Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2026 Jul 27;9(7):e2625410. doi: 10.1001/jamanetworkopen.2026.25410

Cardiovascular Risk Factor Prevalence and Control by Education Strata in Older Adults in 3 Countries

Ralph Lawton 1, David M Cutler 2,3,✉
PMCID: PMC13409021  PMID: 42507441

Key Points

Question

How do cardiovascular risk factor prevalence and control differ over time and by education strata among adults aged 50 years or older in the US, England, and South Korea?

Findings

In this secondary analysis of a serial cross-sectional study of 120 678 adults, the US had the highest prevalence of high cholesterol level, hypertension, and diabetes. Conditional on risk, risk factor awareness, treatment, and control were generally similar or better in the US, and control among those at risk was generally comparable by education strata.

Meaning

This study’s results suggest that differences in risk accumulation, rather than risk factor treatment, primarily contribute to international and educational disparities in cardiovascular outcomes.

This secondary analysis of a cross-sectional study of older adults examines cardiovascular disease risk factor trends in the US, England, and South Korea.

Abstract

Importance

Improvements in cardiovascular disease (CVD) mortality have plateaued in the US and have increased among adults with fewer years of education, a pattern not seen in peer countries. Differences in health system performance managing CVD risk factors may be important in this disparity.

Objective

To compare cardiovascular disease risk factor prevalence and management in the US, England, and South Korea over time and by education strata.

Design, Setting, and Participants

This secondary analysis of a serial cross-sectional study used nationally representative data from adults 50 years or older from 3 countries during years ranging from 1998 to 2020. Data were analyzed from May 2025 to April 2026.

Main Outcomes and Measures

For each risk factor (cholesterol, blood pressure, and diabetes), the share at risk, the care cascade, and the share with uncontrolled risk were measured.

Results

This study assessed data from 120 678 adults (mean [SD] age, 63.5 [9.9] years; No. [weighted %], 54 167 [46%] male and 66 511 [54%] female). For each risk factor, the US (n = 27 327) had higher shares at risk than England (n = 41 986) or South Korea (n = 51 365). For example, England had a smaller proportion at risk: 52% in period 4 (mean [SE], 11.1 [1.6] percentage points [pp] lower than in the US). Cholesterol risk in South Korea increased from 46% in 2005 to 2008 to 51% in 2015 to 2020 (P for trend < .001), with levels ultimately similar to England but 12.1 (1.1) pp below the US. US patients at risk were more likely to be aware, treated, and have controlled risk using medications, resulting in more similar uncontrolled risk across countries. Because of higher control of condition risks in the US, the rates of uncontrolled cholesterol risk in the US and England were both 29% (mean [SE] difference, <0.1 [1.5] pp), and the rate uncontrolled in South Korea was 2.2 (1.0) pp higher than the US despite higher at-risk rates. South Korea achieved rapid gains in treatment and control, especially hypertension (6% controlled in 1998-2001 to 42% in 2005-2008, P for trend < .001) and cholesterol (6% controlled in 2005-2008 to 31% in 2015-2020, P for trend < .001). Across countries, risk was higher among adults without a college degree, but control rates among those at risk were similar by education strata. Diabetes risk increased substantially over time (15% to 22% in the US, 9% to 10% in England, and 13% to 19% in South Korea, P for trend < .001), whereas cholesterol and hypertension risk remained stable or decreased.

Conclusions and Relevance

In this repeated cross-sectional study of adults aged 50 years and older in the US, England, and South Korea, despite higher baseline risk, the US health system achieved comparable or better performance than England and South Korea in managing CVD risk factors. Differences in risk factor treatment did not explain disparities by country or education strata, suggesting that upstream risk accumulation, rather than care delivery, may be associated with increasing disparities in CVD risk.

Introduction

In recent years, life expectancy in the US has decreased substantially relative to other wealthy countries.1,2,3,4 This is partly due to the stagnation in heart disease mortality in the US compared with peer nations.4,5,6 Between 2010 and 2020, for example, age- and sex-adjusted heart disease mortality decreased 4% in the US compared with 21% in the broader Organisation for Economic Co-operation and Development countries.7,8,9,10 Decreases in heart disease mortality have particularly slowed among those with fewer years of education,11,12 for whom heart disease mortality increased between 2010 and 2021.5

Researchers have proposed several theories for why cardiovascular mortality trends have differed across countries and education groups. The literature2,13,14 suggests that health system performance in the US lags behind peer countries; a recent National Academies of Sciences, Engineering, and Medicine report2 concludes that medical care may play a role in increasing midlife mortality rates and socioeconomic status disparities. Prior work15,16,17,18 has examined some parts of cardiovascular disease (CVD) care, and substantial work15,17 examines pieces of the cardiovascular care cascade, often finding them sensitive to policy. However, the difference in risk factor control across countries has not been examined, either overall or for subgroups of the population.

This article compares CVD risk factor trends in the US, England, and South Korea. Heart disease mortality trends differ across the 3 countries. Age-adjusted heart disease mortality decreased by 27% in England and 36% in South Korea from 2010 to 2020 vs only 4% in the US.8 South Korea and England both have sizable mortality disparities by educational level, yet mortality differences between secondary and tertiary educational levels have been stable if not narrowing since 2010.19,20,21 The analysis focuses on the at-risk population share and treatment efficacy for 3 prominent CVD risk factors—high cholesterol level, hypertension, and diabetes—across countries and education strata.

Methods

Data Sources

US data are from the National Health and Nutrition Examination Survey (NHANES, 1999-2020, a biennial cross-sectional survey).22 English data are from the English Longitudinal Study of Aging (ELSA), a panel study representing adults 50 years or older living in England.23 South Korean data are from the Korea National Health and Nutrition Examination Study (KNHANES).24 Because survey years differ, the data are grouped into 4 periods (eTable 1 in Supplement 1). In NHANES, these periods are 1999 to 2002, 2003 to 2008, 2009 to 2014, and 2015 to 2020. In ELSA, these periods are 2004 to 2009, 2012 to 2013, and 2016 to 2019. In KNHANES, these periods are 1998 to 2001, 2005 to 2008, 2009 to 2014, and 2015 to 2020.

Some waves of the ELSA and KNHANES either do not ask all relevant questions or do not measure specific biomarkers. KNHANES did not conduct population-representative glycated hemoglobin (HbA1c) measurements from 2005 to 2010. The 1998 to 2001 KNHANES waves and 2004 to 2005 waves of the ELSA do not ask about cholesterol diagnoses or medication. These waves are excluded from the diabetes and cholesterol analyses, respectively.

The study population was restricted to those 50 years or older who had venous blood measures for cholesterol and HbA1c. No race groups are excluded in analyses. Supplementary analyses use self-reported Black race and Hispanic ethnicity classifications from NHANES. NHANES routinely collects race data to monitor health disparities. Data were analyzed from May 2025 to April 2026. This study used deidentified, publicly available data. Original study teams obtained participant written informed consent. Secondary analysis was determined to not be human participants research by the Harvard University institutional review board. This secondary analysis followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies.

Study Measures

Three cardiovascular disease risk factors were considered: cholesterol level, blood pressure, and HbA1c level. For each, the population proportion at risk is defined as those with a high measured value or a controlled value with self-reported diagnosis.25,26 For cholesterol, at risk is defined as a total cholesterol level greater than or equal to 240 mg/dL or if high-density lipoprotein level is less than 40 mg/dL (to convert to millimoles per liter, multiply by 0.0259). For blood pressure, individuals are at risk if systolic blood pressure is greater than or equal to 140 mm Hg or if diastolic blood pressure is 90 mm Hg or higher. For diabetes, we treat individuals as at risk if their HbA1c level is 6.5% or greater (to convert to proportion of hemoglobin, multiply by 0.01). We focus on consistent risk measures, which have been broadly used across the focal countries in this period.27,28,29,30

Among those at risk, the care cascade is evaluated: the proportion of those at risk who are aware of their risk, the proportion at risk who are receiving medication, and the proportion of those who effectively control risk factors through medication. Control is defined as the opposite of at risk, with the exception that we consider an individual’s blood glucose to be controlled if their HbA1c level is 7.0% or less.31

Statistical Analysis

Age adjustment proceeds in 2 stages, in line with the National Center for Health Statistics approach.32 In the first stage, survey-specific weights are applied, preserving within-cell nonresponse and sampling weights for each survey. For the KNHANES, ELSA, and NHANES waves after 2001 2002, the totals by 5-year age, sex, and college education within each survey year are then calculated. In the second stage, these totals are used to calculate adjustment factors to match the cell-specific population shares in the 2001 to 2002 wave of NHANES, which is a nationally representative sample that, when weighted, reflects the US population at the time. These adjustments are then applied to the survey-specific sample weights, retaining nonresponse and sampling adjustments within each cell but creating a uniform distribution by age, sex, and college education over time and survey. The results for all countries and periods thus match the US population totals in 2001 to 2002.

Within each country × time group, the mean and 95% CI of each risk factor are calculated, using heteroskedasticity-robust SEs.33,34 To test the significance of time trends, ordinary least-squares regressions are estimated with a survey year trend. Survey-weighted, 2-tailed, unpaired t tests were used to compare groups. Cross-country comparisons are reported for the most recent available period for each country. Analyses were conducted in StataMP, version 19.5 (StataCorp), with conventional statistical significance thresholds of 2-sided P < .05.

Results

This study assessed data from 120 678 adults (No. [weighted %] 54 167 [46%] male and 66 511 [54%] female; among adults in the US, No. [weighted %] 5859 [10%] Black, 6177 [9%] Hispanic, 13 119 [75%] White, and 2172 [6%] other [non-Hispanic Asian, multiracial, and other race]), including adults from the US (NHANES, n = 27 327 across all survey waves), England (ELSA, n = 41 986), and South Korea (KNHANES, n = 51 365). All estimates were standardized to the 2001 to 2002 NHANES reference population (mean [SD] age, 63.5 [10.4] years; 46% male; 27% college educated). Detailed sample sizes are in eTable 1 in Supplement 1.

At-Risk Population

Figure 1 shows at-risk prevalences over time by country (eTable 2 in Supplement 1). The US had the highest proportion for all 3 biomarkers in every period. For cholesterol, 61% of individuals in the US in 1999 to 2002 were at risk, increasing to 64% by 2015 to 2020 (P for trend = .01). England had a smaller proportion at risk: 52% in period 4 (mean [SE], 11.1 [1.6] percentage points [pp] lower than in the US; P for trend = .04). Cholesterol risk in South Korea increased from 46% in 2005 to 2008 to 51% in 2015 to 2020 (P for trend < .001), similar to England but 12.1 (1.1) pp below the US.

Figure 1. Bar Graphs Showing Proportion of Adults at Risk of High Cholesterol Levels, Hypertension, and Diabetes by Country.

Three-panel bar charts of adult risk proportions for cholesterol, hypertension, diabetes. Three vertically stacked panels labeled A, B, and C at the upper left of each plot. All panels use pale blue vertical bars with black outlines and black vertical error bars with caps. The vertical axis in all panels is labeled Proportion of adults. Panel A title at top left: Cholesterol. Panel A vertical axis ranges from zero point three to zero point eight with gridlines. Along the horizontal axis are three country groupings labeled below with brackets: U S, England, and Korea. Under U S are four time periods: 1999 to 2002, 2003 to 2008, 2009 to 2014, and 2015 to 2020, with bar heights near zero point six one, zero point six two, zero point six four, and zero point six three. Under England are three periods: 2004 to 2009, 2012 to 2013, and 2016 to 2019, with bar heights near zero point five five, zero point five six, and zero point five two. Under Korea are three periods: 2004 to 2008, 2009 to 2014, and 2015 to 2020, with bar heights near zero point four six, zero point four five, and zero point five one. Panel B title: Hypertension. Vertical axis ranges from zero point three to zero point eight. Under U S are four periods (1999 to 2002, 2003 to 2008, 2009 to 2014, 2015 to 2020) with bar heights near zero point five nine, zero point six one, zero point five nine, and zero point six zero. Under England are three periods (2004 to 2009, 2012 to 2013, 2016 to 2019) with bar heights near zero point five three, zero point five one, and zero point four eight. Under Korea are four periods (1998 to 2001, 2005 to 2008, 2009 to 2014, 2015 to 2020) with bar heights near zero point four eight, zero point four five, zero point four eight, and zero point four seven. Panel C title: Diabetes. Vertical axis ranges from zero to zero point two five. Under U S are four periods (1999 to 2002, 2003 to 2008, 2009 to 2014, 2015 to 2020) with bar heights near zero point one five, zero point one eight, zero point two zero, and zero point two two. Under England are three periods (2004 to 2009, 2012 to 2013, 2016 to 2019) with bar heights near zero point zero nine, zero point one one, and zero point one zero. Under Korea are three periods with labels 1998 to 2001, 2009 to 2014, and 2015 to 2020, with bar heights near zero point one three, zero point one nine, and zero point one nine.

Open in a new tab

Proportion of adults 50 years or older at risk of high cholesterol levels, hypertension, and diabetes by country and period. At risk is defined by biomarker thresholds or self-report of diagnosis or treatment (see Methods). Estimates are survey weighted and reweighted to the 2001 to 2002 US population age, sex, and education strata distribution. Error bars indicate 95% CIs.

Blood pressure risk in the US was relatively stable, between 59% and 62% (P for trend > .99). Risk in England was lower and decreased from 53% in 2004 to 2009 to 48% in 2016 to 2019 (P for trend < .001). Risk in South Korea was relatively constant at approximately 48% (P for trend = .57). Risk in the US remained higher than in South Korea (period 4: mean [SE], 12.8 [1.1] pp) and England (period 4: mean [SE], 12.3 [1.5] pp) for all periods.

Diabetes risk increased in all 3 countries: from 15% to 22% in the US, 9% to 10% in England, and 13% to 19% in South Korea (P for trend < .001). Risk remained highest in the US (period 4: mean [SE], 12.0 [1.0] pp vs England and 3.0 [0.8] vs South Korea).

Higher body mass index (BMI; calculated as weight in kilograms divided by height in meters squared) in the US is consistent with higher at-risk rates. Mean (SE) BMI in the US was 30.0 (0.1) in 2015 to 2020 compared with 28.4 (0.2) in England and 24.2 (0.1) in South Korea. However, diabetes risk was higher in South Korea (19%) than in England (mean [SE], 8.9 [0.7] pp below South Korea), despite lower mean BMI.

Cholesterol Care Cascade

Figure 2 evaluates the cholesterol care cascade. eTable 3 in Supplement 1 shows values. Awareness, treatment, and control increased over time in the US. Awareness of risk increased from 66% to 82% during the study period (P for trend < .001), the proportion of adults receiving medications increased from 30% to 48% (P for trend < .001), and the proportion controlled increased from 19% to 35% (P for trend < .001). With larger improvements in the care cascade than in the at-risk proportion, the share with uncontrolled cholesterol risk decreased from 40% to 29% (P for trend < .001). Holding the population fixed to 2001 to 2002, this would correspond to more than 8 million fewer adults with uncontrolled cholesterol risk.

Figure 2. Bar Graphs Showing Cholesterol Care Cascade by Country.

Three-panel bar charts of cholesterol cascade proportions over time in US, England, Korea. Three vertically stacked panels labeled A, B, and C at the left margin. Each panel contains grouped vertical bar charts with pale blue bars and black error bars. The vertical axis in all panels is labeled Proportion of adults, ranging from zero to one point zero with tick marks at zero point two intervals. Along the horizontal axis, five grouped categories appear in this order: At risk, Aware, Medication, Controlled, and Uncontrolled. Under each category, time-period labels appear as stacked year ranges. A vertical dotted blue line near the right side separates the first four categories from the Uncontrolled category. Panel A title at upper left reads US. Time ranges under each category: 1999 to 2002, 2003 to 2008, 2009 to 2014, 2015 to 2020. At risk bars reach one point zero for all four periods. Aware bars rise from about zero point six six to about zero point eight two across periods. Medication bars are about zero point three zero, zero point four five, zero point five zero, and zero point four eight. Controlled bars are about zero point one nine, zero point three two, zero point three six, and zero point three five. Uncontrolled bars are about zero point four zero, zero point three three, zero point three one, and zero point two nine. Panel B title reads England. Time ranges: 2008 to 2009, 2012 to 2013, 2016 to 2019. At risk bars reach one point zero for all three periods. Aware bars are about zero point five three, zero point five nine, and zero point six six. Medication bars are about zero point three three, zero point three five, and zero point three six. Controlled bars are about zero point two five, zero point two seven, and zero point two eight. Uncontrolled bars are about zero point four two, zero point three five, and zero point two nine. Panel C title reads Korea. Time ranges: 2005 to 2008, 2009 to 2014, 2015 to 2020. At risk bars reach one point zero for all three periods. Aware bars are about zero point two three, zero point four zero, and zero point five six. Medication bars are about zero point one zero, zero point two three, and zero point three nine. Controlled bars are about zero point zero six, zero point one eight, and zero point three one. Uncontrolled bars are about zero point four one, zero point three four, and zero point three one.

Open in a new tab

Cholesterol care cascades for each country over time are plotted. Bars reflect the proportion of adults 50 years or older at risk of high cholesterol levels who meet each stage of the care cascade, with 95% CIs (error bars). The far right shows the proportion of the population with uncontrolled high risk. Estimates use survey weights and are reweighted to match the age, sex, and education strata distribution of the 2001 to 2002 US population. Vertical dashed line separates estimates of the care cascade conducted within the at-risk population from estimates of uncontrolled risk conducted in the overall population.

In England, awareness also increased, from 53% to 66% (P for trend < .001), with little change in medication use or control. Uncontrolled cholesterol risk decreased from 42% to 29%, associated primarily with decreasing at-risk prevalence (P for trend < .001).

South Korea showed significant improvement across all domains between 2005 to 2008 and 2015 to 2020. Awareness increased from 23% to 56%, medication use from 9% to 38%, and the controlled fraction from 6% to 31%. Despite increasing at-risk prevalence, uncontrolled cholesterol risk decreased from 41% to 31% (P for trend < .001 for all). Because of higher control in the US, the rates of uncontrolled cholesterol risk in the US and England were both 29% (mean [SE] difference, <0.1 [1.5] pp), and the rate in South Korea was 2.2 (1.0) pp higher than the US despite higher at-risk rates.

Hypertension Care Cascade

The US hypertension cascade has improved alongside cholesterol (Figure 3; values in eTable 4 in Supplement 1). Awareness, treatment, and control all increased over time. With risk relatively constant and treatment improving, the population share with uncontrolled blood pressure decreased from 38% to 30% (P for trend < .001 for all), corresponding to a decrease of 5 million adults in the 2001 to 2002 population.

Figure 3. Bar Graphs Showing Hypertension Care Cascade by Country.

Three-panel bar charts of hypertension cascade proportions over time in US, England, Korea. Three vertically stacked panels labeled A, B, and C at the upper left of each panel. All panels use the same vertical axis at left labeled Proportion of adults, ranging from 0 to 1 point 0 with tick marks at 0 point 2 intervals. Bars are pale blue with black outlines and black 95 percent confidence interval error bars. Panel A title at upper left: US. Along the horizontal axis, five grouped categories appear left to right: At risk, Aware, Medication, Controlled, Uncontrolled. Under each of the first four categories are four time-period labels: 1999 to 2002, 2003 to 2008, 2009 to 2014, 2015 to 2020. At risk bars are all at 1 point 0. Aware bars are approximately 0 point 74, 0 point 80, 0 point 85, 0 point 82. Medication bars are approximately 0 point 65, 0 point 72, 0 point 79, 0 point 78. Controlled bars are approximately 0 point 33, 0 point 41, 0 point 52, 0 point 48. A vertical blue dotted line appears just before the Uncontrolled group. Uncontrolled bars are approximately 0 point 38, 0 point 34, 0 point 27, 0 point 30. Panel B title: England. Categories are the same, with three time periods under the first four categories: 2004 to 2009, 2012 to 2013, 2016 to 2019. At risk bars are at 1 point 0. Aware bars are about 0 point 71, 0 point 73, 0 point 75. Medication bars are about 0 point 51, 0 point 57, 0 point 56. Controlled bars are about 0 point 28, 0 point 35, 0 point 36. A vertical blue dotted line precedes Uncontrolled. Uncontrolled bars are about 0 point 34, 0 point 30, 0 point 27. Panel C title: Korea. Same categories as panel A with four time periods: 1998 to 2001, 2005 to 2008, 2009 to 2014, 2015 to 2020. At risk bars are at 1 point 0. Aware bars are about 0 point 37, 0 point 74, 0 point 73, 0 point 79. Medication bars are about 0 point 18, 0 point 64, 0 point 66, 0 point 72. Controlled bars are about 0 point 06, 0 point 42, 0 point 44, 0 point 53. A vertical blue dotted line precedes Uncontrolled. Uncontrolled bars are about 0 point 43, 0 point 25, 0 point 26, 0 point 21.

Open in a new tab

Hypertension care cascades for each country over time are plotted. Bars reflect the proportion of adults 50 years or older at risk of hypertension who meet each stage of the care cascade, with 95% CIs (error bars). The far right shows the proportion of the population with uncontrolled high risk. Estimates use survey weights and are reweighted to match the age, sex, and education strata distribution of the 2001-2002 US population. Vertical dashed line separates estimates of the care cascade conducted within the at-risk population from estimates of uncontrolled risk conducted in the overall population.

In England, controlled hypertension increased from 28% to 35% among those at risk. This increase reduced the uncontrolled population risk from 33% to 27% (P for trend < .001).

In South Korea, controlled hypertension improved rapidly. Control increased from 6% in 1998 to 2001 to 42% in 2005 to 2008 and continued increasing (P for trend = .003), eventually exceeding the US in 2015 to 2020 (mean [SE], 4.0 [1.3] pp higher). Uncontrolled hypertension decreased from 43% to 21% (P for trend < .001).

Blood pressure control in the US exceeded that in England and South Korea, substantially narrowing at-risk differences. Although the rate of uncontrolled blood pressure remained a mean (SE) of 3.4 (1.2) pp higher in the US than England and 9.0 (0.9) pp above South Korea, differences were much smaller than the at-risk rate differences of 12.0 (1.5) pp above England and 13.0 (1.1) pp above South Korea.

Diabetes Care Cascade

Figure 4 shows diabetes care cascades across countries, and eTable 5 in Supplement 1 shows values. Diabetes treatment improved in the US but not as much as for cholesterol and blood pressure. Diabetes risk awareness in the US improved from 80% to 86% (P for trend = .003), and medication use increased from 67% to 75% (P for trend = .001). Control increased from 30% to 39% between 1999 to 2002 and 2003 to 2008 and was between 37% and 40% afterwards (P for trend = .17). However, because of large risk increases, the population share with uncontrolled diabetes risk increased from 7% to 12% (P for trend < .001), an increase of approximately 4 million adults relative to the 2001 to 2002 population.

Figure 4. Bar Graphs Showing Diabetes Care Cascade by Country.

Three-panel bar charts of cascade proportions over time for US, England, and Korea. Three vertically stacked panels labeled A, B, and C at the upper left of each panel. Each panel contains grouped vertical bars with thin black outlines and pale blue fill, each bar topped with a black capped error bar. The vertical axis at left in all panels reads Proportion of adults, ranging from zero to one point zero with tick marks at zero point two intervals. In panel A, title text US appears near the upper left. Along the horizontal axis, five grouped categories appear left to right: At risk, Aware, Medication, Controlled, and Uncontrolled. Under each category are four time-period labels: 1999 to 2002, 2003 to 2008, 2009 to 2014, and 2015 to 2020. At risk bars are at one point zero for all four periods. Aware bars are approximately zero point eight, zero point eight, zero point eight three, and zero point eight five. Medication bars are approximately zero point six six, zero point seven zero, zero point seven four, and zero point seven five. Controlled bars are approximately zero point three zero, zero point three nine, zero point four zero, and zero point three seven. A vertical dotted blue line appears between Controlled and Uncontrolled. Uncontrolled bars are approximately zero point zero seven, zero point one one, zero point one two, and zero point one three. In panel B, title text England appears near the upper left. Categories and ordering match panel A, with three time-period labels under each category: 2004 to 2009, 2012 to 2013, and 2016 to 2019. At risk bars are at one point zero for all three periods. Aware bars are approximately zero point seven zero, zero point seven three, and zero point eight six. Medication bars are approximately zero point five eight, zero point six one, and zero point six four. Controlled bars are approximately zero point two seven, zero point two five, and zero point two eight. The same vertical dotted blue line separates Controlled from Uncontrolled. Uncontrolled bars are approximately zero point zero five, zero point zero five, and zero point zero six. In panel C, title text Korea appears near the upper left. Categories match panel A, with three time-period labels under each category: 1998 to 2001, 2011 to 2014, and 2015 to 2020. At risk bars are at one point zero for all three periods. Aware bars are approximately zero point six three, zero point six seven, and zero point seven four. Medication bars are approximately zero point five nine, zero point six one, and zero point seven zero. Controlled bars are approximately zero point three five, zero point three zero, and zero point four zero. The vertical dotted blue line again separates Controlled from Uncontrolled. Uncontrolled bars are approximately zero point zero eight, zero point one one, and zero point one two.

Open in a new tab

Diabetes care cascades for each country over time are plotted. Bars reflect the proportion of adults 50 years or older at risk of diabetes who meet each stage of the care cascade, with 95% CIs (error bars). The far right shows the proportion of the population with uncontrolled high risk. Estimates use survey weights and are reweighted to match the age, sex, and education strata distribution of the 2001 to 2002 US population. Vertical dashed line separates estimates of the care cascade conducted within the at-risk population from estimates of uncontrolled risk conducted in the overall population.

The findings in England were quite different. Awareness improved from 71% to 86% (P for trend < .001), but treatment and control changed little. Given smaller at-risk increases than in the US, rates of uncontrolled diabetes increased slightly from 5% to 6% (P for trend = .01). In South Korea, awareness (mean [SE], +11.3 [3.1] pp) and treatment (mean [SE], +11.3 [3.3] pp) improved, but uncontrolled diabetes risk increased from 7% to 11%, paralleling increases in the at-risk proportion (P for trend = .02). Overall, medical control of diabetes was greater in the US than in England (mean [SE], 9.8 [2.9] pp better) but was similar in South Korea (mean [SE], 2.2 [2.1] pp worse).

Differences by Race in the US

Supplementary analyses by race showed broadly similar cross-country patterns regardless of comparator group (Figure 5; eTable 6 in Supplement 1). Black and Hispanic US residents had higher diabetes risk than White US residents, with somewhat lower control rates.

Figure 5. Bar Graphs Showing At-Risk and Controlled Proportions of Adults by Education Strata.

Six-panel bar charts of risk and uncontrolled risk for cholesterol, hypertension, diabetes. Three rows by two columns of bar charts labeled A, B, and C at the left margin. Each row has a left plot with vertical axis label Proportion of adults at risk and a right plot with vertical axis label Proportion of adults with uncontrolled risk. All plots use vertical bars with black error bars and capped whiskers. Panel A title Cholesterol at upper left. Panel A left plot vertical axis from zero to zero point seven five; horizontal categories, left to right: for U S in dark teal, No college, College, White, Black, Hispanic; for England in orange, No college, College; for Korea in light blue, No college, College. A legend near the top center labels U S, England, Korea with matching color swatches. Approximate at risk proportions: U S No college about zero point six four, U S College about zero point six two, U S White about zero point six four, U S Black about zero point five six, U S Hispanic about zero point six three; England No college about zero point five four and College about zero point four six; Korea No college about zero point five two and College about zero point five. Panel A right plot vertical axis from zero to zero point six; approximate uncontrolled proportions: U S No college about zero point three six, U S College about zero point three one, U S White about zero point three four, U S Black about zero point three eight, U S Hispanic about zero point two nine; England No college about zero point three and College about zero point one seven; Korea No college about zero point three one and College about zero point two nine. Panel B title Hypertension. Panel B left plot vertical axis from zero to zero point seven five; approximate at risk proportions: U S No college about zero point six six, U S College about zero point four six, U S White about zero point five eight, U S Black about zero point seven five, U S Hispanic about zero point six; England No college about zero point five and College about zero point four; Korea No college about zero point five and College about zero point four. Panel B right plot vertical axis from zero to zero point six; approximate uncontrolled proportions: U S No college about zero point four eight, U S College about zero point four eight, U S White about zero point five one, U S Black about zero point four, U S Hispanic about zero point four one; England No college about zero point three eight and College about zero point two eight; Korea No college about zero point five two and College about zero point four nine. Panel C title Diabetes. Panel C left plot vertical axis from zero to zero point four five; approximate at risk proportions: U S No college about zero point two four, U S College about zero point one six, U S White about zero point one nine, U S Black about zero point two nine, U S Hispanic about zero point three; England No college about zero point one two and College about zero point zero seven; Korea No college about zero point two one and College about zero point one five. Panel C right plot vertical axis from zero to zero point six; approximate uncontrolled proportions: U S No college about zero point three six, U S College about zero point three eight, U S White about zero point four, U S Black about zero point three five, U S Hispanic about zero point two eight; England No college about zero point three and College about zero point one six; Korea No college about zero point three nine and College about zero point four.

Open in a new tab

Bar graphs showing proportions of adults 50 years or older at risk for high cholesterol level, hypertension, and diabetes (left) and proportion of those at risk with levels medically controlled (right) by college education strata for latest periods available in each country (2015-2020 for US, 2016-2019 for England, and 2015-2020 for South Korea). Estimates use survey weights and are reweighted to match the age, sex, and education strata distribution of the 2001 to 2002 US population. Error bars indicate 95% CIs.

Risk and Control by Education Strata

Figure 5 and eTable 7 in Supplement 1 show the at-risk proportions and the controlled risk factors by college degree in the most recent period. In all 3 countries and for virtually all measures, at-risk rates were higher for people without a college degree than those with one. The gaps are smallest for cholesterol and largest for blood pressure and diabetes. For example, the mean (SE) difference in at-risk rate for high cholesterol in the US in 2015 to 2020 is 0.5 (2.3) pp, whereas the mean (SE) difference is 19.0 (2.3) pp for hypertension and 7.9 (1.7) pp for diabetes. In contrast, there was no clear pattern of better control by college degree status. In the US, cholesterol control rates were 36% for those without a college and 31% for those with (mean [SE] difference, 4.4 [2.7] pp).

Discussion

Understanding why cardiovascular mortality trends have diverged across countries and education strata is essential to explaining the stagnation and reversal of midlife mortality improvements in the US. Lagging health system performance in the US and socioeconomic gradients in performance are often hypothesized as potential explanations, although few studies2,4,35,36,37,38,39 directly compare health system performance across settings and strata. This study evaluates the performance of health systems and underlying health risk internationally and by education strata for 3 important CVD risk factors, drawing 4 key conclusions.

First, population risk levels in the US are much higher than in comparison countries for all 3 risk factors. Rates of cholesterol and hypertension risk are nearly 10 pp higher in the US than in England or South Korea. Diabetes rates are 12.0 and 3.0 pp over English and South Korean rates, respectively.

Second, risk factor control rates are higher in the US than in comparison countries. Control rates in the US are higher than in England, although South Korea has largely caught up with US control. The net effect is that cross-country differences in uncontrolled risk are much smaller than at-risk differences and often not statistically distinguishable across countries.

Third, although risk rates in all countries are higher for those with less education, control rates by education strata tend to be similar. This pattern is consistent across countries despite the fact that a college degree may capture different segments of the socioeconomic distribution in each setting. In the US, there are no significant differences in risk factor control by education strata.

Fourth, diabetes risk has increased more substantially over time than hypertension and cholesterol risks, particularly in the US and Korea. Although hypertension and cholesterol risk have null to slight increasing trends over time, diabetes rates increased from 15 to 22pp in the US and from 13 to 19pp in South Korea. A common criticism of the US health system is that spending is higher even as health outcomes are worse.2,25,35,40,41,42 This study shows that the US health system appears to be, in relative terms, quite capable of identifying and treating CVD risk factors.

Similar concerns about poor health system performance have been suggested to contribute to health disparities by socioeconomic status.5,43,44 However, findings of similar system performance in risk factor control by education strata across countries suggest the health system’s ability to identify and treat risk is not driving education disparities. Prior work45 has found differences in medication choice or intensity by education strata, but effective control appears to be similar in practice. For both international and education differences, disparities in the at-risk proportion, rather than in treatment conditional on risk, are much more important.

Why does risk vary across countries and education groups? Increasing obesity prevalence is the most important underlying factor influencing blood pressure, cholesterol, and diabetes.46,47,48 However, attributing differences to obesity leaves several facts unexplained. The first is South Korea’s high risks for all 3 risk factors, especially diabetes, despite lower BMI. Disproportionate BMI-conditional risk has been observed in other Asian populations.49,50 In addition, increases in obesity in the US and England have not led to large increases in cholesterol or hypertension risk; only diabetes rates have increased markedly.

A variety of factors could affect the link between BMI and CVD risk. Reduced inflammation across cohorts with lower lifetime infectious disease exposure may contribute to decreasing cholesterol and blood pressure risk.51,52,53 Factors such as air pollution may also shape CVD risks, especially for hypertension.54 Differing diets may also play a role, especially for differences across countries.49,50

We did not use mortality prediction models and therefore cannot formally decompose differential trends in heart disease mortality by country and education group. However, with the exception of hypertension, our results do not show consistently adverse rates of uncontrolled risk factors in the US relative to other countries.

Limitations

This study has several limitations. First, sample sizes within country, period, and education subgroups are smaller, potentially limiting power to detect small differences across groups. Second, we cannot fully capture differences in socioeconomic status across countries. Education strata is highly informative but may reflect different aspects of the SES distribution in different settings, and we avoid examining income gradients due to issues of comparability across countries and reverse-causality concerns with respect to health. Third, measurement tools are not fully harmonized across countries, and some waves omit key components. Fourth, laboratory measures are based on gold standard venous assays, but there may be variation across measuring laboratories. Fifth, although our metrics document aggregate performance in managing population health, we are not able to capture the impact of specific features of each health system. Extending this analysis to other high-income countries with universal coverage, such as Canada and Sweden, would provide additional comparative context. Sixth, we do not investigate potential trends among individuals younger than 50 years.

Conclusions

In this cross-sectional study of the US, South Korea, and England over time, the US health care system was more effective at identifying and treating CVD risk factors than England and South Korea. Thus, primary care access and treatment delivery for CVD risk factors and education strata were not associated with adverse CVD mortality trends in the US compared with these other countries. Understanding changes in the at-risk population, especially its association with obesity, deserves more attention.

Supplement 1.

eTable 1. Sample sizes and data notes

eTable 2. At risk fractions by country over time

eTable 3. Cholesterol cascade by country over time

eTable 4. Hypertension cascade by country over time

eTable 5. Diabetes cascade by country over time

eTable 6. At risk and controlled fractions by race and ethnicity, over time

eTable 7. At risk and controlled fractions by country and education, most recent period

jamanetwopen-e2625410-s001.pdf (635.7KB, pdf)
Supplement 2.

Data Sharing Statement

jamanetwopen-e2625410-s002.pdf (15.8KB, pdf)

References

  • 1.Case A, Deaton A. Mortality and morbidity in the 21st century. Brookings Pap Econ Act. 2017;2017:397-476. doi: 10.1353/eca.2017.0005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Harris KM, Majmundar MK, Becker T, eds. High and Rising Mortality Rates Among Working-Age Adults. National Academies Press; 2021. [PubMed] [Google Scholar]
  • 3.Ho JY, Hendi AS. Recent trends in life expectancy across high income countries: retrospective observational study. BMJ. 2018;362:k2562. doi: 10.1136/bmj.k2562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Mehta NK, Abrams LR, Myrskylä M. US life expectancy stalls due to cardiovascular disease, not drug deaths. Proc Natl Acad Sci U S A. 2020;117(13):6998-7000. doi: 10.1073/pnas.1920391117 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Case A, Deaton A. Accounting for the widening mortality gap between American adults with and without a BA. Brookings Pap Econ Act. 2023;(2):1-44. doi: 10.1353/eca.2023.a935412 [DOI] [Google Scholar]
  • 6.Lopez AD, Adair T. Is the long-term decline in cardiovascular-disease mortality in high-income countries over? evidence from national vital statistics. Int J Epidemiol. 2019;48(6):1815-1823. doi: 10.1093/ije/dyz143 [DOI] [PubMed] [Google Scholar]
  • 7.Murphy SL, Kochanek KD, Xu J, Arias E. Mortality in the United States, 2020. NCHS Data Brief. 2021. Accessed June 9, 2026. https://stacks.cdc.gov/view/cdc/112079 [PubMed]
  • 8.OECD. OECD Data Explorer. 2025. Accessed June 9, 2026. https://www.oecd.org/en/data/datasets/oecd-DE.html
  • 9.Sawyer A, Flagg LA. State declines in heart disease mortality in the United States, 2000-2019. NCHS Data Briefs. 2021. Accessed June 9, 2026. https://stacks.cdc.gov/view/cdc/112339 [PubMed]
  • 10.Yan BW, Arun AS, Curtis LH, et al. JACC data report: cardiovascular disease mortality trends in the United States (1999-2023). J Am Coll Cardiol. 2025;85(25):2495-2498. doi: 10.1016/j.jacc.2025.05.018 [DOI] [PubMed] [Google Scholar]
  • 11.Cutler DM, Long G, Berndt ER, et al. The value of antihypertensive drugs: a perspective on medical innovation. Health Aff (Millwood). 2007;26(1):97-110. doi: 10.1377/hlthaff.26.1.97 [DOI] [PubMed] [Google Scholar]
  • 12.Cutler DM, Ghosh K, Messer KL, Raghunathan T, Rosen AB, Stewart ST. A satellite account for health in the United States. Am Econ Rev. 2022;112(2):494-533. doi: 10.1257/aer.20201480 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Emanuel EJ, Gudbranson E, Van Parys J, Gørtz M, Helgeland J, Skinner J. Comparing health outcomes of privileged US citizens with those of average residents of other developed countries. JAMA Intern Med. 2021;181(3):339-344. doi: 10.1001/jamainternmed.2020.7484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Papanicolas I, Niksch M, Figueroa JF. Avoidable mortality across US states and high-income countries. JAMA Intern Med. 2025;185(5):583-590. doi: 10.1001/jamainternmed.2025.0155 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.McWilliams JM, Meara E, Zaslavsky AM, Ayanian JZ. Differences in control of cardiovascular disease and diabetes by race, ethnicity, and education: U.S. trends from 1999 to 2006 and effects of medicare coverage. Ann Intern Med. 2009;150(8):505-515. doi: 10.7326/0003-4819-150-8-200904210-00005 [DOI] [PubMed] [Google Scholar]
  • 16.Inoue K, Liu M, Aggarwal R, Marinacci LX, Wadhera RK. Prevalence and control of diabetes among US adults, 2013 to 2023. JAMA. 2025;333(14):1255-1257. doi: 10.1001/jama.2024.28513 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Zhou B, Danaei G, Stevens GA, et al. ; NCD Risk Factor Collaboration (NCD-RisC) . Long-term and recent trends in hypertension awareness, treatment, and control in 12 high-income countries: an analysis of 123 nationally representative surveys. Lancet. 2019;394(10199):639-651. doi: 10.1016/S0140-6736(19)31145-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Johnson DY, Marinacci LX, Wadhera RK. Hypertension, diabetes, and high cholesterol awareness among US adults. JAMA Cardiol. 2025;10(8):859-860. doi: 10.1001/jamacardio.2025.1536 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Bahk J, Lynch JW, Khang YH. Forty years of economic growth and plummeting mortality: the mortality experience of the poorly educated in South Korea. J Epidemiol Community Health. 2017;71(3):282-288. doi: 10.1136/jech-2016-207707 [DOI] [PubMed] [Google Scholar]
  • 20.Mackenbach JP, Kulhánová I, Artnik B, et al. Changes in mortality inequalities over two decades: register based study of European countries. BMJ. 2016;353:i1732. doi: 10.1136/bmj.i1732 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Marmot MG, Smith GD, Stansfeld S, et al. Health inequalities among British civil servants: the Whitehall II study. In: Hyman S, ed. Stress and the Brain. Routledge; 2002. [Google Scholar]
  • 22.Centers for Disease Control and Prevention . NHANES questionnaires, datasets, and related documentation. Accessed June 9, 2026. https://wwwn.cdc.gov/nchs/nhanes/Default.aspx
  • 23.Steptoe A, Breeze E, Banks J, Nazroo J. Cohort profile: the English longitudinal study of ageing. Int J Epidemiol. 2013;42(6):1640-1648. doi: 10.1093/ije/dys168 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Kweon S, Kim Y, Jang MJ, et al. Data resource profile: the Korea National Health and Nutrition Examination Survey (KNHANES). Int J Epidemiol. 2014;43(1):69-77. doi: 10.1093/ije/dyt228 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ali MK, Bullard KM, Gregg EW, Del Rio C. A cascade of care for diabetes in the United States: visualizing the gaps. Ann Intern Med. 2014;161(10):681-689. doi: 10.7326/M14-0019 [DOI] [PubMed] [Google Scholar]
  • 26.Gardner EM, McLees MP, Steiner JF, Del Rio C, Burman WJ. The spectrum of engagement in HIV care and its relevance to test-and-treat strategies for prevention of HIV infection. Clin Infect Dis. 2011;52(6):793-800. doi: 10.1093/cid/ciq243 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kang S, Kang SM, Choi JH, et al. 2025 Clinical Practice Guidelines for Diabetes Management in Korea: recommendation of the Korean Diabetes Association. Diabetes Metab J. 2025;49(4):582-783. doi: 10.4093/dmj.2025.0469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Committee for the Korean Guidelines for the Management of Dyslipidemia . 2015 Korean Guidelines for the Management of Dyslipidemia: executive summary (English translation). Korean Circ J. 2016;46(3):275-306. doi: 10.4070/kcj.2016.46.3.275 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Kim KI, Ihm SH, Kim GH, et al. 2018 Korean Society of Hypertension guidelines for the management of hypertension, part III: hypertension in special situations. Clin Hypertens. 2019;25:19. doi: 10.1186/s40885-019-0123-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.NICE. Blood glucose management: type 2 diabetes in adults. 2015. Accessed Jun 9, 2026. https://www.nice.org.uk/guidance/ng28/chapter/Blood-glucose-management#hba1c-measurement-and-targets
  • 31.American Diabetes Association . Standards of medical care in diabetes-2022 abridged for primary care providers. Clin Diabetes. 2022;40(1):10-38. doi: 10.2337/cd22-as01 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.NCHS . age adjustment - health, United States. 2025. Accessed June 9, 2026. https://www.cdc.gov/nchs/hus/sources-definitions/age-adjustment.htm
  • 33.Huber PJ. The behavior of maximum likelihood estimates under nonstandard conditions. In: Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability. Vol 1: Statistics. University of California Press; 1967:221-234. Accessed June 9, 2025. https://projecteuclid.org/ebooks/berkeley-symposium-on-mathematical-statistics-and-probability/Proceedings-of-the-Fifth-Berkeley-Symposium-on-Mathematical-Statistics-and/chapter/The-behavior-of-maximum-likelihood-estimates-under-nonstandard-conditions/bsmsp/1200512988
  • 34.White H. A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica. 1980;48(4):817-838. doi: 10.2307/1912934 [DOI] [Google Scholar]
  • 35.Banks J, Marmot M, Oldfield Z, Smith JP. Disease and disadvantage in the United States and in England. JAMA. 2006;295(17):2037-2045. doi: 10.1001/jama.295.17.2037 [DOI] [PubMed] [Google Scholar]
  • 36.Montez JK, Beckfield J, Cooney JK, et al. US state policies, politics, and life expectancy. Milbank Q. 2020;98(3):668-699. doi: 10.1111/1468-0009.12469 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Glied S, Lleras-Muney A. Technological innovation and inequality in health. Demography. 2008;45(3):741-761. doi: 10.1353/dem.0.0017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Goldman DP, Smith JP. Can patient self-management help explain the SES health gradient? Proc Natl Acad Sci U S A. 2002;99(16):10929-10934. doi: 10.1073/pnas.162086599 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Lleras-Muney A, Lichtenberg FR. The effect of education on medical technology adoption: are the more educated more likely to use new drugs. 2002. Accessed June 9, 2026. https://www.nber.org/papers/w9185
  • 40.Cutler DM, Ly DP. The (paper) work of medicine: understanding international medical costs. J Econ Perspect. 2011;25(2):3-25. doi: 10.1257/jep.25.2.3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Kazemian P, Shebl FM, McCann N, Walensky RP, Wexler DJ. Evaluation of the cascade of diabetes care in the United States, 2005-2016. JAMA Intern Med. 2019;179(10):1376-1385. doi: 10.1001/jamainternmed.2019.2396 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Papanicolas I, Woskie LR, Jha AK. Health care spending in the United States and other high-income countries. JAMA. 2018;319(10):1024-1039. doi: 10.1001/jama.2018.1150 [DOI] [PubMed] [Google Scholar]
  • 43.Dickman SL, Himmelstein DU, Woolhandler S. Inequality and the health-care system in the USA. Lancet. 2017;389(10077):1431-1441. doi: 10.1016/S0140-6736(17)30398-7 [DOI] [PubMed] [Google Scholar]
  • 44.OECD . Health for Everyone? OECD Publishing; 2019. Accessed June 9, 2026. https://www.oecd.org/en/publications/health-for-everyone_3c8385d0-en.html
  • 45.Carter AR, Gill D, Morris R, et al. Educational inequalities in statin treatment for preventing cardiovascular disease: cross-sectional analysis of UK Biobank. medRxiv. Preprint posted online July 17, 2020. doi: 10.1101/2020.06.11.20128116 [DOI]
  • 46.Mensah GA, Wei GS, Sorlie PD, et al. Decline in cardiovascular mortality: possible causes and implications. Circ Res. 2017;120(2):366-380. doi: 10.1161/CIRCRESAHA.116.309115 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Wild SH, Byrne CD. ABC of obesity: risk factors for diabetes and coronary heart disease. BMJ. 2006;333(7576):1009-1011. doi: 10.1136/bmj.39024.568738.43 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Joseph P, Leong D, McKee M, et al. Reducing the global burden of cardiovascular disease, part 1: the epidemiology and risk factors. Circ Res. 2017;121(6):677-694. doi: 10.1161/CIRCRESAHA.117.308903 [DOI] [PubMed] [Google Scholar]
  • 49.Lawton R, Frankenberg E, Seeman T, Karlamangla A, Sumantri C, Thomas D. Explaining adverse cholesterol levels and distinct gender patterns in an Indonesian population compared with the U.S. Econ Hum Biol. 2024;54:101403. doi: 10.1016/j.ehb.2024.101403 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Roth GA, Mensah GA, Johnson CO, et al. ; GBD-NHLBI-JACC Global Burden of Cardiovascular Diseases Writing Group . Global burden of cardiovascular diseases and risk factors, 1990-2019: update from the GBD 2019 Study. J Am Coll Cardiol. 2020;76(25):2982-3021. doi: 10.1016/j.jacc.2020.11.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Crimmins EM, Finch CE. Infection, inflammation, height, and longevity. Proc Natl Acad Sci U S A. 2006;103(2):498-503. doi: 10.1073/pnas.0501470103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Guzik TJ, Nosalski R, Maffia P, Drummond GR. Immune and inflammatory mechanisms in hypertension. Nat Rev Cardiol. 2024;21(6):396-416. doi: 10.1038/s41569-023-00964-1 [DOI] [PubMed] [Google Scholar]
  • 53.Lawton RI. Early-life infectious disease exposure, the “hygiene hypothesis”, and lifespan: evidence from hookworm. Proc Natl Acad Sci USA. 2025;122(35):e2504265122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Sagheer U, Al-Kindi S, Abohashem S, et al. Environmental pollution and cardiovascular disease, part 1 of 2: air pollution. JACC Adv. 2023;3(2):100805. doi: 10.1016/j.jacadv.2023.100805 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement 1.

eTable 1. Sample sizes and data notes

eTable 2. At risk fractions by country over time

eTable 3. Cholesterol cascade by country over time

eTable 4. Hypertension cascade by country over time

eTable 5. Diabetes cascade by country over time

eTable 6. At risk and controlled fractions by race and ethnicity, over time

eTable 7. At risk and controlled fractions by country and education, most recent period

jamanetwopen-e2625410-s001.pdf (635.7KB, pdf)
Supplement 2.

Data Sharing Statement

jamanetwopen-e2625410-s002.pdf (15.8KB, pdf)

Articles from JAMA Network Open are provided here courtesy of American Medical Association

RESOURCES