Continued decline in the incidence of myocardial infarction beyond the COVID-19 pandemic: a nationwide study of the Swedish population aged 60 and older during 2015–2022

During the early phase of the COVID-19 pandemic, the number of hospital admissions for myocardial infarction declined across the world [1‐18]. Despite the vast number of published studies, it is still unknown whether the observed declines reflect a real decrease in the risk of myocardial infarction or merely the fact that fewer patients reached a hospital. Several studies hypothesized that reduced care seeking among individuals who experienced symptoms of myocardial infarction led to lower numbers of patients presenting in hospitals [15‐18]. Evidence for this hypothesis is, however, mixed. Although delays in the care of myocardial infarction were observed in some settings, several European studies did not find significantly increased delays between symptom onset and first medical contact for patients with myocardial infarction [16, 19, 20]. Moreover, since timely treatment of myocardial infarction is essential, delays in care seeking would likely result in increasing case fatality and in growing proportions of patients dying before reaching the hospital. Lower admission rates for cardiovascular diseases were indeed accompanied by increased case fatality during the first pandemic year in the United States [18] but not in several European countries [4, 8, 14, 21]. However, there are also other hypotheses that could explain the declining number of myocardial infarctions during the early pandemic, such as changes in lifestyle factors, stress levels, and environmental exposures [22]. These hypotheses would result not only in a decline in the number of patients presenting at hospitals but also in an overall decline in incidence rates.

In parallel, COVID-19 infection has been linked to an increase in the risk of cardiovascular diseases, including acute myocardial infarction [23]. This suggests that the initial decline in hospital admissions might eventually turn into increasing incidence rates of myocardial infarction in the long run—especially since COVID-19 affected a large share of the population. Moreover, the cardiovascular damage and exacerbation of cardiovascular disease during infection with the SARS-CoV-2 virus [23, 24] might have affected survival rates for patients with myocardial infarction. Nevertheless, to our knowledge, data reflecting case-fatality throughout the pandemic years have not been presented.

Incidence rates –as well as case fatality– declined continuously during the past decades. Most previous studies did not take this trend into account and directly compared data observed in 2020 to an earlier reference period, usually 2019 [3, 4, 6, 7, 9‐11, 13, 16]. In addition, only a few studies have estimated incidence rates based on accurate measures of person-time at risk, considering the increased mortality due to COVID-19 itself. Previous studies were, moreover, often limited to clinical populations [3, 4]. In contrast to administrative registers, the coverage of clinical databases declined during the pandemic [25, 26].

With this study, we explore the seemingly paradoxical scenario in which the number of myocardial infarctions decreased in the early pandemic, but subsequent research revealed an elevated risk of myocardial infarction associated with COVID-19 infection. One could thus hypothesize that the incidence of myocardial infarction increased during the later phases of the pandemic. Here, we present population-wide trends in age-specific incidence rates, case fatality and the proportion of patients hospitalized between the onset of the COVID-19 pandemic and the end of 2022 in Sweden.

The first three waves of the COVID-19 pandemic in Sweden (green) together with the cumulative number of confirmed COVID-19 cases per 1000 inhabitants (gray) are shown in Fig. 1. Since testing capacities were limited in the early pandemic and recommendations to test all suspected cases in laboratories were effectively stopped by the Swedish government during February 2022, the cumulative proportion of the Swedish population affected by the virus likely exceeds the numbers shown in Fig. 1.

The increased risk of myocardial infarction associated with Covid-19 infection, along with the ideas that monitoring of risk factors has been compromised during the pandemic, and that lockdowns have negatively influenced health behaviours, have led to widespread concern about increasing rates of heart disease following the global pandemic [16, 22‐24, 30]. Our results do not support these concerns. Despite the high spread of COVID-19 across the Swedish population, we found that incidence rates of myocardial infarction continued to decline at least until the end of 2022, thus following the long-term downward trend observed already before 2020. While there is an indication that the declining trend may have halted among the oldest men, observed rates still lie well within predicted intervals. Even in Stockholm County, an area in which COVID-19 was already widespread during March and April 2020, when vaccinations were not yet available and medical staff was still inexperienced in treating the virus [31], we found no evidence for increasing rates of myocardial infarction.

Evaluating changes in the incidence of myocardial infarction is challenging, as rates are shaped by a complex interplay of long-term trends, seasonal fluctuations, and changes in the population at risk. Simple comparisons to earlier years can therefore lead to incorrect conclusions and to an overestimation of differences between the pandemic and prepandemic periods. We fitted expected rates for the years 2020 to 2022 based on the previous years’ trends and seasonal variation also considering changes in the composition of the population at risk. Even in these analyses, we found substantially lower incidence rates; approximately 900 fewer events occurred during the first pandemic wave than expected, a number corresponding to 13% fewer than expected myocardial infarctions during this period.

Competing risk of death from COVID-19 is one proposed mechanism behind the declining number of cardiovascular disease events. Severe COVID-19 infections and cardiovascular diseases share common risk factors [22, 30], and it is hence possible that the number of high-risk individuals depleted faster than the total population at risk, thereby not only reducing the total number of myocardial infarctions but also incidence rates. This hypothesis is, however, challenged by consistently lower incidence rates in areas outside of Stockholm County already in March and April 2020. These areas experienced virtually no deaths from COVID-19 in this early phase of the pandemic, yet introduced recommendations for older individuals to stay at home [31]. Furthermore, we analyzed changes in the composition of the population at risk with respect to age, sex, comorbidity, and care status and found no substantial changes during the pandemic.

The etiological mechanisms behind the notable decline in myocardial infarction in the early stage of a global pandemic are intriguing and remain to be studied further. Altered stress levels, lifestyle, and environmental factors, such as reduced air pollution during lockdown, may have contributed to lowering the risk of acute myocardial infarction [22]. While many of these factors operate through long-term accumulation of risk, factors that trigger myocardial infarctions in the short term, such as stress or air pollution, may contribute as well [32, 33]. Research has shown that air pollution can indeed affect the risk of myocardial infarction within weeks, days and even hours of exposure to pollutants [34‐36]. Even despite the comparatively lenient restrictions during the pandemic, Swedish air pollution levels decreased substantially. WHO reported a roughly 30% lower mean annual concentration of NO₂ fine particles and 18% lower concentrations of PM₁₀ and PM_2.5 particles during 2020 compared to 2018–2019 in Stockholm [37].

The absence of higher fatality and of higher proportions of patients dying before receiving care is noteworthy. Clinical processes and staff have been challenged during the pandemic; surgeries have been postponed, and waiting times for patients with many diseases have increased [38]. Indeed, delays in the care pathways of cardiovascular conditions as well as poorer treatment outcomes have been observed in some studies in low- and middle-income countries [16]. For the Swedish setting, the clinical register Swedeheart reported that the time to treatment of acute myocardial infarction had not been prolonged during the pandemic [26]. Reporting to this register is not mandatory and has declined during the pandemic [25, 26], but our study based on nationwide administrative data supports the conclusion that increased pressure on the Swedish health care system has not led to poorer outcomes for patients presenting with acute myocardial infarction.

Our study has several strengths. We use nationwide administrative data on the entire Swedish population, which allow us to derive precise estimates of person-time at risk and incident myocardial infarction. While reporting to clinical registers is prone to be disrupted once clinical processes are challenged and staff shortages occur, reporting to administrative registers is mandatory and has a high priority because it is directly linked to the reimbursement of health care costs. Sensitivity and positive predictive values for myocardial infarction in Swedish inpatient data have been shown to be excellent [28, 29], Specific ICD codes are available to encode a history of myocardial infarction, limiting the probability of misclassifying historical events as incident events. Nevertheless, we cannot rule out some misclassification. Our data did not allow us to identify myocardial infarctions for which patients did not seek any care, and it is further possible that causes of death are misclassified in some instances. However, this would only induce bias if misclassification changed systematically over time. Although one could argue that the accuracy of cause of death assignment has decreased under the pressure of the pandemic, medical scrutiny may have also been promoted by efforts to determine the presence of COVID-19 infection in deceased individuals. Either way, we obtained similar results when excluding data from death records, indicating that misclassification of cause of death cannot explain the pronounced declines in myocardial infarction incidence in Sweden. Finally, it should be noted that our study is limited to ages 60 and above but Swedish authorities reported that 14% of all myocardial infarctions occured in ages below 60 years as of 2022. Younger ages might have adopted different lifestyles than older people during the pandemic, and it is not certain that our findings can be generalized to the younger population.

The incidence of myocardial infarction among individuals aged 60 + in Sweden continued to decrease between 2020 and 2022, despite concerns about an increased incidence of cardiovascular diseases during the COVID-19 pandemic. During the first wave of the pandemic, there was an additional decline in incidence rates. These declines were neither accompanied by increasing case fatality nor by lower shares of patients being hospitalized. Our findings support the conclusion that increased pressure on the Swedish health care system has not led to increased risks or poorer outcomes for patients presenting with acute myocardial infarction. Our work also suggests that the effect of COVID-19 on myocardial infarction risk is not substantial, as we found no increase in incidence at the population level, despite the large share of the population that has been exposed to COVID-19.