Trends in cardiovascular mortality in the United States from 1968 to 2019: analysis of the CDC WONDER database

Abstract

Cardiovascular disease (CVD) is the leading cause of mortality in the USA.1 In the past several years, substantial gains have been made in the prevention and management of CVD.2 Moreover, better control of most CV risk factors has increased life expectancy and reduced age-adjusted mortality rates (AAMR).1 Quantifying the trends in CV mortality over the last five decades years is important to appreciate population-level changes in the prevalence of risk factors, advances in CVD management and to identify groups where these advances have lagged. Here, we describe approximately 50-year trends in age-adjusted CV mortality in the USA. Deaths occurring in the USA from 1968 to 2021 with CVD as an underlying cause were extracted from the Centers for Disease Control and Prevention Wide-Ranging OnLine Data for Epidemiologic Research (CDC WONDER) database.3 Cardiovascular disease deaths were identified in patients ≥25 years of age with International Classification of Diseases-8-clinical modification (ICD-8-CM) codes 390–458 for the years 1968–78, ICD-9-CM codes 390–459 for the years 1979–98, and ICD-10-CM codes I00-I99 after 1999. We used this age cut-off as age-standardization using 18 years as a cut-off was not available. Age-adjusted mortality rates per 100 000 people with 95% confidence intervals (CIs) were reported. The 2000 US standard population was used for age-standardization.4 The Joinpoint Regression Program (Joinpoint V 4.9.0.0 available from the National Cancer Institute) was used to determine trends in mortality from 1968 to 2019. Annual percentage change (APC) with 95% CIs was calculated using the Monte Carlo permutation test among intervals identified by the Joinpoint regression. The weighted averages of the APCs were reported as average APCs (AAPCs) and 95% CIs as a summary of the reported mortality trend across the study period. Additionally, CV AAMRs (only overall) were reported separately for 2020 and 2021 (not included in joinpoint regression analysis). Between 1968 and 2019, there were 48 060 872 deaths due to CV causes. Overall, the AAMR decreased from 1172.5 (1170.1 to 1174.9) in 1968 to 330.8 (330.1 to 331.5) in 2019 with an AAPC of −2.4% (−2.7 to −2.2). The APC in CVD mortality during 2001–10 and 2010–19 was −4.1% (−4.5 to −3.7) and −0.8% (−1.1 to −0.4), respectively (Figure 1A). The AAMR for CVD was 346.0 (345.2 to 346.7) in 2020 and 359.6 (358.8 to 360.3) in 2021. Age standardized cardiovascular mortality rates for the overall population, males, and females (A), and Blacks or AA and Whites (B). Cause-specific cardiovascular mortality for ischaemic heart disease (IHD), acute myocardial infarction (included in IHD), and cerebrovascular diseases are given in (C). The dashed vertical lines represent transition in ICD codes. Between 1968 and 2019, there were in 23 949 606 and 24 111 266 deaths due to CV causes in males and females, respectively. Among males, the AAMRs were higher compared to females and decreased from 1436.9 (1432.8 to 1441.1) in 1968 to 401.6 (400.4 to 402.8) in 2019. Among females, the AAMR decreased from 964.1 (961.2 to 967.0) in 1968 to 271.8 (271.0 to 272.7) in 2019. During the study period, the AAMRs reduced among males and females (AAPC −2.4% [−2.7 to −2.2]) and −2.4% [−2.7 to −2.1], respectively) (Figure 1A). Between 1968 and 2019, there were 42 171 613 and 5 184 681 deaths due to CV causes in White and Black or African Americans (AA) populations, respectively. Among Black or AA individuals, the AAMRs were higher compared to White individuals and decreased from 1306.9 (1298.2 to 1315.6) in 1968 to 426.8 (424.3 to 429.4) in 2019. Among White individuals, the AAMRs decreased from 1161.2 (1158.7 to 1163.7) in 1968 to 326.5 (325.7 to 327.2) in 2019. The CVD mortality decreased among both Black or AA (AAPC −2.2% [−2.3% to −2.0%]) and White individuals (−2.4% [−2.7% to −2.2%]) during the study period (Figure 1B). The AAMR for ischaemic heart disease (IHD) and acute myocardial infarction (AMI) (subgroup of IHD) decreased between 1968 and 2019. The AAMR for cerebrovascular mortality initially declined from 1968 to 2010, and then reached a plateau (APC −0.2% [−0.7 to 0.3]). There was a trend towards higher AAMR for cerebrovascular disease than AMI after 2006 (Figure 1C). In this review of national death certificate data from the USA, we report that the AAMR from CVD have significantly declined from 1968 to 2019, among both sexes and Black or AA and White individuals. Throughout the study period, Black or AA individuals and males demonstrated a higher AAMR than White individuals and females, respectively. This decline in AAMR from CVD appears to have slowed in the last decade prior to COVID-19 pandemic. The AAMR in 2019, 2020, and 2021 was 330.8, 346.0, and 359.6 per 100 000 population, respectively. This increase is likely related to the COVID-19 pandemic and the associated increase in overall age-adjusted CVD mortality. The higher AAMR among Black individuals is well known. Our data suggest that death due cerebrovascular diseases is now more common that AMI. Our results are similar to a recent report from the same database.5 We extend the findings of this study to 1968, to compare contemporary rate of decline with more historic data. While it is encouraging to note the significant decline in AAMR, decrease in the pace of improvement in the last decade is of concern. If true, this is likely due to an increase in the prevalence and possible earlier onset of risk factors such as hypertension, obesity, and diabetes mellitus.6–8 National trends also suggest an increasing incidence of diabetes mellitus and obesity from a young age.9,10 Additionally, there is a high prevalence of physical inactivity and sedentary lifestyle in the USA, an established risk factor for CV mortality.11 The benefits of CV risk reduction with better control of risk factors such as smoking and therapies such as statins may be marred by an increase in these cardiometabolic risk factors.12 Underutilization of guideline-directed medical therapy has been shown in multiple previous studies,13,14 thus employing interventions in this area can help further reduce cardiovascular mortality. Our analyses have limitations. The use of ICD codes and death certificates may lead to misclassification of the cause of death. There is sufficient evidence for the independent association of social determinants of health on CV outcomes,15 but we were not able to adjust for these factors. The improvements in CVD mortality rate seen since 1968 have decelerated after 2010. There is a persistently higher rate among Black individuals and males. AMKM contributed to the conception of the work, acquisition, analysis, interpretation of data, drafted portions of the manuscript and critically revised the manuscript. KG drafted and critically revised the manuscript. SSV provided supervision and critically revised the manuscript. VJ, TSK, ATM, MDS, LAA, and VJ provided critical review. All gave final approval and agree to be accountable for all aspects of work ensuring integrity and accuracy. Trends in age-adjusted cardiovascular mortality in the USA over five decades. The improvement in CVD mortality rate seen since 1968 has decelerated after 2010. Institutional Review Board approval was not sought due to use of publicly available, anonymized data. This research did not receive any funding. The data underlying this article are available at https://wonder.cdc.gov/mortSQL.html.

Topics & Concepts