Advertisement

Socioeconomic outcome and access to care in adults with epilepsy in Sweden: A nationwide cohort study

  • Klara Andersson
    Affiliations
    Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

    Institute of Health and Care Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
    Search for articles by this author
  • Anneli Ozanne
    Affiliations
    Institute of Health and Care Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

    Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden
    Search for articles by this author
  • Anna Edelvik Tranberg
    Affiliations
    Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

    Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden
    Search for articles by this author
  • John E Chaplin
    Affiliations
    Department of Pediatrics, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
    Search for articles by this author
  • Kristian Bolin
    Affiliations
    Department of Economics, Centre for Health Economics, University of Gothenburg, Gothenburg, Sweden
    Search for articles by this author
  • Kristina Malmgren
    Affiliations
    Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

    Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden
    Search for articles by this author
  • Johan Zelano
    Correspondence
    Corresponding author at: Neurology, Blå stråket 7, Sahlgrenska University Hospital, SE41345, Gothenburg, Sweden.
    Affiliations
    Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

    Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden
    Search for articles by this author
Open ArchivePublished:December 03, 2019DOI:https://doi.org/10.1016/j.seizure.2019.12.001

      Highlights

      • Lower socioeconomic status correlates with more severe epilepsy.
      • Access to neurologist is more common in adults with high socioeconomic status.
      • For people with a higher educational level, socioeconomic consequences of epilepsy are present in a lower degree.

      Abstract

      Purpose

      Epilepsy has well-documented associations with low income and low education levels, but the impact of a patient’s socioeconomic standing (SES) on the effects of epilepsy have been less studied.

      Method

      We performed a register-based cross-sectional study and asked if SES was associated with more severe epilepsy or limited access to care in Sweden, where health care is universal, and if socioeconomic outcomes (employment and income) differed for persons with epilepsy (PWE) with different levels of educational attainment. The study cohort consisted of all adult patients with an epilepsy diagnosis in the Swedish patient register in 2000–2015 (n = 126,406) and controls (n = 379,131) matched for age, gender, and place of birth.

      Results

      Somatic and psychiatric comorbidities were more common in PWE, while education and income levels were lower. Among PWE, hospitalizations were more common in persons with lower income or education. Having at least one prescription written by a neurologist in the study period was more common in the high-income and high-education groups. Finally, although low educational attainment was associated with low levels of income and inversely associated with employment in both persons with epilepsy and controls, regression analyses demonstrated that these associations were much more noticeable in cases than controls.

      Conclusions

      We conclude that both the severity and consequences of epilepsy are greater in persons of low SES, even in a country with universal health care. This indicates that universal access may not be sufficient to mitigate socioeconomic inequity in epilepsy.

      Keywords

      1. Introduction

      An association between socioeconomic status (SES) and epilepsy has been demonstrated in both insurance-based and public health-care systems. A US survey in 2010 by the Centre for Disease Control and Prevention demonstrated that the prevalence of active epilepsy is 1% in the US, but 1.9% in low-income US households [
      • Centers for Disease C
      Prevention. Epilepsy in adults and access to care--United States, 2010.
      ]. A geographical correlation between socioeconomic deprivation and prevalence of epilepsy has been disclosed in the UK [
      • Steer S.
      • Pickrell W.O.
      • Kerr M.P.
      • et al.
      Epilepsy prevalence and socioeconomic deprivation in England.
      ]. Data also indicate that SES is associated with the severity of epilepsy. In the US, persons with epilepsy with low SES are more likely to have prolonged seizures, use emergency services more frequently, have reduced quality of life due to epilepsy, and have greater side effects from their antiepileptic medication [
      • Begley C.E.
      • Basu R.
      • Reynolds T.
      • et al.
      Sociodemographic disparities in epilepsy care: results from the Houston/New York City health care use and outcomes study.
      ]. SES may also affect access to epilepsy care. A retrospective cohort study of SES and paediatric epilepsy surgery in Canada indicates that time to epilepsy surgery is influenced by the household income [
      • Rubinger L.
      • Chan C.
      • Andrade D.
      • et al.
      Socioeconomic status influences time to surgery and surgical outcome in pediatric epilepsy surgery.
      ] and in Sweden persons with low SES are more likely to be hospitalized for epilepsy and less likely to have their prescription for antiepileptic drugs (AED) written by a neurologist [
      • Li X.
      • Sundquist J.
      • Sundquist K.
      Socioeconomic and occupational risk factors for epilepsy: a nationwide epidemiological study in Sweden.
      ,
      • Mattsson P.
      • Tomson T.
      • Eriksson O.
      • et al.
      Sociodemographic differences in antiepileptic drug prescriptions to adult epilepsy patients.
      ].
      Epilepsy also has well-documented socioeconomic consequences. Persons with epilepsy have lower income and are more likely to be unemployed [
      • Kobau R.
      • Zahran H.
      • Thurman D.J.
      • et al.
      Epilepsy surveillance among adults--19 states, behavioral risk factor surveillance system, 2005.
      ]. There are also long-term effects on marital status and/or employment [
      • Jennum P.
      • Christensen J.
      • Ibsen R.
      • et al.
      Long-term socioeconomic consequences and health care costs of childhood and adolescent-onset epilepsy.
      ]. Nevertheless, how SES correlates with the health-related consequences of epilepsy has been less studied. Thus, the present study aims to improve available knowledge in the area, using Swedish register data. Moreover, the study also assesses the effect of SES on access to a neurologist, in order to validate previous observations. Investigations on the importance of SES for health-related consequences of epilepsy in Sweden may give valuable information on how far such publicly funded health care systems may reduce the socioeconomic consequences of epilepsy.

      2. Materials and methods

      2.1 Registers

      The study is a national cross-sectional study based on the National Patient Register (NPR). The NPR contains information, e.g. diagnoses, on all cases admitted to inpatient care from 1987 and on all cases seen in hospital-based outpatient care since 2001 (with improved coverage from 2005). For each case in the NPR, three age- and gender-matched controls were randomly selected from the population register per cohort entry date for the case. Information on socioeconomic factors, such as income, educational level and employment status, was collected from the Longitudinal integration database for health insurance and labour market studies (Statistics Sweden, LISA), which contains information for all individuals over 15 years after 2010. Information on AEDs was collected from the Swedish Prescribed Drug Register, which contains information on all prescribed pharmaceuticals dispensed in Sweden (from July 2005, with personal identification numbers). Information on deaths was collected from the Swedish Cause of Death register, to which reporting is also mandatory. Information on epilepsy surgery was collected from the Swedish National Epilepsy Surgery Register, which contains information on all epilepsy surgery procedures performed in Sweden from 1990 [
      • Edelvik A.
      • Rydenhag B.
      • Olsson I.
      • et al.
      Long-term outcomes of epilepsy surgery in Sweden: a national prospective and longitudinal study.
      ]. All data were cross-referenced and anonymized by the National Board of Health and Welfare before we were given access to them.

      2.2 Ascertainment of epilepsy

      We used the recommendations set out in the ILAE epidemiology commission report 2010 [
      • Thurman D.J.
      • Beghi E.
      • Begley C.E.
      • et al.
      Standards for epidemiologic studies and surveillance of epilepsy.
      ]. Therefore, epilepsy was defined as a single ICD-10 code of G40 in the NPR, meeting the recommendation for probable epilepsy. For the purpose of the sensitivity analysis, onset of epilepsy was defined as an ICD-10 code of G40 or ICD-9 345 (not 345Q) occurring at least five years after the establishment of NPR in 1987, to allow five years to detect pre-existing epilepsy.

      2.3 Definitions

      Educational attainment, income, and employment were used as measures of SES according to the variables registered in LISA. Income was categorized as quartiles of income in the control group. AEDs were analysed using ATC-codes. The following AEDs were categorized as “new” AEDs: Oxcarbazepine, Rufinamide, Eslicarbazepine, Lamotrigine, Topiramate, Gabapentin, Levetiracetam, Zonizamide, Pregabalin, Stiripentol, Lacosamide, Retigabine, Perampanel, Brivaracetam. Comorbidities were defined according to diagnosis codes; intellectual disability ICD-9 317–319 or ICD-10 F70-79, psychiatric comorbidity ICD-9 295, 296, 300,311,314,301,299 or ICD-10 F20, F25,F301, F31, F32-39, F40-42, F44-45, F48, F90, F60-62, F69, F84, substance abuse ICD-9 304, 305X or ICD-10 F11-F19, vascular comorbidity ICD-9 390–459 or ICD-10 I10-I99, tumour ICD-9 140–239 or C00-D48, dementia ICD-9 290, 331 or ICD-10 F00-F03.

      2.4 Study cohort

      The NPR contained 126,736 unique individuals that had at least one epilepsy diagnosis (ICD-10 G40) registered after 18 years of age during 2000–2015. For each case, three controls were selected matched for birth year, sex, and residential municipality at the time of diagnosis. The National Board of Health and Welfare removed 330 of the adults with epilepsy prior to delivering the data, since 46 had invalid personal numbers, 264 were not in the register at the time of diagnosis and 20 had no possible controls, resulting in 126,406 adults with epilepsy. For 61 of these adults with epilepsy, fewer than 3 controls were found (Fig. 1).
      Fig. 1
      Fig. 1Selections for sub-analyses. 1) Entire cohort 2) Severity of epilepsy 3) Access to care, Socioeconomic outcomes in epilepsy, 4) Regression analysis.
      *Psychiatric disorder, cerebral vascular comorbidity, brain tumour, substance addiction or dementia.

      2.5 Statistical analyses

      Missing data were considered as such, and no imputations were made. The statistical tests applied were Chi square, Mantel Haenszel, logistic regression and interaction effect. Demographics, socioeconomic variables, and comorbidities were analysed for the entire cohort. We used different selection criteria for different analyses, which are outlined in Fig. 1. For analyses of the association between SES and access to epilepsy care, individuals with intellectual disability or age ≥65 years were excluded. In the analysis of an association between SES and socioeconomic outcomes (low income and employment) of epilepsy, individuals with comorbidities or age ≥65 years were excluded. All analyses were computed in SPSS version 23. All statistical tests were two-sided with alpha = 0.05.

      2.6 Ethical approval

      This study, performed in agreement with privacy legislation in Sweden, was approved by the regional ethics committee in Gothenburg (approval number 839-16).

      3. Results

      3.1 Sociodemographics and comorbidities

      The entire cohort consisted of 126,406 adult persons with an epilepsy diagnosis during 2000–2015 and 379,131 controls. Demographics, comorbidities, and socioeconomic basics are given in Table 1. As expected, somatic and psychiatric comorbidities were more common in cases, while education and income were lower in cases than in controls (chi square, p < 0.001).
      Table 1Study cohort. Demographics, comorbidities, education and income.
      DemographicsEpilepsy (n = 126406)Controls (n = 379131)Proportional difference %p-value
      Age ± sd.63.1 ± 21.963.1 ± 21.90.90
      n(%)n(%)
      Male66907(52.9)200666(52.9)0.99
      Female59499(47.1)178465(47.1)
      Deceased44314(35.1)66508(17.5)17.6<0.001
      Immigrated18606(14.7)61190(16.1)−1.4<0.001
      Single household39771(31.5)120610(31.8)−0.30.02
      Comorbidities
      Intellectual disability3351(2.7)464(0.1)2.6<0.001
      Psychiatric comorbidity18745(14.8)28685(7.6)7.2<0.001
      Substance addiction3745(3.0)2529(0.7)2.3<0.001
      Vascular comorbidity22617(17.9)7133(1.9)16.0<0.001
      Tumour10167(8.0)1426(0.4)7.6<0.001
      Dementia3632(2.9)4592(1.2)1.7<0.001
      Education<0.001
      School ≤ 9 yrs34006(36.3)102060(30.2)1.0
      Highschool ≤ 2 yrs25115(26.8)82410(24.4)2.4
      Highschool = 3 yrs14594(15.6)59303(17.5)−1.9
      University < 3 yrs8890(9.5)38976(11.5)−2.0
      University ≥ 3 yrs10998(11.7)55253(16.3)−4.3
      Income (100 SEK)<0.001
      Level 1 (<1587)34158(42.3)80192(26.1)16.2
      Level 2 (1587–2625)23626(29.3)85695(27.9)1.4
      Level 3 (2625–3740)13526(16.8)69015(22.5)−5.7
      Level 4 (>3740)9374(11.6)71789(23.4)−11.8
      Income per family unit

      (100 SEK)
      <0.001
      Level 1 (<1668)34466(42.7)85488(27.9)14.8
      Level 2 (1668–2356)19465(24.1)77603(25.3)−1.2
      Level 3 (2356–3222)14376(17.8)69652(22.7)−4.9
      Level 4 (>3222)12377(15.3)73948(24.1)−8.8

      3.2 Severity of epilepsy and socioeconomic standing

      To assess the severity of epilepsy in persons with different SES, the number of hospitalizations for epilepsy in 2015 was calculated from the NPR for persons that were alive at the time for data export with active epilepsy (diagnosis registered in the last five years before export, n = 46,176). Hospitalizations for epilepsy were more common in persons with lower income or education (Fig. 2A–B). In addition, treatment with more than one AED in some year between 2010–2015 was also more common in persons with lower income or education (Fig. 2C–D).
      Fig. 2
      Fig. 2Severity of epilepsy; number of hospitalizations and prescribed AEDs, in relation to income quartile and educational level. Mantel- Haenszel p-value < 0.001 for A,B, C and D.

      3.3 Access to neurologist

      The proportion of persons per education or income strata who were prescribed at least one AED in 2010–2015 by a neurologist was assessed in persons of working age alive at the time for data export with an epilepsy diagnosis in the last five years without intellectual disability (n = 31,036). Persons with intellectual disability were excluded because organization of care for this patient group varies and all prescriptions for this patient group may be written by a primary care physician even if a neurologist has been consulted. An additional analysis focused on the type of AED that was prescribed. Having at least one prescription written by a neurologist in 2010–2015 was more commonly found in the high-income and high-education groups. Only 78% of persons in the group with the lowest education had a prescription from a neurologist, compared to 91% in the highest education group. It was more common for persons with lower income levels to have been prescribed new AEDs in 2010–2015 (p < 0.001). This difference was not seen as clearly between educational levels, yet it was significant (Fig. 3). Representation in the national epilepsy surgery register was not significantly different among education groups (1.5% in the group with lowest education compared to 1.5% in the highest education group (p > 0.05)), but significantly more common in the lower income group; 2.1% in income level 1 compared to 1.0 % in income level 4 (p < 0.001).
      Fig. 3
      Fig. 3Access to neurologist in relation to prescriptions from neurologist and of a new type of AED, in relation to income level and educational level. *= Mantel-Haenszel p-value < 0.001 when testing for a significant trend towards the staple furthest to the right.

      3.4 Socioeconomic outcomes with different SES

      The association between epilepsy and two main socioeconomic outcomes –income and employment – was assessed in persons with different levels of educational attainment, to investigate if a higher level of education was protective against poor socioeconomic outcomes of epilepsy. The difference between persons with epilepsy and controls regarding low income or unemployment was more pronounced for persons with low education (Fig. 4). When studying low income and unemployment separately, there was a significant interaction (p < 0.001) effect between epilepsy and educational level i.e. the effect of epilepsy on low income or unemployment is not constant but also depends on the educational level.
      Fig. 4
      Fig. 4Employment and low income per educational attainment in cases and controls of working age and without psychiatric disorder, cerebral vascular comorbidity, brain tumour, substance addiction, or dementia. Mantel Haenszel p-value< 0.001. Interaction effect (epilepsy*educational level) p < 0.001.
      In regression analyses, where persons with potentially confounding comorbidities were excluded and adjustments made for potentially confounding socioeconomic factors (age, gender, and immigration), low educational attainment was associated with low level of income and inversely associated with employment in both persons and controls. Notably, for each educational level, the OR was greater in persons with epilepsy than controls and the difference was more pronounced for lower education levels (Table 2).
      Table 2Logistic regression of associations between education level and income or employment.
      Low income

      OR (95%CI)
      EpilepsyControls
      n32348160173
      crudeadjusted

      +age, gender, immigration
      crudeadjusted

      +age, gender, immigration
      University ≥ 3 yrsRefRefRefRef
      University < 3 yrs1.69 (1.48–1.94)1.74 (1.52–2.00)1.34 (1.26–1.43)1.39 (1.31–1.48)
      High school = 3 yrs1.58 (1.40–1.79)1.73 (1.52–1.96)1.48 (1.40–1.56)1.41 (1.33–1.49)
      High school ≤ 2 yrs1.96 (1.74–2.21)2.25 (1.99–2.55)1.33 (1.26–1.41)1.67 (1.57–1.77)
      School ≤ 9 yrs4.17 (3.73–4.67)4.38 (3.90–4.92)3.07 (2.91–3.23)2.57 (2.43–2.71)
      Employment

      OR (95%CI)
      EpilepsyControls
      n32348160173
      crudeadjusted

      +age, gender, immigration
      crudeadjusted

      +age, gender, immigration
      University ≥ 3 yrsRefRefRefRef
      University < 3 yrs0.72 (0.64–0.80)0.72 (0.64–0.80)0.82 (0.79–0.86)0.81 (0.78–0.85)
      High school 3 yrs0.79 (0.71–0.87)0.60 (0.54–0.66)0.85 (0.81–0.88)0.69 (0.66–0.72)
      High school ≤2 yrs0.37 (0.34–0.41)0.43 (0.39–0.48)0.59 (0.57–0.62)0.67 (0.64–0.70)
      School ≤ 9 yrs0.20 (0.19–0.22)0.16 (0.15–0.18)0.48 (0.46–0.50)0.42 (0.40–0.44)

      3.5 Sensitivity analyses

      To test for the influence of possible confounding factors on the education level the analyses for access to care and socioeconomic outcome were performed on the sub-group of people with epilepsy that had been diagnosed after reaching the age of 30, thus having completed their education. This did not alter our main findings (supplemental Fig. 5).

      4. Discussion

      We used Swedish national registers to investigate associations between SES and access to care and socioeconomic outcomes in persons with epilepsy in Sweden. In agreement with the existing literature low income or education level was associated with more severe epilepsy [[
      • Steer S.
      • Pickrell W.O.
      • Kerr M.P.
      • et al.
      Epilepsy prevalence and socioeconomic deprivation in England.
      ], [
      • Jennum P.
      • Gyllenborg J.
      • Kjellberg J.
      The social and economic consequences of epilepsy: a controlled national study.
      ,
      • Jennum P.
      • Sabers A.
      • Christensen J.
      • et al.
      Welfare consequences for people with epilepsy and their partners: a matched nationwide study in Denmark.
      ,
      • Heaney D.C.
      • MacDonald B.K.
      • Everitt A.
      • et al.
      Socioeconomic variation in incidence of epilepsy: prospective community based study in south east England.
      ]. By contrast, persons with high SES were more likely to be treated by a neurologist [
      • Mattsson P.
      • Tomson T.
      • Eriksson O.
      • et al.
      Sociodemographic differences in antiepileptic drug prescriptions to adult epilepsy patients.
      ]. We also report that epilepsy correlated with low income and lack of employment to a greater degree in persons with a low level of education. To our knowledge, this is the first time an association between SES and the socioeconomic consequences of epilepsy has been demonstrated in a nationwide study. Notable strengths of our investigation include the nationwide scope and unbiased assessment of epilepsy as well as socioeconomic variables.
      As demonstrated in the UK and US, low SES was more common in persons with epilepsy and among persons with epilepsy associated with more severe disease [
      • Centers for Disease C
      Prevention. Epilepsy in adults and access to care--United States, 2010.
      ,
      • Steer S.
      • Pickrell W.O.
      • Kerr M.P.
      • et al.
      Epilepsy prevalence and socioeconomic deprivation in England.
      ], [
      • Ferro M.A.
      A population-based study of the prevalence and sociodemographic risk factors of self-reported epilepsy among adults in the United Kingdom.
      ,
      • Burneo J.G.
      • 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.
      ]. It was also more common for persons with a high SES to be treated by a neurologist, raising questions about equal access to care. One important aspect reducing access to neurologists may be the low number of neurologists in Sweden - approximately 50 per million inhabitants, compared to the recently reported EU median of 79 per million inhabitants [
      • Zelano J.
      • Klecki J.
      • Christensen J.
      • et al.
      The provision of epilepsy care across Europe 2017: a 17-year follow-up survey.
      ]. In the present study, poorer access to neurologist was seen in patients with low income, compared to those with high income. An explanation could be that patients residing in rural areas have more limited access to a Neurologist and also have generally lower income than patients residing in the larger cities. However, more studies would be needed to draw such conclusions, and it would not change the importance of the finding, since the goal is to provide equal care to the whole population.
      Another interesting observation is that although severe epilepsy was more common in persons with a low level of education, being prescribed new AEDs was nearly equally common in all educational levels. This could indicate that newer AEDs are offered earlier as a therapeutic option to persons of high education, although more in-depth studies are needed. Our findings regarding hospitalization and prescriptions by neurologists are in agreement with previous investigations in Sweden [
      • Li X.
      • Sundquist J.
      • Sundquist K.
      Socioeconomic and occupational risk factors for epilepsy: a nationwide epidemiological study in Sweden.
      ], [
      • Mattsson P.
      • Tomson T.
      • Edebol Eeg-Olofsson K.
      • et al.
      Association between sociodemographic status and antiepileptic drug prescriptions in children with epilepsy.
      ].
      Epilepsy was associated with lower levels of employment, in agreement with a nationwide study from Denmark [
      • Jennum P.
      • Gyllenborg J.
      • Kjellberg J.
      The social and economic consequences of epilepsy: a controlled national study.
      ]. In our material, the association between epilepsy and lack of employment or low income was stronger in persons with low educational attainment. This is in agreement with observations from the US that epilepsy affects persons with low SES more [
      • Burneo J.G.
      • 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.
      ,
      • Bautista R.E.
      • Wludyka P.
      Factors associated with employment in epilepsy patients.
      ]. Childhood epilepsy can affect educational attainment [
      • Schneider-von Podewils F.
      • Gasse C.
      • Geithner J.
      • et al.
      Clinical predictors of the long-term social outcome and quality of life in juvenile myoclonic epilepsy: 20-65 years of follow-up.
      ,
      • Geerts A.
      • Brouwer O.
      • van Donselaar C.
      • et al.
      Health perception and socioeconomic status following childhood-onset epilepsy: the Dutch study of epilepsy in childhood.
      ], but unless epilepsy originates from organic injury in early childhood that manifests later in life, this is unlikely to confound our observation, given that similar result was found in our sensitivity analysis where only persons diagnosed with epilepsy after the age of 30 were included. Our findings contrast slightly with a small (n = 63) older study in Sweden, in which there was no negative effect on employment after a first seizure in adults [
      • Lindsten H.
      • Stenlund H.
      • Edlund C.
      • et al.
      Socioeconomic prognosis after a newly diagnosed unprovoked epileptic seizure in adults: a population-based case-control study.
      ]. The authors did not exclude persons with comorbidities, which together with the low sample size limits comparability. One possible explanation is that competition in the labour market may have increased since that study, and perhaps more so for persons with low education. Why the association between epilepsy and unemployment is stronger for persons with low level of education is an important area for future research.
      There are limitations to our study. The register-based method does not allow detailed review of medical records and it is difficult to exclude residual confounding, since epilepsy is a disease with several underdiagnosed comorbidities. Regarding SES, Sweden has a fairly compressed income structure, so education may be a better marker of SES than income. We were also unable to investigate racial/ethnic disparities, since ethnicity is not recorded in Swedish registers.
      On the whole, our findings confirm the well-known association between epilepsy and low SES, which seems to transcend a wide range of health-care systems [
      • Steer S.
      • Pickrell W.O.
      • Kerr M.P.
      • et al.
      Epilepsy prevalence and socioeconomic deprivation in England.
      ], [
      • Li X.
      • Sundquist J.
      • Sundquist K.
      Socioeconomic and occupational risk factors for epilepsy: a nationwide epidemiological study in Sweden.
      ], [
      • Jennum P.
      • Christensen J.
      • Ibsen R.
      • et al.
      Long-term socioeconomic consequences and health care costs of childhood and adolescent-onset epilepsy.
      ], [
      • Heaney D.C.
      • MacDonald B.K.
      • Everitt A.
      • et al.
      Socioeconomic variation in incidence of epilepsy: prospective community based study in south east England.
      ]. Especially in Denmark, large nationwide studies have demonstrated long-term socioeconomic consequences of childhood and adolescent-onset epilepsy, and lower employment rates in persons with epilepsy [
      • Jennum P.
      • Christensen J.
      • Ibsen R.
      • et al.
      Long-term socioeconomic consequences and health care costs of childhood and adolescent-onset epilepsy.
      ], [
      • Jennum P.
      • Sabers A.
      • Christensen J.
      • et al.
      Welfare consequences for people with epilepsy and their partners: a matched nationwide study in Denmark.
      ]. To this, we can add that access to care seems more limited and the socioeconomic consequences of epilepsy more severe for persons with low SES, which agrees with observations from the US, where health care coverage has been more limited [
      • Burneo J.G.
      • 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.
      ]. That socioeconomic disparities in epilepsy transcend health care systems suggests that in countries with insurance-based systems, expansion of health care coverage solely will not close the socioeconomic treatment gap. Conversely, countries with public health care need to implement outreach programs or other measures aimed at mitigating unequal access. More research should also be devoted specifically to the consequences of epilepsy in persons of low SES.

      Funding sources

      This work was supported by the Swedish society for medical research , The Sahlgrenska Academy at the University of Gothenburg through the LUA and the ALF agreement Västra Götalandsregionen, the Swedish government’s Agreement for Medical training and Research, Margarethahemmet foundation, and the patient association NEURO Sweden.

      Transparency document

      Declaration of Competing Interest

      JZ reports outside the submitted work: consultancy fee from the Swedish Medical Products agency, honoraria from the journal Neurologi i Sverige, investigator without personal compensation in clinical trials at Sahlgrenska university hospital sponsored by Bial, SK Life science and GW Pharma.

      Acknowledgements

      Mattias Molin and Per Ekman, Statistiska konsultgruppen, Gothenburg for data management and assistance with statistical analyses.

      Appendix A. Supplementary data

      The following is Supplementary data to this article:

      References

        • Centers for Disease C
        Prevention. Epilepsy in adults and access to care--United States, 2010.
        MMWR Morb Mortal Wkly Rep. 2012; 61: 909-913
        • Steer S.
        • Pickrell W.O.
        • Kerr M.P.
        • et al.
        Epilepsy prevalence and socioeconomic deprivation in England.
        Epilepsia. 2014; 55: 1634-1641
        • Begley C.E.
        • Basu R.
        • Reynolds T.
        • et al.
        Sociodemographic disparities in epilepsy care: results from the Houston/New York City health care use and outcomes study.
        Epilepsia. 2009; 50: 1040-1050
        • Rubinger L.
        • Chan C.
        • Andrade D.
        • et al.
        Socioeconomic status influences time to surgery and surgical outcome in pediatric epilepsy surgery.
        Epilepsy Behav. 2016; 55: 133-138
        • Li X.
        • Sundquist J.
        • Sundquist K.
        Socioeconomic and occupational risk factors for epilepsy: a nationwide epidemiological study in Sweden.
        Seizure. 2008; 17: 254-260
        • Mattsson P.
        • Tomson T.
        • Eriksson O.
        • et al.
        Sociodemographic differences in antiepileptic drug prescriptions to adult epilepsy patients.
        Neurology. 2010; 74: 295-301
        • Kobau R.
        • Zahran H.
        • Thurman D.J.
        • et al.
        Epilepsy surveillance among adults--19 states, behavioral risk factor surveillance system, 2005.
        MMWR Surveill Summ. 2008; 57: 1-20
        • Jennum P.
        • Christensen J.
        • Ibsen R.
        • et al.
        Long-term socioeconomic consequences and health care costs of childhood and adolescent-onset epilepsy.
        Epilepsia. 2016; 57: 1078-1085
        • Edelvik A.
        • Rydenhag B.
        • Olsson I.
        • et al.
        Long-term outcomes of epilepsy surgery in Sweden: a national prospective and longitudinal study.
        Neurology. 2013; 81: 1244-1251
        • Thurman D.J.
        • Beghi E.
        • Begley C.E.
        • et al.
        Standards for epidemiologic studies and surveillance of epilepsy.
        Epilepsia. 2011; 52: 2-26
        • Jennum P.
        • Gyllenborg J.
        • Kjellberg J.
        The social and economic consequences of epilepsy: a controlled national study.
        Epilepsia. 2011; 52: 949-956
        • Jennum P.
        • Sabers A.
        • Christensen J.
        • et al.
        Welfare consequences for people with epilepsy and their partners: a matched nationwide study in Denmark.
        Seizure. 2017; 49: 17-24
        • Heaney D.C.
        • MacDonald B.K.
        • Everitt A.
        • et al.
        Socioeconomic variation in incidence of epilepsy: prospective community based study in south east England.
        BMJ. 2002; 325: 1013-1016
        • Ferro M.A.
        A population-based study of the prevalence and sociodemographic risk factors of self-reported epilepsy among adults in the United Kingdom.
        Seizure. 2011; 20: 784-788
        • Burneo J.G.
        • 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: 2285-2295
        • Zelano J.
        • Klecki J.
        • Christensen J.
        • et al.
        The provision of epilepsy care across Europe 2017: a 17-year follow-up survey.
        Epilepsia Open. 2019; 4: 144-152
        • Mattsson P.
        • Tomson T.
        • Edebol Eeg-Olofsson K.
        • et al.
        Association between sociodemographic status and antiepileptic drug prescriptions in children with epilepsy.
        Epilepsia. 2012; 53: 2149-2155
        • Bautista R.E.
        • Wludyka P.
        Factors associated with employment in epilepsy patients.
        Epilepsy Behav. 2007; 10: 89-95
        • Schneider-von Podewils F.
        • Gasse C.
        • Geithner J.
        • et al.
        Clinical predictors of the long-term social outcome and quality of life in juvenile myoclonic epilepsy: 20-65 years of follow-up.
        Epilepsia. 2014; 55: 322-330
        • Geerts A.
        • Brouwer O.
        • van Donselaar C.
        • et al.
        Health perception and socioeconomic status following childhood-onset epilepsy: the Dutch study of epilepsy in childhood.
        Epilepsia. 2011; 52: 2192-2202
        • Lindsten H.
        • Stenlund H.
        • Edlund C.
        • et al.
        Socioeconomic prognosis after a newly diagnosed unprovoked epileptic seizure in adults: a population-based case-control study.
        Epilepsia. 2002; 43: 1239-1250