Jul 23, 2008 - Children NHS Trust, London, United Kingdom; yMRC Centre of Epidemiology for Child Health, Institute of Child. Health, University .... sions of deprivation (income, employment, health, educa- ..... vals from onset of CSE to calling emergency services, .... to the 3-fold difference in the Richmond, Virginia, USA.
Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
FULL-LENGTH ORIGINAL RESEARCH
Socioeconomic deprivation independent of ethnicity increases status epilepticus risk *yRichard F.M. Chin, *Brian G.R. Neville, yzCatherine Peckham, yAngie Wade, yHelen Bedford, and *zxRod C. Scott for the NLSTEPSS Collaborative Group1 *Neurosciences Unit, Institute of Child Health, University College London, and Great Ormond Street Hospital for Children NHS Trust, London, United Kingdom; yMRC Centre of Epidemiology for Child Health, Institute of Child Health, University College London, London, United Kingdom; zNational Centre for Young People with Epilepsy, Lingfield, Surrey, United Kingdom; and xRadiology and Physics Unit, Institute of Child Health, University College London, London, United Kingdom
SUMMARY Background: A higher incidence of convulsive status epilepticus (CSE) has been reported in nonwhite compared to white populations. Socioeconomic factors can be intricately involved in observed ethnic ‘‘effects,’’ and the importance of socioeconomic status on health conditions is widely recognized. Understanding the effect of socioeconomic factors on CSE would provide insights into etiology and management, leading to the development of novel prevention strategies. Methods: From a population-based UK study on childhood CSE, we tested the hypothesis that socioeconomic deprivation independent of ethnicity increases the risk of childhood CSE. Home postal codes were used to measure the socioeconomic status of the neighborhood in which patients lived relative to that of the borough in which the neighborhood was located. The child’s ethnicity was reported by parent(s). Relationships between socioeconomic status, ethnicity, and
Convulsive status epilepticus (CSE), the most common medical neurologic emergency in childhood, with an incidence of 18–20 first-ever episodes per 100,000 children per year in the multiethnic and socioeconomically diverse population of North London in the United KingAccepted July 23, 2008; Early View publication October 24, 2008. Address correspondence to Richard F.M. Chin, The Wolfson Centre, Mecklenburgh Square, London WC1N 2AP, U.K. E-mail: r.chin@ich. ucl.ac.uk 1 Members of the NLSTEPSS Collaborative Group are listed in the Appendix. Wiley Periodicals, Inc. ª 2008 International League Against Epilepsy
incidence were investigated using Poisson regression analysis. Results: A total of 176 children were enrolled. The incidence of CSE in nonwhite children [18.5, 95% confidence interval (CI) 13.7–23.3/100,000/year] was 1.8 (95% CI 1.3–2.4) times greater than for white children (10.5, 95% CI 7.9–13.1/100,000/ year) (p < 0.0005). Socioeconomic deprivation and Asian ethnicity were independently associated with increased incidence. For each point increase in Index of Multiple Deprivation (IMD) 2004, there was a 1.03 cumulative increased relative risk (95% CI 1.01–1.06, p = 0.007). Asian children were 5.7 times (95% CI 1.7–18.9) more likely than white children to have a first-ever episode of CSE (p = 0.004). Socioeconomic and ethnicity effects were related to etiology of CSE. Interpretation: Ethnic and socioeconomic factors independently affect risk for prolonged febrile seizures and acute symptomatic CSE, but not for other types of childhood CSE. KEY WORDS: Socioeconomic status, Ethnicity, Population-based, Status epilepticus, Children.
dom (Chin et al., 2006) is associated with significant morbidity and mortality (Raspall-Chaure et al., 2006). Previous studies have reported a higher incidence of CSE in nonwhite compared to white populations (DeLorenzo et al., 1996; Coeytaux et al. 2000). If this effect were only genetically determined, there would then be limited possiblity of the development of primary prevention but ethnicity effects can be confounded by socioeconomic factors (Blakely et al., 2006). The importance of socioeconomic status on health is widely recognized (Heaney et al., 2002; Blakely & Wilson, 2006). The government of the United Kingdom and the World Health Organization (WHO) aim to achieve
1022
1023 Deprivation and Status Epilepticus greater allocation of resources to conditions that are more frequent among socioeconomically deprived people in order to reduce inequalities in health (World Health Organization, 1985; Department of Health, 1997). A more detailed understanding of socioeconomic factors associated with the onset of CSE may provide insights into its etiology and facilitate further research on primary prevention strategies. Using data from the North London Convulsive Status Epilepticus in Childhood Surveillance Study (NLSTEPSS), the first prospective population-based study specifically addressing CSE in childhood (Chin et al., 2006), we test the hypothesis that socioeconomic deprivation, independent of ethnicity, increases the risk of having an episode of CSE in childhood.
Subjects and Methods The methodology of case ascertainment, data collection, and database validation in NLSTEPSS have been described previously in detail (Chin et al., 2006). NLSTEPSS was approved by the London Multicentre Research Ethics Committee and local research ethics committees of all hospitals involved in the study. All data were anonymized before analysis. Between 2002 and 2004, clinical and demographic data on episodes of childhood convulsive status epilepticus (a tonic, clonic, or tonic– clonic seizure, or two or more such seizures between which consciousness is not regained lasting at least 30 min) (International League Against Epilepsy, 1993) within North London were collected. Using capture– recapture methodology, a conservative estimate of the case ascertainment in NLSTEPSS was 74–81%. In NLSTEPSS, 176 children [median age 2.1 years (range 0.08–15.98), 95 male] with first-ever episodes of CSE were enrolled. Of the 176 children, 56% (95% CI 48– 63%) were previously neurologically normal (normal neurodevelopment, no history of epilepsy, and no neurologic deficits). One-third (95% CI 25–39%) had prolonged febrile seizures (PFS), and acute symptomatic CSE occurred in 17% (95% CI 12–23%). Remote symptomatic and acute on remote symptomatic CSE each accounted for 16% (95% CI 12–23%) of episodes, and 12% (95% CI 8–18%) of children had idiopathic or cryptogenic epilepsy. In 7% (95% CI 4–12%) of children, the etiology of CSE could not be determined (Chin et al., 2006). North London is composed of 15 boroughs, and from census and local area data, the age distribution, sex, socioeconomic status, and ethnic composition for the resident borough of each child was estimated (Office of National Statistics, http://www.statistics.gov.uk/census2001/cb_3. asp). Available data on individual children with CSE included age, sex, home address postal codes (which were used to measure socioeconomic status of the area in which the child lived), ethnicity, and etiology of CSE.
Socioeconomic status was measured by an index calculated from 2001 census and 2001–2003 local area data on socioeconomic characteristics, the Index of Multiple Deprivation 2004 (IMD 2004). Using children’s home address postal codes, IMD 2004 scores for their local neighborhood were obtained and used as proxy measure for individual socioeconomic status, with increasing scores correlated with worsening socioeconomic status (Office of the Deputy Prime Minister, 2004). The importance and reliability of IMD 2004 scores are reflected in their use by the UK government to better inform policy and help direct resources toward those people living in the most deprived areas (Office of the Deputy Prime Minister, 2004). IMD 2004 scores comprise seven separate dimensions of deprivation (income, employment, health, education, housing, crime, and living environment) whose individual scores have been standardized and combined in a weighted manner to make a single score for the smallest socially homogenous local area, the local super output area (average 1,500 residents and 400 households) (see Table 1) (Office of the Deputy Prime Minister, 2004). In North London, mean IMD 2004 scores of its 15 constituent boroughs ranged from 13.57–45.75 (Office of the Deputy Prime Minister, 2004) and the IMD 2004 scores for super output areas that constitute the 15 boroughs ranged between 2.36 and 76.36. As part of standard hospital admission procedure, parents report their child’s ethnicity based on the four ethnic populations used in the UK census 2001: white, black, Asian, or other nonwhite (includes mixed) (see Table 2 for details of census 2001 description of each ethnic group and percentage of each in North London). Statistics Analyses were conducted in Intercooled Stata 6.0 and SPSS Version 10 (Chicago, IL, U.S.A.). The incidence of CSE was estimated using North London childhood population data from UK Census 2001 as the denominator (Office of National Statistics, 2004a, 2004b), and the average number of incidence cases per year as numerator. We calculated ethnicity-specific incidence rates by sex and etiology. IMD 2004 scores for the local super output area in which each child lived were compared to the mean IMD 2004 score for the borough in which the super output area was located. The relationships between socioeconomic status, ethnicity, incidence of CSE, and seizure type were investigated using Poisson multiple regression analysis.
Results Of the 176 children with first-ever episodes of CSE enrolled in NLSTEPSS, 63 were white, 31 were black, 57 were Asian, and 25 were other nonwhite. The incidence of CSE in nonwhite children [18.5 (95% CI 13.7–23.3) per Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
1024 R. F. M. Chin et al. Table 1. Dimensions of Deprivation for IMD 2004 (Office of the Deputy Prime Minister, 2004) Dimensions of Deprivation for IMD 2004 Weighting Income deprivation
22.5%
Employment deprivation
22.5%
Health deprivation and disability
13.5%
Education, skill and training deprivation
13.5%
Housing and services
9.3%
Crime
9.3%
Living environment deprivation
9.3%
Indicators Adults & children in Income Support households (2001) Adults & children in Income Based Job Seekers Allowance households (2001) Adults & children in Working Families Tax Credit households whose equivalized income (excluding housing benefits) is below 60% of median before housing costs (2001) Adults & children in Disabled Person’s Tax Credit households whose equivalized income (excluding housing benefits) is below 60% of median before housing costs (2001) National Asylum Support Service supported asylum seekers in England in receipt of subsistence only and accommodation support (2002) Unemployment claimant count (JUVOS) of women aged 18–59 and men aged 18–64 averaged over 4 quarters (2001) Incapacity Benefit claimants women aged 18–59 and men aged 18–64 (2001) Severe Disablement Allowance claimants women aged 18–59 and men aged 18–64 (2001) Participants in New Deal for those 18–24 who are not included in the claimant count (2001) Participants in New Deal for 25 + who are not included in the claimant count (2001) Participants in New Deal for Lone Parents aged 18 and over (2001) Years of Potential Life Lost (1997–2001) Comparative Illness and Disability Ratio (2001) Measures of emergency admissions to hospital (1999–2002) Adults younger than 60 with mood or anxiety disorders (1997–2002) Average points score of children at Key Stage 2 (2002) Average points score of children at Key Stage 3 (2002) Average points score of children at Key Stage 4 (2002) Proportion of young people not staying on in school or with school level education above 16 (2001) Proportion of those aged younger than 21 not entering Higher Education (1999–2002) Secondary school absence rate (2001–2002) Proportions of working age adults (aged 25–54) in the area with no or low qualifications (2001) Household overcrowding (2001) LA level percentage of households for whom a decision on their application for assistance under the homeless provisions of housing legislation has been made, assigned to SOAs (2002) Difficulty of Access to owner-occupation (2002) Road distance to GP premises (2003) Road distance to a primary school (2001–2002) Road distance to a Post Office (2003) Road distance to a supermarket or convenience store (2002) Burglary (4 recorded crime offense types, April 2002–March 2003) Theft (5 recorded crime offense types, April 2002 to March 2003, constrained to CDRP level) Criminal damage (10 recorded crime offense types, April 2002 to March 2003) Violence (14 recorded crime offence types, April 2002 to March 2003) Social and private housing in poor condition (2001) Houses without central heating (2001) Road traffic accidents involving injury to pedestrians and cyclists (2000–2002)
IMD 2004 score = Sum of weighting · dimension scores = (0.225 · Income) + (0.225 · Employment) + (0.135 · Education) + (0.135 · Health) + (0.093 · Housing) + (0.093 Crime) + (0.093 Environment).
Table 2. Descriptions of ethnic groups and percentage of total childhood population in North London Ethnicity White Asian Black Other nonwhite
Description British, Irish, other white Indian, Pakistani, Bangladeshi, other Asian Caribbean, African, other black White and black Caribbean mixed, White and black African mixed, white and Asian mixed, White and other nonwhite mixed, Chinese, other nonwhite
Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
% 50 21 18 11
100,000/year] was 1.8 (95% CI = 1.3–2.4) times greater than for white children [10.5 (95% CI 7.9–13.1) per 100,000/year] (p < 0.0005). In a subgroup analysis, black children had a similar incidence [14.6 (95% CI 9.5–19.7) per 100,000/year] to white children, but Asian children [22.7 (95% CI 16.8–28.6) per 100,000/year] and other nonwhite children [22.4 (95% 13.7–26.4) per 100,000/ year] had a higher incidence of CSE. We examined the distribution of etiology, association with epilepsy, seizure type, and duration of CSE across populations. The only difference in seizure characteristics among ethnic populations was in the etiology of CSE. The risk of prolonged febrile seizures, the most common type
1025 Deprivation and Status Epilepticus Table 3. Incidence rates and relative risk of CSE according to ethnicity, sex, and etiology Nonwhite
Sex Male No. of boys First-ever episodes Incidence/100,000a Relative risk (95% CI)b Female No. of girls First-ever episodes Incidence/100,000a Relative risk (95% CI)b All No. of children First-ever episodes Incidence/100,000a Relative risk (95% CI)b Etiology Prolonged febrile seizure No. (%) Incidence/100,000a Relative risk (95% CI)b Acute symptomatic No. (%) Incidence/100,000a Relative risk (95% CI)b Remote symptomatic No. (%) Incidence/100,000a Relative risk (95% CI)b Acute on remote symptomatic No. (%) Incidence/100,000a Relative risk (95% CI)b Idiopathic epilepsy related No. (%) Incidence/100,000a Relative risk (95% CI)b Cryptogenic epilepsy related No. (%) Incidence Relative risk (95% CI)b Unclassified No. (%) Incidence/100,000a Relative risk (95% CI)b
White
All
Black
Asian
Other
All
153,790 34 11.1 1.0
154,366.0 61.0 19.8 1.8 (1.2–2.7)
54,978 16 14.6 1.3 (0.7–2.4)
65,937 30 22.7 2.1 (1.3–3.4)
33,451 15 22.4 2.0 (1.1–3.7)
308,156 95 15.4
146,770 29 9.9 1.0
150,394 52 17.3 1.7 (1.1–2.8)
54,369 15 13.8 1.4 (0.8–2.6)
63,661 27 21.2 2.2 (1.3–3.6)
32,364 10 15.4 1.6(0.8–3.2)
297,164 81 13.6
300,560 63 10.5 1.0
304,760 113.0 18.5 1.8 (1.3–2.4)
109,347 31 14.2 1.4 (0.9–2.1)
129,598 57 22.0 2.1 (1.5–3.0)
65,815 25 19.0 1.8 (1.1–2.9)
605,320 176 14.5
17 (27) 2.8 1.0
39 (35) 6.4 2.3 (1.3–4.0)
6 (19) 2.7 1.0 (0.4–2.5)
21 (37) 8.1 2.9 (1.5–5.4)
12 (48) 9.1 3.2 (1.5–6.8)
56 4.6
13 (21) 2.2 1.0
17 (15) 2.8
5 (16) 2.3 1.1 (0.4–3.0)
7 (12) 2.7 1.2(0.5–3.1)
5 (20) 3.8 1.8 (0.6–4.9)
30 2.5
13 (21) 2.2 1.0
16 (14) 2.6 1.3 (0.6–2.7)
4 (13) 1.8 0.8 (0.3–2.6)
10 (18) 3.9 1.8 (0.8–4.1)
2 (8) 1.5 0.7 (0.2–3.1)
29 2.4
10 (16) 1.7 1.0
18 (16) 3.0 1.8 (0.8–3.8)
8 (26) 3.7 2.2 (0.9–5.6)
8 (14) 3.1 1.8 (0.7–4.7)
2 (8) 1.5 0.9 (0.2–4.2)
28 2.3
5 (8) 0.8 1.0
13 (12) 2.1 2.6 (0.9–7.2)
5 (16) 2.3 2.7 (0.8–9.5)
5 (9) 1.9 2.3 (0.7–8.0)
3 (12) 2.3 2.7 (0.7–11.5)
18 1.5
1 (2) 0.2 1.0
2 (2) 0.3 1.9 (0.2–21.7)
0 (0) 0.0 –
2 (4) 0.8 4.6 (0.4–51.1)
0 (0) 0.0 –
3 0.2
4 (6) 0.7 1.0
8 (7) 1.3 1.9 (0.6–6.6)
3 (10) 1.4 2.1 (0.5–9.2)
4 (7) 1.5 2.3 (0.6–9.3)
1 (4) 0.8 1.1 (0.1–10.2)
12 1.0
CI, confidence interval. a Crude incidence rate. b White served as reference group.
of CSE was three times higher in Asian children (2.9, 95% CI 1.5–5.4) and other nonwhite children (3.2, 95% CI 1.5–6.8) compared to white children, but black children (1.0, 95% CI 0.4–3.0) and white children had a similar risk (see Table 3 for details on ethnicity specific incidence according to sex and etiology). There was a significant relationship between socioeconomic status and ethnicity. When socioeconomic status
was included in the model, socioeconomic deprivation and Asian ethnicity were independently associated with increased incidence of CSE. For one-point increase in the Index of Multiple Deprivation 2004, there was a cumulative increased relative risk of CSE of 1.04 (95% CI 1.01– 1.06, p = 0.005). Therefore, a 30-point difference in IMD 2004 scores would increase the risk of CSE 3.2 fold (1.04 to the power of 30). After adjusting for socioeconomic Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
1026 R. F. M. Chin et al. Table 4. Poisson multiple regression of incidence of CSE according to prolonged febrile seizures, acute symptomatic or unprovoked afebrile CSE, ethnicity and IMD 2004 scores of socioeconomic status Nonprolonged Febrile Seizures (n = 120) Acute Symptomatic (n = 30)
PFS (n = 56) Variable IMD 2004 Ethnicity White Black Asian Other
Unprovoked (n = 90)
All (n = 120)
All (n = 176)
RR (95% CI)
p
RR (95% CI)
p
RR (95% CI)
p
RR (95% CI)
p
RR (95% CI)
p
1.05 (1.02–1.09)
0.005
1.07 (1.00–1.14)
0.04
0.98 (0.94–1.04)
0.65
1.01 (0.99–1.04)
0.3
1.04 (1.01–1.06)
0.005
1.00 (ref) 2.2 (0.21–24.7) 13.5 (2.80–65.2) 1.29 (0.11–14.6)
– 0.51 0.001 0.21
1.00 (ref) 0.000 (0.000–32.1) 0.003 (0.000–6431) 1.07 (0.002–561.8)
– 0.17 0.44 0.98
1.00 (ref) 1.2 (0.79–18.4) 1.3 (0.16–10.4) 1.5 (0.11–19.3)
– 0.89 0.80 0.77
1.00 (ref) 0.59 (0.43–8.13) 1.66 (0.24–11.3) 2.03 (0.20–20.9)
– 0.69 0.60 0.59
1.00 (ref) 1.22 (0.21–7.14) 5.74 (1.7–18.9) 1.60 (0.29–8.6)
– 0.83 0.004 0.55
PFS, prolonged febrile seizures; RR, relative risk. Note that the confidence intervals in subgroups are very wide, reflecting the small group numbers and consequent imprecision of estimates.
status, Asian children were 5.7 times (95% CI 1.7–18.9) more likely to have a first-ever episode of CSE than white children (p = 0.001), but there was no difference in the incidence of CSE between black and other nonwhite children compared to white children (see Table 4). There were no significant interactions between these variables. On analysis according to CSE type, there was an association between socioeconomic deprivation (relative risk 1.05, 95% CI 1.02–1.1.09, p = 0.001) and Asian ethnicity (relative risk 13.5, 95% CI 2.80–65.2, p = 0.02) and the incidence of prolonged febrile seizures. Children who lived in socioeconomically deprived areas were also associated with an increased risk of acute symptomatic CSE (relative risk 1.07, 95% CI 1.00–1.14, p = 0.04). There were no socioeconomic or ethnicity effects on the incidence of afebrile unprovoked CSE (see Table 4). Because previous studies have demonstrated that families of lower socioeconomic status have delayed access to acute medical care (Committee on Pediatric Emergency Medicine 2000; Feinberg et al., 2002), we conducted post hoc regression analyses of the relationship between intervals from onset of CSE to calling emergency services, from calling emergency services to arrival of emergency medical personnel to the patient, and socioeconomic status. Both intervals were independent of patients’ socioeconomic status (p = 0.4 and 0.8, respectively). We also examined the relationship between IMD 2004 scores and etiologies, and there was no difference in mean IMD 2004 scores among etiologies (Kruskal-Wallis ANOVA chisquare = 2.4, d.f. = 4, p = 0.7; see Fig. 1).
Discussion In this prospective population-based study, socioeconomic status, independent of ethnicity, was associated with the development of prolonged febrile seizures and acute symptomatic CSE but not other types of childhood Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
Figure 1. Etiology of convulsive status epilepticus according to IMD 2004 score. Epilepsia ILAE
CSE. This observation is important as it opens a potential new line of research to identify factors associated with socioeconomic status that increase the risk of prolonged febrile seizures and acute symptomatic CSE and might lead to primary prevention strategies. Lower socioeconomic status is associated with an increased prevalence of a wide range of other conditions such as epilepsy, asthma, diabetes, depression, cancer, and meningococcal disease (Heaney et al., 2002; Basagana et al., 2004; Feldman et al., 2004; Williams et al., 2004). Our findings demonstrate an association between a higher incidence of CSE and lower socioeconomic status that is limited to prolonged febrile seizures and acute symptomatic CSE. Therefore, an important consideration is whether social drift or social causation explains the association.
1027 Deprivation and Status Epilepticus In social drift the circumstances of the individual’s social class do not cause the onset of the CSE but rather an individual’s deteriorating health results in a downward shift in social class. In our study, 56% of children were previously neurologically normal before their episode of CSE, and thus their general care provision would have been expected to be similar to other members of their community. Therefore, it is unlikely that social drift would explain the observed association. Children who had an episode of CSE have been shown to have medium- and/or long-term subsequent medical, cognitive and behavioral difficulties, although the extent of this at a populationbased level remains unclear (Raspall-Chaure et al., 2006). It is thus possible that children who have had an episode of CSE could experience subsequent social drift because of the demands on their family in caring for any subsequent difficulties. It is more likely that there is a social causation for the observed increased incidence of prolonged febrile seizures and acute symptomatic CSE in children living in socieconomically deprived areas. Overcrowding, poor nutrition, infections, reduced access to health care, and delayed recognition and treatment of illnesses are more common in socioeconomically deprived groups (Chaudhuri, 2004; Stevens et al., 2006) and may provide plausible explanations for a higher incidence of prolonged febrile seizures and acute symptomatic CSE in such children. Although results from our study support previous observations that the incidence of CSE in nonwhite populations is higher than in white populations (DeLorenzo et al., 1996; Coeytaux et al., 2002), this was apparent only in the Asian population in NLSTEPSS. This is in stark contrast to the 3-fold difference in the Richmond, Virginia, USA study, which did not adjust for socioeconomic status and has the highest reported incidence of status epilepticus (DeLorenzo et al., 1996). Richmond has the highest percentage of people living in poverty relative to the other municipal localities of Virginia (21.4% compared to 6.2%) (http://www.richmondgov.com/departments/budget/ docs/Map_Poverty_Data.pdf). Thus it is possible that the high incidence of CSE reported in the Richmond study and attributed to the racial composition of the population (57% black, 38% white) was confounded by the socioeconomic composition of Richmond. In NLSTEPSS, after adjusting for socioeconomic status, Asian children but not black or other nonwhite children had a significantly higher incidence of CSE compared to white children, but the effect of socioeconomic deprivation on incidence of CSE was lower. We found no associations between socioeconomic deprivation or ethnicity on the incidence of afebrile unprovoked CSE. These findings are consistent with those of previous studies, which failed to find an association between socioeconomic deprivation and epilepsy in children (Hesdorffer et al., 2005; Reading et al., 2006).
In NLSTEPSS, prolonged febrile seizure was the most common etiology of CSE, with Asian children having a 3fold higher incidence than white children. When adjusted for socioeconomic deprivation, we found strong independent associations between socioeconomic and Asian ethnicity and the incidence of prolonged febrile seizures. Together these data suggest that genetic factors play a role in the occurrence of CSE and lend support to the hypothesis that there is a genetic predisposition to prolonged febrile seizures (Iwasaki et al., 2002; Tsai et al., 2002; Kanemoto et al., 2003). We used official census 2001 data released in 2004 after revisions of the original data (Office of National Statistics, 2004a, 2004b). The original data were inaccurate because individuals from hard-to-count areas were missed and the data were inflated by 2%, ‘‘imputed’’ to account for uncounted individuals. These changes had a disproportionate effect upon returns from hard-to-count areas such as inner city, ethnically diverse, and socially deprived areas (Office of National Statistics, 2001). Consequently, the final official census count for these areas was higher than the original, and our findings on the effect of ethnicity and socioeconomic deprivation are likely to be an underestimate of the true effect. In the absence of patient level data, we used IMD2004 scores, which have been shown to provide a valid proxy for a population weighted measure of socioeconomic deprivation of local super output areas, but which is still not an individual measure (Strong et al., 2006) and is a limitation of our study. A measure of socioeconomic deprivation that is assigned to a group of people is an aggregate measure of the experience of that group. Individuals within the group may experience very different levels of deprivation, and an association seen at the group level may thus not apply at the individual level. The assumption that it does apply is referred to as the ecological fallacy, which is always an important consideration in studies examining data at the local area or group level (Greenland & Robins, 1994). Whether they reflect individual level associations or not, area level associations remain interesting and important, and in many instances an ecological rather than individual level analysis is more appropriate (Smith et al., 1998; Pearce, 2000). Because socioeconomic deprivation has been demonstrated to be an independent risk factor for poor health at both the area and the individual level (Smith et al., 1998), inclusion of ecological analyses, such as ours, are necessary if we are to fully understand the relationship between socioeconomic factors and childhood CSE. We conclude that ethnicity and socioeconomic status contribute independently to the incidence of prolonged febrile seizures and acute symptomatic CSE but not for other types of childhood CSE. Research on the ethnic and specific socioeconomic factors that predispose to prolonged febrile seizures and acute symptomatic CSE is Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
1028 R. F. M. Chin et al. needed to develop preventative and new treatment strategies for seizures in childhood including CSE.
Acknowledgments We thank all medical, nursing, and administrative staff who reported to NLSTEPSS. We are also thankful to Richard Lynn from the British Paediatric Surveillance Unit whose help was invaluable in the development and implementation of the active regional surveillance card scheme used in NLSTEPSS. Rod Scott was supported by the Wellcome Trust. NLSTEPSS was funded by an anonymous donor to the Institute of Child Health. This work was undertaken at GOSH/UCL Institute of Child Health, which received a proportion of funding from the Department of Health’s NIHR Biomedical Research Centres funding scheme. The Centre for Paediatric Epidemiology and Biostatistics also benefits from funding support from the Medical Research Council in its capacity as the MRC Centre of Epidemiology for Child Health. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. None of the authors has any conflict of interest to disclose.
References Basagana X, Sunyer J, Kogevinas M, Zock JP, Duran-Tauleria E, Jarvis D, Burney P, Anto JM. (2004) Socioeconomic status and asthma prevalence in young adults: the European Community Respiratory Health Survey. Am J Epidemiol 160:178–188. Blakely T, Wilson N. (2006) Shifting dollars, saving lives: what might happen to mortality rates, and socio-economic inequalities in mortality rates, if income was redistributed? Soc Sci Med 62:2024– 2034. Blakely T, Fawcett J, Hunt D, Wilson N. (2006) ‘‘What is the contribution of smoking and socioeconomic position to ethnic inequalities in mortality in New Zealand?’’. Lancet 368:44–52. Chaudhuri N. (2004) Interventions to improve children’s health by improving the housing environment. Rev Environ Health 19:197– 222. Chin RF, Neville BG, Peckham C, Bedford H, Wade A, Scott RC. (2006) Incidence, cause, and short-term outcome of convulsive status epilepticus in childhood: prospective population-based study. Lancet 368:222–229. Coeytaux A, Jallon P, Galobardes B, Morabia A. (2000) Incidence of status epilepticus in French-speaking Switzerland: (EPISTAR). Neurology 55:693–697. Coeytaux A, Jallon P, Galobardes B, Morabia A. (2002) Incidence of status epilepticus in French-speaking Switzerland: (EPISTAR). Neurology 55:693–697. Committee on Pediatric Emergency Medicine. (2000) Access to pediatric emergency medical care. Pediatrics 105:647–649. DeLorenzo RJ, Hauser WA, Towne AR, Boggs JG, Pellock JM, Penberthy L, Garnett L, Fortner CA, Ko D. (1996) A prospective, population-based epidemiologic study of status epilepticus in Richmond, Virginia. Neurology 4:1029–1035. Department of Health. (1997) The new NHS: modern and dependable. Stationery Office, London. Feinberg E, Swartz K, Zaslavsky A, Gardner J, Walker DK. (2002) Family income and the impact of a children’s health insurance program on reported need for health services and unmet health need. Pediatrics 109:E29. Feldman L, McMullan C, Abernathy T. (2004) Angina and socio-economic status in Ontario: how do characteristics of the county you live in influence your chance of developing heart disease? Can J Public Health 95:228–232. Greenland S, Robins J. (1994) Invited commentary: ecologic studies – biases, misconceptions, and counterexamples. Am J Epidemiol 139:747–760.
Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
Heaney DC, MacDonald BK, Everitt A, Stevenson S, Leonardi GS, Wilkinson P, Sander JW. (2002) Socioeconomic variation in incidence of epilepsy: prospective community based study in south east England. BMJ 325:1013–1016. Hesdorffer DC, Tian H, Anand K, Hauser WA, Ludvigsson P, Olafsson E, Kjartansson O. (2005) Socioeconomic status is a risk factor for epilepsy in Icelandic adults but not in children. Epilepsia 46:1297–1303. International League Against Epilepsy. (1993) Guidelines for epidemiologic studies on epilepsy. Epilepsia 34:592–596. Iwasaki N, Nakayama J, Hamano K, Matsui A, Arinami T. (2002) Molecular genetics of febrile seizures. Epilepsia 43(Suppl. 9):32–35. Kanemoto K, Kawasaki J, Yuasa S, Kumaki T, Tomohiro O, Kaji R, Nishimura M. (2003) Increased frequency of interleukin-1beta-511T allele in patients with temporal lobe epilepsy, hippocampal sclerosis, and prolonged febrile convulsion. Epilepsia 44:796–799. Office of the Deputy Prime Minister. (2004) The English Indices of Deprivation 2004: Summary (revised). Office of the Deputy Prime Minister, London. Office of National Statistics. (2001) Census based local authority population studies. National Statistics Online – Census 2001 – Census news. Accessed on March 21, 2008. Office of National Statistics (2004a) Census 2001. Standard tables for wards in England and Wales (DVD). Office of National Statistics, London. Office on National Statisitics. (2004b) Small Area Statistics. http:// neighbourhood.statistics.gov.uk. Accessed on September 2, 2007. Pearce N. (2000) The ecological fallacy strikes back. J Epidemiol Community Health 54:326–327. Raspall-Chaure M, Chin RF, Neville BGR, Scott RC. (2006) Outcome of paediatric convulsive status epilepticus: a systematic review. Lancet Neurol 5:769–779. Reading R, Haynes R, Beach R. (2006) Deprivation and incidence of epilepsy in children. Seizure 15:190–193. Smith GD, Hart C, Watt G, Hole D, Hawthorne V. (1998) Individual social class, area-based deprivation, cardiovascular disease risk factors, and mortality: the Renfrew and Paisley Study. J Epidemiol Community Health 52:399–405. Stevens GD, Seid M, Mistry R, Halfon N. (2006) Disparities in primary care for vulnerable children: the influence of multiple risk factors. Health Serv Res 41:507–531. Strong M, Maheswaran R, Pearson T. (2006) A comparison of methods for calculating general practice level socioeconomic deprivation. Int J Health Geogr 5:29. Tsai FJ, Hsieh YY, Chang CC, Lin CC, Tsai CH. (2002) Polymorphisms for interleukin 1 beta exon 5 and interleukin 1 receptor antagonist in Taiwanese children with febrile convulsions. Arch Pediatr Adolesc Med 156:545–548. Williams CJ, Willocks LJ, Lake IR, Hunter PR. (2004) Geographic correlation between deprivation and risk of meningococcal disease: an ecological study. BMC Public Health 4:30. World Health Organization. (1985) Targets for all. World Health Organization, Copenhagen.
Appendix. The NLSTEPSS Collaborative Group The members of the NLSTEPSS Collaborative Group are Richard Chin, Rod Scott, Brian Neville, Helen Bedford, Catherine Peckham, Angie Wade (Institute of Child Health), Simon Roth (Barnet General Hospital), Ruby Schwartz (Central Middlesex Hospital), Jacqueline Taylor (Chase Farm Hospital), Ed Abrahamson (Chelsea and Westminster Hospital), Mark Kenny, Mark Peters (Great Ormond Street Hospital for Children and Children’s Acute Transport Service), Shane Tibby (Guy’s Hospital), Adnan
1029 Deprivation and Status Epilepticus Manzur (Hammersmith Hospital), Rajiv Sood (Homerton Hospital), Elaine Hughes (King’s College London), MAS Ahmed (King George Hospital), Satheesh Mathew (Newham General Hospital), Arvind Shah (North Middlesex Hospital), Warren Hyer (Northwick Park Hospital), Michael Greenberg (Royal Free Hospital), Adelaida Mar-
tinez (Royal London Hospital), Sophie kellet (St. George’s Hospital), Simon Nadel, Ian Maconochie (St. Mary’s Hospital), Adrian Goudie, John Jackman (St. Thomas’ Hospital), Mark Gardiner (University College London), Anthony Cohn (Watford General Hospital), Corina O’Neill (Whipps Cross Hospital), and Andrew Robins (Whittington Hospital).
Epilepsia, 50(5):1022–1029, 2009 doi: 10.1111/j.1528-1167.2008.01796.x
