Decline of Coronary Heart Disease Mortality in Finland During 1983 to 1992: Roles of Incidence, Recurrence, and Case-Fatality
The FINMONICA MI Register Study
Background The rate of coronary heart disease (CHD) mortality in eastern Finland has been the highest in the world. The official mortality statistics suggest, however, that it has declined by 60% during the past 20 years. The aim of the present study was to examine the contributions of incidence, recurrence, and case fatality of coronary events to the trends in CHD mortality in three areas of Finland.
Methods and Results Population-based myocardial infarction registers have been operating in the provinces of North Karelia and Kuopio in eastern Finland and the Turku/Loimaa area in southwestern Finland from 1983 to 1992. During this 10-year period, each suspected coronary event in persons 35 to 64 years of age was evaluated for registration. Of these, 13 566 fulfilled the criteria of myocardial infarction or coronary death. Almost one fourth (22.4%) of the coronary events were sudden, out-of-hospital deaths. Among men, the average change in mortality was −7.1% per year (95% confidence interval, −8.4% to −5.8%) in North Karelia, −5.0% per year (−7.0% to −3.0%) in Kuopio, and −4.9% per year (−8.2% to −1.6%) in Turku/Loimaa. Among women, the corresponding changes were −5.6% (−11.1% to −0.1%), −4.4% (−8.1% to −0.7%), and −8.1% (−13.0% to −3.2%). In eastern Finland, the decline in CHD mortality was due to a decline in recurrent coronary events but also in the incidence of first coronary events, whereas in southwestern Finland, the decline in case-fatality rate had the major role.
Conclusions The decline in CHD mortality rate in Finland appears to be the result of a successful combination of primary and secondary prevention measures and improvements in acute coronary care.
In Finland, as in many Western countries, the rate of coronary heart disease (CHD) mortality has declined considerably during the past 20 years.1 This decline has been accompanied by simultaneous decreases of the main CHD risk factors,2 and it has been shown that the decrease in risk factors almost completely explains the decrease in CHD mortality in the 1970s and early 1980s, but since the mid 1980s, the CHD mortality has declined more than could be predicted from changes in risk factor levels.3 This has led to speculation that during the 1970s the decline in CHD mortality rate was mainly due to the primary prevention, but during the 1980s improved treatment and secondary prevention have increasingly contributed to the decline.
In principle, the decrease in CHD mortality rate in a defined population can be due to a decreased incidence of first coronary events, decreased recurrence of coronary events, decreased case-fatality rates, or combination of all these. The changes in the incidence of first coronary events reflect mainly the effects of primary prevention, whereas decreased recurrence of coronary events can be attributed to secondary prevention, and the changes in case-fatality rates are dependent on acute coronary care and the severity of disease. We have used the final 10-year results of the FINMONICA Myocardial Infarction (MI) Register to examine the roles of incidence, recurrence, and case-fatality rates of coronary events in the decline of CHD mortality rates in Finland.
The FINMONICA MI Register is the Finnish contribution to the MI registers of the WHO MONICA Project (multinational MONItoring of trends and determinants of CArdiovascular disease).4 5 Registration methods and 5-year trends in coronary events have been described previously.6 7 The 10-year period of the FINMONICA Project covers the years 1983 to 1992. The monitored areas were the provinces of North Karelia and Kuopio in eastern Finland and the Turku/Loimaa area in southwestern Finland. Populations of the study areas remained stable, but the annual population counts were obtained from the Central Statistical Office of Finland. In the middle of the registration period in 1988 there was within the age range of 35 to 64 years 33 611 men and 31 916 women in North Karelia, 48 635 men and 47 798 women in Kuopio, and 37 625 men and 40 420 women in Turku/Loimaa area.
Each event suspected to be a coronary death or nonfatal MI occurring in permanent residents of the study areas was evaluated for registration. Main sources for case-finding were the hospital admission lists and death certificates of the area. Data for persons 35 to 64 years of age were included in this report. According to the MONICA protocol,5 8 suspected coronary events were classified into the following diagnostic categories: (1) definite MI, (2) possible MI or coronary death, (3) ischemic cardiac arrest with successful resuscitation, (4) no MI, and (5) fatal unclassifiable events. There were only 77 fatal unclassifiable events during the study period, and in the analyses they were combined with the category of possible MI. The diagnostic classification was based on symptoms, serial Minnesota coding of ECGs, cardiac enzymes, and, in fatal cases, autopsy findings and history of CHD. All relevant hospital documents and autopsy reports were available for the investigators of the FINMONICA MI Register performing the classification.
In the registration of hospitalized patients, the FINMONICA MI Register used primarily the “hot pursuit” approach. This means that the registration was initiated soon after the patient's admission to a hospital. Specially trained nurses interviewed the patient and collected information from the hospital records. These data were then transferred to the investigators of the local registration team working in the main hospital of each area. They carried out the diagnostic classification described above. Every 3 months, the local registration teams obtained a list of hospital discharges with CHD diagnoses (International Classification of Diseases [ICD] codes 410 to 414) in their areas. Hospital records of events occurring on this list but not in the MI register were checked and evaluated for registration. ICD 8 was used in Finland until 1986 and ICD 9 since 1987.
Information on out-of-hospital deaths was primarily based on the review of death certificates in each area. A list of deaths in which CHD (ICD codes 410 to 414) or sudden death (codes 795 in ICD 8 and 798 in ICD 9) appeared as an underlying or other cause of death was prepared. Copies of death certificates, medico-legal documents, and possible autopsy reports then were obtained from the authorities of the provincial health administration and evaluated for registration. The autopsy rate was 47.6% for men and 41.4% for women with definite or possible fatal MI in North Karelia. The respective proportions were 49.2% and 47.7% in Kuopio and 78.0% and 71.2% in the Turku/Loimaa area. The autopsies were directed particularly to cases of out-of-hospital death, in which clinical information was scarce.
The local registration teams sent their data periodically to the coordinating center at the National Public Health Institute, Helsinki. There the data were checked for logical errors and corrected together with the registration teams when necessary. Annually, the completeness of the registration of coronary deaths was checked with record linkage of the MI register data with the National Death Register through the use of the personal ID number unique to every citizen in Finland. At the end of the 10-year period, the completeness of the registration of nonfatal events was once more checked with record linkage of the MI register data with the National Hospital Discharge Register for ICD codes 410 to 411. Information on events found in these national registers but not in the FINMONICA MI Register was sent to the local registration teams, which evaluated the cases according to the MONICA protocol.
Adherence to the protocol and standardized coding was supervised by organization of regular meetings for the registration teams 2 to 3 times per year. In these meetings, international test cases distributed by the MONICA Quality Control Centre for event registration in Dundee, Scotland, and test ECGs distributed by the Quality Control Centre for ECG coding in Budapest, Hungary, were discussed and coded. In addition, FINMONICA has organized national test sets both for coronary events and ECGs to monitor the performance of registration teams.
In the present report, coronary event rates are presented according to the “MONICA Definition 4,”9 which has been developed largely on the basis of Finnish experience.6 7 10 It has been shown to give consistent results in different monitoring areas in Finland and to correspond closely to the clinical diagnosis of MI.10 It includes fatal definite MIs, fatal possible MIs, nonfatal definite MIs, and nonfatal probable MIs. The diagnostic criteria for each of these categories have been described in detail.5 9 The incidence of coronary events (first-ever events) was determined by asking the patient whether he or she has had previous MIs and by checking the hospital records. In cases of out-of-hospital death, the medicolegal documents and previous hospital records were checked. Sometimes relatives were also interviewed. Only in 5 events in North Karelia, 22 events in Kuopio, and 10 events in Turku/Loimaa was no information available on the history of previous MIs. These were almost exclusively out-of-hospital deaths. In the analyses, these cases with unknown history of previous MI were combined with the category “no previous MI.”
The statistical computations were carried out with the use of SAS.11 Coronary event rates were expressed per 100 000 persons and age-standardized according to the direct method using 5-year age groups and the world standard population as the standard.12 The 28-day case-fatality rate was expressed as the proportion of fatal events from all events in percents. It was age-standardized according to the practice of the WHO MONICA Project through the use of age distribution of pooled MI patients in the WHO MONICA MI registers as the standard.5 The calculation of trends in event rates and the contributions of changes in incidence, recurrence, and case-fatality to the changes in CHD mortality was carried out as described in “Appendix 1.”
In all, 25 155 suspected coronary events were registered during the 10-year study period. Of these, 13 566 fulfilled the criteria of MI or sudden coronary death and were included in the analyses. Almost one fourth (22.4%) of the included coronary events were sudden, out-of-hospital deaths. The distribution of the included coronary events by area, sex, and year is given in Table 1.⇓ It also shows the age-standardized attack rates of coronary events and their trends in the study areas. The attack rates declined significantly both in men and women of North Karelia and Kuopio. Smaller and nonsignificant declining trends were observed in men and women of the Turku/Loimaa area. CHD mortality rates declined significantly in all three areas in both sexes (Figure).⇓
The incidence of first coronary events declined significantly in North Karelia and Kuopio among men and in North Karelia among women (Table 2).⇓ The declines in the mortality from an incident coronary event were similar to the declines in the incidence of first events in all three areas among men and in North Karelia and Kuopio among women, suggesting that the changes in case-fatality of first coronary events did not add much to the decline in the incidence. Only in women of Turku/Loimaa was the decline in mortality from incident coronary events steeper than the decline in the incidence, suggesting improved case-fatality rates. It should be noted, however, that sudden out-of-hospital deaths from incident coronary events also declined steeply in Turku/Loimaa (Table 2).⇓
Approximately 64% of all coronary events were incident and 36% were recurrent events. The age-standardized occurrence of recurrent coronary events declined very steeply in the populations of North Karelia and Kuopio, as did the mortality from recurrent events (Table 3).⇓ In Turku/Loimaa, the occurrence of recurrent events declined less than in the other areas, but nevertheless the mortality from recurrent events declined steeply also in Turku/Loimaa.
The 28-day case-fatality rates in the beginning of the study period varied among men from 43.6% in Turku/Loimaa to 30.4% in Kuopio (Table 4).⇓ Among women, rates varied from 42.4% in Turku/Loimaa to 19.4% in Kuopio. The case-fatality of recurrent coronary events was higher than that of incident events. Changes in the total 28-day case-fatality rate remained modest during the study period. Only in men of North Karelia and in men and women of Turku/Loimaa, who had the highest levels to start with, were significant declines observed. The trends in survival until the age of 65 years of patients who had survived their first MI were estimated with the use of a model described in “Appendix 1.” This long-term survival improved considerably during the study period. The annual average changes among men were +4.3% in North Karelia, +1.3% in Kuopio, and +4.5% in Turku/Loimaa. Among women, the corresponding annual average changes were +0.4% in North Karelia, +0.95% in Kuopio, and +2.4% in Turku/Loimaa.
Calculations on the different components of mortality reduction during the 10-year period showed that among men of North Karelia and Kuopio, the largest part of the reduction of coronary mortality was due to the decline in the occurrence of recurrent coronary events (Table 5).⇓ Declines in the incidence of first coronary events also contributed considerably, whereas the improvement in case-fatality rate contributed only 11.9% in men of North Karelia. In men of Kuopio, changes in incidence and recurrence predicted a negative contribution of case-fatality to the reduction of CHD mortality, which is consistent with the fact that the case-fatality rate increased in men of Kuopio from 30.4% to 33.8%. In men of Turku/Loimaa, the improvement in case-fatality rate contributed the major part to the reduction in CHD mortality rate. Decline in the occurrence of recurrent coronary events also contributed considerably in Turku/Loimaa (29.1%), but the role of the decline in incidence was smaller. Among women, the reduction of coronary mortality rate varied from 32 deaths per 100 000 in Kuopio to 19 deaths per 100 000 in Turku/Loimaa. Because of small numbers, the percentage contributions are unstable in women. Nevertheless, decline in the incidence seems to have played the major role in North Karelia and decline in case-fatality rate in Turku/Loimaa. Decline in the occurrence of recurrent coronary events contributed considerably in all areas.
At the beginning of the 10-year monitoring period of the WHO MONICA Project, North Karelia in eastern Finland had the highest CHD mortality rate of all participating populations among men, and Kuopio ranked very high.5 Our present results demonstrate steep declines in the CHD mortality rate in all FINMONICA areas during 1983 to 1992. Declining trends have been reported from many Western countries, but those in eastern Finland are among the steepest in the world.13 14 15 16 For some time, it has been a matter of scientific debate whether the decrease in the coronary mortality rate is mainly due to decreased incidence or decreased case-fatality rate.16 17 18 19 The rationale is that the changes in incidence are considered to reflect the effect of primary prevention, whereas changes in case-fatality rate are mainly attributed to the changes in acute coronary care.
In Finland, the contributions of incidence and case-fatality rate were previously assessed in the context of the MI registers organized by the European Office of the WHO in the 1970s.20 21 The conclusion then was that the decline in CHD mortality rate, which took place in Finland during 1970 to 1977, was mainly due to the declining incidence of coronary events and thus due to primary prevention, while the changes in case-fatality rate played a smaller role.20 21 Since that time, however, considerable development has taken place in the field of acute coronary care; for example, coronary care units became widely used in the latter half of the 1970s. Thus, their main impact on the in-hospital mortality occurred before the FINMONICA period. During the 1980s, treatment practices have been monitored by the acute coronary care surveys of the FINMONICA MI Registers, which have revealed marked changes in the treatment of MI as well as in the treatment of symptomatic CHD in general.22 Most notably, the use of thrombolytic treatment for definite MIs has increased from almost zero to 37% to 47%, depending on study area, during 1986 to 1992. The use of acetyl salicylic acid also has increased significantly. β-Blockers were commonly used in 1986, when the first acute coronary care survey of FINMONICA was carried out. More recently, the use of ACE inhibitors after MI has increased considerably. Thus, there are reasons to believe that in addition to the favorable changes in risk factors,2 treatment has played a role during our study period.
Our findings demonstrate that the decline in CHD mortality rate during the period of 1983 to 1992 in the FINMONICA areas was due to the combined effects of declines in incidence, recurrence, and case fatality. In particular, the role of declined occurrence of recurrent coronary events was prominent in each area and in both sexes. This is a novel finding, since previous research in this field has concentrated on incidence and case fatality. It should be noted that we examined the role of recurrent coronary events in two different ways. First, the trends in the occurrence of recurrent coronary events and the trends in the mortality from recurrent coronary events were examined per population (Table 3).⇑ Thus, the observed reductions can be due to either declined recurrence of coronary events among those who have experienced at least one previous event or declined pool size of persons having had a previous coronary event in the population. The latter is a function of changes in incidence and case fatality of first coronary events. Second, using statistical modeling, we estimated trends in long-term survival among patients who had survived their first coronary event. These results complement each other and show that the prognosis of persons who have survived their first coronary event has clearly improved.
The concept that the declined recurrence of coronary events plays a role in the decline of CHD mortality rate is further supported by the follow-up results of our team showing that the 1-year prognosis of patients who have survived their first MI has improved significantly during the FINMONICA period.23 In the present data, the recurrent coronary events contributed 36% of the total attack rate and 41% of the CHD deaths. Accordingly, there is potential to reduce CHD mortality rate by secondary prevention measures, which reduce the recurrence of coronary events. Actually, the potential of secondary prevention is even larger than can be concluded from the numbers above because a considerable proportion of patients who experience their first MI already have had symptomatic CHD before their coronary event. The recent results of the Scandinavian Simvastatin Survival Study,24 results of several trials on ACE inhibitors,25 and other research26 27 28 suggest that the prognosis of patients who have survived their first MI can be further improved.
Except in Turku/Loimaa, the role of case fatality remained modest. This was somewhat disappointing, since in clinical trials thrombolytic treatment has resulted in the improvement in case fatality of MI.29 A possible explanation is that the frequency of thrombolytic treatment in 1992 was still insufficient to make its effects clearly discernible at the population level. Interestingly, however, a recent study from the Minnesota Heart Survey16 reported a significant improvement in the 28-day case-fatality rate of hospitalized MIs between 1985 and 1990, although the use of thrombolytic treatment was not more common than in the FINMONICA areas. The improvement in long-term prognosis of post-MI patients in the FINMONICA areas was consistent with the findings of the Minnesota group and can be explained in part by the effects of thrombolytic treatment, which reduce the extent of myocardial damage. It should be kept in mind also that almost a quarter of all coronary events and two thirds (67%) of CHD deaths in the present population-based study were sudden, out-of-hospital deaths. These people can be reached by primary prevention measures but a considerable proportion of them potentially also by secondary prevention measures, since in our data 36% of persons who died out-of-hospital had had a previous MI and a further 25% had symptomatic CHD.
Our present findings are the result of examination of more than 25 000 suspected coronary events according to a standardized protocol during 10 years. We also had the unique opportunity to check the completeness of registration by computerized record linkage with the National Death Register and the National Hospital Discharge Register, covering all deaths and hospitalizations in Finland. Therefore, we are confident that our registration completely covers both fatal and hospitalized nonfatal coronary events and that the main findings are not biased by missing data or misclassification of events. There were no major changes in emergency medical services or in the frequency of autopsies during the study period. However, hospital admission practices may have changed toward the end of our study, since rapid hospitalization of patients with chest pain was emphasized with the adoption of thrombolytic treatment. Changed hospitalization policy may have led to a change of the clinical spectrum of hospitalized MIs to a milder direction, as has been observed in earlier analyses of the FINMONICA data.30 We have, however, taken this into account by calculating the event rates according to the MONICA Definition 4, which also includes milder events (nonfatal probable MIs). Therefore, it is not likely that the change in the clinical spectrum of MIs has biased the trends in attack rates or incidence rates.
On the basis of the National Death Register and the National Hospital Discharge Register, it seems to us that our findings on trends in CHD mortality and attack rate of MI in the FINMONICA areas are fairly well generalizable to other parts of Finland. Little is known, however, on trends in the incidence and case fatality of MI outside the FINMONICA areas. It seems plausible that also in other parts of eastern and northern Finland, where the risk factor levels have been high, primary and secondary prevention may have played the major role in the decline in CHD mortality rate. In southwestern Finland, where the risk factor levels have been lower, decline in case-fatality rates has had a more important contribution. The generalizability of our results to countries other than Finland is more complex. An observational study such as ours does not justify firm conclusions on causality even within one country, and inferences regarding the situation in other countries must be made with caution. We believe, however, that our results have relevance and could stimulate similar research in other countries where the CHD mortality trends have been decreasing.
These final 10-year results of the FINMONICA MI Register Study have confirmed the steep decline of CHD mortality rates in the monitored areas. This decline is mainly the result of decreased incidence and recurrence of coronary events in eastern Finland, whereas the decreased case-fatality rate was the largest contributor in southwestern Finland. The prominent role of decreased occurrence of recurrent coronary events was consistent in all three areas and in both sexes, suggesting effective secondary prevention. Thus, the favorable development evidently was due to a successful combination of primary and secondary prevention and improvements in acute coronary care.
The trends in event rates and case-fatality rates were determined, as described previously,7 with the use of regression analysis with the natural logarithm of rate as dependent variable and year as independent variable, ie,Ln(rate)|<|=|>|a|<|+|>|b(year)The regression coefficient of year (b) multiplied by 100 gives the average annual change in percents.
We had particular interest in trends in incidence of first coronary events (I1), trends in 28-day survival from first coronary events (S1), and trends in survival until age 65 years of patients who had survived their first coronary event (S2). We can calculate I1 and S1 directly from the data but not S2. If none of the variables involved change in time, we have the approximate equationI_|<|1|>||<|-|>|M|<|=|>|I_|<|1|>||<|\times|>|S_|<|1|>||<|\times|>|S_|<|2|>|where M denotes the coronary mortality from first and recurrent events. After log-transformation, this equation takes the formLn(I_|<|1|>||<|-|>|M)|<|=|>|Ln(I_|<|1|>|)|<|+|>|Ln(S_|<|1|>|)|<|+|>|Ln(S_|<|2|>|)If we calculate trends for (I1−M), I1, S1, and S2, as described above, we geta|<|+|>|b(year)|<|=|>|(a_|<|1|>||<|+|>|a_|<|2|>||<|+|>|a_|<|3|>|)|<|+|>|(b_|<|1|>||<|+|>|b_|<|2|>||<|+|>|b_|<|3|>|) (year)where b, b1, b2, and b3 are the trends in (I1−M), I1, S1, and S2, respectively. Ignoring the intercepts, b3 can be now calculated asb_|<|3|>||<|=|>|b|<|-|>|b_|<|1|>||<|-|>|b_|<|2|>|This gives an estimate of the later survival of patients who have survived their first coronary event. It can be shown that for this approximate estimate, we no longer need to assume that I1, M, S1, and S2 are constant over time. All that is needed is that the trends are not very large and that most of those who die of a recurrent event die within a few years from the first event. The latter assumption is valid because the upper age limit of our registration was 65 years, and in this age range most of the coronary events occur in persons who are around 60 years of age or over. An additional assumption that we are making is that the mortality from causes other than CHD is small and does not change substantially over time. Previous literature from Finland31 and our own unpublished data indicate that this assumption can be made.
The proportion of mortality rate reduction as the result of the decline in the incidence of first coronary events, decline in the occurrence of recurrent coronary events in population, and change in 28-day case-fatality rate was calculated as follows. First, the reduction of age-standardized mortality rate during the study period was calculated with the use of a regression model as described above. For example, for men in North Karelia it was 253 (per 100 000). Then, the incidence of first coronary events in the beginning and at the end of the study period was estimated with the use of the regression model. For men in North Karelia, these were 772 and 513 (per 100 000), respectively. The 28-day case-fatality rate of incident coronary events, again calculated with the use of regression-based smoothing, was 38.2% in the beginning of the study period. Assuming that the case-fatality rate of incident coronary events remained unchanged throughout the study period, we then estimated the number of reduced deaths (per 100 000 men) due to the reduced incidence of first coronary events to be (0.382×772)−(0.382×513)=99.9, ie, 39.5% of all reduced deaths. The number of reduced deaths due to the reduction in the occurrence of recurrent coronary events was calculated analogously to that of the incidence, assuming that the case-fatality rate of recurrent coronary events remained unchanged. For men in North Karelia, the equation was (0.464×601)−(0.464×260)=122, ie, 48.6%. The remaining part of the mortality reduction, which was not due to the changes in incidence or recurrence, was considered to be due to the changes in case-fatality rate and was obtained by subtraction: 100−39.5−48.6=11.9%, or 30.2 of the total 253 reduced deaths. This is consistent with the fact that according to direct modeling, the case-fatality rate of all coronary events declined during the study period among men in North Karelia from 41.7% to 36.7%.
The FINMONICA MI Register Study Group
North Karelia. Harri Mustaniemi, Matti Ketonen, Paula Korhonen, Eira Laiste.
Kuopio. Heikki Miettinen, Seppo Lehto, Pertti Paloma¨ki, Kalevi Pyo¨ra¨la¨, Seija Apell, Marjatta Selka¨inaho, Pirkko Koljonen.
Turku. Matti Arstila, Tapio Vuorenmaa, Pirjo Immonen-Ra¨iha¨, Anu Mononen.
Loimaa. Esko Kaarsalo, Matti Niemela¨, Aino-Maija Kantee.
National Public Health Institute, Helsinki. Markku Ma¨ho¨nen, Veikko Salomaa, Matti Romo, Pekka Puska, Jorma Torppa, Jaakko Tuomilehto.
The FINMONICA MI Register received funding from the Academy of Finland and from the Finnish Heart Foundation.
Members of the FINMONICA MI Register Study Group are listed in “Appendix 2.”
- Received April 22, 1996.
- Revision received July 30, 1996.
- Accepted July 31, 1996.
- Copyright © 1996 by American Heart Association
Vartiainen E, Puska P, Jousilahti P, Korhonen HJ, Tuomilehto J, Nissinen A. Twenty-year trends in coronary risk factors in North Karelia and in other areas in Finland. Int J Epidemiol.. 1994;23:495-504.
Vartiainen E, Puska P, Pekkanen J, Tuomilehto J, Jousilahti P. Changes in risk factors explain changes in mortality from ischaemic heart disease in Finland. BMJ.. 1994;309:23-27.
Tunstall-Pedoe H, Kuulasmaa K, Amoyel P, Arveiler D, Rajakangas A-M, Paja˛k A. Myocardial infarction and coronary deaths in the World Health Organization MONICA Project: registration procedures, event rates, and case-fatality rates in 38 populations from 21 countries in four continents. Circulation.. 1994;90:583-612.
Tuomilehto J, Arstila M, Kaarsalo E, Kankaanpa¨a¨ J, Ketonen M, Kuulasmaa K, Lehto S, Miettinen H, Mustaniemi H, Paloma¨ki P, Puska P, Salomaa V, Torppa J, Vuorenmaa T. Acute myocardial infarction (AMI) in Finland: baseline data from the FINMONICA AMI register in 1983-1985. Eur Heart J.. 1992;13:577-587.
Salomaa V, Arstila M, Kaarsalo E, Ketonen M, Kuulasmaa K, Lehto S, Miettinen H, Mustaniemi H, Niemela¨ M, Paloma¨ki P, Pyo¨ra¨la¨ K, Torppa J, Tuomilehto J, Vuorenmaa T. Trends in the incidence of and mortality from coronary heart disease in Finland, 1983-1988. Am J Epidemiol.. 1992;136:1303-1315.
WHO MONICA Project. MONICA Manual. Revised ed. Geneva, Switzerland: WHO, Cardiovascular Diseases Unit; 1990.
WHO MONICA Project, prepared by Salomaa V, Dobson A, Miettinen H, Rajakangas A-M, Kuulasmaa K. Mild myocardial infarction: a classification problem in epidemiologic studies. J Clin Epidemiol. In press.
Users Guide: Statistics. Version 6, 4th ed, Vol 2. Cary, NC: SAS Institute Inc. 1989:846.
Waterhouse J, Muir C, Corree P, eds. Cancer Incidence in Five Continents. Vol III (IARC Scientific Publication No. 15). Lyon, France: International Agency for Research on Cancer; 1976.
Thom TJ. International mortality from heart disease: rates and trends. Int J Epidemiol. 1989;18(suppl 1):S20-S28.
Jackson R, Stewart A, Beaglehole R. Trends in coronary heart disease mortality and morbidity in Auckland, New Zealand, 1974-1986. Int J Epidemiol.. 1990;19:279-283.
Burke GL, Sprafka JM, Folsom AR, Luepker RV, Norsted SW, Blackburn H. Trends in CHD mortality, morbidity and risk factor levels from 1960 to 1986: the Minnesota Heart Survey. Int J Epidemiol. 1989;18(suppl 1):S73-S81.
Beaglehole R. Medical management and the decline in mortality from coronary heart disease. BMJ.. 1986;292:33-35.
Goldman L, Cook F, Hashimoto B, Stone P, Muller J, Loscalzo A. Evidence that hospital care for acute myocardial infarction has not contributed to the decline in coronary mortality between 1973-1974 and 1978-1979. Circulation.. 1982;65:936-942.
Pohjola-Sintonen S, Siltanen P, Haapakoski J, Romo M. The declining trend in the incidence of acute myocardial infarction in Finland: the results of Helsinki Coronary Register during 1970-1977. Eur Heart J.. 1985;6:834-839.
Pohjola-Sintonen S, Siltanen P, Haapakoski J, Romo M. The mortality and case fatality rates in acute myocardial infarction in Finland: the results of Helsinki Coronary Register during 1970-1977. Eur Heart J.. 1987;8:354-359.
Miettinen H, Ma¨ho¨nen M, Salomaa V, Ketonen M, Arstila M, Immonen-Ra¨iha¨ P, Kaarsalo E, Lehto S, Mustaniemi H, Niemela¨ M, Paloma¨ki P, Puska P, Pyo¨ra¨la¨ K, Torppa J, Tuomilehto J, Vuorenma T. Syda¨ninfarktiin sairastuneiden potilaiden la¨a¨kehoitoka¨yta¨nto¨ Suomessa vuosina 1986-1992: FINMONICA tutkimuksen tuloksia. Suomen La¨a¨ka¨rilehti.. 1996;6:543-552.
Niemela¨ M, Salomaa V, Arstila M, for the FINMONICA AMI Register Study Group. One-year prognosis of first acute myocardial infarction: results from the FINMONICA AMI Register Study during 1983-89. Eur Heart J. 1994;15(suppl 510):2657. Abstract.
Yusuf S, Wittes J, Friedman L. Overview of results of randomized clinical trials in heart disease, I: Treatments following myocardial infarction. JAMA.. 1988;60:2088-2093.
Eccles M, Bradshaw C. Use of secondary prophylaxis against myocardial infarction in the north of England. BMJ.. 1991;302:91-92.
Ha¨ma¨la¨inen H, Luurila OJ, Kallio V, Knuts L-R, Arstila M, Hakkila J. Long-term reduction in sudden deaths after a multifactorial intervention programme in patients with myocardial infarction: 10-year results of a controlled investigation. Eur Heart J.. 1989;10:55-62.
Fibrinolytic Therapy Trialists' Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet.. 1994;343:311-322.
Salomaa V, Miettinen H, Paloma¨ki P, Arstila M, Mustaniemi H, Kuulasmaa K, Tuomilehto J, for the FINMONICA AMI Register Study Group. Diagnostic features of acute myocardial infarction: changes over time from 1983 to 1990: results from the FINMONICA AMI Register Study. J Intern Med.. 1995;237:151-159.
Pohjola S, Siltanen P, Romo M. Five-year survival of 728 patients after myocardial infarction: a community study. Br Heart J.. 1980;43:176-183.