Disparities in Epilepsy Diagnosis and Management in High-Income Countries

Abstract

Purpose of Review

Currently, an estimated 3.4 million people in the United States live with epilepsy. Previous studies have identified health disparities associated with race/ethnicity, socioeconomic status (SES), sex, insurance status, and age in this population. However, there has been a dearth of research addressing these disparities. We performed a literature review of articles published between 2010 and 2020 pertaining to health disparities in people with epilepsy (PWE), identified key factors that contribute to gaps in their care, and discussed possible solutions.

Recent Findings

Health disparities in prevalence, treatment access, time to diagnosis, health care delivery and engagement, and clinical outcomes were identified among individuals who were either of low SES, rural-based, uninsured/underinsured, older patients, patients of color, or female sex.

Summary

Disparities in care for PWE continue to persist. Greater priority should be placed on addressing these gaps intricately tied to sociodemographic factors. Reforms to mitigate health disparities in PWE are necessary for timely diagnosis, effective treatment, and positive long-term outcomes.

Introduction

The CDC estimated that over 3.4 million people are living with “active epilepsy” (i.e., physician-diagnosed epilepsy requiring medication and/or having at least one seizure in the past year) in the United States.¹ People with epilepsy (PWE) were more likely to report cognitive deficits, mental illness, and difficulties with their activities of daily living than those without epilepsy.^2-4 PWE suffer higher morbidity and mortality rates compared with the general population and experience delayed or inadequate treatment.^5,6 Health disparities are pervasive in the US health care system in regard to sociodemographic factors, sexual orientation, sex, geographic location, and disability status.¹ Previous systematic reviews have been conducted to identify disparities in epilepsy care according to various sociodemographic factors both within the United States and globally. A systematic review of global epilepsy treatment gaps conducted by authors of a study⁷ in 2010 examined 68 reviews and identified treatment gaps of over 50% in low- and middle-income countries (LMICs). Countries classified as high-income countries had less than a 10% gap in treatment; however, treatment gaps had significant variations across and within the individual countries. Notably, there were significantly higher treatment gaps in rural areas compared with nonrural areas. The International League Against Epilepsy conducted a robust systematic review in 2009 examining 278 studies in Canada, the United States, and the English-speaking Caribbean.⁸ This study examined disparities in epilepsy care according to socioeconomic status (SES), race/ethnicity, age, and education. The authors also compared disparities in people with epilepsy compared with both healthy individuals and individuals with chronic illnesses. The authors identified disparities in surgical outcomes, medical treatment outcomes, epidemiology, antiseizure medication (ASM) adherence, access to medical and surgical care, and rates of epilepsy surgery according to multiple sociodemographic factors. This review aims to re-examine barriers to epilepsy treatment (sociodemographic factors, geographic location, and disability status) in developed countries, to explore disparities associated with factors that have been poorly explored in previous reviews like sex and urban vs rural disparities, to assess changes that have occurred because previous reviews, and to touch on proposed avenues that may lead to better seizure control and quality of life (QOL).

Methods

Search Strategy

We queried the PubMed database for all articles published between the years 2010–2020 using the key search terms: ([epilepsy disparities] AND [patients with epilepsy] OR [PWE] OR [people with epilepsy] AND [developed nations] OR [developing countries] OR [high-income countries] OR [HIC]).

Study Selection

We included all articles focused on epilepsy disparities in countries with comparable economic systems with the United States, including those within Canada, the United Kingdom, Germany, and Norway. Exclusion criteria applied to articles focused on epilepsy disparities in LMICs, psychogenic nonepileptic seizures, and topics not related to the review's purpose. We also excluded articles published in non-English languages. All authors independently reviewed potentially eligible titles and abstracts and contributed to the review process, conducted between October 2021 and February 2022. Disagreements between the reviewers were infrequent and resolved by majority consensus. Where applicable, articles not meeting the inclusion criteria were excluded from the study. A visual of our search strategy is displayed in the Figure, while a summary of key findings from previous studies is displayed in the Table.

Figure. Search Strategy Flow Diagram.

Table.

Summary of Epilepsy Disparities Reported in Previous Studies

Year	Country of origin	Study design	Study population (n)	Male (%)	Mean age (y)	Key Finding(s)
20109	The United States	Survey	111	36.4	41.2	Compared with White individuals, AA individuals had higher seizure frequency and scores on the BMQ indicating a higher mistrust of medications
201010	The United States	Retrospective study	5,779	46.1	38.1	<10% of patients with TLE receive ATL. Younger age and private insurance are independent predictors of receiving ATL, and AA race independently predicts decreased likelihood of receiving ATL
2010e5	The United Kingdom	Retrospective study	846	NS	NS	Whereas 43.7% of adults younger than 20 years were using the specialist service, only 2.3% of people with epilepsy who were 85 years and older had been referred
201111	The United States	Retrospective study	108	54.6	42.3	Compared with White individuals, AA individuals have significantly poorer AED adherence, as measured by the MPR.
201112	The United States	Prospective study	566	44.7	42	Low SES patients was associated with increased ED and general practitioner visits, greater likelihood of having uncontrolled seizures, drug-related side effects, to be stigmatized, and have a poorer QoL
201213	The United States	Retrospective study	Patients receiving lobectomy (n = 6,653)	5.9	30.6	White patients were more likely to have surgery than racial minorities, and privately insured individuals were more likely to receive lobectomy than those with Medicaid or Medicare
			Patients not receiving lobectomy (n = 105,373)	94.1	31.9
201314	The United States	Survey	12,894	44.4	NS	Lifetime prevalence was 1.53% overall; 0.77% in White patients, 2.13% in AA individuals, and 3.4% in those with less than a high school diploma. Prevalence of active epilepsy was 0.79% and followed similar subgroup comparisons as lifetime prevalence
201315	The United States	Cross-sectional study	311	41.8	46	PWE received 40.9% of quality improvement recommended care. AA individuals were more likely to receive ≥50% of QI recommended care compared with White individuals. AA individuals scored significantly worse than White individuals for 2 patient-reported measures: perceived racial/ethnic disparities and difficulties getting follow-up appointments
2013e1	The United States	Survey	White individuals (n = 113)	30.1	42.9	Annual household income, receiving disability benefits, and number of AEDs being currently used significantly distinguished White individuals from AA individuals. Furthermore, AA individuals reported higher utilization of religion, denial, emotional support, positive reframing, and planning as coping reactions compared with White individuals
			AA individuals (n = 70)	28.6	43
2013e2	The United States	Retrospective study	213	45.1	37	Patients of Asian/Pacific Islander or AA race were significantly less likely to pursue surgical treatment of epilepsy compared with White individuals. Limited English proficiency was also significantly associated with lower odds of surgery
201316	The United States	Retrospective study	Prevalence sample (n = 684,516)	31.2	NS	The incidence and prevalence of epilepsy were significantly higher in men, in older people, in AA individuals, and in people with pre-existing disability and/or comorbid conditions
			Incidence sample (n = 80,478)	28.5
201317	The United States	Retrospective study	115,632	51	NS	Characteristics such as age, race/ethnicity, insurance status, and the presence of comorbid conditions were associated with disparities in access to specialized care in PWE.
2014e4	The United Kingdom	Retrospective study	15	40	67a	Patients interviewed did not cite difficulties in accessing hospitals, reluctance to attend clinics, older characteristics as limiting factors for older adults wanting to attend a specialist epilepsy clinic. However, there was concern for negative biases among professionals
201418	The United States	Retrospective study	760,117b	50.3	39.8	The incidence of SE was higher among AA individuals, compared with White individuals and other races. Mortality, however, was lower among AA individuals compared with White patients and other races. AA men had the highest incidence, relatively younger age of onset, and the lowest mortality
201419	Germany	Retrospective study	43,712	49.8	53.5	AED treatment differed significantly among adults with epilepsy in Germany. Sex, insurance type, and place of residence strongly influenced AED administration
201420	The United States	Retrospective study	Hispanic (n = 66,663)	44.9	Without epilepsy (28)	Charges for medical services provided to Hispanic PWE between the ages of 18 and 49 were significantly less than those for non-Hispanic patients with epilepsy. Attributable factors included differences in insurance status, setting of care, and total number of procedures
					With epilepsy (28)
			Non-Hispanic (69,157)	55.1	Without epilepsy (47)
					With epilepsy (48)
201421	The United States	Retrospective study	223	44.4	36.9	AA and Asian/PI patients, as well as those with limited English proficiency, have significantly longer times to ATL after presurgical evaluation
201522	The United States	Retrospective study	Hispanics (n = 38)	36.8	43.44	Depression scores were significantly higher in Hispanic PWEs than in United States–born PWEs. Hispanic patients were also found to be receiving significantly less AEDs compared with their United States–born peers
			United States–born (n = 47)	34	40.72
201523	The United States	Cross-sectional study	NS	54.2	NS	An increment in the rates of epilepsy surgeries was noted across all age groups, in boys and girls, all races, and all payer types. The rate of increase was lowest in AA individuals and in children with public insurance
201524	The United States	Retrospective study	695	ECoG patients (53)	25.9	White race, private insurance coverage, and large, urban, and academic hospitals were significantly were associated with higher ECoG procedures compared with AA patients, government insurance coverage, and smaller, rural, and private practice institutions
				Patients with epilepsy (49.6)	30.6
2016e3	Canada	Retrospective study	10,661	51	47	Within 2 years of being defined as medically intractable, only 1.2% of the study population underwent epilepsy surgery. Death occurred in 12% of those with medically intractable epilepsy. Those who underwent the procedure were younger and had fewer comorbidities compared with those who did not
201625	The United States	Retrospective study	67,733	49.4	NS	The prevalence of epilepsy in the sample population was 0.69%, with higher rates in Hispanic than in non-Hispanic children and adolescents. A child or adolescent with epilepsy had an additional $9,103.25 per year in associated medical costs relative to children and adolescents who did not have epilepsy
201626	Canada	Retrospective study	284	57	10.57	Patients in the lowest income quintile had a significantly higher TTS relative to the highest income quintile and had significantly lower odds of an improvement in seizure frequency relative to the highest income quintile
201627	The United States	Retrospective study	Deceased PWE (n = 21, 451)	51.8	61.8	Deceased PWE were more likely to be rural residents, AA, older than age 45 years, Medicare insured, in the middle-income group, and have 5 or more comorbid conditions compared with living PWE.
			Living PWE (n = 41, 343)	50.9	35.1
201728	The United States	Retrospective study	3240	35.7	NS	TLE diagnosis was significantly greater for AA individuals than White individuals. Women were disproportionately represented in the study population, and AA women carried the most statistical weight for the TLE prevalence difference
201729	The United States	Retrospective study	Prevalent epilepsy group (n = 36,912)	38.4	NS	Monotherapy is common across all racial/ethnic groups of older adults with new-onset epilepsy, older AEDs are commonly prescribed, and treatment is frequently delayed
			New epilepsy group (n = 3,706)	35.1
			Medicare random sample (n = 633,710)
201730	The United States	Retrospective study	Prevalent epilepsy group (n = 36,912)	38.4	NS	Beneficiaries without neurology care in deductible drug benefit phase or in high poverty areas were less likely to have QUIET-9 concordant care. Enzyme-induced AED use is high and concordance with recommendations low, among all racial/ethnic groups of older adults with epilepsy
			New epilepsy group (n = 3,706)	35.1
201731	The United States	Retrospective descriptive study	Kid's inpatient database (n = 6,672,829)	4.5	47.9	In both pediatric and adult admissions, there was an over-representation of Whites and underrepresentation of AA individuals, which persisted after stratifying by socioeconomic status. Female patients were underrepresented in epilepsy surgery, but gender disparities were partially explained by differences in socioeconomic status
			National Inpatient Sample (n = 29,963,139)	58.2	41.1
201732	The United States	Retrospective study	208	51.8	10a	AA patients and hemispherectomy were independently associated with NSQIP-defined complications. Patients undergoing hemispherectomy were also at a significantly higher risk of unplanned readmission after pediatric epilepsy surgery
201833	The United States	Retrospective study	1,964	NS	NS	ICM complication rates were comparable with, if not lower than, standard resective surgery. Disparities in access to ICM exist, with AA individuals and those with Medicaid significantly less likely to undergo ICM.
201934	The United States	Retrospective study	385	43	22	AA patients with TLE were more likely to be female, have seizure onset in adulthood, and have normal MRIs compared with White counterparts with TLE.
201935	The United States	Retrospective study	31,523	51	7a	Prevalence rates of childhood epilepsy were not significantly different across races/ethnicities
201936	The United States	Retrospective study	24,159	47.7	NS	There was a trend toward a decrease in the use of a VNS among adult patients with refractory epilepsy. Results also suggested that AA patients with refractory epilepsy were less likely to receive a VNS independently of other variables
201937	The United States	Retrospective study	776	54	6.5	After controlling for confounding socioeconomic and demographic factors, children of Hispanic ethnicity experienced a reduced likelihood of drug-responsive epilepsy and had longer median time to remission compared with White patients
2019e12	Norway	Survey	1,182	31.5	41.8	>90% of respondents wanted general information about epilepsy, and over 75% wanted information on more specific issues, like epilepsy surgery. Obtaining information about epilepsy surgery and neurostimulation was significantly associated with male sex
202038	The United States	Retrospective study	14,337	48.6	44	The overall median duration from first seizure code to epilepsy diagnosis code was 19.0 mo, and 56.0% filled an AED prescription. Some minorities were more likely to follow pathways with increased durations and delay to diagnosis, and the duration to diagnosis varied significantly across the care pathway
202039	The United States	Retrospective study	NS	NS	NS	Disparities in patients receiving epilepsy surgery by race/ethnicity were observed during 2010–2012, but they were not after 2013. Children with higher household income levels had consistently higher proportions of receiving epilepsy surgery than those with lower levels
202040	The United States	Retrospective study	AA (n = 58)	42	37.9	AA patients had a weaker relationship between lacosamide daily dose and serum levels as compared with White patients and required a higher lacosamide dose per weight to achieve similar levels
			White individuals (n = 35)	40	48.6

Abbreviations: AA = African American; AED = antiepileptic drug; ATL = anterior temporal lobectomy; BMQ-S = Beliefs About Medicines Questionnaire–Specific; CI = confidence interval; ECoG = electrocorticography; ICM = intracranial monitoring; MPR = medication possession ratio; NS = not specified; NSQIP = National Surgical Quality Improvement Program–Pediatric; OR = odds ratio; PWE = people with epilepsy; QoL = quality of life; QUIET-9 = Quality Indicator for Epilepsy Treatment 9; RR = relative risk; SE = status epilepticus; TLE = temporal lobe epilepsy; TTS = time to surgery.

^aMedian age.

^bRefers to patient discharges with potential inclusion of patients with multiple hospital discharges during the study duration

Standard Protocol Approvals, Registrations, and Patient Consents

This review did not involve a research study with human participants. Therefore, no patient consent was obtained, nor specific standard protocols were followed.

Data Availability

All data sources are easily accessible from the PubMed database and can be found through key search terms described above and/or respective sources cited in this review.

Results

We identified 327 articles using the PubMed database. Fifty-nine duplicates and 94 articles (including nonrelevant titles and those published in languages other than English) were identified and excluded from our review. An additional 53 articles were excluded during abstract screening. Of the remaining 121 records, studies occurring in LMICs (n = 48), not discussing epilepsy disparities (n = 27), or addressing psychogenic, nonepileptic seizures (n = 8) were identified and excluded. The remaining 38 articles met the inclusion criteria and were included in the final review.

Disparities Associated With Socioeconomic Status

SES describes the societal standing of an individual or group of individuals. It is assessed using various factors (e.g., level of education, income, and occupation) and is commonly involved in the development of health care disparities experienced by socially vulnerable groups.⁵ SES is an important determinant of health across all ages, particularly in children and adolescents.⁶ Low SES children are more than twice as likely to experience acute and chronic illnesses.^6,41,42 Increased prevalence of epilepsy in children with low SES was consistent across the existing literature, indicating that SES is intricately intertwined with epilepsy.⁴³ Barriers to optimal care and outcomes in PWE have been especially noted in individuals with low SES.⁴⁴ For this review, low SES is defined as an annual household income below 138% of the Federal Poverty level based on household size, which is the income level established by the United States for Medicaid qualification.

Nearly one-third of PWE have seizures refractory to medications, and low SES PWE face obstacles to optimal epilepsy care because of income limitations, inadequate insurance, and other factors.^45,46 These obstacles may be exacerbated by geographic location because rural areas often have lower SES, poorer transportation, and limited facilities compared with urban areas.^47,48 Analysis of global epilepsy treatment identified significantly higher gaps in care in both rural and low-income areas.⁷ Individuals receiving routine care in metropolitan epilepsy centers have greater access to specialized epilepsy care than PWE living in rural areas.⁴⁸ For example, epilepsy surgery is an especially effective treatment in refractory epilepsy, with up to 60% of good surgical candidates seeing seizure cessation.^45,49-51 However, low SES PWE are less likely to undergo epilepsy surgery, more likely to experience a lengthier duration between seizure and surgical operation, and have a lower probability of experiencing improvement in seizure frequency after surgery.^13,26,31,52 Rural PWE have decreased access to centers performing surgical evaluation and epilepsy surgery compared with their urban counterparts.^17,53,54

Socioeconomic factors such as the inability to afford medication can contribute to ASM nonadherence and poor seizure control.^54,55 A large percentage of PWE enrolled in Medicaid are African American persons (AAs), unemployed, and/or have low-income levels in addition to having low educational attainment.^27,56 Distrust of medical institutions, negative connotations of epilepsy surgery, lower health literacy, and exaggeration of epilepsy surgery risks all contribute to poor utilization of epilepsy surgery as a treatment modality for refractory epilepsy in low-income communities.¹⁶ Assessing mortality in PWE revealed that deceased individuals with epilepsy were more likely to be rural AA individuals younger than 45 years, Medicare insured, or living in zip codes where the annual median income was less than $36,000.^27,56 These individuals were less likely to be taking an ASM compared with higher SES individuals.¹⁹ Adults reporting active epilepsy were more likely to be Medicaid insured, unemployed, disabled, and experiencing barriers to health care access, such as inability to afford treatment and transportation issues, compared with adults reporting no history of epilepsy.⁵⁷ Furthermore, the CDC found that nearly half of adults with active epilepsy and seizures had an annual family income of less than $25,000.¹⁶

Insurance coverage also plays an integral role in access to specialized epilepsy care and positive outcomes achieved. PWE who were either uninsured or enrolled in Medicaid/Medicare had decreased access to specialized epilepsy care and were significantly less likely to undergo video-EEG monitoring (VEEG) for their epilepsy compared with PWE from higher SES groups.¹⁷ VEEG monitoring is essential to determine whether a person with epilepsy is a surgical candidate; therefore, decreased utilization of VEEG in uninsured or publicly insured patients may contribute to disparities in surgical intervention in these groups. Moreover, rural PWE with public health insurance were more likely to receive older ASMs, suggesting that these patients may not be receiving treatment reflecting optimal current care.¹⁶ An analysis of older PWE on Medicare using the Quality Indicator for Epilepsy Treatment 9 (QUIET-9) found that only 60% of these patients were treated according to the standards of care set by the QUIET-9.³⁰ Lack of concordance with the QUIET-9 standards was especially noted when PWE were paying for their medications out of pocket. This finding demonstrates the negative impact that poor drug coverage and low SES have on PWE regarding receiving appropriate ASM treatment and improved epilepsy outcomes.

A review of the US Medicare administrative claims database found that older adults with new-onset epilepsy were more likely to have delayed care and to be treated initially with older ASMs, particularly for lower-income individuals.²⁹ Intracranial monitoring (ICM) is an important tool used to evaluate PWE when their epileptogenic foci cannot be characterized using other techniques; however, children with Medicaid were significantly less likely to undergo ICM, which may factor into the decreased utilization of epilepsy surgery in managing these individuals.³³

People with uncontrolled seizures are at increased risk of negative health outcomes, have higher health care utilization, and spend more on health care.¹⁶ Owing to decreased access, engagement, and disparities experienced by low SES PWE, this group has poorer epilepsy outcomes and experiences more complications compared with higher SES PWE. PWE from low SES groups are also more likely to use emergency rooms, experience more hospitalizations, and are less likely to see a specialist for their epilepsy care.⁵⁶ Similarly, they had lower ASM compliance and poorer QOL, experienced more stigma regarding their diagnosis, and had a higher rate of drug-related adverse effects and uncontrolled seizures compared with their counterparts.^12,58

Disparities Associated With Race/Ethnicity

Increased epilepsy prevalence in people of color has been well documented in the literature. Nearly three-fold higher rates of both active epilepsy and age-adjusted lifetime epilepsy have been observed in AA persons compared with White patients.¹⁴ AA patients also have been found to have a higher rate of temporal lobe epilepsy (TLE)²⁸ and a significantly higher incidence but lower age of onset.¹⁸ A study analyzing the Medical Expenditure Panel Survey found the prevalence of epilepsy in a sample of over 60,000 kids and adolescents was 0.69%, with higher rates in Hispanic patients than in non-Hispanic patients.²⁵ However, an analysis within a Minnesota county demonstrated that epilepsy was higher in non-Hispanic patient, White patients, and AA patients when compared with Hispanic patients.³⁵ However, these contradicting results may have been due to differences in patient sampling.

Several studies have assessed whether people of color suffer more severe epilepsy than White PWE. One study on a state level found mortality increases in AA PWE compared with White PWE²⁷; however, a separate national study found lower mortality among AA patients with status epilepticus compared with White patients and other races.¹⁸ These conflicting results were likely due to the differing type of epilepsy and scale of the study affecting the observed association between race and epilepsy. Compared with White patients, AA patients and/or Hispanic patients⁵³ and children³⁹ had more frequent hospitalization rates and emergency room visits. Hispanic patients younger than 18 years had a longer length of hospital stay.²⁰

Lack of early diagnosis or intervention can lead to increased seizure frequency. One crucial factor is the time to diagnosis. Ethnically minoritized communities had a significantly longer time between initial seizure onset and diagnosis.³⁸ Similarly, only about half of newly diagnosed American Indian patients with epilepsy or Alaska Native patients with epilepsy encountered a neurologist within 30 days before diagnosis or 60 days after diagnosis. AA patients were least likely to see a neurologist.³⁰ In addition, the etiology of epilepsy in these communities may be underdiagnosed. One study found that AA patients with TLE were more likely to have a normal MRI interpretation³⁴ while concurrently having a significantly higher prevalence for TLE in another study of the same research group.²⁸ The authors attributed this inconsistency to possible differences in the skill of the physician, facilities, and/or the diagnostic machinery. Even after presurgical evaluation, people of color and those with limited English proficiency have shown significantly longer duration to a lobectomy compared with their White counterparts.²¹

Another racial disparity that has been well-reported is the lack of access to specialized care.²⁶ One study found that Hispanic PWE with less extensive health insurance coverage, Medicaid, or uninsured status had decreased access to treatment.²² There has also been a significantly lower rate of therapeutic surgery,¹³ particularly AA patients.^{10,19,23,36,59} Although the overall rate of surgeries for children with epilepsy increased between 1997 and 2009, this increase was lowest in AA children.²³ Even before surgery, AA patients were significantly less likely to receive electrocorticography (ECoG) than White patients.²⁴ Regarding higher intensity care, such as emergency department visits, AA patients were also found to have higher-quality indicators of recommended care (a composite measure for overall technical processes of care including first seizure assessment, initial treatment, chronic care, etc).¹⁵ The authors suggested this finding is likely due to a higher rate of hospitalizations from severe seizures leading to more intensive care and higher quality indicators of recommended care. This finding further highlights the possible utilization of emergency departments as primary epilepsy care. After the enactment of the Affordable Care Act in 2010, racial disparities in access to surgical treatment were found to have been reduced.⁶⁰ In another study conducted in 2017, there were no major racial/ethnic differences in the ASM prescription pattern in their cohort.²⁹ Therefore, equalizing health care access may play a significant role in mitigating these disparities.

Disparities in treatment outcomes also persist. Overall, Hispanic PWE had a reduced likelihood of achieving seizure freedom.³⁷ Some studies had found that AA patients had a higher rate of postsurgical complications³² and postoperative mortality,⁵³ but these results have been challenged.³² AA persons were also found to have higher perceived racial/ethnic disparities and patient-perceived difficulty obtaining follow-up appointments.¹⁵ The authors attributed patients' ability to obtain follow-up appointments to discontinuity of care, ineffective patient-provider communication, and perceived racial/ethnic disparities. Hispanic PWE also reported more seizure-related worries and higher depression scores, while receiving less treatment for mental health.²² Biological considerations may also play a factor in the lack of effectiveness of treatments. In one study, AA patients were found to have lower lacosamide levels compared with White patients after adjusting for age, sex, weight, and daily dose.⁴⁰ Thus, inadequate access and lower quality care may underlie the ineffectiveness of ASMs and surgical resection in achieving seizure freedom.

Personal characteristics also lead to differences in epilepsy treatment, including distrust of traditional medicine. AA patients had lower scores on a “Beliefs About Medicines” questionnaire corresponding to a higher distrust of medicine, which was associated with higher seizure frequency.⁹ AA patients also had lower adherence to ASMs, as measured by the medication possession ratio,¹¹ which could correlate to distrust of medicine. Outside of traditional medical strategies, AA PWE seem to use more engagement-type coping reactions such as religion or positive reframing but had more denial of their seizure status compared with their White counterparts.^e1 Patients with limited English proficiency may have limited knowledge about epilepsy with resultant distrust of medical services.^e2 Language barriers between the patient and physician contribute to less adequate care or instruction, corresponding to a lower patient rating of quality of care. Furthermore, miscommunication during an epilepsy presurgical evaluation could lead to patients' undervaluing the importance of the surgery. Thus, special consideration must be considered when counseling patients with limited English proficiency, and professional translators should be used when available.

Disparities Associated With Age

Although the utilization of epilepsy surgery in children has increased over the past 2 decades, the surgery rate remains below the expected level. This disparity persists in epilepsy surgery between children covered by private and public insurance.²³ In their study examining national trends of epilepsy surgery usage within the pediatric population between 1997 and 2009, the authors of a study found that the rate of increase was lowest among AAs, children younger than 1 year, and those with public health insurance.²³ In another study focused on refractory TLE and the utilization of anterior temporal lobectomy (ATL) on epilepsy outcomes, lower rates of ATL for older PWE with nonprivate insurance were also identified.³⁶ While high-quality studies addressing disparities in the treatment of older PWE and a lack of timely referral to epilepsy specialty centers exist, these reports are few compared with smaller studies described in the literature. In a retrospective cohort study focused on disparities in surgery among patients with intractable epilepsy, only 124 of 10,661 participants had undergone epilepsy surgery within the following 2 years. Patients receiving surgery were significantly younger than those who did not. Using a multivariable model, the authors found that older age with high comorbidity index was associated with a lower incidence of receiving surgery.^e3 These findings are reinforced by a second study that reported higher rates of ECoG among younger patients compared with the general epilepsy population (25.9 years vs 30.6 years).²⁴ In a third study assessing gaps in access to specialized epilepsy care in a large cohort of participants stratified by sex, race, insurance status, and comorbid conditions, adults older than 35 years were less likely to undergo ECoG monitoring.⁵⁰

Age disparities also exist in other countries with similar economies, such as the United Kingdom. In a small UK study analyzing 15 patients' (ranging from 49 to 84 years) perceptions of the referral of older adults (50 years and older) to a specialist epileptic clinic, 4 reasons were cited as contributing factors to the lack of referral: gaps in referral knowledge, complex differential diagnosis, unclear referral pathway, and duration of the time since onset.^e4 Although the outcomes of this study were limited by its sample size, the study suggested that providers may have varying assumptions on epilepsy management in older PWE.

The significance of timely referral and access to an appropriate-level treatment center was again emphasized in a second UK article addressing equitable access to specialty epilepsy services by older adults. In this study, the authors examined the epilepsy care records of 846 adults (aged 18 years and older) with the assistance of a prior epidemiologic study to estimate the proportion of local patients accessing their service in different age ranges.^e5 They found that 43.7% of adults younger than 20 years were using their regional specialist service compared with only 2.3% of adults aged 85 years and older. The authors therefore inferred that general practitioners and emergency room physicians were less likely to refer older patients to specialty centers.^e5

Disparities Associated With Sex

Some epilepsy syndromes may have sex-related incidence and prevalence.^e6 More women than men are diagnosed with idiopathic generalized epilepsy. Overall, focal epilepsies are more frequent in men, possibly attributable to a higher incidence of head injury because of participation in contact sports and dangerous occupations.^e7 In a study reviewing the incidence of concussion in 8 contact sports, the authors found that among team sports for high school male individuals, ice hockey players had the highest incidence of concussion. By contrast, soccer players had the lowest incidence of these injuries. At the professional level, similar concussion rates were found in both ice hockey and rugby players.^e7 These findings lend support to the notion that because male individuals are more likely to participate in contact sports, they are more likely to experience concussions and possibly epileptic complications. Conversely, women have been reported to experience cryptogenic focal-type epilepsies more frequently than men despite the fact that the incidence of epilepsy is generally higher in male persons.^e8 Female patients with catamenial epilepsy (CE), defined as a complex neuroendocrine condition with seizures centered around the perimenstrual period, experience a higher frequency of epilepsy during menses compared with epileptic women without CE.^e10 This condition is reported to affect 70% of women with epilepsy.^e9,e10 Changes in seizure sensitivity may also occur during times of puberty, pregnancy, and menopause.^e11

In one study based on a German epilepsy population, the authors observed that compared with women, men were less likely to receive lamotrigine and more likely to receive carbamazepine.¹⁹ However, additional factors, including insurance type and place of residence (urban vs rural and West vs East), influenced the type of ASM prescribed. According to the authors, the explanation for this finding in Germany remains unclear. In addition to the sex-based discrepancies that exist with ASM prescriptions, the same has been reported regarding undergoing epilepsy surgery, with women undergoing operative management less frequently than men.³¹

There are limited studies eliciting PWEs' perspectives and assessing whether these patients received sufficient information regarding epilepsy and available treatment options from their providers. A questionnaire found that 90% of PWE wanted more general information about the condition.^e12 Among female respondents, 78.6% wanted information about contraception and pregnancy. Of these, 65.2% reported having received information on this topic.^e12 Male individuals were more likely to have obtained information about epilepsy surgery, vagus nerve stimulation, nutrition, and weight.^e12 Despite the potential benefits, epilepsy surgery remains largely underutilized,⁴⁴ likely because it is a limited and expensive procedure. This study provided insights into the importance of the interaction between health care providers and PWE.

Discussion

Despite the decade-long gap between our review and the most recently reported review of epilepsy disparities,⁸ our findings provide great evidence that suboptimal epilepsy care persist in the United States and other developed countries with comparable economies because of sociodemographic and geographic factors.

With approximately 13.4% of the US population living below the federal poverty line and 19.3% of the population living in areas designated as rural^e13, economic disparities contribute significantly to diminished outcomes for patients. We identified persistent disparities that exist in epilepsy treatment for both patients with lower SES and patients living in rural areas. Poverty and insurance status were associated with less access to procedures needed for surgical evaluation, referral for epilepsy surgery, delayed treatment, and diminished outcomes compared with individuals with higher SES or living in urbanized areas. Rural residents had diminished access to health information compared with urban residents, even after accounting for race, income, and education levels.^e14 Low SES was associated with nonadherence to ASM regimens, older ASMs that may complicate care for the older patients because of more toxicity and drug interactions, and greater distrust in medical institutions.

Multiple studies have identified effective programs for addressing care disparities experienced by impoverished and rural communities.^45,e15 Epilepsy self-management combines the 3 patient-controlled aspects of epilepsy care, which are treatment management (taking medications as prescribed and communicating with health providers), seizure management (identifying and avoiding seizure triggers), and lifestyle management (getting adequate sleep and reducing stress). There is little support for SES-related disparities in self-management so that latter improvement strategies are consistent across all patient populations.⁴⁵ Project Access is a program that provides grants to both state and local agencies to improve young PWEs' quality of care, access to early diagnosis, and coordinated care, among other services. A recent review of this program found that education, outreach, community support, and professional guidance resulted in increased access to epilepsy care and better outcomes for these patients.^e15 In addition, the CDC has developed the “Managing Epilepsy Well Network,” which offers several evidence-based self-management programs and tools to help PWE improve their medication adherence, behaviors, and health outcomes.^e16 Finally, given that a significant portion of the extremely poor live in rural areas, increased care burdens for rural individuals compound barriers to optimal epilepsy care. Thus, improving access to epilepsy centers would greatly address the barriers experienced by low SES PWE.

Structural factors attributed to the area in which a patient receives care (such as lower quality care and less access to treatment facilities, neurologists, and/or medications) are problems that many people of color experience. In addition, patient-specific factors, such as less trust in the medical system or poorer adherence to medications, could contribute to higher hospitalization rates. These patient groups experience longer times to diagnosis, more delays in treatment, increased distrust of the medical community, and reduced access to surgery. AA PWE and Hispanic PWE and those with limited English proficiency are at risk for worse treatment and outcomes compared with White PWE. Several factors can play a role in explaining these results. First, the cost of health care can prevent access to adequate diagnostic tools and treatment because these communities are more likely to experience low SES. Second, distrust in the medical profession can also prolong the time before patients seek medical attention, reducing adherence to ASMs and willingness to pursue treatment. However, even when the latter patients seek out treatment, they are less likely to have access to a neurologist, have less accessibility to follow-up meetings, and have more surgical complications. Thus, in almost all aspects of epilepsy workup, treatment, and outcomes, underprivileged PWE endure worse statistics than their counterparts. Despite some discrepancies in previous literature, we identified a need for interventions that ensure adequate care is provided to these populations.

Mending the distrust between these patient groups and the medical profession is a crucial first step. There is a long history of structural racism and injustice in the medical institution, in which the US Public Health Service Syphilis study is likely the most notable example involving African American individuals.^e16 From 1932 to 1972, syphilis treatment was intentionally held without informed consent to study the “natural progression of untreated syphilis in African American men.” During the COVID pandemic, there were also numerous reports from AA individuals of mistreatment from health care providers.^e16 Therefore, there persists a need for a more culturally sensitive physician workforce, with an increased representation of people of color that have previously been underrepresented in community and academic hospitals. Patients also tend to feel more comfortable and adhere to medications prescribed by physicians who share similarities with them.^e16,e17,e18 Patient education regarding the signs of worsening epilepsy and options for transport to treatment facilities could significantly affect PWE outcomes. Patients must understand how detrimental an untreated or poorly treated seizure disorder can be to their QOL.

Outside of patient distrust, epilepsy disparities such as higher surgical complications and lower achievement of seizure freedom in minority PWE calls for more research into why these treatments are less beneficial in minoritized patient groups. Retrospective analyses of these outcomes are needed to determine why these patient groups experience an increased risk for complications. The factors involved may include the number of comorbidities, quality of the facility, tools used during surgical operation and during postoperative recovery, and effectiveness of postoperative instructions. Similarly, for seizure freedom, a review of whether physicians recommend different ASMs, monopharmacological or polypharmacological treatment, and surgical resection vs neuromodulation or ablation at equal rates, irrespective of a patient's race, is needed. Nonetheless, region-specific and national database studies can provide ample insight into the factors contributing most to the result reported in this review.

Several studies suggest an age discrepancy in surgical and medical treatment exists in the United States. Furthermore, delayed referrals to specialty centers add to this issue. This disparity also exists in countries outside the United States, such as the United Kingdom, where cost is not a factor. However, it is possible that the limited availability of neurologists and specialty centers in the United Kingdom contributes to this disparity. Furthermore, neurologists' assumptions regarding older PWEs' willingness to attend their appointments and the impact of epilepsy on their overall health have also been reported to undermine timely aid for these patients, thus jeopardizing their QOL. Conscious efforts to minimize implicit biases on the part of health professionals may serve as a partial solution to this age disparity and offer older PWE better long-term health outcomes.

There is growing evidence indicating that sex is an important variable in the type and prevalence of epilepsy experienced by PWE. A 2009 study by Reddy showed that different effects of sex hormones could lead to significant differences between male patients and female patients in treatment, prognosis, and social impact.^e10 These differences can range from the incidence of various seizure types to the over-representation and/or underrepresentation of epilepsy surgery. Solutions to reducing sex disparities could include a combination of approaches that include patient-centered education, traditional healer-centered education, expansion of public access to ASMs, and health system reforms to lessen the out-of-pocket expense for PWE.

The results of our research must be interpreted within the context of the study design. First, the search strategy included only articles written in the English language to avoid any possibility of misinformation in text translation. However, we recognize that this likely limited the number of otherwise eligible studies that could have been included in this review. Second, our search was limited to the PubMed database, which potentially would restrict the number of studies included. Third, our review focused solely on epilepsy disparities in patients within developed countries. Therefore, the findings reported in our study may not be generalizable to patients living in LMICs. Finally, owing to the largely retrospective nature of the studies included in our review, there is an increased potential for selection bias and confounding.

Epilepsy disparities in developed countries, including the United States, affect patient access, diagnosis and treatment timelines, surgery usage, medication adherence, and prognosis. These disparities are more pronounced when analyzed by SES, race, age, and sex. Solutions involve multifaceted approaches like government funding to enhance care access, community education initiatives, addressing health care provider biases, and self-management programs. Increasing the number of minority physicians can reduce implicit racial biases. Impact of institutional racism and physician mistreatment on epilepsy disparities requires deeper exploration, which may help identify and facilitate new avenues for a more culturally sensitive workforce.

Ongoing discussions on affordable health care insurance are crucial for ensuring equitable coverage regardless of race, SES, sex, or age. Promoting access to trusted community-based providers and specialists is vital, particularly through safety-net programs like Medicare, Medicaid, and the Children's Health Insurance Program. Thus, easier access to these sources should be promoted while also seeking ways to facilitate ongoing care for these patients.

The current literature lacks substantial support for age-related and sex-related disparities' underlying causes, demanding increased attention and practical improvements. Research into key factors contributing to these disparities, especially among women and older populations, is necessary. National studies evaluating intervention effectiveness and larger prospective studies are essential to gather specific data and devise targeted solutions.

Appendix. Authors

Name	Location	Contribution
Joseph S. Miller, BA	Ohio University Heritage College of Osteopathic Medicine	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data
Folake Oladele, BS	Memorial Sloan Kettering Cancer Center	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data
Darrian McAfee, BA	University of Maryland School of Medicine	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data
Christopher O. Adereti, MD	Ross University School of Medicine	Drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; analysis or interpretation of data
William H. Theodore, MD	National Institute of Neurological Disorders and Stroke, National Institutes of Health	Drafting/revision of the manuscript for content, including medical writing for content; study concept or design; analysis or interpretation of data
Elizabeth O. Akinsoji, MD, MPH	Department of Neurology, Emory University School of Medicine	Drafting/revision of the manuscript for content, including medical writing for content; study concept or design; analysis or interpretation of data

Study Funding

The authors report no targeted funding.

Disclosure

The authors report no relevant disclosures. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.