Trends in Fatal and Nonfatal Strokes Among Persons Aged 35 to ≥85 Years During 1991–2002 in Finland
Background and Purpose— Declining trends in the incidence and mortality of stroke have been observed in Finland since the beginning of the 1980s until 1997. In this study we analyzed the trends in fatal and nonfatal strokes in Finland during 1991–2002.
Methods— The Finnish Hospital Discharge Register was linked to the National Causes of Death Register to produce a Cardiovascular Disease Register, which includes data on 410 760 cerebrovascular events (International Statistical Classification of Diseases, 10th Revision [ICD-10] codes I60–I69) in patients aged ≥35 years in 1991–2002.
Results— Age-standardized incidence of first-ever stroke (ICD-10 codes I60–I64, excluding I63.6) per 100 000 persons declined during 1991–2002 annually by 2.2% (95% CI, −2.4% to −1.9%) among men and 2.5% (−2.8% to −2.2%) among women aged 35 to 74 years. In patients aged 75 to 84 years, the change in the incidence of first-ever stroke was −2.6% per year (−3.0% to −2.2%) among men and −3.2% per year (−3.5% to −2.9%) among women. A similar trend was observed also in the oldest age group, in patients aged ≥85 years. Among patients aged 35 to 74 years, the 28-day case fatality of first-ever stroke declined annually by 3.2% (−3.9% to −2.5%) among men and by 3.0% (−3.8% to −2.2%) among women. A significant decrease was found in the 28-day case fatalities of all subtypes of stroke in this age group.
Conclusions— The favorable development in stroke incidence, mortality, and case-fatality has continued in Finland during 1991–2002.
Finland used to have one of the highest reported incidences of stroke in the world.1 However, declining trends in the incidence and mortality of stroke have been observed since the beginning of the 1980s until 1997.2–4
The Finnish Cardiovascular Disease Register (CVDR) is a countrywide database5 that compiles data from the records of the National Causes of Death Register, National Hospital Discharge Register, and Social Insurance Institution to provide data on mortality and morbidity of coronary heart disease (CHD) and cerebrovascular diseases in Finland.6 In the international context, the European joint project EUROCISS has recommended similar registers to be launched in other member states of the European Union.7
We have previously reported data on the validity of the diagnosis of CHD events and on the annual trends in fatal and nonfatal CHD events based on the CVDR project.8,9 In the present study we analyzed trends in nonfatal and fatal cerebrovascular events in patients aged ≥35 years in Finland during a 12-year period from 1991–2002. The database is uniquely large and contains every stroke event leading to hospitalization or death in all age groups in Finland.
Subjects and Methods
Data on cerebrovascular diseases in 1991–2002 in Finland were obtained from the Finnish Hospital Discharge Register and the Causes of Death Register. These administrative registers cover the whole country, comprising information on all hospitalizations and deaths of permanent residents of Finland. The Hospital Discharge Register is maintained by the National Research and Development Centre for Welfare and Health and the Causes of Death Register by Statistics Finland. Both registers use the International Classification of Diseases (ICD). In Finland, the 10th version of the ICD (International Statistical Classification of Diseases, 10th Revision [ICD-10]) has been in use since January 1, 1996.
Data on hospitalizations due to cerebrovascular diseases in the Hospital Discharge Register were linked to the Causes of Death Register with the use of personal identification codes. Deaths and hospitalizations due to cerebrovascular diseases with ICD-10 codes I60–I69 were included. Case selection in the Hospital Discharge Register included the main diagnosis and the additional diagnoses. From the Causes of Death Register, the data covered deaths with cerebrovascular diseases as the underlying or the immediate cause of death or as the contributing cause of death.
One person could have had multiple events during 1 year. However, only 1 event was registered if the time interval between 2 hospitalizations or hospitalization and death was ≤28 days. The events were classified according to the year in which the event began.
The calculation of incidence rates of first-ever strokes included events with ICD-10 codes I60–I64 (excluding I63.6, cerebral infarction due to venous thrombosis). A case was classified as the first for a patient if no prior hospitalizations with ICD-10 codes I60–I64 (excluding I63.6) were found during the preceding 7 years. If any previous stroke-related hospitalizations were found, the case was recorded as a recurrent event.
Attack and mortality rates and case fatalities were calculated separately for ischemic strokes (ICD-10 codes I63–I64), for subarachnoid hemorrhage (SAH) (I60), and for intracerebral hemorrhage (ICH) (I61). The attack rates include both first and recurrent stroke attacks. The attack rates were also calculated for the whole category of cerebrovascular diseases (I60–I69) because in the administrative data that category gives the most robust and most comprehensive picture of the burden caused by these diseases to society. The ICD-9 codes corresponding to ICD-10 codes used are given at http://www.ktl.fi/cvdr/.
Even though the study period covered the years 1991–2002, the calculations of the 28- to 365-day case fatalities were performed for years 1991–2001 because the follow-up data for the year 2002 in this setting would have been incomplete. The 2002 data also include patients hospitalized because of stroke at the turn of the year 2002.
The register linkage was based on the personal identification codes. In the registers, 0.1% of personal identification codes were flawed. The CVDR data, together with a more detailed description of the methodology, are available at http://www.ktl.fi/cvdr/.
Stroke incidence, mortality, and attack rates were expressed per 100 000 persons and age-standardized to the European standard population. The annual population counts for the denominators were obtained from the National Population Register. The case fatality figures were age-standardized according to the practice of the World Health Organization (WHO) Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) Project with the use of weights 1/23 (for the group aged 35 to 44 years), 3/23 (45 to 54 years), 7/23 (55 to 64 years), and 12/23 (65 to 74 years).
Trends in incidence and mortality rates were estimated with the use of log-linear regression models with year as an independent variable assuming Poisson distribution for the annual numbers of events within age groups. To avoid the effects of random fluctuations, the event rates presented in the tables are smoothed with Poisson regression. The statistical analyses were performed with software from the SAS Institute.10
The CVDR includes data on 410 760 cerebrovascular events (ICD-10 codes I60–I69) registered in Finland in 1991–2002 among patients aged ≥35 years. Of these events, 201 277 (49%) occurred among men and 209 483 (51%) among women. During this time period there were 131 375 events of stroke (ICD-10 codes I60–I64, excluding I63.6) among men and 146 485 among women aged ≥35 years. Of the strokes among men, 57% (74 917 events) were first-ever attacks. The respective percentage among women was 62% (90 449 events).
The age-standardized attack rates of all cerebrovascular diseases (ICD-10 codes I60–I69) decreased significantly during the study period. The annual change in the attack rate was −2.6% (95% CI, −3.6% to −1.7%) among men and −2.5% (−3.0% to −2.0%) among women in the group aged 35 to 74 years. Similar decreasing trends were observed in patients aged ≥75 years (data not shown).
The age-standardized incidence of first-ever stroke (ICD-10 codes I60–I64, excluding I63.6) decreased significantly in the group aged 35 to 74 years from 392/100 000 to 303/100 000 (rates are 3-year moving averages) among men and from 227/100 000 to 175/100 000 among women (Figure 1). The incidence of first-ever stroke declined significantly in all 5 university hospital districts of Finland (data not shown). A declining trend was also observed in the attack rates combining first and recurrent strokes; the annual change was −1.6% (95% CI, −1.8% to −1.4%) among men and −1.9% (−2.2% to −1.7%) among women in the group aged 35 to 74 years.
Similarly to the younger population, the incidence of first-ever stroke declined among men aged 75 to 84 years by 2.6% (−3.0% to −2.2%) per year from 2517/100 000 in 1991 to 1808/100 000 in 2002 and among women of this age group by 3.2% (−3.5% to −2.9%) per year from 2241/100 000 in 1991 to 1561/100 000 in 2002. The incidence of first-ever stroke also showed a declining trend among both men and women aged ≥85 years, with an annual change of −1.1% (−1.7% to −0.3%) and −1.9% (−2.3% to −1.5%), respectively.
Declining trends were observed in attack rates of ischemic stroke and of ICH, whereas the attack rate of SAH remained stable among men and increased among women in the group aged 35 to 74 years (Table 1⇓). Table 2 shows the respective rates for patients aged ≥75 years. However, calculation of trends in SAH attack and mortality for these elderly patients was not meaningful because of too few cases during the study period.
During the study period, mortality decreased in all subtypes of stroke in both sexes in patients aged 35 to 74 years (Table 1⇑). The mortality of ischemic stroke decreased in patients aged ≥75 years (Table 2). The mortality of ICH remained stable among men aged ≥75 years, decreased among women aged 75 to 84 years, and increased among women aged ≥85 years (Table 2).
Both 28-day and 28- to 365-day case fatalities of first-ever strokes decreased significantly in both sexes in patients aged 35 to 74 years (Figures 2 and 3⇓). In patients aged 75 to 84 years, the 28-day case fatality decreased annually by 3.7% (−4.5% to −2.9%) among men and by 3.5% (−4.1% to −3.0%) among women. In patients aged ≥85 years, the 28-day case fatality decreased annually by 3.2% (−4.4% to −2.0%) among men and 1.9% (−2.5% to −1.3%) among women.
In patients aged 35 to 74 years, a declining trend was observed in 2-, 28-, and 28- to 365-day case fatalities in all subtypes of stroke, except in the 28- to 365-day case fatality of ICH among both sexes (Table 3). Additionally, the 28- to 365-day case fatality of SAH remained stable among women (Table 3).
In the present study we analyzed trends in fatal and nonfatal cerebrovascular events in Finland during the 12-year period of 1991–2002. Our findings, which show declining rates for stroke incidence and mortality in patients aged 35 to 74 years, are in agreement with the results from the FINSTROKE Study covering the group aged 25 to 74 years and certain areas of Finland during the years 1983–1997.3 In the present study attack rates declined for ischemic stroke and ICH but were unchanged for SAH among men and increased among women in patients aged 35 to 74 years. Congruent findings have recently been reported from the United Kingdom,11 whereas a Swedish study arrived at contrary results.12 The declining trend in the mortality rate of stroke has nearly leveled off in the United States, where the annual change in death rate for stroke diminished from −3.5% in 1980–1990 to only −0.7% in 1990–1997.13
Since the risk of stroke increases with age, the disease particularly affects the elderly. Similarly to the younger population in our study, the incidence of first-ever strokes decreased annually in both sexes aged ≥75 years, a finding that supports earlier regional data from Finland.4 However, the declining trends were observed specifically in ischemic strokes; the attack rates of ICH remained stable or increased in this age group.
In the study by Sivenius et al,3 there was a trend for declining 28-day case fatality of all strokes in patients aged 25 to 74 years during 1983–1997. Because of larger numbers, we were able to elaborate the finding and demonstrated a significant decrease in 28-day case fatalities of all subtypes of stroke in patients aged 35 to 74 years. This decline may be explained not only by improved acute treatment of stroke14 but also by the decreasing severity of stroke events.15 Sivenius et al3 did not find any change in the case fatality of ischemic stroke, unlike in our study, in which we observed significant decreases in 2-, 28-, and 28- to 365-day case fatalities of ischemic stroke among both sexes in the group aged 35 to 74 years.
The 28-day case fatality of stroke in patients aged 75 to 99 years did not change significantly in the FINSTROKE Study, even though there was a nonsignificant declining trend.4 In our study the 28-day case fatality of stroke declined among men and women aged ≥75 years. The longer follow-up period and larger number of patients in our study may explain the different results between these 2 studies.
The occurrence of SAH shows large variations between countries, and Finland has been reported to have the highest attack rates within 11 populations according to the WHO MONICA Project.16 Earlier studies from the beginning of the 1990s did not find any change in the case fatality of SAH in Finland.15,17 A later report from 1988–1997 observed a slightly diminishing trend in men.3 We found a significant decrease in 2-, 28-, and 28- to 365-day case fatalities of SAH in men and a smaller decrease among women in 2- and 28-day case fatalities.
A recent report from Sweden observed an increasing incidence of stroke among the population aged 30 to 65 years.12 There were also differences in the incidence of stroke between different geographic areas of Sweden. In our study we found similar declining trends in all 5 university hospital districts across Finland.
The strength of the present study is that the dataset is uniquely large, with >400 000 cerebrovascular events. It covers all of Finland and all age groups, and the dataset is freely available on the Internet. Despite these advantages, some limitations should be taken into account. The source registers used in compiling the CVDR were collected for administrative rather than for scientific purposes. Surveillance using administrative data, such as the Hospital Discharge Register and the Causes of Death Register, usually slightly underestimates the total stroke occurrence. This is true especially for ischemic strokes, and thus the numbers of ischemic strokes may have been underestimated in the present study. The feasibility and reliability of Finnish administrative registers for epidemiological studies on stroke have previously been verified.18,19 Because these validation studies were performed before the 10th version of the ICD was introduced in Finland, we have no validation data for the era of ICD-10. However, we have recently validated the CHD part of the CVDR, and the results indicate that the administrative registers collected with the 10th version of the ICD can be used as they were used previously, with some caution, for epidemiological studies.8 Furthermore, as shown by Sivenius et al,3 by the year 1997 >97% of the stroke diagnoses in Finland were verified by either CT, MRI, or necropsy.
In conclusion, the favorable development in stroke incidence, mortality, and case fatality has continued in Finland. The incidence and mortality of first-ever stroke decreased significantly among Finns aged ≥35 years. Moreover, our study showed that it is feasible to use the CVDR for monitoring the occurrence of stroke in Finland. The EUROCISS Project has suggested that similar population-based registers on coronary and cerebrovascular diseases using consistent classifications should be established in several European countries to improve comparability of the epidemiological data between different countries.7
- Received September 23, 2004.
- Revision received October 29, 2004.
- Accepted November 3, 2004.
Tuomilehto J, Rastenyte D, Sivenius J, Sarti C, Immonen-Räihä P, Kaarsalo E, Kuulasmaa K, Narva EV, Salomaa V, Salmi K, Torppa J. Ten-year trends in stroke incidence and mortality in the FINMONICA Stroke Study. Stroke. 1996; 27: 825–832.
Thorvaldsen P, Asplund K, Kuulasmaa K, Rajakangas AM, Schroll M. Stroke incidence, case fatality, and mortality in the WHO MONICA project: World Health Organization Monitoring Trends and Determinants in Cardiovascular Disease. Stroke. 1995; 26: 361–367.
Sivenius J, Tuomilehto J, Immonen-Räihä P, Kaarisalo M, Sarti C, Torppa J, Kuulasmaa K, Mähönen M, Lehtonen A, Salomaa V, for the FINSTROKE Study. Continuous 15-year decrease in incidence and mortality of stroke in Finland: the FINSTROKE Study. Stroke. 2004; 35: 420–425.
Laatikainen T, Pajunen P, Pääkkönen R, Keskimäki I, Hämäläinen H, Rintanen H, Niemi M, Moltchanov V, Salomaa V. National Cardiovascular Disease Register, statistical database. Available at: http://www.ktl.fi/cvdr/URN:NBN:fi-fe20041365. Accessed June 24, 2004.
Mähönen M, Salomaa V, Keskimäki I, Moltchanov V. The feasibility of routine mortality and morbidity register data linkage to study the occurrence of acute coronary heart disease events in Finland: the Finnish Cardiovascular Diseases Registers (CVDR) Project. Eur J Epidemiol. 2000; 16: 701–711.
EUROCISS Working Group. Coronary and cerebrovascular population-based registers in Europe: are morbidity indicators comparable? Results from the EUROCISS Project. Eur J Public Health. 2003; 13: 55–60.
Pajunen P, Koukkunen H, Ketonen M, Jerkkola T, Immonen-Räihä P, Kärjä-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. The validity of the Finnish Hospital Discharge Register and Causes of Death Register data on coronary heart disease. Eur J Cardiovasc Prev Rehabil. 2005; In press.
Pajunen P, Pääkkönen R, Juolevi A, Hämäläinen H, Keskimäki I, Laatikainen T, Moltchanov V, Niemi M, Rintanen H, Salomaa V. Trends in fatal and non-fatal coronary heart disease events in Finland during 1991–2001. Scand Cardiovasc J. 2005; In press.
SAS Institute Inc. Users Guide, Statistics, Version 8. Vol 1. Cary, NC: SAS Institute; 1999.
Rothwell PM, Coull AJ, Giles MF, Howard SC, Silver LE, Bull LM, Gutnikov SA, Edwards P, Mant D, Sackley CM, Farmer A, Sandercock PA, Dennis MS, Warlow CP, Bamford JM, Anslow P, for the Oxford Vascular Study. Change in stroke incidence, mortality, case-fatality, severity, and risk factors in Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study). Lancet. 2004; 363: 1925–1933.
Medin J, Nordlund A, Ekberg K. Increasing stroke incidence in Sweden between 1989 and 2000 among persons aged 30 to 65 years: evidence from the Swedish Hospital Discharge Register. Stroke. 2004; 35: 1047–1051.
Cooper R, Cutler J, Desvigne-Nickens P, Fortmann SP, Friedman L, Havlik R, Hogelin G, Marler J, McGovern P, Morosco G, Mosca L, Pearson T, Stamler J, Stryer D, Thom T. Trends and disparities in coronary heart disease, stroke, and other cardiovascular diseases in the United States: findings of the national conference on cardiovascular disease prevention. Circulation. 2000; 102: 3137–3147.
Lindsberg PJ, Soinne L, Roine RO, Salonen O, Tatlisumak T, Kallela M, Häppölä O, Tiainen M, Haapaniemi E, Kuisma M, Kaste M. Community-based thrombolytic therapy of acute ischemic stroke in Helsinki. Stroke. 2003; 34: 1443–1449.
Numminen H, Kaste M, Aho K, Waltimo O, Kotila M. Decreased severity of brain infarct can in part explain the decreasing case fatality rate of stroke. Stroke. 2000; 31: 651–655.
Ingall T, Asplund K, Mähönen M, Bonita R. A multinational comparison of subarachnoid hemorrhage epidemiology in the WHO MONICA stroke study. Stroke. 2000; 31: 1054–1061.
Immonen-Räihä P, Mähönen M, Tuomilehto J, Salomaa V, Kaarsalo E, Närvä EV, Salmi K, Sarti C, Sivenius J, Alhainen K, Torppa J. Trends in case-fatality of stroke in Finland during 1983 to 1992. Stroke. 1997; 28: 2493–2499.
Mähönen M, Salomaa V, Keskimäki I, Moltchanov V, Torppa J, Molarius A, Tuomilehto J, Sarti C, for the FINMONICA Stroke Register Study Group. The feasibility of combining data from routine hospital discharge and causes-of-death registers for epidemiological studies on stroke. Eur J Epidemiol. 2000; 16: 815–817.