GBD 2019 was an integrated surveillance system that used comparative risk assessment (CRA) to estimate attributable mortality, years of life lost (YLLs), years of life lived with disability (YLDs), and disability-adjusted life-years (DALYs) at the global and regional levels for 204 countries and territories from 1990 to 2019. The Global Health Data Exchange (GHDx) query tool (https://ghdx.healthdata.org/gbd-results-tool) was used to obtain data on yearly metabolsim-related numbers and age-standardized rates (ASR) of CVDs deaths, DALYs, YLLs, YLDs by location, age, sex, and socioeconomic status from 1990 to 2019.

In this study, CVDs consisted of ischemic heart disease, aortic aneurysm, rheumatic heart disease, stroke, hypertensive heart disease, nonrheumatic valvular heart disease, cardiomyopathy and myocarditis, atrial fibrillation and flutter, aortic aneurysm, endocarditis, peripheral arterial disease and other cardiovascular and circulatory diseases. The metabolic risk factors included high fasting plasma glucose (FBG), high low-density lipoprotein cholesterol (LDL-C), high systolic blood pressure (SBP), high body-mass index (BMI), and kidney dysfunction.

The global burden of CVDs attributable to metabolic risk factors is represented by the number and ASR and their 95% uncertainty intervals (95% UI) of deaths, DALYs, YLLs, and YLDs. The age-standardized rate (ASR) is a measure of the rate a population would have if it had a standard age structure. Standardization is required when comparing populations of different ages because age has a substantial impact on the risk of dying from CVDs. The ASR was calculated as follows: ASR=∑i=1Aaiwi∑i=1Awi×100,000, where ai is specific age ratio, wi is the number (or weight) of the selected standard population,and 100,000 indicates per 100,000 population (15).

EAPC is a concise and frequently utilized metric of the ASR trend over a given time period. With the following equation: y=α+βx+ε, the natural logarithm of the regression line is fitted to the ASR, where y=ln⁡(ASR), and x=calendaryear. Using the linear regression model, the estimated annual percentage change (EAPC) and its 95% confidence interval (CI) were calculated as 100×[exp⁡(β)−1] (16). An EAPC of 0, a positive EAPC and a negative EAPC show that the rate is constant, has a downward trend, or has an upward trend over time, respectively; the greater the absolute value of the EAPC is, the more rapidly the rate changes over time. The sociodemographic index (SDI), which ranges from 0 (worst) to 1 (best), is a lagged distribution of the total fertility rate of the population under 25, the average education level of adults, and per capita income indices. The EAPC and its 95% confidence interval (CI) were used to quantify temporal trends (17).

In order to understand the future trends of the CVDs burden attributed to metabolic risk factors, we used the Autoregressive integrated moving average (ARIMA) model to make predictions in this study. ARIMA is a commonly used time series model, where p, d, and q are the order of autoregression (AR), difference and partial autoregression (MA) required to make the data stationary, respectively. The ARIMA model consists of the following three main components: AR refers to a model that shows a changing variable that regresses on its own lagged, or prior, values, integrated (I) represents the differencing of raw observations to allow for the time series to become stationary, MA incorporates the dependency between an observation and a residual error from a moving average model applied to lagged observations (18). The ARIMA model is built using “auto.arima” functions from the “forecast” and “tseries” packages. The optimal model and parameters are selected in accordance with the Akaike Information Criterion (AIC) and Bayesian Criterion (BIC). The Ljung-Box test is conducted on the model's residual sequence. If P > 0.05, the test is passed, indicating that the data are not white noise and that the ARIMA model is well-fitting; otherwise, the model is remodeled (19). The constructed model was then used to predict the number and ASRs of deaths, DALYs, YLDs, and YLLs per year until 2029. All statistical analyses were carried out using R (version 4.1.3).

High SBP, followed by high LDL-C, high FBG, high BMI, and renal dysfunction, were the metabolic risk factors with the biggest effects on CVDs (Figure 1). Elevated SBP was linked to all types of CVDs and causes 43.2 percent of deaths number (10.0 million) and 42.2 percent of DALYs number (213.9 million) attributable to metabolic abnormalities in the CVDs in 2019. The second risk factor with the largest effects was high LDL-C and was responsible for 19.0 percent of metabolic-related CVDs mortality and 19.5 percent of DALYs number. Increased BMI was closely related to cardiovascular disorders such stroke, ischemic heart diseases, hypertensive heart diseases, and AF and caused 14.0% of deaths and 17.1% of DALYs number (Supplementary Table S1).

From 1990 to 2019, the global burden of CVDs attributed to metabolic risk factors continued to rise (Figure 2). Specifically, the numbers of deaths, DALYs, YLDs, and YLLs increased by 59.3%, 51.0%, 104.55%, and 47.8%, respectively. CVDs deaths and DALYs number attributable to metabolic risk factors reached 13.7 million and 192.7 million, respectively, in 2019. The ASR of CVDs attributable to metabolic risk factors showed a significant downward trend after taking into account the effects of increasing population as well as aging. The ASRs for deaths, DALYs, YLDs, and YLLs decreased from 252.9 [95% UI (230.0, 274.6)] to 176.1 [95% UI (156.2, 192.4)], 4,965.8 [95% UI (4,602.7, 5,357.3)] to 3,573.4 [95% UI (3,240.3, 3,870.5)], 278.9 [95% UI (199.9, 359.0)] to 274.9 [95% UI (197.2, 349.1)], and 4,686.9 [95% UI (4,341.0, 5,048.4)] to 3,298.6 [95% UI (3,007.1, 3,572.7)] per 100,000 people, respectively (Table 1). The EAPCs of deaths and DALYs ASR were −1.4 (−1.4,1.3) and −1.2 (−1.3,−1.2), respectively (Supplementary Table S2).

From 1990 to 2019, the change in CVDs death and DALYs ASRs attributable to metabolic risk factors showed significant regional differences globally (Figure 3). The Eastern Mediterranean region had the greatest death and DALYs ASR in both 1990 and 2019, while the Americas had the lowest (Figure 3). The ASR of deaths has decreased by more than 100 per 100,000 in several regions (Central Europe, Australasia, Tropical Latin America, Western Europe, High-Income Asia Pacific), but many other regions (Central Latin America, East Asia, Andean Latin America, South Asia, Southeast Asia) had decreases in the death ASR of less than 50 per 100,000, and even increases were observed in Oceania and Central Asia.

According to the calculation of EAPCs, most regions had negative (declining) ASRs for deaths and DALYs (Figure 4). The area with the greatest reduction in deaths was High-income Asia Pacific [−4.0 (−4.2, −3.8)], while the area with the greatest reduction in DALYs was Australasia [−4.0 (−4.2, −3.7)].The three countries with the largest reductions of death ASRs were Bahrain, Czechia and Estonia. Uzbekistan, Tajikistan, and Azerbaijan had the largest increases. As for DALYS, the three countries with the largest decreases were Bahrain, Czechia and Mauritius, whereas Uzbekistan, Tajikistan and the Philippines had the largest increases (Supplementary Table S3).

The numbers of deaths and DALYs grew dramatically with increasing age. In 1990, the numbers of deaths and DALYs peaked in the 75–79 and 65–69 age groups, respectively. By 2010 and 2019, the peaks had shifted to the 80–84 and 70–74 age groups, respectively (Figure 5).

The burden of CVDs attributable to metabolic risk factors exhibited significant sex differences and changes over time. In 1990, there were more deaths number among women than men (4.34 vs. 4.27 million), a difference that lasted for five years, and in 1995, for the first time, there were more deaths among men than women (4.73 vs. 4.72 million), and this trend remained so until 2019. Since 1990, the number of DALYs and YLLs was consistently higher in men than in women, while the number of YLDs was lower than that in women. From 1990 to 2019, the ASRs for deaths, DALYs, and YLLs were consistently higher for men than for women, while YLDs were consistently lower.

The age-standardized death and DALYs rates for CVDs attributable to metabolic factors varied widely with the SDI (Supplementary Table S4). As the SDI rose from 1990 to 2019 (0.5–0.7), the global ASR for deaths (from 252.9 to 176.1 per 100,000 persons) and DALYs (from 4,965.8 to 3,554.5 per 100,000 persons) fell steadily. Almost all regions, with the exception of Oceania and Central Asia, experienced a decline in the ASR for deaths and DALYs as the SDI rose. The intermediate SDI sector had the highest burden of CVDs, whereas the high SDI segment had the lowest burden, with both death and DALYs ASR significantly declining over the past several decades. (Supplementary Table S2 and Figure 6).

According to our predictions, the number of CVDs deaths, DALYs, YLDs and YLLs attributable to metabolic risk factors will continue to increase globally in the coming years for both men and women (Figure 7). The number of deaths will increase by 20.4%, from 13.7 million in 2019 to 16.5 million in 2029. The number of DALYs will also increase (11.7%) from 290.95 million to 324.8 million (Supplementary Table S5).

Similar to the past 30 years, the adjusted ASR of deaths, DALYs, YLDs, YLLs will gradually decrease globally except for female YLLs (Supplementary Figure S1). The ASR of deaths will drop from 176.1 per 100,000 in 2019 to 149.6 per 100,000 in 2029 (3,573.5–3,101.1 for DALYs, 3,298.6–2,827.5 for YLLs), but the ASR of YLDs will increase from 274.9 to 279.6 per 100,000 (Supplementary Table S5). There will be a clear gender difference in the change in YLDs; that for males will decrease from 265.6 to 261.0 per 100,000, but the change in YLDs for females will increase significantly (from 282.4 to 295.2 per 100,000).