Racial and socioeconomic disparities in the advanced treatment of medically intractable pediatric epilepsy

Sandeep KandregulaDepartment of Neurosurgery, LSU Health Shreveport; and

Search for other papers by Sandeep Kandregula in
jns
Google Scholar
PubMed
Close
 MBBS
,
Danielle TerrellDepartment of Neurosurgery, LSU Health Shreveport; and

Search for other papers by Danielle Terrell in
jns
Google Scholar
PubMed
Close
 MD
,
Robbie BeylDepartment of Statistics, Pennington Biomedical Research Center, Baton Rouge, Louisiana

Search for other papers by Robbie Beyl in
jns
Google Scholar
PubMed
Close
 PhD
,
Anne FreelinDepartment of Neurosurgery, LSU Health Shreveport; and

Search for other papers by Anne Freelin in
jns
Google Scholar
PubMed
Close
 BS
,
Bharat GuthikondaDepartment of Neurosurgery, LSU Health Shreveport; and

Search for other papers by Bharat Guthikonda in
jns
Google Scholar
PubMed
Close
 MD
,
Christina NotarianniDepartment of Neurosurgery, LSU Health Shreveport; and

Search for other papers by Christina Notarianni in
jns
Google Scholar
PubMed
Close
 MD
, and
Jamie TomsDepartment of Neurosurgery, LSU Health Shreveport; and

Search for other papers by Jamie Toms in
jns
Google Scholar
PubMed
Close
 MD
Free access

OBJECTIVE

Racial and ethnic disparities in healthcare have gained significant importance since the Institute of Medicine published its report on disparities in healthcare. There is a lack of evidence on how race and ethnicity affect access to advanced treatment of pediatric medically intractable epilepsy. In this context, the authors analyzed the latest Kids’ Inpatient Database (KID) for racial/ethnic disparities in access to surgical treatment of epilepsy.

METHODS

The authors queried the KID for the years 2016 and 2019 for the diagnosis of medically intractable epilepsy.

RESULTS

A total of 29,292 patients were included in the sample. Of these patients, 8.9% (n = 2610) underwent surgical treatment/invasive monitoring. The mean ages in the surgical treatment and nonsurgical treatment groups were 11.73 years (SD 5.75 years) and 9.5 years (SD 6.16 years), respectively. The most common insurance in the surgical group was private/commercial (55.9%) and Medicaid in the nonsurgical group (47.7%) (p < 0.001). White patients accounted for the most common population in both groups, followed by Hispanic patients. African American patients made up 7.9% in the surgical treatment group compared with 12.9% in the nonsurgical group. African American (41.1%) and Hispanic (29.9%) patients had higher rates of emergency department (ED) utilization compared with the White population (24.6%). After adjusting for all covariates, the odds of surgical treatment increased with increasing age (OR 1.06, 95% CI 1.053–1.067; p < 0.001). African American race (OR 0.513, 95% CI 0.443–0.605; p < 0.001), Hispanic ethnicity (OR 0.681, 95% CI 0.612–0.758; p < 0.001), and other races (OR 0.789, 95% CI 0.689–0.903; p = 0.006) had lower surgical treatment odds compared with the White population. Medicaid/Medicare was associated with lower surgical treatment odds than private/commercial insurance (OR 0.603, 0.554–0.657; p < 0.001). Interaction analysis revealed that African American (OR 0.708, 95% CI 0.569–0.880; p = 0.001) and Hispanic (OR 0.671, 95% CI 0.556–0.809; p < 0.001) populations with private insurance had lower surgical treatment odds than White populations with private insurance. Similarly, African American patients, Hispanic patients, and patients of other races with nonprivate insurance also had lower surgical treatment odds than their White counterparts after adjusting for all other covariates.

CONCLUSIONS

Based on the KID, African American and Hispanic populations had lower surgical treatment rates than their White counterparts, with higher utilization of the ED for pediatric medically intractable epilepsy.

ABBREVIATIONS

APR-DRG = All Patient Refined Diagnosis Related Group; DBS = deep brain stimulation; ED = emergency department; KID = Kids’ Inpatient Database; LITT = laser interstitial thermal therapy; RNS = responsive neurostimulation; SEEG = stereo-EEG; VNS = vagus nerve stimulation.

OBJECTIVE

Racial and ethnic disparities in healthcare have gained significant importance since the Institute of Medicine published its report on disparities in healthcare. There is a lack of evidence on how race and ethnicity affect access to advanced treatment of pediatric medically intractable epilepsy. In this context, the authors analyzed the latest Kids’ Inpatient Database (KID) for racial/ethnic disparities in access to surgical treatment of epilepsy.

METHODS

The authors queried the KID for the years 2016 and 2019 for the diagnosis of medically intractable epilepsy.

RESULTS

A total of 29,292 patients were included in the sample. Of these patients, 8.9% (n = 2610) underwent surgical treatment/invasive monitoring. The mean ages in the surgical treatment and nonsurgical treatment groups were 11.73 years (SD 5.75 years) and 9.5 years (SD 6.16 years), respectively. The most common insurance in the surgical group was private/commercial (55.9%) and Medicaid in the nonsurgical group (47.7%) (p < 0.001). White patients accounted for the most common population in both groups, followed by Hispanic patients. African American patients made up 7.9% in the surgical treatment group compared with 12.9% in the nonsurgical group. African American (41.1%) and Hispanic (29.9%) patients had higher rates of emergency department (ED) utilization compared with the White population (24.6%). After adjusting for all covariates, the odds of surgical treatment increased with increasing age (OR 1.06, 95% CI 1.053–1.067; p < 0.001). African American race (OR 0.513, 95% CI 0.443–0.605; p < 0.001), Hispanic ethnicity (OR 0.681, 95% CI 0.612–0.758; p < 0.001), and other races (OR 0.789, 95% CI 0.689–0.903; p = 0.006) had lower surgical treatment odds compared with the White population. Medicaid/Medicare was associated with lower surgical treatment odds than private/commercial insurance (OR 0.603, 0.554–0.657; p < 0.001). Interaction analysis revealed that African American (OR 0.708, 95% CI 0.569–0.880; p = 0.001) and Hispanic (OR 0.671, 95% CI 0.556–0.809; p < 0.001) populations with private insurance had lower surgical treatment odds than White populations with private insurance. Similarly, African American patients, Hispanic patients, and patients of other races with nonprivate insurance also had lower surgical treatment odds than their White counterparts after adjusting for all other covariates.

CONCLUSIONS

Based on the KID, African American and Hispanic populations had lower surgical treatment rates than their White counterparts, with higher utilization of the ED for pediatric medically intractable epilepsy.

Epilepsy affects nearly 70 million people worldwide of all ages.1 Based on the 2015 Centers for Disease Control and Prevention statistics, nearly 3 million adults and 470,000 children are diagnosed with active epilepsy in the US. Based on the National Survey of Children’s Health in 2007, active epilepsy among children was estimated at around 6.3/1000, and the lifetime prevalence was estimated at around 10.2/1000.2 However, nearly 70% of epileptic patients could be seizure free if appropriate antiseizure medications were used. Epilepsy is medically refractory in nearly 30% of patients, of whom 20% may be surgical candidates.1 Earlier randomized controlled trials reported good seizure freedom in adults and children after resection of the temporal lobe in medically intractable epilepsy.35 This is particularly important in the case of children where early surgery and seizure freedom can significantly improve cognitive development.4 Recent data have shown increased utilization of epilepsy surgery in children with medically intractable epilepsy because of increased comprehensive epilepsy centers and expansion of presurgical evaluation.6

Although there has been an increase in the number of surgeries, several barriers, including patient-related factors, family- or physician-related factors, and health system factors such as insurance and epileptologist availability, exist for accessing advanced treatment for epilepsy.711 Among these, racial and socioeconomic disparities deserve special attention. The complex interaction of race/ethnicity with other socioeconomic factors can heavily influence access to care, especially in subspecialty care. Since the report "Unequal treatment: confronting racial and ethnic disparities in health care" from the Institute of Medicine was published in 2003, there has been an increase in awareness of disparities in healthcare.12

Prior studies on adults with medically refractory epilepsy have reported that minority populations, particularly African American and Hispanic populations, people with limited English proficiency, and people with Medicaid insurance have lower odds of access to surgical treatment.8,1317 In this context, we sought to analyze the racial/ethnic and socioeconomic disparities and the interaction effect of race and insurance in accessing the advanced treatment of pediatric medically intractable epilepsy using the Kids’ Inpatient Database (KID).

Methods

We queried the KID for the years 2016 and 2019. The KID releases data files once every 3 years. It comprises most of the inpatient hospitalizations of children across the US. The KID includes 80% of the pediatric discharges in the US, with data from more than 4000 US hospitals. The data were queried to diagnose medically intractable epilepsy using the ICD-10 diagnostic codes. After extraction of the hospitalizations with diagnosis, data were queried for the presence of the procedures stereo-EEG (SEEG), deep brain stimulation (DBS), responsive neurostimulation (RNS), vagus nerve stimulation (VNS), resection of the focal lesion, and laser interstitial thermal therapy (LITT) using the ICD-10 procedural codes (Table 1). Inclusion criteria were patients who were younger than 20 years and admitted with a primary diagnosis of medically intractable epilepsy. Demographics such as age, sex, race/ethnicity, insurance status, and the median income of the zip code were collected. Race was recorded into four different categories: White, African American, Hispanic, and other. Other races included Asian Americans, Native Americans, and Pacific Islanders. The payer/insurance status includes Medicare, Medicaid, private/commercial, and uninsured. Comorbidity status was assessed using the All Patient Refined Diagnosis Related Group (APR-DRG) risk of severity and mortality. The APR-DRG is a standardized method of assessing the comorbidities and risk of mortality based on the diagnosis and clinical condition of the patient.18,19 The Healthcare Cost and Utilization Project data provided this information for each patient. The location of the patient was provided based on the 6-category classification of the National Center for Health Statistics.

TABLE 1.

Details of the ICD-10 codes used for the data extraction

Diagnosis/ProcedureICD-10 Code
Medically intractable epilepsyG40.019, G40.119, G40.219, G40.319, G40.A19, G40.B19, G40.419, G40.509, G40.804, G40.814, G40.824, G40.834, G40.89, G40.919
Temporal sclerosisG93.81
DBS00H00MZ, 00H03MZ, 00H04MZ
RNS0NH00NZ
SEEG00H002Z, 00H032Z, 00H042Z
LITTD0Y1KZZ, D0Y0KZZ
Resection00B70ZZ, 00B03ZZ, 00B04ZZ, 00B73ZZ, 00B74ZZ, 00500ZZ, 00504ZZ, 00503ZZ, 00570ZZ, 00573ZZ, 00574ZZ, 00B00ZZ, 00B60ZZ, 00B63ZZ, 00B64ZZ

Statistical Analysis

All categorical variables were described as proportions and frequencies and continuous variables as mean with SD or median with the IQR as appropriate. Groups were analyzed using the chi-square test for the categorical variables and the one-way ANOVA/Student t-test for the continuous variables. All variables were assessed for normality using the Shapiro-Wilk test. Logistic regression analysis was performed to find the predictors of access to the surgical procedure in medically intractable epilepsy. Because of the deficient proportion of Medicare patients, Medicaid and Medicare were clubbed together for the regression analysis. The Hosmer-Lemeshow test was used to assess the fitness of the model. Missing values for all variables were handled using the multiple imputation method. The patient location was coded into metropolitan and nonmetropolitan based on the National Center for Health Statistics scheme. The metropolitan population included central counties, fringe counties, counties in metropolitan areas of 250,000–999,999, and counties in metropolitan areas of 50,000–249,999 population. Micropolitan counties and not-metropolitan or micropolitan counties were classified as a single group (nonmetropolitan group).

Furthermore, an interaction analysis was performed to evaluate the effect of race/ethnicity and insurance on the outcome variable. Insurance was dichotomized into private/commercial and not-private insurance for the interaction analysis. Odds ratios along with 95% confidence intervals were presented. A p value < 0.05 was considered statistically significant. All statistical analyses were performed using SAS software version 9.4 (SAS Institute).

Results

Surgical Treatment Versus Nonsurgical Treatment

A total of 29,292 patients were included in the sample; 8.9% (n = 2610) of the patients underwent surgical treatment/invasive monitoring. The mean age in the surgical treatment group was 11.73 years (SD 5.75 years) and that in the nonsurgical treatment group was 9.6 years (SD 6.16 years). The most common insurance in the surgical group was private/commercial (55.9%) and Medicaid in the nonsurgical group (47.7%) (p < 0.001). White patients were the most common population in both groups, followed by Hispanic patients. African American patients represented 7.9% of the surgical treatment group compared with 12.9% in the nonsurgical group. The mean hospital charges in the surgical treatment group were $207,668 (SD $155,980) and $39,368 (SD $69,524) in the nonsurgical group (p < 0.001). Large bedsize hospitals were the most common hospitals (68.6% vs 68.7%). Urban teaching hospitals were the most common hospitals in both groups (Table 2).

TABLE 2.

Demographic details based on the surgical treatment for medically intractable epilepsy

VariableSurgical Treatment (n = 2610)Nonsurgical Treatment (n = 26,682)p Value
Mean age (SD), yrs11.73 (5.75)9.55 (6.16)<0.001
Female sex46% (1200)47.9% (12,781)
Insurance<0.001
 Medicaren < 100.3% (93)
 Medicaid35.8% (934)47.7% (12,731)
 Private/commercial55.9% (1458)44.9% (11,990)
 Uninsured8.1% (211)7% (1869)
Race<0.001
 White59.5% (1552)50.3% (13,430)
 African American7.9% (206)12.9% (3452)
 Hispanic16.8% (438)20.7% (5512)
 Asian/Pacific Islander4% (103)3.1% (827)
 Native American0.7% (17)0.6% (154)
 Other5% (130)6.4% (1721)
 Missing6.2% (163)5.9% (1585)
APR-DRG risk of severity<0.001
 1 (minor loss of function, no comorbidities)16.5% (430)2.2% (584)
 2 (moderate loss of function)65.4% (1707)2.4% (637)
 3 (major loss of function)15.4% (402)93.2% (24,862)
 4 (extreme loss of function)2.7% (70)2.2% (600)
APR-DRG risk of mortality<0.001
 1 (minor likelihood of dying)79.9% (2084)80.4% (21,457)
 2 (moderate likelihood of dying)15.3% (399)15.2% (4056)
 3 (major likelihood of dying)3.9% (102)2.8% (750)
 4 (extreme likelihood of dying)0.9% (25)1.6% (419)
Median income of zip code<0.001
 $1–49,99920.1% (6751)25.8% (511)
 $50,000–64,99922.5% (6195)23.7% (573)
 $65,000–85,99928.3% (6728)25.7% (719)
 ≥$86,00029.1% (6500)24.8% (738)
Patient location<0.001
 Central counties of metro areas ≥1 million population30.3% (777)36.2% (9565)
 Fringe counties of metro areas ≥1 million population26.7% (684)25.7% (6784)
 Counties in metro areas of 250,000–999,999 population19% (488)18.1% (4777)
 Counties in metro areas of 50,000–249,999 population9.8% (250)7.8% (2047)
 Micropolitan counties9% (230)7.1% (1874)
 Not metropolitan or micropolitan counties5.3% (136)5.1% (1354)
 Missing1.7% (44)1.1% (282)
Total charges (SD), $207,668 (155,980)39,368 (69,524)<0.001
Hospital bedsize0.05
 Small7.8% (205)9.1% (2429)
 Medium23.5% (614)22.2% (5924)
 Large68.6% (1791)68.7% (18,329)
Hospital teaching status<0.001
 Ruraln < 100.5% (130)
 Urban nonteachingn < 104.4% (1173)
 Urban teaching99.6% (2610)95.1% (25,379)
Hospital control<0.001
 Government/federal9.8% (256)10.8% (2871)
 Private nonprofit89.2% (2327)86% (22,959)
 Private profit1% (27)3.2% (851)
Hospital region<0.001
 Northeast15.6% (407)21.1% (5620)
 Midwest28.4% (742)25.2% (6724)
 South30.7% (802)31.1% (8292)
 West25.2% (659)22.7% (6047)

Boldface type indicates statistical significance. Values are presented as the percentage of patients (number of patients) unless stated otherwise. Variables with fewer than 10 patients are represented as n < 10 based on Healthcare Cost and Utilization Project guidelines.

In the nonsurgical group, 93.2% belonged to grade 3 of the APR-DRG risk of severity, whereas in the surgical treatment group, 65.4% belonged to grade 2. Higher-income quartiles (by zip code) had higher surgical treatment rates (29.1% vs 24.8%). In the surgical group, 30.3% lived in the central counties of metropolitan areas, whereas 36.2% in the nonsurgical treatment group lived in these areas. Complete details are provided in Table 2.

Differences Based on Race/Ethnicity

The most common insurance in the White population was private/commercial (57.8%) and Medicaid for the African American (64.8%) and Hispanic (66%) populations. Private insurance was also the most common in the population of other races (48.1%). Higher-income quartiles (3rd and 4th) accounted for 57% of the White population and 59.5% of the population of other races, but it accounted for only 31.6% of the African American population and 41.9% of the Hispanic population (Table 3). Additionally, African American (41.1%) and Hispanic (29.9%) patients had higher rates of emergency department (ED) utilization compared with White patients (24.6%) (Table 3).

TABLE 3.

Demographic details of the patients with medically intractable epilepsy by race/ethnicity

VariableWhite (n = 14,983)African American (n = 3658)Hispanic (n = 5950)Other (n = 4701)
Age (SD), yrs9.9 (6.15)9.6 (6.33)9.6 (6.08)9.2 (6.11)
Female sex48.2% (7223)45.5% (1663)48.5% (2885)47% (2211)
Insurance
 Medicare0.3% (47)0.8% (29)0.3% (16)n < 10
 Medicaid35.7% (5353)64.8% (2369)66% (3929)42.8% (2014)
 Private/commercial57.8% (8667)28.7% (1049)24.7% (1468)48.1% (2263)
 Uninsured6.1% (915)5.8% (210)9% (537)8.9% (417)
APR-DRG risk of severity
 1 (minor loss of function, no comorbidities)4% (594)2.7% (99)2.7% (163)3.4% (158)
 2 (moderate loss of function)9.2% (1377)5.3% (193)6.7% (398)8% (376)
 3 (major loss of function)84.8% (12,698)89.8% (3284)87.7% (5218)86.4% (4063)
 4 (extreme loss of function)2.1% (314)2.2% (81)2.9% (171)2.2% (104)
APR-DRG risk of mortality
 1 (minor likelihood of dying)80.6% (12,080)80.6% (2947)77.9% (4636)82.5% (3878)
 2 (moderate likelihood of dying)15% (2254)14.6% (535)17.2% (1024)13.6% (641)
 3 (major likelihood of dying)2.9% (429)2.9% (106)3% (179)2.9% (137)
 4 (extreme likelihood of dying)1.5% (219)1.9% (69)1.9% (112)0.9% (44)
Median income of zip code
 $1–49,99918.9% (2799)46.1% (1663)32.9% (1926)19.6% (874)
 $50,000–64,99924% (3555)22.3% (803)25.2% (1475)20.9% (934)
 $65,000–85,99927.8% (4108)19% (683)25.5% (1493)26% (1163)
 ≥$86,00029.2% (4323)12.6% (456)16.4% (962)33.5% (1497)
Patient location
 Central counties of metro areas ≥1 million population23.3% (3486)49.8% (1821)55.2% (3282)37.3% (1753)
 Fringe counties of metro areas ≥1 million population28.1% (4205)21.4% (783)18.6% (1099)29.4% (1381)
 Counties in metro areas of 250,000–999,999 population19.9% (2977)17.2% (629)17.1% (1010)13.8% (649)
 Counties in metro areas of 50,000–249,999 population10.6% (1585)4.7% (170)4.3% (257)6.1% (285)
 Micropolitan counties10.1% (1507)4.3% (157)2.9% (173)5.7% (267)
 Not metropolitan or micropolitan counties7.7% (1147)2.5% (90)1.5% (91)3.4% (162)
 Missing0.5% (76)n < 100.7% (40)4.3% (203)
VNS1.6%1.3%1.4%1.4%
RNS0.5%0.2%0.5%0.3%
SEEG2.8%1.3%1.8%2.9%
DBS1.6%0.8%1.4%1.3%
LITT0.6%0.3%0.3%0.4%
Resection5.3%2.4%3.5%4.7%
ED utilization for seizures24.6% (3685)41.1% (1466)29.9% (1781)22.3% (1235)
Hospital bedsize
 Small10.7% (1608)11.3% (414)5.8% (345)5.7% (267)
 Medium21.8% (3265)21.8% (796)23.5% (1401)22.9% (1075)
 Large67.5% (10,110)66.9% (2448)70.7% (4204)71.4% (3358)
Hospital teaching status
 Rural0.6% (91)0.3% (11)n < 100.6% (27)
 Urban nonteaching4.1% (619)3.8% (140)4.9% (290)2.8% (131)
 Urban teaching95.3% (14,274)95.9% (3506)95% (5655)96.6% (4542)
Hospital control
 Government/federal9.9% (1484)15.4% (564)11.9% (709)7.9% (370)
 Private nonprofit87.7% (13,140)78.5% (2871)85.2% (5069)89.5% (4206)
 Private profit2.4% (359)6.1% (222)2.9% (172)2.7% (125)
Hospital region
 Northeast18.1% (2710)21.8% (798)19.7% (1173)28.6% (1345)
 Midwest33.7% (5055)23.9% (23.9)10.5% (623)19.4% (913)
 South29.9% (4484)43.9% (1607)28.8% (1714)27.4% (1288)
 West18.2% (2734)10.3% (377)41% (2441)24.5% (1153)

Values are presented as the percentage of patients (number of patients) unless stated otherwise. Variables with fewer than 10 patients are represented as n < 10 based on Healthcare Cost and Utilization Project guidelines.

Factors Predicting Surgical Treatment

After adjusting for all covariates, the odds of surgical treatment increased with increasing age (OR 1.06, 95% CI 1.053–1.067; p < 0.001). Sex did not predict surgical treatment rates (female sex: OR 0.926, 0.854–1.003; p = 0.06). African American patients (OR 0.513, 95% CI 0.443–0.605; p < 0.001), Hispanic patients (OR 0.681, 95% CI 0.612–0.758; p < 0.001), and patients of other races (OR 0.789, 95% CI 0.689–0.903; p = 0.006) had lower surgical treatment odds compared with White patients (Fig. 1). Patients with Medicaid/Medicare had lower surgical treatment odds compared with those with private/commercial insurance (OR 0.603, 0.554–0.657; p < 0.001). Rural and urban nonteaching hospitals had lower surgical treatment odds than urban teaching hospitals (OR 0.08, 95% CI 0.045–0.148; p < 0.001) (Table 4). The odds of surgical treatment increased with an increase in income quartile (OR 1.13, 95% CI 1.095–1.177; p < 0.001).

FIG. 1.
FIG. 1.

Forest plot showing the odds ratios of insurance and race/ethnicity in predicting the surgical treatment for pediatric medically intractable epilepsy. The reference line for race indicates White population and private/commercial payer for insurance.

TABLE 4.

Factors predicting the surgical treatment of pediatric medically intractable epilepsy

FactorOR95% CIp Value
Age1.061.053–1.067<0.001
APR-DRG severity0.080.076–0.089<0.001
APR-DRG mortality1.010.945–1.0830.72
Median income zip code quartiles1.141.101–1.183<0.001
Female (ref: male)0.9260.854–1.0030.06
Race
 African American (ref: White)0.5130.443–0.605<0.001
 Hispanic (ref: White)0.6810.612–0.758<0.001
 Other (ref: White)0.7890.689–0.9030.006
Insurance
 Medicaid/Medicare (ref: private/commercial)0.6030.554–0.657<0.001
 Uninsured (ref: private/commercial)0.9290.798–1.0820.34
Hospital bedsize
 Small bed (ref: large)0.8630.742–1.0030.05
 Medium bed (ref: large)1.060.963–1.1670.232
 Rural & urban nonteaching (ref: urban teaching)0.0810.045–0.148<0.001
Nonmetropolitan (ref: metropolitan)1.1771.048–1.322 0.0059

Boldface type indicates statistical significance.

Interaction of Race/Ethnicity and Insurance

Interaction analysis revealed that African American (OR 0.708, 95% CI 0.569–0.880; p = 0.001) and Hispanic (OR 0.671, 95% CI 0.556–0.809; p < 0.001) patients with private insurance have lower surgical treatment odds than White patients with private insurance. Similarly, African American patients, Hispanic patients, and patients of other races with nonprivate insurance also had lower surgical treatment odds than their White counterparts (Fig. 2) after adjusting for all other covariates.

FIG. 2.
FIG. 2.

Forest plot showing the interaction of race/ethnicity and insurance in predicting the surgical treatment of pediatric medically intractable epilepsy. The reference line for race indicates White population.

Discussion

We start our discussion by acknowledging limitations of the data. First and foremost, coding errors are common in national administrative databases. The measurement and categorization of race and ethnicity in the national administrative database are complex. Race and ethnicity are complex, fluid social constructs that can change with the context, place, and time. Several studies have shown the discordance between self-reported race/ethnicity and the information recorded in the administrative data, making it challenging to generalize the results. Second, a statistical association with a particular variable may not imply causation. These statistical associations can be due to unmeasured confounding factors (disease-related factors influencing the decision) in the data. Because of limitations of the study design, we could not account for access to epilepsy care/epilepsy center availability within the patient location, compliance with antiepileptic drugs, knowledge of and attitudes toward the disease, education, and employment status. Some of the disparities may also be due to the availability of epileptic surgeons or referral patterns to neurologists. This information cannot be accounted for through the KID. Data missing with respect to race/ethnicity was another important limitation of our study. Approximately 6% of the population’s race-related data were not available. However, all missing data were handled using a multiple imputation model, which was a standardized method. With these limitations in mind, we would like to discuss our study’s main findings.

Our findings suggest that there are racial and socioeconomic disparities in access to surgical treatment of medically intractable epilepsy in the pediatric population. The African American and Hispanic populations had lower surgical treatment rates than the White population and higher rates of ED utilization. Along with this, patients with Medicaid/Medicare had lower rates of surgical treatment. Our analysis reveals a dynamic interaction between race/ethnicity and insurance coverage. Black and Hispanic patients had lower surgical treatment rates regardless of insurance status—private or public—in comparison with White patients. This is an important finding, demonstrating that insurance alone—which typically improves patient access to care—does not erase the systemic barriers to healthcare that people of color encounter in the American healthcare system. Other studies in different specialties have shown that the treatment refusal rates are higher in African Americans. This phenomenon stems from different perspectives on healthcare and the deep mistrust that Black Americans have of the healthcare system because of incidents like the forced medical experimentation of enslaved Black Americans, the Tuskegee trials, the treatment of Henrietta Lacks, and forced sterilization centers that persisted for much of the 20th century. Physician-related factors, such as communication skills, implicit biases, and cultural beliefs, also play a role in perpetuating disparities.7,10 Comorbidities have an inverse relationship with surgical treatment rates, and older children were more likely to undergo surgery than younger ones. One could argue that the differences in race/ethnicity were primarily due to differences in the prevalence rates of epilepsy by race. However, there is ample evidence suggesting that the epilepsy rates are similar among White, Black, and other minority populations in the US.2024 Comorbidities have an inverse relationship with surgical treatment rates, and older children were more likely to undergo surgery than younger ones.

A recent National Health Interview Survey reported that adults with epilepsy were more likely to have Medicaid and Medicare insurance than private insurance compared with adults without epilepsy, indicating the importance of insurance in access to care for epilepsy.7 Also, based on Centers for Disease Control and Prevention statistics, age-adjusted prevalence rates of non-Hispanic White patients (1.1%) and non-Hispanic Black patients (1.2%) have higher rates of active epilepsy compared with the Hispanic population.

A study from Canada reported similar findings with respect to health resource utilization in minority populations. Jetté et al. reported that aboriginal patients in the Calgary Health Region were less likely to visit a neurologist (OR 0.3, 95% CI 0.2–0.6) and more likely to use the ED than nonaboriginal patients.25 These findings mirror the Black and Hispanic populations in the US.

Schiltz et al., in their study of 115,632 patients with intractable seizures, reported that the Black (OR 0.54, 95% CI 0.47–0.60) and Hispanic (OR 0.87, 95% CI 0.80–0.95) populations had lower rates of access to epilepsy care after adjusting for all potential covariates, including the availability of epilepsy care centers by linking the area resource file to the National (Nationwide) Inpatient Sample data.8 Another interesting finding in their study was that availability of epilepsy centers did not influence access to specialized care in the case of patients with private insurance; however, it negatively predicted access to specialized care for Medicare and Medicaid insurance.

In their study, Sharma et al. reported no racial disparities in access to surgical treatment of temporal lobe epilepsy in the adult population based on the National (Nationwide) Inpatient Sample (2012–2013).13 However, our study has significant methodological differences from theirs. First, we included all the medically intractable epilepsy diagnoses, including temporal sclerosis. Second, our outcome variable was any surgical treatment (e.g., DBS, RNS, LITT, and resection). The main limiting factor in their study was the exceptionally low sample of African American patients (n = 11), which they acknowledged.

Similarly, another study by Burneo et al. highlighted the disparities in surgical treatment rates by race and socioeconomic status.15 Contrary to other epidemiological studies, this was based on a single epilepsy center in the US. In an unadjusted analysis, African American patients and those with Medicaid insurance had lower rates of surgery than White patients. However, in multivariate analysis, after adjusting for socioeconomic status, education, and bitemporal epileptogenesis, race/ethnicity failed to predict the surgical treatment access, with the limiting factor of the study being a small sample size (n = 100). In addition, all patients underwent presurgical evaluation and were deemed to be candidates for surgical intervention. It is unclear why Black patients had lower rates of surgery in their study.

All of these studies compared only one modality as their outcome (such as video-EEG or resection). Our study is unique because it focuses on any specialized surgical care (such as LITT, DBS, SEEG, RNS, and resection), which represents a more complete picture of the surgical care of patients with epilepsy. Another important finding in our study is the association of income with surgical treatment rates. The interaction between poverty and epilepsy is complex. Elliott et al. showed that individuals living in poverty were less likely to report taking medications for epilepsy, signifying the cost of medications as one of the most critical barriers to epilepsy care.26 A recent National Health Interview Survey showed that 27.9% of patients reported that they could not afford to pay their medical bills.7 In our study, the interaction analysis revealed that patients in higher income quartiles have higher surgical treatment rates irrespective of the insurance coverage.

Nathan and Gutierrez detailed the factors that play a vital role in surgical treatment disparities as FACETS:10 fear of treatment, access to care, communication barriers, education level, patient’s trust in the physician, and social support. Interaction of these factors can contribute to disparities in epilepsy care. Because of potential limitations of the administrative database, our study could not address the factors mentioned above, stressing the need for future studies to focus on these aspects. Taken together, our findings give a bird’s-eye view of surgical epilepsy care in the US by population. However, there are several limitations of the national database to validate these results. In the current literature with the given limitations, these are the most relevant findings we could associate with socioeconomic disparities.

Conclusions

Based on the KID, African American and Hispanic populations had lower surgical treatment rates than their White counterparts, with higher utilization of the ED for seizures.

Acknowledgments

This study was supported in part by grant no. U54 GM104940 from the National Institute of General Medical Sciences of the National Institutes of Health, which funds the Louisiana Clinical and Translational Science Center.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Toms, Kandregula. Acquisition of data: Kandregula. Analysis and interpretation of data: Kandregula, Beyl. Drafting the article: Kandregula, Terrell, Freelin. Critically revising the article: Toms, Kandregula, Terrell, Freelin, Guthikonda, Notarianni. Reviewed submitted version of manuscript: Toms, Kandregula, Terrell, Freelin, Guthikonda, Notarianni. Approved the final version of the manuscript on behalf of all authors: Toms. Statistical analysis: Kandregula, Beyl. Study supervision: Guthikonda, Notarianni.

References

  • 1

    Zack MM, Kobau R. National and state estimates of the numbers of adults and children with active epilepsy—United States, 2015. MMWR Morb Mortal Wkly Rep. 2017;66(31):821825.

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Russ SA, Larson K, Halfon N. A national profile of childhood epilepsy and seizure disorder. Pediatrics. 2012;129(2):256264.

  • 3

    Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med. 2001;345(5):311318.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4

    Dwivedi R, Ramanujam B, Chandra PS, et al. Surgery for drug-resistant epilepsy in children. N Engl J Med. 2017;377(17):16391647.

  • 5

    Engel J Jr, McDermott MP, Wiebe S, et al. Early surgical therapy for drug-resistant temporal lobe epilepsy: a randomized trial. JAMA. 2012;307(9):922930.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Pestana Knight EM, Schiltz NK, Bakaki PM, Koroukian SM, Lhatoo SD, Kaiboriboon K. Increasing utilization of pediatric epilepsy surgery in the United States between 1997 and 2009. Epilepsia. 2015;56(3):375381.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    Tian N, Kobau R, Zack MM, Greenlund KJ. Barriers to and disparities in access to health care among adults aged ≥18 years with epilepsy—United States, 2015 and 2017. MMWR Morb Mortal Wkly Rep. 2022;71(21):697702.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    Schiltz NK, Koroukian SM, Singer ME, Love TE, Kaiboriboon K. Disparities in access to specialized epilepsy care. Epilepsy Res. 2013;107(1-2):172180.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9

    Burneo JG, Jette N, Theodore W, et al. Disparities in epilepsy: report of a systematic review by the North American Commission of the International League Against Epilepsy. Epilepsia. 2009;50(10):22852295.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10

    Nathan CL, Gutierrez C. FACETS of health disparities in epilepsy surgery and gaps that need to be addressed. Neurol Clin Pract. 2018;8(4):340345.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    Wiebe S. Epilepsy: Does access to care influence the use of epilepsy surgery?. Nat Rev Neurol. 2016;12(3):133134.

  • 12

    Nelson A. Unequal treatment: confronting racial and ethnic disparities in health care. J Natl Med Assoc. 2002;94(8):666668.

  • 13

    Sharma K, Kalakoti P, Henry M, et al. Revisiting racial disparities in access to surgical management of drug-resistant temporal lobe epilepsy post implementation of Affordable Care Act. Clin Neurol Neurosurg. 2017;158:8289.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14

    McClelland S III, Guo H, Okuyemi KS. Racial disparities in the surgical management of intractable temporal lobe epilepsy in the United States: a population-based analysis. Arch Neurol. 2010;67(5):577583.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Burneo JG, Black L, Knowlton RC, Faught E, Morawetz R, Kuzniecky RI. Racial disparities in the use of surgical treatment for intractable temporal lobe epilepsy. Neurology. 2005;64(1):5054.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16

    Burneo JG, Black L, Martin R, et al. Race/ethnicity, sex, and socioeconomic status as predictors of outcome after surgery for temporal lobe epilepsy. Arch Neurol. 2006;63(8):11061110.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17

    Burneo JG, Shariff SZ, Liu K, Leonard S, Saposnik G, Garg AX. Disparities in surgery among patients with intractable epilepsy in a universal health system. Neurology. 2016;86(1):7278.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18

    Santos JV, Viana J, Pinto C, et al. All Patient Refined-Diagnosis Related Groups’ (APR-DRGs) Severity of Illness and Risk of Mortality as predictors of in-hospital mortality. J Med Syst. 2022;46(6):37.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19

    McCormick PJ, Lin HM, Deiner SG, Levin MA. Validation of the All Patient Refined Diagnosis Related Group (APR-DRG) Risk of Mortality and Severity of Illness modifiers as a measure of perioperative risk. J Med Syst. 2018;42(5):81.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20

    Murphy CC, Trevathan E, Yeargin-Allsopp M. Prevalence of epilepsy and epileptic seizures in 10-year-old children: results from the Metropolitan Atlanta Developmental Disabilities Study. Epilepsia. 1995;36(9):866872.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    Annegers JF, Dubinsky S, Coan SP, Newmark ME, Roht L. The incidence of epilepsy and unprovoked seizures in multiethnic, urban health maintenance organizations. Epilepsia. 1999;40(4):502506.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22

    Kobau R, DiIorio CA, Price PH, et al. Prevalence of epilepsy and health status of adults with epilepsy in Georgia and Tennessee: Behavioral Risk Factor Surveillance System, 2002. Epilepsy Behav. 2004;5(3):358366.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23

    Strine TW, Kobau R, Chapman DP, Thurman DJ, Price P, Balluz LS. Psychological distress, comorbidities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia. 2005;46(7):11331139.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24

    Hussain SA, Haut SR, Lipton RB, Derby C, Markowitz SY, Shinnar S. Incidence of epilepsy in a racially diverse, community-dwelling, elderly cohort: results from the Einstein aging study. Epilepsy Res. 2006;71(2-3):195205.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25

    Jetté N, Quan H, Faris P, et al. Health resource use in epilepsy: significant disparities by age, gender, and aboriginal status. Epilepsia. 2008;49(4):586593.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26

    Elliott JO, Lu B, Shneker BF, Moore JL, McAuley JW. The impact of ‘social determinants of health’ on epilepsy prevalence and reported medication use. Epilepsy Res. 2009;84(2-3):135145.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Collapse
  • Expand
  • View in gallery
    FIG. 1.

    Forest plot showing the odds ratios of insurance and race/ethnicity in predicting the surgical treatment for pediatric medically intractable epilepsy. The reference line for race indicates White population and private/commercial payer for insurance.

  • View in gallery
    FIG. 2.

    Forest plot showing the interaction of race/ethnicity and insurance in predicting the surgical treatment of pediatric medically intractable epilepsy. The reference line for race indicates White population.

  • 1

    Zack MM, Kobau R. National and state estimates of the numbers of adults and children with active epilepsy—United States, 2015. MMWR Morb Mortal Wkly Rep. 2017;66(31):821825.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2

    Russ SA, Larson K, Halfon N. A national profile of childhood epilepsy and seizure disorder. Pediatrics. 2012;129(2):256264.

  • 3

    Wiebe S, Blume WT, Girvin JP, Eliasziw M. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N Engl J Med. 2001;345(5):311318.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4

    Dwivedi R, Ramanujam B, Chandra PS, et al. Surgery for drug-resistant epilepsy in children. N Engl J Med. 2017;377(17):16391647.

  • 5

    Engel J Jr, McDermott MP, Wiebe S, et al. Early surgical therapy for drug-resistant temporal lobe epilepsy: a randomized trial. JAMA. 2012;307(9):922930.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Pestana Knight EM, Schiltz NK, Bakaki PM, Koroukian SM, Lhatoo SD, Kaiboriboon K. Increasing utilization of pediatric epilepsy surgery in the United States between 1997 and 2009. Epilepsia. 2015;56(3):375381.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7

    Tian N, Kobau R, Zack MM, Greenlund KJ. Barriers to and disparities in access to health care among adults aged ≥18 years with epilepsy—United States, 2015 and 2017. MMWR Morb Mortal Wkly Rep. 2022;71(21):697702.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8

    Schiltz NK, Koroukian SM, Singer ME, Love TE, Kaiboriboon K. Disparities in access to specialized epilepsy care. Epilepsy Res. 2013;107(1-2):172180.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9

    Burneo JG, Jette N, Theodore W, et al. Disparities in epilepsy: report of a systematic review by the North American Commission of the International League Against Epilepsy. Epilepsia. 2009;50(10):22852295.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10

    Nathan CL, Gutierrez C. FACETS of health disparities in epilepsy surgery and gaps that need to be addressed. Neurol Clin Pract. 2018;8(4):340345.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11

    Wiebe S. Epilepsy: Does access to care influence the use of epilepsy surgery?. Nat Rev Neurol. 2016;12(3):133134.

  • 12

    Nelson A. Unequal treatment: confronting racial and ethnic disparities in health care. J Natl Med Assoc. 2002;94(8):666668.

  • 13

    Sharma K, Kalakoti P, Henry M, et al. Revisiting racial disparities in access to surgical management of drug-resistant temporal lobe epilepsy post implementation of Affordable Care Act. Clin Neurol Neurosurg. 2017;158:8289.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14

    McClelland S III, Guo H, Okuyemi KS. Racial disparities in the surgical management of intractable temporal lobe epilepsy in the United States: a population-based analysis. Arch Neurol. 2010;67(5):577583.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15

    Burneo JG, Black L, Knowlton RC, Faught E, Morawetz R, Kuzniecky RI. Racial disparities in the use of surgical treatment for intractable temporal lobe epilepsy. Neurology. 2005;64(1):5054.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16

    Burneo JG, Black L, Martin R, et al. Race/ethnicity, sex, and socioeconomic status as predictors of outcome after surgery for temporal lobe epilepsy. Arch Neurol. 2006;63(8):11061110.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17

    Burneo JG, Shariff SZ, Liu K, Leonard S, Saposnik G, Garg AX. Disparities in surgery among patients with intractable epilepsy in a universal health system. Neurology. 2016;86(1):7278.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18

    Santos JV, Viana J, Pinto C, et al. All Patient Refined-Diagnosis Related Groups’ (APR-DRGs) Severity of Illness and Risk of Mortality as predictors of in-hospital mortality. J Med Syst. 2022;46(6):37.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19

    McCormick PJ, Lin HM, Deiner SG, Levin MA. Validation of the All Patient Refined Diagnosis Related Group (APR-DRG) Risk of Mortality and Severity of Illness modifiers as a measure of perioperative risk. J Med Syst. 2018;42(5):81.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20

    Murphy CC, Trevathan E, Yeargin-Allsopp M. Prevalence of epilepsy and epileptic seizures in 10-year-old children: results from the Metropolitan Atlanta Developmental Disabilities Study. Epilepsia. 1995;36(9):866872.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    Annegers JF, Dubinsky S, Coan SP, Newmark ME, Roht L. The incidence of epilepsy and unprovoked seizures in multiethnic, urban health maintenance organizations. Epilepsia. 1999;40(4):502506.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22

    Kobau R, DiIorio CA, Price PH, et al. Prevalence of epilepsy and health status of adults with epilepsy in Georgia and Tennessee: Behavioral Risk Factor Surveillance System, 2002. Epilepsy Behav. 2004;5(3):358366.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23

    Strine TW, Kobau R, Chapman DP, Thurman DJ, Price P, Balluz LS. Psychological distress, comorbidities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia. 2005;46(7):11331139.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24

    Hussain SA, Haut SR, Lipton RB, Derby C, Markowitz SY, Shinnar S. Incidence of epilepsy in a racially diverse, community-dwelling, elderly cohort: results from the Einstein aging study. Epilepsy Res. 2006;71(2-3):195205.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 25

    Jetté N, Quan H, Faris P, et al. Health resource use in epilepsy: significant disparities by age, gender, and aboriginal status. Epilepsia. 2008;49(4):586593.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26

    Elliott JO, Lu B, Shneker BF, Moore JL, McAuley JW. The impact of ‘social determinants of health’ on epilepsy prevalence and reported medication use. Epilepsy Res. 2009;84(2-3):135145.

    • Crossref
    • Search Google Scholar
    • Export Citation

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 487 487 58
PDF Downloads 391 391 66
EPUB Downloads 0 0 0