SEARCH

SEARCH BY CITATION

Keywords:

  • Epilepsy;
  • Incidence of epilepsy;
  • Regional changes;
  • Secular trends;
  • Epidemiology;
  • Antiepileptic drug

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Purpose:  Regional variations and temporal trends in the incidence of new-onset epilepsy are clinically important and may offer clues on how to prevent epilepsy.

Methods:  We examined regional differences and secular trends in the incidence of new-onset epilepsy in the Finnish population based on the nationwide full-refundable antiepileptic drug registry and the population registry in the years 1986–2008.

Key Findings:  The overall incidence of epilepsy was significantly higher in eastern Finland than in middle [risk ratio (RR) 1.08 (95% confidence interval, CI 1.05–1.12)), p < 0.0001] and western Finland [RR 1.32 (1.30–1.35), p < 0.0001] but it was declining from 1986 to 2008 in all regions [RR 0.83 (0.81–0.84), p < 0.0001]. The mean annual decline was 0.6%. Although the incidence of epilepsy was falling from 1986 to 2008 in childhood [annual decline 1.9%, RR 0.80 (0.75–0.86), p < 0.0001] and in middle age [annual decline 0.8%, RR 0.88 (0.84–0.93), p < 0.0001], it increased significantly in the elderly (age 65 years or older) in all of Finland [annual increase 3.5%, RR 1.25 (1.18–1.33), p < 0.0001], and particularly in east versus west Finland [RR 1.48 (1.42–1.55), p < 0.0001]. As a result, starting with the year 2000, the incidence rate of epilepsy was higher in the elderly than in children for all of Finland.

Significance:  In view of the falling incidence of epilepsy in childhood and middle-age in all of Finland from 1986 to 2008, the significant increase in the incidence of epilepsy in the elderly is of concern. The regional increase of epilepsy may offer clues for allocating resources and, possibly, population epileptogenesis between west and east Finland and for strategies to prevent epilepsy in the elderly.

Although an increase in the incidence of new-onset epilepsy has been reported in the elderly (Hauser et al., 1993; Sillanpää et al., 2006), several studies suggest that the incidence of epilepsy is falling in the general population (Cockerell et al., 1995), particularly in the middle-aged (Sillanpää et al., 2006) and in children (Hauser et al., 1993; Annegers et al., 1995; Sillanpää et al., 2006). In a recent nationwide study from Finland (Sillanpää et al., 2006), a significant decline in the incidence of epilepsy was found in the total population, particularly in children and the middle-aged, but an increase in the elderly. This study did not examine regional differences and covered a follow-up of only 17 years. However, there are no nationwide long-term studies to examine temporal changes in the incidence of new-onset epilepsy over >20 years for each age group. Although a prevalence study based on the hospital discharge registries suggested that the epilepsy incidence is highest in eastern Finland and lowest in western Finland, with middle Finland in between (Sillanpää & Koivusilta, 1986), we found no nationwide reports from Finland or elsewhere that examined regional differences in the incidence of new-onset epilepsy over an extended observation period.

Finding regional differences in the incidence of epilepsy would not only be clinically important for public health planning but may also provide clues for research on epileptogenic mechanisms that either expose to or protect from the development of epilepsy. Because the age-related incidence of epilepsy seems to subject to secular trends, as reviewed above, we undertook a large nation-wide long-term study of geographic differences in the incidence of new epilepsy in Finland in different age groups.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Full-refundable drug registry

This study was based on the registry data of the total Finnish population (approximately 5,300,000 in 2008) from 23 years (1986–2008). Since 1967, 3 years after that the National Sickness Insurance scheme came into force, a registry has been kept in the Social Insurance Institution (SII) on patients granted full-refundable drugs, including antiepileptic drugs (AEDs), for people with long-term diseases. The right for this special refund is based on clinical diagnosis and the statement given on a semistructured form by a trained adult or pediatric neurologist. In Finland, the density of trained neurologists in the population is approximately 0.9 per 10,000, distributed reasonably well geographically in the country. All neurologists have passed a national written examination, and have been authorized and listed in the national computerized registry. They are working in both public and private sectors and in both outpatient and inpatient care.

The right to receive full-refundable standard AEDs is granted by the SII for noninstitutionalized patients. The request requires filling out of a semistructured two-part form, the first part by the patient on his or her subjective medical problems and the second part completed by an authorized physician. After its existence for >20 years, the SII register can be considered to have good coverage of all patients with epilepsy and generally to work well and reliably (Gissler et al., 1995; Heinonen et al., 1997; Pietilä et al.,1997; Rapola et al., 1997; Pajunen et al., 2005a). However, patients who are permanently institutionalized and, during stay in institution, present with newly onset epilepsy are beyond the refundable drugs system, only because they get their AEDs free-of-charge in the institution. On the other hand, there is a current trend to deinstitutionalize patients and to register institutionalized patients who start to have seizures in the general health care system, providing fully refundable AEDs and entry into SII, these patients are then eligible for full refunding of AEDs. The following AEDs are fully refunded: carbamazepine, clobazam, clonazepam, fosphenytoin, gabapentin, lamotrigine, levetiracetam, oxcarbazepine, nitrazepam, phenobarbital, phenytoin, pregabalin, rufinamide, sodium valproate, stiripentol, tiagabine, topiramate, vigabatrin, and zonisamide.

Patients

The SII registry covers all the age groups nationwide in Finland from 1986 to 2008. For the present study, both the population data and the SII registry data were linked and grouped according to three geographic areas: western, middle, and eastern Finland. For the study, the SII data were available by year of entering AED treatment, age group, gender, and geographic or catchment area of hospital district. The data were linked from hospital discharge registry, fully refundable AED registry, and population statistics data registry. Epileptic seizures and syndromes were defined according to the contemporary guidelines of the International League Against Epilepsy (ILAE) (Commission on revised classification of seizures, 1981; Commission on revised classification of epilepsy, 1989). The study subjects were grouped by age as follows: children (0–15 years), middle-aged (16–64), and elderly (65+). The study design is described in detail previously (Sillanpää et al., 2006).

Validity of registries

The present data were derived from the structured forms of medical statements based on the assessment by authorized neurologists for fully refundable AEDs. The original database is on the national Drug Prescription Registry at the Social Insurance Institution (SII) established in 1967. The law-based registry, data collection principles, record structure, and inclusion criteria were the same for all Finnish citizens and remained unchanged during the whole study period, except for additional approval of new AEDs. The administrative registries have been well accepted by the Finnish population (Heinonen, 1997). Although some underestimation in registry studies may occur (Immonen-Räihä et al., 2003), the resulting medical statements are commonly regarded as valid and reliable sources of information in Scandinavia (Juul-Jensen & Foldspang, 1983; Gissler et al., 1995; Heinonen, 1997; Pietilä et al., 1997; Rapola et al., 1997; Pajunen et al., 2005a). In support of validity, no significant difference was found either between the hospital discharge registry data and an ad hoc study of acute myocardial infarction (Mähönen et al., 1995), or between the hospital discharge registry data and the cause-of-death registry data (Pietilä et al., 1997).

Finally, reliable data may be obtained by comparing registries with clinical records. According to the SII data, the mean annual age-adjusted incidence of active epilepsy in 1975–1979 was 24/100,000 in an epidemiologic study (Keränen et al., 1989); however, the hospital discharge data showed the significantly higher rate of 61.5/100,000 in the same geographic area in 1986 (unpublished data on file). Among approximately 8,000 Finnish children, the cumulative incidence of epilepsy was higher in ad hoc disease data than in routinely collected data (10.8/1,000 vs. 6.8/1,000, p < 0.01) (Gissler et al., 2000). Similarly, a comparative study showed the incidence of epilepsy based on questionnaire data to be higher than that based on the SII data (Gissler et al., 2000).

Statistical analysis

The data were analyzed using Poisson regression analysis in which the incidence of epilepsy per 100,000 persons was the response variable, and geographic area, gender, age, and year of onset of epilepsy were explanatory variables. We determined the annual percent decline, which is defined as the mean decline per year of the total decline from the total follow-up period. Gender and classified age were defined as grouping factors and year was defined as a quantitative (continuous) variable. Three separate analyses were carried out in a similar way for the three geographic areas and different ranges of age: (1) for patients of all ages classified as children (0–14), the middle-aged (15–64), and the elderly (65+ years); (2) for pediatric patients aged 0–14 classified into 5-year age groups; and (3) for geriatric patients aged 65 years or older, classified into 5-year age groups. First a full model with all main effects and all interactions of the three explanatory variables was fitted into the data. If the interaction of the three variables was nonsignificant it was left out and the two-factor interactions were tested. The final model consisted of the significant main and intersection effects. The results were quantified by calculating the risk ratios (RRs) with 95% confidence intervals (95% CIs). p-values of <0.05 were considered statistically significant. Statistical computation was done using the procedure GENMOD in the SAS system for Windows, release 9.1.3/2004 (SAS Institute Inc., Cary, NC, U.S.A.).

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Secular trends in the incidence of epilepsy in Finland (1986–2008)

Table 1 exhibits the numbers of newly diagnosed cases, total population, and incidence rates of epilepsy by year, age group, and gender in the Finnish population based on the SII data from 1986 to 2008. The incidence of epilepsy was declining from 1986 to 2008 in all regions [RR 0.83 (95% CI 0.81–0.84), p < 0.0001]. The mean annual decline was 0.6%. The incidence remained, throughout the observation period, 1.2-fold [RR 1.21 (95% CI 1.19–1.23), p < 0.0001] higher in men than women in all age groups and all regions, with the most marked difference in eastern Finland. In the beginning of the study, compared with western Finland, the incidence of epilepsy was higher in middle [RR 1.08 (1.05–1.12), p < 0.0001] and eastern Finland [RR 1.32 (1.30–1.35), p < 0.0001]. Except for random variation, differences between regions were noted throughout the observation period. In the year 2000, the incidence rate of epilepsy in the elderly permanently exceeded that of children.

Table 1.   Incidence per 100,000 of epilepsy by year in western, middle, and eastern Finland
Year0–15 years16–64 years65+ yearsTotal
EastMiddleWestEastMiddleWestEastMiddleWestEastMiddleWest
19861851747803572541,38166432866084712,447
153,990111,447742,816489,757311,0072,483,07588,23754,675487,233735,284477,1293,713,124
120.1156.1105.072.981.755.674.878.658.782.798.765.9
19871721506683461971,25266392545843862,174
156,442111,499742,198486,781310,6862,490,51289,91255,753494,374733,135477,9383,727,084
109.9134.590.071.163.450.373.470.051.479.780.858.3
19881321286233172191,20169522695183992,093
156,475111,987746,643483,107310,1242,495,61292,04457,030502,767731,626479,1413,745,022
84.4114.383.465.670.648.175.091.253.570.883.355.9
19891591076223352041,25175422605693532,133
157,325113,200751,431480,266310,3192,497,92593,97758,042510,545731,568481,5613,759,901
101.194.582.869.865.750.179.872.450.977.873.356.7
19901491276063381701,29876503095633472,213
158,017113,456756,804478,540311,3932,506,14896,20759,312518,249732,764484,1613,781,201
94.3111.980.170.654.651.879.084.359.676.871.758.5
19911771487153381981 59977473945923932,708
158,142113,794759,922478,693313,1292,519,81498,70060,550526,281735,535487,4733,806,017
111.9130.194.170.663.263.578.077.674.980.580.671.2
19921201005103251591 18388463615333052,054
157,587113,747762,897479,244314,8372,532,087100,71961,639533,678737,550490,2233,828,662
76.187.966.967.850.546.787.474.667.672.362.253.6
19931621055573411981,273145583616483612,191
156,439113,495765,446478,935315,6382,541,656102,73862,926541,228738,112492,0593,848,330
103.692.572.871.262.750.1141.192.266.787.873.456.9
1994135785232791501,10987553605012831,992
155,161112,912766,902475,811316,4922,544,514105,31764,401550,640736,289493,8053,862,056
87.069.168.258.647.443.682.685.465.468.057.351.6
1995141835922681621,21883613614923062,171
153,139112,771770,397473,356317,9092,557,311107,76165,786559,624734,256496,4663,887,332
92.173.676.856.651.047.677.092.764.567.061.655.8
19961561045773331741,287120463326093242,196
150,599111,897771,134470,638318,9582,566,728109,73467,022567,386730,971497,8773,905,248
103.692.974.870.854.650.1109.468.658.583.365.156.2
19971381066423091641,276138573855853272,303
147,746111,014770,269467,127319,8602,579,998111,34467,996574,346726,217498,8703,924,613
93.495.583.366.151.349.5123.983.867.080.665.558.7
19981451015733061951,200126753995773712,172
143,819109,560766,018463,563320,9452,598,316112,26368,687579,247719,645499,1923,943,581
 100.892.274.866.060.846.2112.2109.268.980.274.355.1
Continued
199998955052701611,057118713504863271,912
139,852108,351761,430460,223322,2032,613,772113,69569,593585,230713,770500,1473,960,432
70.187.766.358.750.040.4103.8102.059.868.165.448.3
200096654702681561,056135643984992851,924
136,006107,698756,979456,066324,3382,625,420115,39070,770592,313707,462502,8063,974,712
70.660.462.158.848.140.2117.090.467.270.556.748.4
200189764702371471,102113834044393061,976
132,722107,162754,150452,100326,4272,638,062117,07271,913599,833701,894505,5023,992,045
67.170.962.352.445.041.896.5115.467.462.560.549.5
2002105755272411421,056130604254762772,008
129,615106,699752,345449,237327,3712,646,234118,61673,147608,478697,468507,2174,007,057
81.070.370.053.643.439.9109.682.069.868.254.650.1
200366694752031661,107126924613953272,043
127,096106,029749,151446,140327,6472,641,696121,77075,630626,917695,006509,3064 017,764
51.965.163.445.550.741.9103.5121.673.556.864.250.8
200479674252131381,058116744274082791,910
124,681106,272748,448444,643328,3872,653,435122,96776,184631,671692,291510,8434,033,554
63.463.056.847.942.039.994.397.167.658.954.647.4
200589894952541731,210172965845153582,289
122,146106,299745,391443,371330,3602,666,994124,05177,124639,853689,568513,7834,052,238
72.983.766.457.352.445.4138.7124.591.374.769.756.5
200684835222311501,235146855404613182,297
119,427106,378741,809439,429329,5952,671,536127,52680,057660,967686,382516,0304,074,312
70.378.070.452.645.546.2114.5106.281.767.261.656.4
2007103874532631691,246155875705213432,269
117,171106,271738,831439,035331,3562,692,606127,72780,591666,740683,933518,2184,098,177
87.981.961.359.951.046.3121.4108.085.576.266.255.4
200879524632561671,285188917115233102,459
114,977106,052735,887437,560332,4872,707,224129,28882,155680,479681,825520,6944,123,590
68.749.062.958.550.247.5145.4110.8104.576.759.559.6

The overall incidence of epilepsy, including male and female patients of all ages, declined significantly from 1986 to 2008 in all three study regions: in western [annual decline 0.4%, RR 0.94 (95% CI (0.89–0.99), p = 0.0243], middle [annual decline 1.7%, RR 0.85 (95% CI 0.80–0.91), p < 0.0001], and eastern Finland [annual decline 0.3%, RR 0.93 (0.87–0.99), p = 0.0207]. The decline was most pronounced in the region of middle Finland (Fig. 1).

image

Figure 1.   Decline in the overall annual incidence of epilepsy in western 0.4%, middle 1.7%, and eastern Finland 0.3%.

Download figure to PowerPoint

Age-related differences

In addition, we found substantial age-related differences (Fig. 2, Table 2). The overall incidence of epilepsy in Finland declined from 1986 to 2008 in children [annual decline 1.9%, RR 0.80 (95% CI 0.75–0.86), p < 0.0001] and the middle-aged [annual decline 0.8%, RR 0.88 (95% CI 0.84–0.93), p < 0.0001], whereas it increased in the 65+ group [annual increase 3.5%, RR 1.25 (1.18–1.33), p < 0.0001]. Furthermore, we found regional differences in the incidence of epilepsy for each age group.

image

Figure 2.   Changes between age groups in the incidence of epilepsy by region and gender from 1986 to 2008 in Finland. Differences in the incidence of epilepsy by region and gender in 1986 and in 2008 are shown in western, middle, and eastern Finland for each age group (A, B, C). RR, risk ratio; 95% CI, 95% confidence interval; and p-values for 1986 and 2008 are provided in Table 2.

Download figure to PowerPoint

Table 2.   Changes between age groups in the incidence of epilepsy by region and gender from 1986 to 2008 in Finland
Corresponding Figure 2 panel19862008
RR (95% CI)p-valueRR (95% CI)p-value
  1. RR, risk ratio; 95% CI, 95% confidence interval; and p-values for 1986 and 2008.

A. 0–15 years    
 Eastern versus middle0.85 (0.77–0.94)0.00231.12 (0.99–1.26)0.0768
 Eastern versus western1.21 (1.12–1.30)<0.00011.12 (1.03–1.23)0.0109
 Middle versus western1.41 (1.30–1.53)<0.00011.01 (0.91–1.11)0.8831
B. 16–64 years    
 Eastern versus middle1.09 (1.01–1.17)0.02731.19 (1.09–1.29)<0.0001
 Eastern versus western1.38 (1.31–1.45)<0.00011.24 (1.17–1.31)<0.0001
 Middle versus western1.27 (1.19–1.35)<0.00011.04 (0.98–1.12)0.1959
C. 65+ years    
 Eastern versus middle1.05 (0.91–1.21)0.52431.13 (1.01–1.27)0.0298
 Eastern versus western1.47 (1.33–1.62)<0.00011.50 (1.39–1.62)<0.0001
 Middle versus western1.40 (1.24–1.58)<0.00011.32 (1.20–1.46)<0.0001

Compared with western Finland, the risk for epilepsy among children was significantly higher in eastern Finland both in 1986 [RR 1.21 (95% CI 1.12–1.30), p < 0.0001] and 2008 [RR 1.12 (95% CI 1.03–1.23), p < 0.0109]. Among the middle-aged, both in 1986 and 2008, the risk for the onset of epilepsy was highest in eastern, second highest in middle, and lowest in western Finland (see Fig. 2). Similarly, among the elderly, the increase in the incidence rate was significantly higher in eastern and middle Finland versus western Finland in 1986 and middle versus west, both in 1986 and 2008. More specifically, a significant 1.5-fold increase of epilepsy could be demonstrated in elderly men and women (65+ years), in eastern versus western Finland [RR 1.48 (95% CI 1.42–1.55), p < 0.0001].

Interaction analysis of regional and age-related differences in the incidence of epilepsy

In the interaction analysis of the whole database, using the division into three geographic regions and the age classification of 0–15, 16–64, and 65 years or older, significant interactions existed between gender, age, and year (p = 0.0030), between age and year (p < 0.0001), between gender and year (p = 0.0052), between gender and age (p = 0.0026), between region and year (p = 0.0031), and between region and age (p < 0.0031). The annual trend was different in the age groups; the trends in age groups differed from each other in both genders; annual trend was different in the three regions; and the trends in the three age groups differed similarly from each other in all regions. However, there was no significant interaction either between the four explanatory variables (p = 0.6252), between region, gender, and year (p = 0.6003), between region, gender, and age (p = 0.5403), or between region and gender (p = 0.6032).

Regional-, gender-, and age-related differences in the incidence of epilepsy

The overall incidence of epilepsy was significantly higher in eastern Finland than in middle [RR 1.08 (95% CI 1.05–1.12), p < 0.0001] and western Finland [RR 1.32 (95% CI 1.30–1.35), p < 0.0001]. Figure 3 shows the regional changes in the incidence of epilepsy in male and female patients by age. Although the incidence had significantly declined in children [RR 0.80 (95% CI 0.75–0.86), p < 0.0001] and middle-aged [RR 0.88 (95% CI 0.84–0.93), p < 0.0001], the incidence in the elderly was significantly higher in eastern versus western Finland in both 1986 [RR 1.47 (95% CI 1.33–1.62), p < 0.0001] and 2008 [RR 1.50 (95% CI 1.39–1.62), p < 0.0001] (Table 3). However, the increase was not significant either in men for eastern versus western Finland or in women for eastern versus middle Finland.

image

Figure 3.   Changes between genders in the incidence of epilepsy by region and age group from 1986 to 2008 in Finland. Differences in the incidence of epilepsy by region and gender in 1986 and in 2008 in western, middle, and eastern Finland (A, B, C, D, E, F). RR, risk ratio; 95% CI, 95% confidence interval; and p-values for 1986 and 2008 are provided in Table 3.

Download figure to PowerPoint

Table 3.   Changes between genders in the incidence of epilepsy by region and age group from 1986 to 2008 in Finland
Corresponding Figure 3 panel19862008
RR (95% CI)p-valueRR (95% CI)p-value
  1. RR, risk ratio; 95% CI, 95% confidence interval; and p-values for 1986 and 2008.

A. Female, 0–15 yearsa    
 Eastern versus middle0.83 (0.72–0.96)0.01081.06 (0.89–1.27)0.5010
 Eastern versus western1.23 (1.11–1.37)0.00021.08 (0.95–1.24)0.2315
 Middle versus western1.49 (1.32–1.67)<0.00011.02 (0.88–1.17)0.7891
B. Male, 0–15 years    
 Eastern versus middle0.88 (0.77–1.01)0.06511.17 (0.99–1.38)0.0707
 Eastern versus western1.18 (1.07–1.31)0.00151.16 (1.03–1.32)0.0162
 Middle versus western1.35 (1.20–1.50)<0.00011.00 (0.87–1.14)0.9539
C. Female, 16–64 years    
 Eastern versus middle1.22 (1.01–1.25)0.04151.15 (1.02–1.30)0.0250
 Eastern versus western1.37 (1.27–1.48)<0.00011.21 (1.12–1.32)<0.0001
 Middle versus western1.22 (1.11–1.34)<0.00011.05 (0.95–1.17)0.2966
D. Male, 16–64 years    
 Eastern versus middle1.06 (0.96–1.17)0.25441.22 (1.09–1.36)0.0006
 Eastern versus western1.38 (1.29–1.48)<0.00011.26 (1.17–1.36)<0.0001
 Middle versus western1.31 (1.20–1.42)<0.00011.03 (0.94–1.13)0.4795
E. Female, 65+ years    
 Eastern versus middle1.06 (0.86–1.31)0.54981.12 (0.96–1.32)0.1441
 Eastern versus western1.42 (1.23–1.64)<0.00011.73 (1.55–1.93)<0.0001
 Middle versus western1.33 (1.12–1.59)0.00191.54 (1.34–1.76)<0.0001
F. Male, 65+ years    
 Eastern versus middle1.02 (0.84–1.25)0.79851.15 (0.97–1.35)0.1005
 Eastern versus western1.47 (1.28–1.68)<0.00011.29 (1.16–1.44)<0.0001
 Middle versus western1.43 (1.21–1.70)<0.00011.13 (0.98–1.30)0.0948

Incidence of epilepsy in Finnish subpopulations

Western Finnish subpopulation

In the western Finnish subpopulation, significant interactions existed between age and year (p < 0.0001) and between gender and age (p < 0.0001). The annual trend was different in the age groups and the trends in age groups differed similarly from each other in both genders. However, there was no significant interaction between age and gender and year (p = 0.2815) or between gender and year (p = 0.3927).

Middle Finnish population

In the subpopulation of people living in middle Finland, there was no significant interaction between age and gender and year (p = 0.0501) or between gender and year (p = 0.1142). On the other hand, significant interactions existed between age and year (p < 0.0001) and between gender and age (p < 0.0001). The annual trend was different in the age groups and the trends in age groups differed similarly from each other in both genders.

Eastern Finnish population

Interaction analysis within the three variables among people in eastern Finland showed a significant interaction between age and gender and year (p = 0.0113). The annual trend was different in the age groups and the trends in age groups differed from each other in both genders.

Exploring possible birth cohort effects

To study a possible birth cohort effect on the incidence of epilepsy, the age-specific incidence rates were determined in different birth cohorts (Fig. 4). The incidence rates were similar for each birth cohort in the three regions with one exception. The exception was the oldest birth cohort 1897–1930 corresponding to the age group of >80 years. Although all other birth cohorts showed an incline in the incidence of epilepsy in the elderly, the 1897–1930 birth cohort demonstrated either a decline (western region) or a flat incidence rate (middle and eastern region). The effect of the oldest cohort, however, does not negate the fact that, in the elderly, the incidence of epilepsy is generally higher than in younger adults (Fig. 4).

image

Figure 4.   Age-specific incidence rates of epilepsy plotted against birth cohorts by region in 1986–2008.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Our nationwide registry study has three main results. One, during the follow-up period of 23 years, the overall incidence of epilepsy significantly decreased 14% in the Finnish total population. The decrease in the incidence could be seen in western, middle, and eastern Finland.

Although the incidence of epilepsy decreased from 1986 to 2008 by 45% in the young and by 19% in the middle-aged, it increased in the elderly by 80%. Two, the incidence of epilepsy was throughout the observation period 1.2-fold higher in men than women in all age groups and all regions, with the most marked difference in eastern Finland. Three, we found substantial regional differences in the incidence of epilepsy. A significant 1.5-fold increase of epilepsy could be demonstrated in elderly men and women (65+ years) in eastern versus western Finland, whereas the increase was not significant for either men in middle versus western Finland or for women in eastern versus middle Finland. Therefore, middle Finland constitutes a kind of an intermediate zone between western and eastern Finland in terms of significant changes in the incidence of epilepsy. Starting with the year 2000, the incidence rate of epilepsy is higher in the elderly than in children for all of Finland. In the following section we discuss our finding of regional differences in the incidence of epilepsy in more detail.

Regional differences in the incidence of epilepsy

Compared with western Finland, eastern Finland had a significantly higher prevalence of and mortality from cardiovascular diseases (Keys et al., 1958; Pyörälä et al., 1985) as well as their risk factors (Juonala et al., 2004, 2005). Therefore, it seems possible that the higher rate of overt cardiovascular disease as well as risk factors for cardiovascular disease in eastern versus western Finland may have contributed to the higher incidence of epilepsy in eastern Finland. Although we have no evidence for a birth cohort effect, it is noteworthy that the risk factors for cardiovascular diseases were more common and more severe in eastern than western Finland in the 1960s and early 1970s until, in 1972, an intervention program against cardiovascular risk factors was started in eastern Finland (Puska et al., 1976; Puska, 2010). This program may have reduced the incidence of cerebrovascular events and of subsequent epilepsy in the middle-aged but may have come too late for the elderly.

In that respect it may be worthwhile to note that vascular factors have been suggested to contribute to epileptogenesis in symptomatic epilepsy, at least in experimental models of epilepsy (Ndode-Ekane et al., 2010). Furthermore, recessive diseases, typical of the Finnish population (Finnish Disease Heritage) are significantly more common in eastern than western Finland (Norio, 2003).

There is only one study on the regional differences of the occurrence of epilepsy in Finland (Sillanpää & Koivusilta, 1986). Based on nationwide hospital discharge registry data from the years 1977 and 1982, the proportion of the same-age people hospitalized for primary epileptic seizures was shown greater in eastern than in western Finland (120–168/100,000 vs. 79–100/100,000). Several confounding factors of the study include lower population density and longer distances from hospitals in eastern Finland, which are likely to have increased the hospitalization rate and worsened the discharge rate. The mean number of hospitalization days varied between 10.0 and 27.2 days in western Finland, but between 12.9 and 99.6 days in eastern Finland.

Secular trends in the incidence of epilepsy

As noted briefly in the introduction, there are few studies on secular trends in the incidence of epilepsy in the general population. Cockerell et al.(1995) reported a prevalence of active epilepsy in a small population of 6,000 persons to have significantly declined from 5.3/1,000 in 1983 to 4.3/1,000 in 1993. In the Rochester study (Annegers et al., 1995; Hauser et al., 1993), the incidence of one or more unprovoked cryptogenic epileptic seizures declined particularly in the childhood age group 0–9 years from 85/100,000 in 1935–1944 to 60/100,000 in 1975–84. A random sample study of the year 1986 birth cohort (M. Sillanpää, unpublished data), followed for 15 years in southwestern Finland, obtained the cumulative incidence of 150 per 100,000 children aged 4–15 years. This rate is of the same order as the incidence of epilepsy (158/100,000) of the 5–15-year-olds of the present study. As noted in the introduction, a recent nationwide study (Sillanpää et al., 2006) reported a significant decline in the incidence of epilepsy in the total population, particularly in children and the middle-aged, but an increase in the elderly. The novelty of the present study is that we examined the secular trends of the incidence over all age groups and in different regions of one country over 23 years.

The reasons for the decline of the incidence of childhood epilepsy are unknown but may include better obstetric and neonatal care (Hull & Dodd, 1992) and widespread national vaccination programs that prevent encephalitic brain injuries. Although mortality of very preterm infants is declining, there may be an increase in morbidity. During a 10-year period, in very preterm infants (below 25 gestational weeks), no changes in the rates of cerebral haemorrhages or seizures could be found (Lundqvist et al., 2009). Severe postnatal injuries, often followed by epilepsy (Annegers et al., 1998; Christensen et al., 2009), have been decreasing over the time period studied. In Finland, the annual incidence of head injuries decreased from 250–120 per 100,000 in boys and from 150–100 per 100,000 in girls (Parkkari et al., 2000).

Changes in treatment practices might also have contributed to the decline shown in the present registry data. The previous practice of regarding cases with a single seizure, particularly status epilepticus, as epilepsy and in need of drug therapy, may have evolved into a more restrictive use of long-term AED treatment in these cases. The number of pediatric neurologists who currently take care of most children with epilepsy in Finland prefer to have a longer follow-up of the patient with seizures than before to ascertain the diagnosis of epilepsy warranting full-refundable drugs. Some patients may then be left unmedicated. There is also a greater awareness of certain benign epileptic syndromes in infancy and childhood not necessarily needing long-term AED treatment at all. In few cases, some patients may have been administered a novel antiepileptic drug as a first-line drug without reimbursement. Although each of these particular factors may play a marginal role, they may jointly affect the incidence rate.

Incidence of epilepsy in the elderly

According to our study, the elderly are now at higher risk of epilepsy than 20 years ago. Although some studies still report lower rates in the elderly than in children (Forsgren, 1992), the increased rate of epilepsy is now widely recognized (Keränen et al., 1989; Hauser et al., 1993). An increasing trend in the elderly has been reported from France (Loiseau et al., 1990), the United States (Hauser et al., 1993; Annegers et al., 1995), Iceland (Olafsson et al., 1996), and Estonia (Oun et al., 2003).

Reasons for the rising incidence rate of epilepsy in the elderly are largely unclear. In all industrialized countries including Finland aging of people and increasing survival of people with underlying medical conditions such as stroke place them at higher risk than before. Stroke and other vascular catastrophies are the most common risk factor for epilepsy in the elderly (Camilo & Goldstein, 2004). Annegers et al. (1995) attributed an increased incidence in the elderly of longer survival after stroke. In accordance, reports from several Scandinavian countries indicate decreasing incidence, mortality and case fatality rates of stroke (Sivenius et al., 2004). According to the Finnish nationwide registry data (Pajunen et al., 2005b), the incidence of first-ever stroke declined from 1991 to 2002 both in men and in women aged 75–84 years and in those 85 years and older. Municipal health care, responsible for institutionalized patients with epilepsy (who are not included in the SII files), has increasingly moved these patients to open care, in part, to become eligible for fully refundable antiepileptic medication from the SII system. Although no data are available, that may possibly have, to a minor extent, contributed to the increase in the incidence of epilepsy in the elderly in registry-based data.

Cerebral tumors are another major cause of epilepsy in the elderly occurring in 11% (Sander et al., 1990; Forsgren et al., 1996). Similarly, a severe brain injury is associated with late seizures in 12% of adults (Annegers et al., 1998). The incidence of dementing diseases is increasing with aging and with time. A diagnosis of Alzheimer’s disease and a diagnosis of other dementia, classified according to the criteria of DSM-III, were associated with at least sixfold risk for epilepsy (Hesdorffer et al., 1996). Improving skills in identification and ascertainment of epilepsy due to an increased awareness by neurologists about the condition in the elderly may be an additional factor for the increase.

Finally, it is worthwhile to note that the age-specific incidences for different birth cohorts (Fig. 4) are similar in all the three regions, indicating no specific birth cohort effect on the incidence, with one exception. The exception is the oldest age group in the birth cohort of 1897–1930. The 1897–1930 birth cohort showed either a decline or a flat incidence rate in those beyond 80 years of age. In the light of other findings of our study, the most obvious explanation is a shorter life time of that cohort. Consequently, the patients in the oldest cohort at risk did not survive long enough to get epilepsy in their life time. The lower incidence of epilepsy in this subgroup, however, did not change the fact that the elderly people are at higher risk of epilepsy than other adults.

Study limitations

The SII data do not include a small minority of people (3% in 2003, data on the SII file) with epilepsy, who although eligible, elect not to utilize the fully refundable drug system. On the other hand, a small minority of patients with a misdiagnosis of epilepsy has been granted refundable medication (data on file). As an additional limitation of the registry, it does not cover people who are living in institutions for the mentally retarded or in nursing homes. Although there is uncertainty about the incidence of epilepsy among institutionalized patients or among residents of nursery homes, we estimate that the SII registry covers 97% of all persons with new-onset epilepsy (see Data S1).

Implications and conclusions

The two main outcomes of our study are that the incidence of epilepsy is much higher in east versus west Finland and that the incidence of epilepsy has increased in the elderly and declined in the young have implications for epilepsy research and for public health. We are not aware of publications of regional differences of the incidence of epilepsy elsewhere. Therefore, it is difficult as of now to extrapolate our findings to other countries. The increase in the incidence of epilepsy in the elderly is alarming and makes prevention of epilepsy in the aging population a top priority for experimental and clinical epilepsy research. On a more optimistic note, we found that pediatric epilepsy is declining. Identifying the cause(s) of the apparent increase in population epileptogenesis in one region versus another region is also a high priority for epilepsy research. Identifying the mechanism(s) for the decline in epilepsy may provide new tools for efforts to completely cure epilepsy. In conclusion, our study shows a janus-like view of epilepsy, with a decline of pediatric epilepsy and an increase in epilepsy for the rapidly increasing elderly population.

Acknowledgments

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

We like to thank Olli Kaleva, B.A. for the computations and Ilona Autti-Rämö, M.D., Ph.D. for reviewing the text and useful advice.

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

None of the authors has any conflict of interest to disclose. 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.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information
  • Annegers JF, Hauser WA, Lee JR, Rocca WA. (1995) Secular trends and birth cohort effects in unprovoked seizures: Rochester, Minnesota 1935–1984. Epilepsia 36:575579.
  • Annegers JF, Hauser WA, Coan SP, Rocca WA. (1998) A population-based study of seizures after traumatic brain injuries. N Engl J Med 338:2024.
  • Camilo O, Goldstein LB. (2004) Seizures and epilepsy after ischemic stroke. Stroke 35:17691775.
  • Christensen J, Pedersen MG, Sidenius P, Olsen J, Vestergaard M. (2009) Long-term risk of epilepsy after traumatic brain injury in children and young adults: a population-based cohort study. Lancet 373:11051110.
  • Cockerell OC, Eckle I, Goodridge DM, Sander JW, Shorvon SD. (1995) Epilepsy in a population of 6000 re-examined: secular trends in first attendance rates, prevalence, and prognosis. J Neurol Neurosurg Psychiatry 58:570576.
  • Commission on Revised Classification of Epilepsy (1989). (1989) Proposal for revised classification of epilepsies and epileptic syndromes. Commission on classification and terminology of the international league against epilepsy. Epilepsia 30:389399.
  • Commission on Revised Classification of Seizures. (1981) Proposal for revised clinical and electroencephalographic classification of epileptic seizures. From the commission on classification and terminology of the international league against epilepsy. Epilepsia 22:489501.
  • Forsgren L. (1992) Prevalence of epilepsy in adults in northern Sweden. Epilepsia 33:450458.
  • Forsgren L, Bucht G, Eriksson S, Bergmark L. (1996) Incidence and clinical characterization of unprovoked seizures in adults: a prospective population-based study. Epilepsia 37:224229.
  • Gissler M, Teperi J, Hemminki E, Meriläinen J. (1995) Data quality after restructuring a national medical registry. Scand J Soc Med 23:7580.
  • Gissler M, Järvelin MR, Hemminki E. (2000) Comparison between research data and routinely collected register data for studying childhood health. Eur J Epidemiol 16:5966.
  • Hauser WA, Annegers JF, Kurland LT. (1993) Incidence of epilepsy and unprovoked seizures in Rochester, Minnesota: 1935–1984. Epilepsia 34:453468.
  • Heinonen R. (1997) Finns accept the registration of their own data (in Finnish). Tietoyhteiskuntafoorumi 4:2528.
  • Hesdorffer DC, Hauser WA, Annegers JF, Kokmen E, Rocca WA. (1996) Dementia and adult-onset unprovoked seizures. Neurology 46:727730.
  • Hull J, Dodd KL. (1992) Falling incidence of hypoxic-ischaemic encephalopathy in term infants. Br J Obstet Gynaecol 99:386391.
  • Immonen-Räihä P, Sarti C, Tuomilehto J, Torppa J, Lehtonen A, Sivenius J, Mononen A, Narva EV. (2003) Eleven-year trends of stroke in Turku, Finland. Neuroepidemiology 22:196203.
  • Juonala M, Viikari JS, Hutri-Kähönen N, Pietikäinen M, Jokinen E, Taittonen L, Marniemi J, Rönnemaa T, Raitakari OT. (2004) The 21-year follow-up of the cardiovascular risk in young Finns study: risk factor levels, secular trends and east-west difference. J Int Med 255:457468.
  • Juonala M, Viikari JS, Kähönen M, Taittonen L, Rönnemaa T, Laitinen T, Mäki-Torkko N, Mikkilä V, Räsänen L, Åkerblom HK, Pesonen E, Raitakari OT. (2005) Geographic origin as a determinant of carotid artery intima-media thickness and brachial artery flow-mediated dilation: the cardiovascular risk in young Finns study. Arterioscler Thromb Vasc Biol 25:392398.
  • Juul-Jensen P, Foldspang A. (1983) Natural history of epileptic seizures. Epilepsia 24:297312.
  • Keränen T, Riekkinen PJ, Sillanpää M. (1989) Incidence and prevalence of epilepsy in adults in eastern Finland. Epilepsia 30:413421.
  • Keys A, Karvonen MJ, Fidanza F. (1958) Serum cholesterol studies in Finland. Lancet II:175178.
  • Loiseau J, Loiseau P, Guyot M, Duche B, Dartigues JF, Aublet B. (1990) Survey of seizure disorders in the French southwest. I. Incidence of epileptic syndromes. Epilepsia 31:391396.
  • Lundqvist P, Källén K, Hällström I, Westas LH. (2009) Trends in outcomes for very preterm infants in the southern region of Sweden over a 10-year period. Acta Paediatr 98:648653.
  • Mähönen M, Miettinen H, Pyörälä K, Molarius A, Salomaa V, Kuulasmaa K. (1995) Hospital discharge register data in the assessment of trends in acute myocardial infarction. FINMONICA AMI register study team. Ann Med 27:547554.
  • Ndode-Ekane XE, Hayward N, Gröhn O, Pitkänen A. (2010) Vascular changes in epilepsy: functional consequences and association with network plasticity in pilocarpine-induced experimental epilepsy. Neuroscience 166:312332.
  • Norio R. (2003) Finnish disease heritage II: population prehistory and genetic roots of finns. Hum Genet 112:457469.
  • Olafsson E, Hauser WA, Ludvigsson P, Gudmundsson G. (1996) Incidence of epilepsy in rural Iceland: a population-based study. Epilepsia 37:951955.
  • Oun A, Haldre S, Magi M. (2003) Incidence of adult epilepsy in Estonia. Acta Neurol Scand 108:245251.
  • Pajunen P, Koukkunen H, Ketonen M, Jerkkola T, Immonen-Räihä P, Karja-Koskenkari P, Mähönen M, Niemelä M, Kuulasmaa K, Palomäki P, Mustonen J, Lehtonen A, Arstila M, Vuorenmaa T, Lehto S, Miettinen H, Torppa J, Tuomilehto J, Kesäniemi YA, Pyörälä K, Salomaa V. (2005a) The validity of the Finnish hospital discharge register and causes of death register data on coronary heart disease. Eur J Cardiovasc Prev Rehabil 12:132137.
  • Pajunen P, Pääkkönen R, Hämäläinen H, Keskimäki I, Laatikainen T, Niemi M, Rintanen H, Salomaa V. (2005b) Trends in fatal and nonfatal strokes among persons aged 35 to > or =85 years during 1991–2002 in Finland. Stroke 36:244248.
  • Parkkari J, Kannus P, Niemi S, Koskinen S, Palvanen M, Vuori I, Järvinen M. (2000) Childhood deaths and injuries in Finland in 1971–1995. Int J Epidemiol 29:516523.
  • Pietilä K, Tenkanen L, Mänttäri M, Manninen V. (1997) How to define coronary heart disease in register-based follow-up studies: experience from the Helsinki heart study. Ann Med 29:253259.
  • Puska P. (2010) From Framingham to North Karelia: from descriptive epidemiology to public health action. Prog Cardiovasc Dis 53:1520.
  • Puska P, Koskela K, Pakarinen H, Puumalainen P, Soininen V, Tuomilehto J. (1976) The North Karelia Project: a programme for community control of cardiovascular diseases. Scand J Soc Med 4:5760.
  • Pyörälä K, Salonen JT, Valkonen T. (1985) Trends in coronary heart disease mortality and morbidity and related factors in Finland. Cardiology 72:3551.
  • Rapola JM, Virtamo J, Korhonen P, Haapakoski J, Hartman AM, Edwards BK, Heinonen OP. (1997) Validity of diagnoses of major coronary events in national registers of hospital diagnoses and deaths in Finland. Eur J Epidemiol 13:133138.
  • Sander JW, Hart YM, Johnson AL, Shorvon SD. (1990) National general practice study of epilepsy: newly diagnosed epileptic seizures in a general population. Lancet 336:12671271.
  • Sillanpää M, Koivusilta L. (1986) General hospital care in Finland I. hospital care for patients with epileptic seizures in 1977 and 1982. Reports from the University Departments of Public Health, M 93: 176.
  • Sillanpää M, Kälviäinen R, Klaukka T, Helenius H, Shinnar S. (2006) Temporal changes in the incidence of epilepsy in Finland: nationwide study. Epilepsy Res 71:206215.
  • Sivenius J, Tuomilehto J, Immonen-Räihä P, Kaarisalo M, Sarti C, Torppa J, Kuulasmaa K, Mähönen M, Lehtonen A, Salomaa V. (2004) Continuous 15-year decrease in incidence and mortality of stroke in Finland: the FINSTROKE study. Stroke 35:420425.

Supporting Information

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. Disclosure
  8. References
  9. Supporting Information

Data S1. SII registry covers 97% of all persons with new-onset epilepsy.

FilenameFormatSizeDescription
EPI_3186_sm_DataS1.doc40KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.