Ronald D Gregor

Decreasing Mortality from Acute Myocardial Infarctions: Effect of Attack Rates and Case Severity

Mortality from myocardial infarction (MI) has declined in many countries and the reasons for the decline have not been fully quantified. We used the database of the Halifax County MONICA Project to test the hypothesis that the decline of in-hospital mortality from MI can be explained by a trend toward less severe disease as opposed to improved treatment. During the study period 1984-1993, 14,130 people aged 25-74 had been admitted to hospital with suspected MI. Of these, 3774 were diagnosed as definite MI by standardized criteria (480 fatal). For each patient, clinical history, serial cardiac enzymes, and ECG treatment regimen during hospital stay were extracted from patient charts. Survival status 28 days after onset of symptoms was determined. A severity index predicting 28-day case fatality was derived from health status at admission time. During the study period the rate of definite MI in the MONICA target population showed a general downward trend from 221 to 179 per 100,000/year (p = 0.0002). The severity index increased during the observation time (p < 0.0001), predicting 25% higher mortality. Case fatality fluctuated, but showed a marginally significant decline. We conclude that part of the decreased in-hospital mortality from MI is due to lower attack rates. The remainder occurred despite increased case severity and is possibly due to improved in-hospital treatment.

Trends in event rate and case fatality of patients hospitalized with myocardial infarction between 1984 and 2001.

Between 1984 and 1993, prevalence and case fatality of hospitalized acute myocardial infarction (AMI) had declined in the population of Halifax County. We aimed to determine whether these trends continued into the 21st century by investigating patient characteristics, treatment methods, and fatality for hospital admissions of residents of Halifax County, aged 25-74, during 1984-1989 (period 1), 1990-1993 (period 2), and 1998-2001 (period 3) and diagnosed as AMI that were extracted from databases for the Halifax County MONICA and ICONS (Improving Cardiovascular Outcomes in Nova Scotia) Studies. Trends in patient characteristics and treatment methods were assessed by chi2 statistics. Their association with 28-day fatality was determined by logistic regression. Event rate declined during 1984-1993 but not into 1998-2001 (p = 0.206). Compared with 1990-1993, fewer AMI patients during 1998-2001 were > or = 55 years (73.3% vs. 69.9%), cigarette smokers (49.8% vs. 42.9%), had a history of myocardial infarction (28.9% vs. 24.9%), and had an admission heart rate >100 (34.8% vs. 17.4%). Additionally, more patients had a history of diabetes (22.5% vs. 28.1%). Case fatality declined progressively over the 3 study time periods (16.6%, 13.1%, and 9.4%, respectively). Changes also occurred in prevalence of Killip class 4 status during admission (20.2%, 10.3%, and 13.3%, respectively), use of thrombolysis (9.0%, 30.9, and 32.6%, respectively), and percutaneous coronary intervention (PCI) (4.3%, 11.2%, and 22.4%, respectively) in the different periods. Significant associations were found between case fatality and patient history of diabetes, history of MI, age, elevated admission heart rate, Killip class 4 impairment, thrombolysis, and PCI. The ICONS registry of hospitalized acute myocardial infarctions was used to compare case fatality during 1998-2001 with that reported by the Halifax County MONICA Project for 1984-1993. Whereas the population rate of myocardial infarctions had declined between 1984-1993 but not subsequently, case fatality declined significantly throughout the study period. The continued decline in case fatality is likely explained by changes in patient profile on presentation and medical therapies, including the increased use of thrombolysis and PCI.

An Epidemiological Model for Coronary Heart Disease (CHD)

Mortality from CHD has been on the increase in most countries until in the late sixties the trend reversed in a few countries such as the USA, Canada, and Australia (Cooper 1978, Nicholls 1981, Epstein 1979). Ever since then CHD mortality has been on the decline in these countries. Since it is not known whether the incidence of CHD has declined or whether the prognosis of patients with CHD has improved, several studies have been initiated to determine the cause of the trend reversal. The data emanating from these studies will be difficult to interpret unless one is cogniscant of the inherent dynamics of the CHD system. The aim of this communication is to propose a dynamic model for the CHD system and to perform some preliminary system analysis. The benefits to be derived from such a model extend beyond the interpretation of mortality observations. It should be particularly useful in simulating intervention programs, since they are not only difficult to conduct but also require a very long observation period.