based services to increase accessibility for patients who lack the resources to attend in-person visits. Given the misinformation and stigma associated with FS, clinicians may also consider increasing the amount of time spent with patients explaining their FS diagnosis and providing concrete next steps (i.e., treatment recommendations). Finally, systems may consider the importance of both training and integrating pediatric psychologists into settings in which FS are diagnosed.

The present study is the first to assess the effects of race and socioeconomic status on time to treatment and diagnosis of pediatric FS in the United States. Other strengths include a racially and age diverse sample, and EEG-confirmed FS diagnosis. Limitations include uneven sample among comparison groups. However, a power analysis indicated that a sample size of N=13 per group—which was the smallest sample size to produce significant results—achieved adequate power. Further, the present study was unable to assess the effects of ethnicity on time to diagnosis and treatment of pediatric functional seizures due to ethnic homogeneity of the sample. While data on racism and discrimination were not considered, the present study aimed to assess disparities in time to treatment and diagnosis. However, future research should consider the role of these factors in this relationship. Further, the present study was unable to assess potential overlap in the effects of annual household income and educational attainment on delayed diagnosis and treatment. However, measures of socioeconomic status typically combine multiple statistics, and both annual household income and educational attainment have demonstrated strong validity and reliability.⁴¹ Additionally, the present study utilized self-reported date of FS onset, date of FS diagnosis, and date of FS treatment for participants who received treatment prior to enrolling in the study. As the present study utilized a cutoff of $80,000 annual household income, it is important to also note that our findings may be exacerbated among individuals on the lower end of this financial spectrum. Finally, generalizability is limited due to geographical differences in access to services and healthcare availability both worldwide and within the United States. For example, video-EEG—the diagnostic standard for PNES—is widely available in tertiary care centers across the United States but limited in other countries (i.e., Zambia, China, Georgia).⁴² Thus, these results may be exacerbated in countries where access is limited. Conversely, these results may be exacerbated for individuals with lower income in the United States as universal healthcare is not available. Within the United States, these results may be short in comparison to the general population given there is bias in that all participants had video-EEG confirmed FS.

In conclusion, the present study found that adolescents with lower annual household income (<$79,999) experienced longer duration between FS onset and first treatment as compared with adolescents with greater annual household income (>$80,000). Further, adolescents of mothers and fathers with lesser educational attainment (≤Associate’s Degree) experienced longer duration between FS onset and first treatment as compared with adolescents of mothers and fathers with greater educational attainment (≥ Bachelor’s degree). Adolescents of mothers with lesser educational attainment also experienced longer duration between FS onset and diagnosis. Finally, adolescents with at least one parent with a graduate degree experienced decreased time between FS onset and diagnosis and FS onset and treatment. Overall, these findings emphasize disparities in the diagnosis and treatment of health conditions based on socioeconomic status which may contribute to poorer prognosis, lower quality of life, and increased financial strain.¹⁴ In an effort to eliminate these disparities, future research should assess the mechanisms by which these relationships occur.

Highlights.

• Teens with less parent education had more time between symptom onset and diagnosis

• Teens with less parent education had more time between symptom onset and treatment

• Teens with lower household income had more time between symptom onset and treatment

• Additional research is needed to understand and eliminate these disparities