Social inequalities in mild and severe myocardial infarction: how large is the gap in health expectancies?

Acute myocardial infarction (MI) is one of the major causes of hospital admissions leading often to long-term impairments and premature death in the affected population [1‐4]. MI often triggers a sudden deterioration in cardiovascular health resulting in increased risks of further cardiovascular acute events like recurrent MIs or stroke, poor quality of life and impaired functional health [5‐11]. Although considerable progress has been made in treatment and prevention, population ageing and declining mortality after MI are assumed to increase the burden of MI globally [1, 12]. There is a large body of research that confirms the close link between SES and MI incidence and mortality risks [13‐22]. However, so far research investigating social inequalities in MI incidence and mortality with respect to the severity of MI is still insufficient. In particular, little is known about how socioeconomic status (SES) affects the average lifetime spent free of MI and after mild or severe MI.

Major factors contributing to social inequalities in MI incidence and mortality risks lie in the increased prevalence of cardiovascular risk factors and harmful health behaviours in lower SES groups [20, 23, 24]. In the German population these inequalities were found in both, the prevalence of risk factors (e.g. hypertension, hyperlipidaemia, high Body Mass Index, and diabetes) [25‐27] and risky health behaviour (e.g. smoking, physical inactivity) [28‐30] which contribute to the substantial inequalities in MI incidence and mortality risks [31‐34]. The prevalence of cardiovascular risk factors and harmful health-related behaviour also differs significantly between men and women. This applies in particular to smoking, unhealthy nutrition, hazardous alcohol consumption and high blood pressure, which are more prevalent in men than in women [35‐37], while no clear gender differences were found for exercise and in the prevalence of dyslipidaemia and diabetes mellitus [36, 38, 39]. Moreover, some studies suggest that social inequalities in cardiovascular risk factors and harmful health-related behaviour might be stronger in women than in men, e.g. in smoking and diabetes mellitus [24, 36]. Studies also indicate that MIs are often severe in individuals with low SES and cause high mortality [17, 40, 41], increased risks of hospital readmission and of substantial and permanent deterioration in cardiovascular health (especially chronic heart failure, recurrent MI) [19, 21, 42‐44]. However, so far little is known on the gap in average lifetime affected by the consequences of MI between SES groups and how inequalities in life years differ with respect to MI severity.

The aim of the study is to shed light on the interplay between MI incidence and mortality in different SES groups and on the resulting disparities in health expectancies among the German population, i.e. the number of life years free of MI and that affected by mild or severe MI. In addition, gender differences in health expectancies and the related social inequalities are studied. This retrospective cohort study is based on health insurance claims data and investigates social inequalities in MI incidence and mortality with respect to MI severity among the elderly population based on the data of a large German health insurance provider. In order to investigate the length of life free of MI and affected by MI at population level, it is crucial to analyse this interplay between MI incidence and mortality without and after MI: While higher MI incidence rates may increase the expected number of life years after MI in individuals with low SES, increased mortality risks may counterbalance this tendency by shortening the length of survival after MI. In contrast, individuals with high SES may have lower incidence rates, but also lower mortality after MI. Since the extent of the impairments in daily life depends on MI severity, this study focuses in particular on inequalities in MI severity and in the length of life affected by mild or severe MI.

In more detail, the study addresses the following research questions:

Between 2008 and 2017, a total of 1,253,083 insured individuals aged 50 years and above were included in the analyses. Among the study population, a total of 62,676 MI incident cases were observed (64% mild MIs), 31,615 deaths after MI (48% after mild MI), and 276,582 deaths without MI (Table 1). Since the age range is restricted to the elderly, a major share of the study population is retired, leading to low proportions of individuals with higher incomes. This applies especially to women. Detailed information on the number of events, exposures in person-years, and the number of individuals by gender, income group and type of transition can be found in Table 1.

The study shows that differences between income groups in MI incidence and mortality risks are considerable, leading to substantial inequalities in the expected length of life free of MI and after MI. The analyses also reveal that inequalities vary in level according to the severity of MI and gender. Inequalities in MI incidence tend to be stronger in men than in women and are most pronounced in severe MI incidence. Similar patterns were also found for mortality risks without MI and after mild MI, while income inequalities in the risk of death after severe MI were weaker in men than in women. These different patterns of inequalities between men and women are also reflected in the number of life years free of MI and affected by MI. While clear inequalities were found in life years free of MI for both genders, the gap between the low and higher-income group was more than twice as high for men as for women. Men with low incomes are even doubly disadvantaged, as they can expect not only fewer healthy years but also less years after mild MI due to higher mortality risks. In contrast, no significant inequalities in life years after mild and severe MI were found in women and in men after severe MI. While some studies reported a substantial gap in total life expectancy between income and educational groups among the German population [50‐52], research on social inequalities in health expectancies is still very limited. This holds especially true for studies focusing on MI and other specific diseases. This may be due to the limited availability of data that provide information on SES, incidence and mortality. In order to analyse how many years of total life expectancy can be attributed to years free of MI and years after disease onset, the interplay in inequalities between MI incidence and mortality with and without MI has been studied in detail. While the high MI incidence risks of mild and severe MI among men and women with low incomes increased the expected number of life years after MI, this effect was counterbalanced by the comparably higher level of mortality after MI in the low-income group. This may also explain why the number of life years after severe MI did not differ by income group. For higher- income men, this even leads to spending more life years after mild MI than low-income men. In contrast, inequalities in incidence rates and mortality rates without MI have the same effect on MI-free life years: Both, increased incidence and increased mortality rates without MI, reduce MI-free life years in the low-income group.

Our results are consistent with previous studies that found substantial social inequalities in MI incidence and mortality risk [13‐19, 31‐34]. Moreover, our findings on income inequalities in MI severity support earlier findings reporting more severe MI events in individuals with low SES [17, 19, 21, 40‐44]. Compared to the magnitude of inequalities in incidence and mortality without and after mild MI, the inequalities in death after severe MI were weaker, but still evident. This holds especially true for men. Given the high case fatality after a severe MI, this was to be expected. Sharing risk factors and similar pathogenetic processes, substantial income inequalities in life years free of stroke were also found for Germany in elderly women, and even more pronounced in elderly men [53].

Some studies suggest that the extent of inequalities in cardiovascular risk factors and harmful health behaviour may differ between genders, with inequalities being more pronounced in women than in men [24, 36]. For Germany, research on gender differences in health inequalities in risk factors and health behaviour is still limited but there is some evidence that inequalities in diabetes prevalence tend to be stronger in women than in men [36]. This contrasts with our findings which show greater inequalities in MI incidence and mortality among men than among women and underlines the importance of further research on how inequalities in risk factors for cardiovascular diseases affect inequalities in myocardial infarction incidence.

While inequalities in cardiovascular risk factors and health-related behaviour are often cited as most important factors in social inequalities in MI incidence [20, 23, 24], research is still needed on the underlying factors causing social inequalities in mortality after MI. In this context, there is some evidence that SES is an important predictor of how people can cope with the MI and its consequences, and of how they are able to adapt to the disease. In the context of a healthcare system granting nearly equal access to healthcare services for all individuals, the reasons for these mortality inequalities may lie in disparities in the adherence to long-term secondary and tertiary prevention and to changes in life style (e.g. [13, 32, 40]. Since income determines the financial resources available for daily living [49], it can be assumed that life style factors play an important role and foster inequalities in death after MI. Due to the relatively low mortality after mild MI found in our study (Additional file 1, Fig. S2), the probability of dying from causes other than MI can be assumed to be quite high compared to mortality after severe MI. This assumption may also be supported by the fact that the magnitude of inequalities between the income groups in mortality after mild MI lies between that of death without MI and that after severe MI. However, it remains unclear why inequalities in mortality after severe MI are higher in women than in men. These uncertainties indicate that further research is needed on the underlying mechanisms that cause these differences in mortality after acute myocardial infarction.

The study shows how income inequalities in MI incidence and mortality result in gaps in the length of life unaffected and affected by the consequences of MI. We found that income inequalities were most pronounced in MI-free life years. Due to greater inequalities in MI incidence and mortality, the gap in life years free of MI between income groups is considerably wider among men than among women. Furthermore, our findings suggest that inequalities in the years of life after MI, which are assumed to be spent in impaired health, are limited. Marked inequalities in life years spent after MI were only found for men after mild MI with a higher number of years for men with high incomes while no differences were found in life years after severe MI where impairments are likely to be strongest. Our findings underline the importance of public health efforts aimed at reducing health inequalities in cardiovascular diseases and related risk factors, promoting healthier lifestyles and strengthening secondary and tertiary prevention, with particular emphasis on men and women with high-risk profiles for MI incidence and cardiovascular mortality the most disadvantaged groups. Further research is needed to reveal the underlying factors explaining the SES differences in MI incidence and mortality with respect to MI severity.