Burden and attributable risk factors of non-communicable diseases and subtypes in 204 countries and territories, 1990–2021: a systematic analysis for the global burden of disease study 2021

Abstract

Background:

Understanding based on up-to-date data on the burden of non-communicable diseases (NCDs) is limited, especially regarding how subtypes contribute to the overall NCD burden and the attributable risk factors across locations and subtypes. We aimed to report the global, regional, and national burden of NCDs, subtypes, and attributable risk factors in 2021, and trends from 1990 to 2021 by age, sex, and socio-demographic index (SDI).

Materials and methods:

We used data from the Global Burden of Disease Study 2021 to estimate the prevalence, deaths, and disability-adjusted life years (DALYs) for NCDs and subtypes, along with attributable risk factors. Estimates were presented with 95% uncertainty intervals (UI). Relationships between NCD DALYs and SDI across regions and countries were estimated using smoothing splines models.

Results:

In 2021, NCDs accounted for 7.3 trillion global cases, 43.8 million deaths, and 1.73 billion DALYs. Global age-standardized rates showed NCD prevalence at 91 034.0, deaths at 529.7, and DALYs at 20 783.0 per 100 000 population, with changes of −0.1%, −27.9%, and −19.4% from 1990, respectively. Subtypes with the highest age-standardized DALYs were cardiovascular disease (5056), neoplasms (2954), and other NCDs (1913 per 100 000 population), with diabetes and kidney diseases increasing by 25.6% since 1990. Regionally, Oceania had the highest age-standardized DALYs (28 782.0) in 2021, while Southern Sub-Saharan Africa saw the largest increase (+8.0%) since 1990. Nationally, Nauru reported the highest age-standardized DALYs (42 754.3), with Lesotho experiencing the largest increase since 1990 (+38.4%). Cardiovascular diseases had the highest age-standardized DALYs among subtypes across 16 of 21 regions and 159 of 204 countries. Key risk factors globally were high systolic blood pressure (contributing to 12.8% of age-standardized DALYs), dietary risks (10.0%), and tobacco usage (9.9%), with the most significant increase in high body-mass index (+57.8%). High systolic blood pressure was the biggest attributable risk factor for NCDs in 9 regions and 101 countries. Age-standardized data reveal higher NCD prevalence in women and greater mortality and DALYs in men, with DALYs spiking post-45 for both sexes. Men have higher DALYs attributed to most risk factors, excluding those from unsafe sex, intimate partner violence, low physical activity, and high body-mass index. Age-standardized DALYs of NCDs generally decline with the SDI spectrum. Dominant NCD risk factors follow gender-age and SDI-based trajectories.

Conclusion:

Despite declining age-standardized prevalence, death rates, and DALYs for NCDs, they remain a major health issue. Emphasis on managing cardiovascular diseases, cancers, diabetes, kidney diseases, and mental disorders is essential. The burden of NCDs is more severe in low-SDI countries and among males. Prevention efforts should prioritize blood pressure control, dietary improvements, and tobacco reduction, tailoring interventions according to gender-age-based and SDI-development-based trajectories of dominant risk factors.

Introduction

Non-communicable diseases (NCDs) accounted for 74% of all deaths globally in 2017, approximately 41 million annually, rising to 42 million by 2019^[1–3]. These diseases, primarily influenced by lifestyle, environmental, and genetic factors, are generally incurable but manageable through early detection, lifestyle changes, and medical interventions^[1]. The Sustainable Development Goal (SDG) 3.4 aims to reduce premature mortality from NCDs by one-third by 2030. However, by 2019, the age-standardized mortality rate for NCDs had decreased by only 3.7%^[4], indicating significant delays in meeting SDG 3.4.

The challenge of meeting SDG 3.4 is further compounded by the diverse nature of NCDs, requiring coordinated approaches across all subtypes. Effective health policies and optimal resource allocation among these subtypes are critical, but the best strategies for achieving these remain unclear. For example, regions with a high burden of cardiovascular diseases might benefit from allocating more resources to these areas to accelerate progress toward SDG 3.4. Conversely, regions where cancer poses a greater threat may require a different focus for resource allocation. Moreover, the leading risk factors for regions or countries that share the leading diseases may vary – for instance, respiratory diseases may be driven by air pollution in some areas and smoking in others^[5], necessitating tailored proportional allocation of resources to both the leading disease and the leading risk factors.

Given the complexity of NCDs, it is crucial to understand their spectrum and their major risk factors, to allocate resources proportionally, and to address disparities in age, gender, and socioeconomic status at a granular level. Previous research using Global Burden of Disease (GBD) data from 2017 and 2019 has highlighted disease burdens in specific regions or subtypes^[3,6–10] and identified key risk factors^[11–14], underscoring the need for systematic comparisons across subtypes and risk factors.

This study conducts a detailed assessment of the global burden of NCDs, their subtypes, and attributable risk factors in 2021 while tracing trends back to 1990. It is structured around four key questions: (1) What are the current global, regional, and national burdens of NCDs, and how have they evolved since 1990? This information will help in ranking and prioritizing subtypes for targeted interventions. (2) Which risk factors significantly contribute to the NCD burden, and how have these changed over time? This information can facilitate the prioritization of mitigation strategies by location. (3) How do these risk factors influence the burden of various NCD subtypes across different locations in 2021? This analysis can identify leading local risk factors for each subtype. (4) What impacts do age, gender, and SDI have on the burden of NCDs and their risk factors? Insights from this can enable more precise and proportionate resource allocation at the population level. Addressing these aims can help optimize resource allocation and strategy effectiveness for combating NCDs and achieving SDG 3.4.

Materials and methods

Overview

The GBD 2021 study estimated the burden of 371 diseases and injuries and 88 risk factors from 1990 to 2021 in 204 countries and territories and 21 regions^[15–17]. Detailed methodologies were documented in prior publications^[15–17]. Estimates for both fatal and non-fatal outcomes are available online (https://vizhub.healthdata.org/gbd-results/). The reporting of this study was in line with the STROCSS criteria^[18].

Non-communicable diseases

NCDs were classified within the GBD framework as Level 1 causes, encompassing 12 Level 2 subtypes: neoplasms, cardiovascular diseases, chronic respiratory diseases, digestive diseases, neurological disorders, mental disorders, substance use disorders, diabetes and kidney diseases, skin diseases, sense organ diseases, musculoskeletal disorders, and other non-communicable diseases^[17]. These included 106 Level 3 diseases such as breast cancer, stroke, and asthma (Table S1, http://links.lww.com/JS9/D773). Our analysis mainly targeted Level 1 NCDs and delved into the 12 Level 2 subtypes. Their International Classification of Disease (ICD)-9 and ICD-10 codes are detailed in Table S2 (http://links.lww.com/JS9/D773).

Fatal estimates

Mortality data for NCDs were sourced from vital registration, verbal autopsy, surveillance, cancer registries, and police records. Verbal autopsies were notably excluded from child-level cause fatality estimates. Data were standardized for coding system variations, representativeness, completeness, age, sex, and misclassification of maternal and HIV/AIDS deaths. For NCDs (Level 1), an aggregation approach was applied. For Level 2 subtypes, including cardiovascular, chronic respiratory, digestive diseases, and skin/subcutaneous diseases, various linear mixed-effect models and spatiotemporal Gaussian process regression models were created using the Cause of Death Ensemble model (CODEm) framework, accounting for location-specific covariates. Other Level 2 subtypes employed aggregation. CoDCorrect analysis was applied to ensure internal consistency of CODEm results. Multiplication of the estimated number of deaths by the standard life expectancy at the age of death resulted in years of life lost (YLL)^[15].

Non-fatal estimates

Nonfatal estimates, covering incidence, prevalence, and years lived with disability (YLD), were derived from systematic reviews, surveys, disease registries, and estimation of hospital envelope. After data adjustment, estimation of prevalence and incidence by cause and sequela was performed using DisMod-MR 2.1, a Bayesian meta-regression method, and included incorporation of severity distributions, disability weights, and comorbidity adjustment of the sequela. YLDs were estimated by combining prevalence and incidence of causes and sequela with levels of severity related to disability using disability weights, while adjusting for comorbidity. DALYs for GBD 2021 combined cause-specific mortality and non-fatal health loss, adding YLLs to YLDs across all demographics and locations, representing the total health burden. Uncertainty between YLLs and YLDs was treated as independent. DALYs were determined through 1000 simulation draws, summing YLLs and YLDs, with 95% Uncertainty Intervals (UIs) based on the 25th and 975th draws^[15].

GBD risk factor hierarchy

The GBD categorized risk factors into four levels: behavioral, environmental/occupational, and metabolic factors, totaling 88 factors^[16]. Our analysis targeted 18 Level-2 factors for NCDs, including high systolic blood pressure, tobacco, dietary risks, high fasting glucose, body-mass index (BMI), air pollution, high LDL cholesterol, kidney dysfunction, alcohol use, occupational risks, non-optimal temperature, drug use, other environmental factors, low physical inactivity, unsafe sex, childhood abuse and bullying, intimate partner violence, and child and maternal malnutrition.

Risk estimates

For each risk factor, the Comparative Risk Assessment (CRA) framework assessed the attributable burden, defined as the reduction potential in the current burden if past exposure shifted to the theoretical minimum risk exposure level (TMREL). This involved: (1) estimating relative risks (RR) for risk-outcome pairs, (2) calculating exposure levels, (3) establishing the TMREL, (4) determining population attributable fractions, (5) estimating RR-weighted exposure prevalence (summary exposure value), and (6) aggregating risks considering mediation effects^[16].

Statistical analysis

We decomposed absolute numbers and age-standardized rates (ASRs) for point prevalence, mortality, and DALYs of NCDs and their 12 subtypes, using GBD 2021’s global standard population metrics, reported per 100 000 population with 95% uncertainty intervals (UIs)^[17]. The 95% UIs were estimated from the 25th and 975th values of 1000 simulation draws. Additionally, the proportions of DALYs attributable to each Level 2 risk factor were reported, with their definitions and relative risks for NCDs detailed previously^[16]. Age and sex patterns in the burden of NCDs and risk factors were also compared. Additionally, we analyzed the relationship between NCDs burden (DALYs) and Socio-Demographic Index (SDI) across 21 regions and 204 countries using Smoothing Splines models^[19]. SDI, a measure of socio-demographic progress ranging from 0 to 1, incorporates income per capita, average years of education, and the total fertility rate^[20]. For analyzing SDI associations with risk factors, the 204 GBD countries and territories were categorized into quintiles according to their SDI. All statistical analyses were conducted in R software (Version 4.3.2).

Patient and public involvement

The study relied on publicly available aggregate data; thus, no patients were involved in setting the research question or the outcome measures, nor were they involved in the design or implementation of the study.

Results

The results align with the research questions and are divided into four sections: (1) NCD and subtype disease burden and trends; (2) risk factors for NCD DALYs and their trends; (3) risk factors for NCD subtypes in 2021; and (4) disparities in NCD burden and risk factors by age, gender, and SDI. Flowchart in Fig. 1 illustrates the structure and content of the results.

Figure 1.

Flowchart of results presentation. O-TFS: Others are detailed in tables, figures, and supplementary materials; ASR: Age-standardized rate; NCDs: Non-communicable diseases; DALYs: Disability-adjusted life years; SDI: Socio-demographic index.

Disease burden and trend changes for NCDs and their subtypes

Global level

For NCDs, age-standardized prevalence, death, and DALYs have all decreased since 1990. In 2021, there were 7255.1 million global NCD cases, with an age-standardized prevalence of 91 034.0 per 100 000 population – a decrease of 0.1% since 1990. NCDs caused 43.8 million deaths, with an age-standardized death rate of 529.7 per 100 000 population, dropping by 27.9% since 1990. The global DALYs for NCDs were 1727.2 million, with an age-standardized rate of 20 783.0 per 100 000 population (a decrease of 19.7% since 1990) (see Table 1).

Table 1.

Global and regional prevalent cases, deaths, and disability adjusted life years (DALYs) for non-communicable diseases in 2021, and percentage change in age-standardized rates (ASRs) per 100 000 population from 1990 to 2021

Location name	Prevalence			Deaths			DALYs
	2021 counts (millions, 95% UI)	2021 ASRs (per 100 000, 95% UI)	Percentage change in ASRs from 1990	2021 counts (millions, 95% UI)	2021 ASRs (per 100 000, 95% UI)	Percentage change in ASRs from 1990	2021 counts (millions, 95% UI)	2021 ASRs (per 100 000, 95% UI)	Percentage change in ASRs from 1990
Global	7255.1 (7201.5, 7305.5)	91 034 (90 296.7, 91 725.5)	−0.1(−0.1, −0.1)	43.8 (41.8, 45.9)	529.7 (506.6, 554.5)	−27.9(−27.8, −27.9)	1727.2 (1537.9, 1941.6)	20 783 (18 495.7, 23 367.9)	−19.7(−19.7,−19.7)
High-income Asia Pacific	172.5 (171.6, 173.4)	86 477.7 (85 607.1, 87 284.9)	−1.6(−1.6,−1.6)	1.5 (1.5, 1.6)	265.6 (262.4, 270.6)	−45.1(−45.1,−45.1)	45.6 (39.9, 52.2)	13 606.2 (11 472.1, 16 075.5)	−26.5(−26.5,−26.6)
High-income North America	345.1 (342.9, 346.9)	89 610.4 (88 719.2, 90 394.6)	0.7(0.7,0.7)	2.9 (2.9, 2.9)	429.9 (425.4, 434.9)	−24.5(−24.5,−24.5)	108.2 (95, 122.6)	21 223.9 (18 282.4, 24 475.4)	−6.2(−6.2,−6.2)
Western Europe	409.7 (407.7, 411.8)	88 970.4 (88 186.6, 89 757.7)	−0.6(−0.6,−0.6)	3.7 (3.7, 3.8)	344.3 (340.9, 348.2)	−42.3(−42.3,−42.3)	114.4 (100.8, 130.6)	16 386 (13 867.1, 19 332.9)	−24.8(−24.8,−24.9)
Australasia	28.6 (28.4, 28.8)	88 923.7 (88 062.1, 89 793.9)	1.5(1.5,1.5)	0.2 (0.2, 0.2)	319.3 (317.4, 322.6)	−45(−45,−45)	6.9 (6, 7.9)	16 223.9 (13 875.9, 19 059.1)	−26.2(−26.2,−26.3)
Andean Latin America	60.7 (60.1, 61.3)	91 595.4 (90 643.8, 92 538.6)	−0.8(−0.8,−0.8)	0.2 (0.2, 0.3)	399.1 (340, 466)	−25.9(−25.9,−26)	11.4 (9.7, 13.4)	18 147.1 (15 494.7, 21 304.3)	−21.5(−21.5,−21.5)
Tropical Latin America	213 (211.3, 214.5)	92 104.1 (91 169.4, 92 940.1)	−0.3(−0.3,−0.3)	1.1 (1.1, 1.1)	441.9 (432, 452.2)	−32.8(−32.8,−32.8)	51.3 (45, 58.9)	20 665.5 (18 126.3, 23 793.9)	−19(−19,−19.1)
Central Latin America	229 (226.9, 230.9)	89 371.6 (88 485.3, 90 226.9)	−0.9(−0.9,−0.9)	1.2 (1, 1.3)	477.3 (434.2, 520.8)	−18.7(−18.7,−18.7)	52 (45.7, 60.1)	20 527.1 (18 079.5, 23 682.3)	−9.9(−9.9,−9.9)
Southern Latin America	62.4 (61.8, 63)	89 663.2 (88 600.6, 90 808.2)	0.1(0.1,0.1)	0.4 (0.4, 0.4)	416.3 (407.3, 425.5)	−37.7(−37.7,−37.7)	14.6 (12.8, 16.6)	18 471.3 (16 089.5, 21 274)	−23.8(−23.7,−23.8)
Caribbean	44.4 (44.1, 44.7)	92 617.5 (91 903.7, 93 314.7)	−0.1(−0.1,−0.1)	0.3 (0.3, 0.3)	533.1 (477, 589.6)	−21.2(−21.1,−21.2)	11.5 (10.1, 13.3)	22 661.3 (19 853.5, 26 043.5)	−11.1(−11.1,−11.1)
Central Europe	108.5 (107.8, 109.2)	90 032.9 (88 925.3, 91 064.1)	−0.8(−0.8,−0.8)	1.3 (1.2, 1.4)	562.4 (532.4, 592.1)	−35.6(−35.6,−35.6)	37.3 (33.6, 41.5)	20 150.6 (17 825.1, 22 828.4)	−26.6(−26.6,−26.7)
Eastern Europe	193.5 (192.1, 194.8)	89 592.1 (88 523.2, 90 554.4)	−0.2(−0.2,−0.2)	2.4 (2.2, 2.5)	679 (632.2, 728.7)	−21.5(−21.5,−21.6)	73.2 (66.1, 80.9)	24 276.2 (21 810.5, 27 106.2)	−13.6(−13.6,−13.6)
Central Asia	86.4 (85.3, 87.4)	90 246.5 (89 088.2, 91 283.5)	−0.3(−0.3,−0.3)	0.5 (0.4, 0.5)	673.6 (620.3, 727.2)	−16.5(−16.5,−16.6)	20 (17.7, 22.7)	23 272.2 (20 729.5, 26 152.5)	−13.6(−13.6,−13.6)
North Africa and Middle East	569.4 (563.9, 574.9)	91 525.9 (90 681, 92 385.1)	−0.5(−0.5,−0.5)	2.5 (2.2, 2.7)	641.3 (586.9, 696.2)	−27.4(−27.4,−27.5)	121.5 (105.6, 140.7)	23 557.8 (20 805.9, 26 871.5)	−23.4(−23.3,−23.4)
South Asia	1694.9 (1681.3, 1708.5)	91 570 (90 852.5, 92 285.3)	−0.4(−0.4,−0.4)	8.1 (7.6, 8.7)	612.1 (571.8, 655.6)	−10.2(−10.2,−10.2)	369.8 (325, 419.6)	23 033.1 (20 460.8, 25 878.8)	−12.2(−12.1,−12.2)
Southeast Asia	641.1 (635.1, 646.8)	90 854 (89 888.6, 91 756.9)	−1.3(−1.3,−1.3)	3.6 (3.3, 3.9)	618.4 (565, 664)	−13.3(−13.3,−13.4)	151 (134, 168.6)	22 417.4 (20 018.7, 24 936.7)	−12.1(−12.1,−12.1)
East Asia	1363.6 (1354, 1372.6)	88 893.2 (87 922.3, 89 785.8)	−0.7(−0.7,−0.7)	11 (9.4, 12.6)	571.3 (492.7, 650.2)	−40.2(−40.1,−40.2)	362.1 (313.1, 416.2)	18 780 (16 242.7, 21 623.7)	−33.8(−33.7,−33.8)
Oceania	12.5 (12.4, 12.7)	91 920.6 (90 931.9, 92 912.7)	−0.5(−0.5,−0.5)	0.1 (0, 0.1)	838.7 (725.7, 965.2)	−12.5(−12.5,−12.5)	2.8 (2.4, 3.3)	28 781.7 (24 958.5, 32 977.5)	−8.2(−8.2,−8.2)
Western Sub-Saharan Africa	441.6 (436.9, 446.1)	93 228.5 (92 574, 93 863.5)	−0.9(−0.9,−0.9)	1.2 (1, 1.3)	610.4 (536.3, 687.8)	−12.2(−12.2,−12.3)	73.4 (62.6, 85)	22 915.1 (19 954.4, 26 276.3)	−11.1(−11,−11.1)
Eastern Sub-Saharan Africa	381.4 (377.3, 385.5)	92 372.8 (91 690.8, 93 030.4)	−1.6(−1.6,−1.6)	1 (0.9, 1.2)	635.3 (580.1, 702)	−19.4(−19.4,−19.5)	63.1 (54.7, 72.5)	23 728.4 (20 941.5, 26 902.3)	−17.8(−17.8,−17.8)
Central Sub-Saharan Africa	125 (123.5, 126.5)	93 831.3 (93 054.1, 94 595.6)	−1.5(−1.5,−1.5)	0.4 (0.3, 0.4)	758.5 (618.3, 910.8)	−11.6(−11.6,−11.6)	21.1 (17.8, 24.8)	26 105 (22 085.7, 30 543.3)	−12.4(−12.4,−12.4)
Southern Sub-Saharan Africa	71.8 (71, 72.5)	89 811.8 (88 885.8, 90 698.4)	−0.3(−0.3,−0.3)	0.4 (0.3, 0.4)	684.4 (650.7, 717.7)	16.6(16.6,16.6)	16.1 (14.3, 18.1)	24 499.3 (22 034.8, 27 320.8)	8(8,7.9)

95% UI.95% uncertainty intervals.

For subtypes, the top three age-standardized prevalent subtypes globally, in 2021, were other non-communicable diseases (68 383.1 per 100 000 population), neurological disorders (35 333.9), and digestive diseases (28 513.2). Since 1990, diabetes and kidney diseases have seen the most significant rise (+22.8%), while chronic respiratory disease has experienced the largest decrease (−27.1%) (see Fig. S1 for additional subtypes and trend changes, http://links.lww.com/JS9/D773). The top three age-standardized death subtypes globally, in 2021, were cardiovascular disease (235.18), neoplasms (116.49), and chronic respiratory diseases (53.56) per 100 000 population. Since 1990, the biggest increase was observed in diabetes and kidney diseases (+14.8%), while the biggest decrease was shown in chronic respiratory diseases (−36.7%) (see Fig. S2, http://links.lww.com/JS9/D773 for additional subtypes and trend changes). The top three age-standardized DALYs subtypes globally, in 2021, were cardiovascular disease (5056), neoplasms (2954), and other non-communicable diseases (1913) per 100 000 population. Since 1990, diabetes and kidney diseases have seen the biggest increase (1450, +25.6%), followed by mental disorders (1909, +9.4%) while the biggest decrease was observed in chronic respiratory disease (−37.6%) (see Fig. 2 for additional subtypes and trend changes).

Figure 2.

Global and regional age-standardized disability-adjusted life years (DALYs) for Non-Communicable Diseases (NCDs) and its subtypes in 2021, with trends from 1990. Top row in each cell shows 2021 age-standardized DALY values; bottom row indicates changes from 1990. Black labels highlight increasing trend, white labels denote decreases. X-axis labels (NCD subtypes) are ordered from left to right by descending global age-standardized DALYs in 2021. Y-axis labels (regions) are ordered bottom to top by decreasing NCD age-standardized DALYs in 2021. NCDs: Non-communicable diseases; DALYs: Disability-adjusted life years.

Regional level

For NCDs, regional age-standardized NCD prevalence rates in 2021 varied from 86 477.7 in high-income Asia Pacific to 93 831.3 in Central Sub-Saharan Africa per 100 000 population. Between 1990 and 2021, Australasia saw the largest increase (+1.5%), while high-income Asia Pacific and Eastern Sub-Saharan Africa saw the most significant decrease (−1.6%) (see Table 1 and Fig. S1 for other regions and other trends, http://links.lww.com/JS9/D773). Regional age-standardized death rates, in 2021, ranged from 265.64 in high-income Asia Pacific to 838.70 in Oceania. Southern Sub-Saharan Africa experienced the highest mortality increase (+16.6%), while high-income Asia Pacific saw the greatest decrease (−45.1%) (refer to Table 1 and Fig. S2 for other regions and other trends, http://links.lww.com/JS9/D773). Regional age-standardized DALYs in 2021 varied from 13 606.2 in high-income Asia Pacific to 28 781.7 in Oceania, with Southern Sub-Saharan Africa showing the largest increase (+8.0%) and East Asia the most significant decrease (−33.8%) (see Table 1 and Fig. 2 for other regions and other trends).

For subtypes, for age-standardized prevalence in 2021, other non-communicable diseases were the leading subtype across all 21 regions. Between 1990 and 2021, the largest rise was in Oceania’s diabetes and kidney diseases (+45.3%), with the biggest drop in high-income Asia Pacific’s chronic respiratory diseases (−50.3%) (see Fig. S1 for other subtypes and trends, http://links.lww.com/JS9/D773). For age-standardized death in 2021, cardiovascular disease had the highest rates across 18 regions, except in high-income Asia Pacific, Australasia, and Western Europe, where neoplasms led. From 1990 to 2021, the largest rise was in Central Europe’s mental disorders (+16 544.8%), while the most significant fall was in East Europe’s chronic respiratory diseases (−69.6%) (see Fig. S2 for other subtypes and trends, http://links.lww.com/JS9/D773). For age-standardized DALYs in 2021, cardiovascular disease topped the subtype list in most regions, except in high-income North America, Western Europe, Andean Latin America, and South Latin America, where neoplasms topped. Notably, Australasia saw mental disorders lead. The greatest increase was in high-income North America’s substance use disorders (+206.2%), with the largest decrease in East Asia’s chronic respiratory diseases (−66.5%) (see Fig. 2 for other subtypes and trends).

National level

For NCDs, national age-standardized prevalence of NCDs in 2021 ranged from 85 826.3 per 100 000 population in Singapore to 93 965.0 in Nigeria (Table S3 and Fig. 3A). From 1990 to 2021, Australia (+1.84%) experienced the largest increase in ASR prevalence, while Madagascar saw the most significant decrease (−3.83%) (see Table S3 for other countries and trends, http://links.lww.com/JS9/D773). National age-standardized death rates in 2021 varied from 213.3 in San Marino to 1348.4 in Nauru (Table S4 and Fig. S3, http://links.lww.com/JS9/D773). The most notable increase was in Lesotho (+47.9%), with the Republic of Korea recording the largest decrease (−62.0%) (see Table S4 for other countries and trends, http://links.lww.com/JS9/D773). National age-standardized DALYs in 2021 spanned from 12 051.7 in Singapore to 42 754.3 in Nauru (Table S5 and Fig. S4, http://links.lww.com/JS9/D773). Between 1990 and 2021, Lesotho (+38.4%) had the highest increase, whereas the Maldives had the largest decrease (−44.9%) (see Table S5 for other countries and trends, http://links.lww.com/JS9/D773).

Figure 3.

National age-standardized prevalence in 2021 (A), global Disability-adjusted life years (DALYs) by age and sex in 2021 (B), and regional DALY rates by sociodemographic index (SDI) from 1990 to 2021 (C). (B) Bars represent DALYs, and lines depict age-standardized DALYs per 100 000 population. Error bars and dashed lines indicate the 95% uncertainty intervals. DALYs: Disability-adjusted life years; SDI: Socio-demographic index (c). The 31 points represent each region’s annual age-standardized DALY rates from 1990 to 2021. A solid line illustrates expected values based on sociodemographic indices and disease rates across all regions during this period. Regions plotted above the line have a higher-than-expected disease burden (e.g., Oceania), while those below the line have a lower-than-expected burden (e.g., Andean Latin America). DALYs: Disability-adjusted life years.

For subtype, regarding the age-standardized prevalence in 2021, the other non-communicable diseases were the leading subtype across all 204 countries. From 1990 to 2021, Mongolia saw the largest increase in substance use disorder prevalence (+71.7%), while Japan experienced the largest decrease in chronic respiratory diseases (−56.2%) (see Fig. S5 for other subtypes and trends, http://links.lww.com/JS9/D773). For the age-standardized death rates, cardiovascular diseases were the leading subtype in 175 countries, with neoplasms leading in 29 countries. The most significant increase was in Poland for mental disorders (+114 524.9%), and the largest decrease was in New Zealand for mental disorders (−94.0%) (see Fig. S6 for other subtypes and trends, http://links.lww.com/JS9/D773). For the age-standardized DALYs, cardiovascular diseases were predominant in 159 countries, neoplasms in 31 countries, diabetes and kidney diseases in 6 countries, mental disorders in 6 countries, and musculoskeletal disorders in 2 countries. The highest increase was observed in the United States of America for substance use disorders (+216.7%), while Singapore saw the largest decrease in chronic respiratory diseases (−74.4%) (see Fig. S7 for other subtypes and trends, http://links.lww.com/JS9/D773).

Attributable risk factors for NCDs and trend changes

Global level

In 2021, the top three risk factors were high systolic blood pressure (12.8%: contributing to 12.8% of NCD age-standardized DALYs), dietary risks (10.0%), and tobacco use (9.9%). Non-leading risk factors can be referred to in Figure S8 (http://links.lww.com/JS9/D773). Between 1990 and 2021, the most significant risk increase was seen in high body mass index (+57.8%), while the largest decrease was in child and maternal malnutrition (−52.5%) (other trend changes can be referred to in Fig. S8, http://links.lww.com/JS9/D773).

Regional level

In 2021, high systolic blood pressure was the leading risk factor for NCD age-standardized DALYs in nine regions, except for East Asia, Western Europe, and high-income Asia Pacific (where tobacco led), South Asia and Oceania (where air pollution led), the Caribbean, Central Latin America, and Andean Latin America (where high fasting plasma glucose led), Tropical Latin America, Southern Latin America, high-income North America, and Australasia (where high BMI led). Non-leading risk factors can be referred to in Figure S8 (http://links.lww.com/JS9/D773). From 1990 to 2021, high-income North America saw the largest increase in risk from drug use (+337.9%), while Australasia experienced the greatest decrease in risk from child and maternal malnutrition (−90.0%) (other trend changes can be referred to in Fig. S8, http://links.lww.com/JS9/D773).

National level

Across the 204 countries’ NCD age-standardized DALYs, high systolic blood pressure, high fasting plasma glucose, tobacco, high BMI, air pollution, and dietary risks dominated in 101, 46, 30, 15, 10, and 2 countries, respectively. Non-leading risk factors can be referred to in Figure S9 (http://links.lww.com/JS9/D773). The largest increase was seen in child and maternal malnutrition in Georgia by 180 853.2%, while the biggest decrease was seen in non-optimal temperature in El Salvador by −235.8%. Other trend changes can be referred to in Figure S9 (http://links.lww.com/JS9/D773).

Attributable risk factors for NCD subtypes in 2021

Global level

Global leading risk factors attributed to each subtype’s age-standardized DALYs in 2021 were high systolic blood pressure for cardiovascular diseases (49.7%), tobacco use for neoplasms (19.5%), alcohol use for other non-communicable diseases (0.2%), childhood sexual abuse and bullying for mental disorders (5.0%), occupational risks for musculoskeletal disorders (9.6%), high fasting plasma glucose for diabetes and kidney diseases (76.1%) and neurological disorders (4.8%), air pollution for chronic respiratory diseases (34.2%), alcohol use for digestive diseases (20.2%), occupational risk for sense organ diseases (10.0%), and alcohol use for substance use disorders (51.4%). The non-leading risk factors for each subtype can be referred to in Figure 4.

Figure 4.

Global and regional proportion of age-standardized disability-adjusted life years (DALYs) attributable to risk factors for non-communicable diseases (NCDs) and their subtypes in 2021. Risk factors for each NCD subtype at a specific location are ranked left to right in decreasing order, with the leading risk factor labeled by its DALY percentage. X-axis labels (NCD subtypes) are ordered from left to right by descending global age-standardized DALYs in 2021. Y-axis labels (regions) are ordered bottom to top by decreasing NCD age-standardized DALYs in 2021. NCDs: Non-communicable diseases; DALYs: Disability-adjusted life years.

Regional level

The leading risk factors for each subtype across 21 regions varied: Cardiovascular diseases–high systolic blood pressure in 21 locations (range: 37.8%–58.1% DALYs from cardiovascular diseases attributed to this factor); Neoplasms–tobacco in 16 locations (12.4%–27.4%) and unsafe sex in 5 locations (6.6%–16.1%); Other non-communicable diseases–child and maternal malnutrition in 21 locations (0.0%–0.2%); Mental disorders–childhood sexual abuse and bullying in 21 locations (1.8%–7.1%); Musculoskeletal disorders–high BMI in 11 locations (9.4%–13.5%), occupational risks in 9 locations (7.6%–21.4%), and tobacco in 1 location (13.5%); Diabetes and kidney diseases–high fasting plasma glucose in 21 locations (65.7%–93.1%), also affecting neurological disorders in 20 locations (2.8%–6.5%) and 1 location with high BMI (3.8%); Chronic respiratory diseases–tobacco in 15 locations (14.6%–43.1%) and air pollution in 6 locations (11.8%–52.7%); Digestive diseases–high alcohol use in 20 locations (10.0%–34.4%) and high BMI in 1 location (4.3%); Sense organ diseases–occupational risks in 20 locations (5.5%–14.0%) and air pollution in 1 location (8.2%); and Substance use disorders–high alcohol use in 18 locations (50.2%–78.4%) and drug use in 3 locations (65.4%–83.7%). The non-leading risk factors and their ranking for each subtype are detailed in Figure 4.

National level

The leading risk factors for each subtype across 204 countries varied: Cardiovascular diseases–high systolic blood pressure in 202 countries (range: 35.7%–64.4% of DALYs from cardiovascular diseases attributed to this factor) and air pollution in 2 countries (36.5%–40.9%); Neoplasms–tobacco in 143 countries (6.9%–33.0%), unsafe sex in 53 countries (6.8%–18.9%), dietary risks in 5 countries (6.3%–7.9%), and high BMI in 3 countries (6.1%–9.5%); Other non-communicable diseases–child and maternal malnutrition in 204 countries (0.0%–0.4%); Mental disorders–childhood maltreatment in 202 countries (1.4%–10.5%) and intimate partner violence in 2 countries (1.6%–1.9%); Musculoskeletal disorders–high BMI in 108 countries (7.5%–17.4%), occupational risks in 80 countries (7.5%–30.6%), and tobacco in 16 countries (9.8%–17.2%); Diabetes and kidney diseases–high fasting plasma glucose in 203 countries (61.4%–96.3%) and kidney dysfunction in 1 location (64.1%); Neurological disorders – high fasting plasma glucose in 188 countries (2.3%–11.0%), high BMI in 11 countries (2.6%–5.2%), and alcohol use in 5 countries (3.0%–4.6%); Chronic respiratory diseases–tobacco in 122 countries (9.7%–43.5%), air pollution in 80 countries (10.7%–64.1%), and high body-mass index in 2 countries (13.1%–13.3%); Digestive diseases–alcohol use in 176 countries (3.0%–42.3%), high body-mass index in 20 countries (2.8%–11.7%), and drug use in 8 countries (5.7%–14.9%); Sense organ diseases–occupational risks in 198 countries (3.9%–19.5%), air pollution in 4 countries (5.2%–10.7%), and high fasting plasma glucose in 2 countries (6.0%–6.2%); and Substance use disorders–alcohol use in 162 countries (50.3%–91.8%) and drug use in 42 countries (50.9%–89.7%). The non-leading risk factors and their ranking for each subtype are shown in Figure S10 (http://links.lww.com/JS9/D773).

Sex, age, and SDI disparities

Disparities in disease burden

For NCD burden, overall, age-standardized prevalence, deaths, and DALYs all increase with age, with prevalence being higher in women and deaths and DALYs being higher in men. In 2021, the age-standardized prevalence of NCDs increased with age for both genders, being consistently higher in females until age 60, when the gap began to close (Fig. S11, http://links.lww.com/JS9/D773). Age-standardized deaths from NCDs also rose with age, more sharply after 70, and were consistently higher in males, with a widening gender disparity beginning from age 50–54 (Fig. S12, http://links.lww.com/JS9/D773). Age-standardized DALYs climbed with age for both genders, with a notable widening gap favoring males after age 45. This male predominance tapers after 90, whereas the steady rise continues in females (Fig. 3B).

Age-standardized DALYs generally declined with higher SDI among regions, although there were minor increases at SDIs of 0.4, 0.55, and 0.7. Western Europe, Australasia, high-income Asia Pacific, Central Latin America, Andean Latin America, and Western sub-Saharan Africa showed lower-than-expected age-standardized DALYs for their SDI. In contrast, Oceania, North Africa and the Middle East, Eastern Europe, and high-income North America reported higher-than-expected rates (Fig. 3C). National age-standardized DALYs in 2021 slightly increased with socioeconomic progress up to an SDI of 0.5, then significantly decreased with further SDI improvements. Somalia, Nauru, and the United States of America experienced higher-than-expected burdens, while Niger, Australia, and Singapore had lower burdens (Fig. S13, http://links.lww.com/JS9/D773).

Disparities in risk factors

Men had higher NCD age-standardized DALY percentages than women for most attributable risk factors across all age groups in 2021, except for unsafe sex, intimate partner violence, low physical activity, and high BMI. Additionally, each age’s leading risk factor can differ by sex. Specifically, for ages 1–9, kidney dysfunction posed the highest risk for both genders, although men faced additional environmental risks from ages 5–9. For ages 10–19 and 20–29, the predominant risks were childhood sexual abuse and bullying, and drug use, respectively, affecting both sexes. Between ages 30–74, sex disparities were more pronounced: men consistently faced the highest risk from tobacco, whereas women had a higher BMI until age 54, with the leading risk shifting to dietary risks at ages 45–49 and high systolic blood pressure until age 74. After age 75, both sexes exhibited the highest risk from high systolic blood pressure (Fig. 5).

Figure 5.

Percentage of disability-adjusted life years (DALYs) from non-communicable diseases (NCDs) attributable to level 2 risk factors, by age and sex, in 2021. Top row in each cell displays women’s percentage of DALYs attributed to a specific risk factor; the bottom row shows the percentage difference between sexes (men vs. women). Black labels indicate higher percentages in men; white labels denote lower. Blue rectangles highlight the top risk factor for women at each age stage, and yellow rectangles for men. X-axis labels (risk factors) are ordered from left to right by decreasing percentage of NCD age-standardized DALYs in 2021 at the global level. NCDs: Non-communicable diseases; DALYs: Disability-adjusted life years.

In both low and low-middle SDI regions, air pollution emerged as the dominant risk factor for age-standardized DALYs. In contrast, high systolic blood pressure was the primary risk factor in both middle and high-middle SDI regions. For high SDI regions, tobacco was the highest risk factor. Notably, drug use stood out in high SDI regions, exceeding other risks by over four-fold (Fig. S14, http://links.lww.com/JS9/D773).

Discussion

In 2021, NCDs caused 7.3 trillion cases, 43.8 million deaths, and 1.73 billion DALYs worldwide. Major contributors to age-standardized DALYs were cardiovascular diseases, neoplasms, and other NCDs, with sharp increases in diabetes and kidney diseases since 1990 and recent spikes in mental disorders, highlighting urgent health priorities. Key risk factors included high blood pressure, poor diets, tobacco use, and rising high BMI, indicating critical intervention areas. Additionally, regionally and nationally, cardiovascular diseases dominated in 16 regions and 159 countries, followed by neoplasms in 4 regions and 31 countries. Diabetes and kidney diseases were the leading subtypes in 6 countries; mental disorders in 1 region and 6 countries; and musculoskeletal disorders in 2 countries. Leading risk factors were high systolic blood pressure (9 regions, 101 countries), high fasting plasma glucose (3 regions, 46 countries), tobacco use (3 regions, 30 countries), high BMI (4 regions, 30 countries), air pollution (2 regions, 15 countries), and dietary risks (2 countries). These variations call for locally-adapted strategies. Moreover, gender disparities reveal that while women have a higher NCD prevalence, men experience greater mortality and DALYs, suffering more from all factors except unsafe sex, intimate partner violence, low physical activity, and high BMI. Each gender has unique risk factor trajectories over their lifespan, necessitating tailored prevention approaches. Generally, increasing SDI correlates with decreasing age-standardized DALYs, though variations with specific risk factors highlight the need for customized interventions.

The NCD prevalence rate decreased for the first time since 1990 in 2021^[21], possibly reflecting the impact of long-term public health initiatives. Yet, interpreting this trend requires caution, as COVID-19 likely limited healthcare access and changed health-seeking behaviors, resulting in underdiagnoses, delayed treatments, and increased mortality among those with existing NCDs, thus skewing NCD prevalence. Therefore, it is premature to conclusively attribute this decline to successful health strategies, with ongoing monitoring needed to determine if this is a lasting trend or a temporary pandemic-related anomaly. Notably, since 2015, the global age-standardized mortality rate for NCDs has only fallen by 5.7%^[4], well below the SDG 3.4 target. As the global and aging populations grow^[22], the NCD challenge will require sustained or increased attention. To address this, we advocate for strengthening healthcare infrastructure by expanding long-term care facilities, improving access to specialized and preventive care, enhancing routine screenings for early detection, promoting community-based health programs, and leveraging technology and innovation for more effective chronic disease management. Additionally, our study underscores the importance of proactive NCD prevention, focusing on major risk factors such as high systolic blood pressure, dietary risks, tobacco use, and high BMI. While previous research emphasized tobacco, alcohol, and blood pressure^[23,24], we highlight the increasing significance of dietary risks and high BMI, with dietary risks now the second leading risk factor. To foster healthier eating habits, we recommend implementing nutritional education programs, regulating food marketing, subsidizing healthy foods and taxing unhealthy ones, enhancing food labeling, and supporting school-based interventions.

For NCD subtypes, diabetes and kidney diseases have seen significant increases in age-standardized DALYs, yet global awareness and primary care detection of these conditions remain low^[25,26]. Their advanced stages cause substantial burdens^[27], underscoring the need for intensive health policy interventions. Future efforts should prioritize enhancing public awareness campaigns, integrating routine screening for these diseases into primary care, training healthcare professionals in early detection and management, and improving access to effective treatments. Notably, the age-standardized DALY rates for mental disorders remained stable from 1990 to 2019^[28], but experienced a 9.4% increase in 2021. This sharp rise highlights the urgent need to bolster mental health systems by increasing funding, integrating mental health care into primary services, expanding access to professionals, launching public awareness campaigns to reduce stigma, and developing community-based interventions. Moreover, the increasing global prevalence of musculoskeletal and skin diseases demands targeted action. Incorporating musculoskeletal disease management into national NCD policies can address existing oversights, while ultraviolet risk assessments and educational programs are critical for combating these conditions^[29,30]. Additionally, strategies should include regulating occupational exposure locally and supporting global initiatives to reduce ozone-depleting substances. Despite these emerging concerns, cardiovascular diseases and neoplasms remain major contributors to the global burden, necessitating sustained efforts.

Health inequalities are evident from the over two-fold regional and three-fold country variations in age-standardized DALYs. Nearly half of the countries and regions primarily face cardiovascular diseases, while others contend with neoplasms, diabetes, kidney diseases, mental disorders, or musculoskeletal disorders. Risk factors vary as well, with high systolic blood pressure prevalent in nearly half, and others affected by high fasting plasma glucose, tobacco, high BMI, air pollution, and dietary risks. These disparities underscore the need for location-specific strategies to effectively allocate resources to address leading NCD subtypes and their associated risk factors, which can differ even within the same leading subtype. For instance, in the United States and Solomon Islands, where cardiovascular disease is prevalent, health expenditures should particularly target this subtype. However, preventive efforts should focus on the specific dominant risk factors–high blood pressure in the United States and air pollution in the Solomon Islands. Likewise, in the United Kingdom and Peru, where cancer is the leading NCD, resources can be appropriately allocated. Nonetheless, the focus on risk factors should differ: targeting tobacco use in the United Kingdom and unsafe sex in Peru.

Gender disparities in NCDs show that women have a higher age-standardized prevalence, while men experience higher age-standardized mortality and DALYs. These gender differences extend beyond biological aspects, heavily influenced by gender-specific social determinants and behaviors^[31]. Men are more affected by risk factors such as tobacco, alcohol, and unhealthy diets, whereas women incur more DALYs from unsafe sex, intimate partner violence, low physical activity, and high BMI. To address these issues, gender-responsive health education and services are essential. For men, gender-responsive health education should prioritize reducing tobacco and alcohol consumption and promoting healthy diets. For women, the emphasis should be on increasing physical activity and managing high BMI. Addressing disparities in intimate partner violence for women involves strengthening legal frameworks, supporting survivors, and promoting education on healthy relationships. Similarly, to reduce disparities related to unsafe sex among women, enhancing access to sexual education and contraceptive services, as well as integrating sexual and reproductive health into primary care, are critical steps. Additionally, age-specific interventions are critical for each gender. For children under 10, nutritional guidance is essential to prevent kidney issues, with male children facing other environmental risks between the ages of 5-9. Adolescents (10–19) need mental health support and protection against sexual abuse and bullying. Young adults (20–29) should be targeted for drug abuse prevention and treatment. Adults (30–74) require monitoring for tobacco use in males and for blood pressure, dietary risks, and BMI in females. For seniors (over 75), community programs to manage high systolic blood pressure are vital. Therefore, it is imperative to implement and support age-specific health policies that accommodate this age-gender trajectory and allocate resources efficiently. Simultaneously, collecting and analyzing disaggregated data by age and gender will help pinpoint trends and fine-tune local policies effectively.

With rising SDI, age-standardized DALYs for cardiovascular and chronic respiratory diseases increase, but for neoplasms, they decrease^[32]. Overall, we found DALYs for NCDs tend to decrease as SDI rises, with lower SDI regions exhibiting higher disease burdens. These regions should prioritize sustainable, cost-effective strategies such as health education, behavior change initiatives, and integrative medicine. Additionally, employing affordable screening tools and forging partnerships with international organizations for resource support are crucial for NCD management. Different SDI levels present distinct primary risk factors. Air pollution critically affects low and low-middle SDI areas, significantly impacting mortality and life expectancy through diseases such as ischemic heart disease, COPD, cerebrovascular disease, and lung cancer^[33]. Globally, nearly two-thirds of avoidable life expectancy losses from anthropogenic air pollution are attributed to fossil fuel use, highlighting the urgent need for pollution control in these regions^[34]. Key measures include promoting clean cooking and heating, improving public transport, reducing vehicle and industrial emissions, expanding renewable energy, strengthening waste management, and minimizing fossil fuel use. Conversely, high systolic blood pressure is a dominant risk factor in middle and high-middle SDI regions, while high SDI regions face significant challenges from tobacco use. Addressing hypertension requires expanding healthcare access and reinforcing public health campaigns promoting healthy lifestyles. Unlike natural pollutants like eolian dust, the tobacco epidemic is man-made and preventable^[35]. Strengthening tobacco marketing and sales regulations, implementing economic disincentives, enforcing public smoking bans, and offering free or subsidized cessation programs can significantly reduce its impact.

Limitations and future directions

However, the study faces data-related and methodological limitations inherent in the GBD estimation, including the reliance on models in areas without comprehensive vital records, affecting data accuracy. Data sparsity, unreliability, and variations in reporting quality–particularly in western, eastern, southern, and central sub-Saharan Africa and South Asia–along with differences in healthcare infrastructure and cultural factors across countries, and varying definitions of risk factors, may impact local estimation accuracy and global aggregation, requiring caution in interpretation. CODEm-related challenges, such as outlier management, hinder mortality rate precision. The methodology for evaluating the combined effects of risk factors, such as diet, might oversimplify complex interactions. Data scarcity on factors such as environmental pollutants, violence, and the neglect of social determinants such as education and poverty further limit validity. The attributable DALYs were estimated for individual risk factors, but the findings’ applicability may be constrained by the coexistence and interplay of multiple risk factors within populations. Future studies should investigate coexisting risk profiles and develop tailored interventions for specific regions. Additionally, two analytical limitations exist. Percentage change is straightforward but misses temporal dynamics and inflection points. Future research should use advanced methods like time-series analysis for specific regions to identify key shifts and link trends with major global events or health interventions for deeper insights. Smoothing spline models capture overall trends but may miss abrupt changes; future studies could explore alternative methods to provide deeper insights into such variations. Moreover, several scope and generalizability limitations exist. Our study has mapped NCDs and level 2 subtypes comprehensively but has not covered specific level 3 diseases like stroke or COPD, nor has it examined risk factors beyond level 2. Future research should analyze the burden of these specific diseases and broader risk factors. Future research should explore the underlying factors contributing to regional and national disparities in NCD burdens, such as policy, environmental, genetic, and healthcare-related variables, to inform effective health policies and interventions. While our global analysis highlights overarching trends and extreme cases, it lacks the specificity needed for localized strategies. Future research should build on these findings to conduct region-specific reviews, commentaries, or research and develop tailored public health recommendations.

Conclusion

Despite declining age-standardized prevalence, mortality, and DALY rates since 1990, NCDs continue to pose a significant health challenge, revealing a substantial lag in achieving SDG 3.4 and highlighting the need for more effective preventive strategies. This is particularly crucial for key NCD subtypes such as cardiovascular diseases, neoplasms, and increasingly prevalent conditions like diabetes, kidney diseases, and mental disorders. Addressing major risk factors–high blood pressure, diet, and tobacco use–is essential for reducing the NCD burden. Regional and national health inequities in NCDs and subtype burdens and their risk factors emphasize the need for region-specific or collaborative approaches to effectively combat the global NCD challenge. Addressing local diversity requires interpreting findings within specific contexts and tailoring public health initiatives to local factors. Future reviews, commentaries, and research should build on the findings from each location to develop actionable strategies that address specific needs effectively. Gender and SDI health disparities exist in disease burden; the identified gender-age-based and SDI-development-based trajectories of dominant risk factors should guide targeted interventions and policy formulation.

Ethical approval

This study involves the analysis of publicly available, de-identified data from the Global Burden of Disease Study. Given the secondary nature of the data, which does not involve direct interaction with human participants or access to identifiable personal information, ethical approval and informed consent were not required for this analysis.

Consent

As this study uses open-access data, patient consent is not required.

Sources of funding

All the authors declare to have received no financial support or sponsorship for this study.

Author’s contribution

J.L. analyzed the data and drafted the manuscript. D.Y.T.F. is the principal investigator of the study and is responsible for its overall conduct. D.Y.T.F., V.P., P.M.D., Y.S., and N.N.C. critically reviewed and approved the manuscript, and assume responsibility for its contents.

Conflicts of interest disclosure

All the authors declare to have no conflicts of interest relevant to this study.

Research registration unique identifying number (UIN)

Not applicable.

Guarantor

Daniel Yee Tak Fong.

Provenance and peer review

Externally peer-reviewed.

Data availability statement

The data used for the analyses in the study are publicly available at https://ghdx.healthdata.org/gbdresults-tool.

Presentation

None.