Background: Cardiovascular diseases (CVD) constitute one of the most significant causes of morbidity and are among the three most frequent causes of death in industrial countries. They comprise coronary heart disease (CHD), cerebrovascular disease and peripheral artery disease (PAD).
Cholesterol lowering pharmacological interventions (statins) have been increasingly used to prevent CVD. In a previous study, the potential population health gains from 11 years of statin use in secondary prevention in Austria were calculated. The model used was based on the efficacy of statins derived from clinical studies. The results need to be verified with observed CVD epidemiology.
Research question: The study aims to analyse whether the expected health gains from statins in secondary prevention of cardiovascular diseases (derived from modelling) can also be observed in Austrian cardiovascular disease epidemiology.
Method: A literature search was conducted to identify international epidemiological trends of CVD and the statins’ role in these.
Expected health gains from statins in Austria were derived from an earlier study.
Austrian cardiovascular disease epidemiology was analysed descriptively based on administrative hospital data (as proxy for morbidity) and on mortality data. Statistical tests are used to analyse the influence of statins on Austrian CVD mortality.
International trends: Western Europe has experienced downward trends in age-adjusted CVD mortality. Addressing risk factors seems to have a greater impact on the trend than improving treatment. However, the trend towards increasing prevalence of some risk factors (e.g. obesity) may offset some of the positive effects.
Despite the benefits attributed to lowering cholesterol with statins, tackling other risk factors (e.g. smoking) and implementing specific types of treatment/ management (e.g. hospital resuscitation after myocardial infarction) has shown a greater effect on reduced mortality than the use of statins.
Expected health gains in Austria: In roughly 600,000 patients who took statins between 1996 and 2006 about 26,600 cases of MI and roughly 10,200 fatal CHDs (mostly MIs) seem to have been avoided or postponed. The effect on (fatal) stroke was low. Moreover, about 7,000 fewer revascularisation interventions have been estimated compared to not taking statins
However, the model demonstrated that about 68,000 CVD cases or fatal events and 230,000 revascularisation interventions still occurred in spite of statin treatment.
Observed CVD trends in Austria: From the age of 50 and over, discharge rates for those who were hospitalized for CVD rose steadily. Discharges of MI patients rose from 21,218 in 1996 to 37,064 in 2006 (+75 %). For angina pectoris, discharges also rose slightly, while in the case of stroke, data are unclear. In terms of revascularisation, the number of percutaneous coronary interventions (PCI) has also risen considerably (from 5,506 in 1995 to 16,153 in 2005/+ 194 %), while the number of conducted coronary artery bypass grafting (CABG) slightly decreased. Absolute numbers are in any case higher for men than for women.
Age standardized CVD mortality rate fell by 59 % between 1970 and 2005. The decline is similar for men and women. Twenty percent of all deaths and 40 % of all cardiovascular deaths are due to CHD. Most of the mortality cases in CVD occur above the age of 65.
Contrasting the results
While model outputs demonstrated decreasing numbers of CVD cases and revascularisation procedures when statin takers were compared with those not taking statins, hospital discharges indicate an overall increase in CVD morbidity.
On the contrary, CVD mortality rates per 100,000 person years were decreasing in both the model as well as observed mortality when compared to the reference year 1994. The trend between model results and observed mortality is particularly consistent for reductions in CHD and MI mortality rates.
Furthermore, the average decline in relative mortality changes per year and in annual mortality rates per 100,000 is stronger in the period after the launch of statins than before. The difference is statistically significant for both males and females with regard to CHD mortality overall and to MI mortality in particular, but not in the case of cerebrovascular deaths.
Sensitivity analysis: Observed CVD mortality was contrasted with model mortality outputs according to varying gender and age distribution of statin takers in the model. While CVD mortality rate reductions in model and observed mortality were more consistent in sensitivity analysis, CHD mortality rate reduction was higher in sensitivity analysis and, thus, less consistent with observed mortality.