Comparative analysis of metabolic risks and related non-communicable disease burden across sociodemographic index regions from 1990 to 2021

Abstract

Objective

To analyse the trends and differences of metabolic risks related non-communicable diseases (NCDs) globally and across various sociodemographic index (SDI) regions from 1990 to 2021.

Design

Observational study.

Setting

The data of global and all SDI regions were obtained from the Global Burden of Disease Study 2021 database.

Main outcome measures

Age-standardised deaths rate and disability-adjusted life-years (DALYs) percentage of NCDs attributable to metabolic risk were analysed worldwide and across SDI regions from 1990 to 2021. Besides, we analysed the regional trends of metabolic risk summary exposure value (SEV). A detailed analysis of the specific disease burden of various body systems caused by metabolic risks was also provided.

Results

From 1990 to 2021, all regions showed increased SEV for metabolic risks, with the most notable growth in middle (total change: +84.8%, 95% uncertainty intervals (95% UI) 72.6% to 96.0%) and low-middle SDI regions (+84.5%, 95% UI 70.5% to 95.5%). While high SDI regions had the highest SEV (29.916 in 2021, 95% UI 26.968 to 32.237), they experienced significant reductions in mortality (−49.6%, 95% UI –53.5% to −46.2%) attributable to metabolic risks. DALYs associated with metabolic risks, particularly high body mass index (HBMI), increased notably, especially in less-developed regions (DALYs of NCDs due to HBMI in low-middle SDI regions: +99.9%, 95% UI 76.4% to 114.5%). In these areas, notable DALY rises were observed for conditions including cardiovascular diseases, digestive diseases and particularly metabolic risk-related cancers (>80.0%), reflecting a distinctive shift in the distribution of NCD burdens related to metabolic risks.

Conclusions

The rising metabolic risk exposure and associated NCD burden, particularly in less developed regions, underscores the need for targeted public health interventions to mitigate these growing health challenges.

Strengths and limitations of this study.

• This study uses comprehensive data from the Global Burden of Disease Study 2021.

• We analysed trends in specific non-communicable diseases (NCDs) burden for various body systems associated with metabolic risks.

• Evaluates trends across five sociodemographic index regions and highlights regional disparity in NCD burden due to metabolic risks.

• Urban–rural differences were not taken into account in the study.

• The study relies on secondary data, which may have inconsistencies.

Introduction

Metabolic risk (MR) factors have emerged as critical determinants of global health, significantly contributing to the prevalence and severity of non-communicable diseases (NCDs) such as cardiovascular diseases (CVDs), diabetes, certain cancers and chronic respiratory conditions. These risk factors encompass high body mass index (HBMI), elevated blood pressure, high blood glucose levels and abnormal lipid profiles and have become increasingly prevalent over the past three decades.¹,⁵ The Global Burden of Disease (GBD) Study 2021 revealed a steady increase of 1.50% annually in MRs exposures between 1990 and 2021.⁵ The global rise in MRs is closely tied to lifestyle changes, urbanisation and economic development, which have collectively altered dietary habits and physical activity patterns, particularly in low-income and middle-income countries undergoing social transformation.

The economic and social burdens imposed by MRs are profound, straining healthcare systems and economies worldwide.^{6 7} As NCDs driven by these risks account for a significant portion of morbidity and mortality, understanding the dynamics of MR factors across different regions and sociodemographic groups is crucial. Previous research has highlighted substantial disparities in the burden of NCDs attributable to MRs. High sociodemographic index (SDI) countries reported an annual reduction of type 2 diabetes mellitus (T2DM)-associated death rates by 1.76%, while middle SDI countries experienced an increase of 0.49%.⁸ Similarly, Hu et al found that in 2019, the highest age-standardised rate of mortality attributable to MRs was in low SDI regions, contrasted by the lowest in high SDI regions.⁹ More alarmingly, it has been reported that the insufficient control of MR factors in less developed regions has led to a majority of ischaemic heart disease (IHD) related deaths shifting from developed to developing countries.³ These results suggest that MRs’ impact varies widely by region and over time.

Despite the recognition of the critical role MRs play in shaping health outcomes, there remains a gap in comprehensive, region-specific analyses that track these changes over extended periods. The existing research can be roughly divided into three categories. The first category focuses on the impact of MRs on all-cause mortality and disability.^{9 10} The second category is burden analysis for specific risk factor and specific diseases, such as cancer burden analysis attributed to HBMI.¹¹ The third category involves analysing the burden of metabolic diseases themselves, such as type 1 diabetes.¹² And on this basis, analysis for specific ages and regions.^{10 13 14} Overall, there is less focus on NCDs as a whole, and few studies have revealed the spectrum of affected diseases in detail. The GBD Study 2021 offers a robust dataset that spans from 1990 to 2021,⁵ providing an unparalleled opportunity to systematically explore the evolution of MRs and their impact on NCDs across various SDI regions.

This study aims to analyse the differences and changes in the impact of MRs on NCDs across different SDI regions over the past 30 years. By assessing age-standardised summary exposure value (SEV), mortality and disability-adjusted life-years (DALYs) attributable to MRs, we seek to uncover trends and shifts in the burden of these risks. Our goal is to provide insights that can inform public health strategies and policies tailored to the unique challenges and experiences of different regions. Understanding these patterns is essential for developing targeted interventions that address the specific needs of diverse populations. As MRs continue to rise, particularly in less-developed regions, identifying effective prevention and mitigation strategies is imperative. This study’s findings are expected to offer valuable guidance for policy-makers and public health practitioners aiming to reduce the economic and social burdens of NCDs and improve health outcomes globally.

Method

Data source

Data were downloaded from the results tool of GBD 2021: https://vizhub.healthdata.org/gbd-results/. The GBD 2021 study covers the burden of 459 diseases, injuries and risk factors in 204 countries and territories worldwide from 1990 to 2021, including major diseases and injuries such as CVDs, cancers, respiratory diseases and traffic accidents, as well as risk factors including behavioural risks, environmental risks and MRs. The data are disaggregated by age, gender and time periods. There are multiple burden indicators: deaths, years of life lost (YLLs), years lived with disability (YLDs), DALYs, prevalence, incidence, healthy-adjusted life expectancy and SEV for risk factors.^{4 5}

Definition and explanation of indicators

GBD 2021 estimates the relative risk (RR) between risk factors and outcomes based on published systematic reviews and meta-analyses and calculates the SEV for each risk factor through RR weighting. SEV represents the population’s exposure level to a particular risk factor, ranging from 0 (no risk) to 100 (highest risk level).⁵ The estimation of the disease burden attributable to MR factors involves a comprehensive process that quantifies deaths and DALYs attributable to risk factors. DALYs are the sum of YLLs and YLDs. One DALY can be interpreted as 1 year of healthy life lost.¹⁵

GBD uses the Cause of Death Ensemble Model (CODEm) to calculate cause-specific mortality. CODEm integrates multiple statistical models, systematically evaluates various covariate combinations, and then combines these results to estimate the number of deaths for each cause in specific locations, ages, genders and years.¹⁶ GBD study establishes causality between diseases and risks using the theoretical minimum risk exposure level (TMREL). TMREL represents the hypothetical minimum level of exposure to a specific risk factor at which the health damage associated with that risk factor would be minimised.¹⁷ Based on large-scale population surveys and published exposure data for each risk factor, a Bayesian network meta-regression model (DisMod-MR 2.1) and Spatio-temporal Gaussian process regression model are applied to aggregate the data and determine TMREL.^{18 19} Subsequently, the population attributable fraction (PAF) is used to quantify deaths and DALYs. PAF represents the proportion of the expected reduction in disease burden to the current disease burden when the population’s risk exposure level is reduced to TMREL. PAF is used to estimate attributable DALYs and deaths by gender, age, location and year.^{20 21}

SDI is a compound metric of sociodemographic development state strongly correlated with health outcomes, where 0 represents the minimum level of development and 1 represents the maximum. The 204 countries and territories were categorised according to SDI quintile into five groups: low SDI, low-middle SDI, middle SDI, high-middle SDI and high SDI.¹⁸ Age-standardised rate was calculated according to the global standard population, which is necessary when comparing populations from different locations or a sample population over time.²² The 95% uncertainty interval (UI) reflects the accuracy of the estimates, and when the 95% UI does not include 0, the estimate is considered statistically significant.¹⁰

GBD 2021 assessed a series of risk factors under three categories: behavioural risks, environmental/occupational risks and MRs. MRs include high fasting plasma glucose (HFPG), high low-density lipoprotein (LDL) cholesterol (HLDL), high systolic blood pressure (HSBP), HBMI, low bone mineral density and kidney dysfunction. The overall effect of MRs is obtained through comprehensive calculation. Detailed definition information and burden assessment procedures are provided in the original GBD studies.^{4 5}

Data analysis

This study extracted age-standardised SEV and the burden of NCDs attributable to MRs from the GBD database for the global and five SDI regions from 1990 to 2021, including age-standardised death rates and DALYs percentages. Stratified by SDI regions, specific analyses were conducted on: (1) Trends in age-standardised SEV for total MRs and specific risk factors. (2) Trends in age-standardised death rates and DALYs percentages for NCDs attributable to total MRs and specific risk factors. (3) Attribution proportions and evolution of total MRs and specific risk factors in overall NCDs DALYs and DALYs for specific diseases, including CVDs, diabetes, chronic kidney diseases, neoplasms, digestive diseases, respiratory diseases, neurological diseases, eye diseases and musculoskeletal diseases. Additionally, we compared rapidly developing countries with slowly developing developed ones based on the growth rate of per capita gross domestic product (GDP). This study used R V.4.2.2 for data processing and analysis.

Results

Age-standardised SEV

From 1990 to 2021, there was an increasing trend in MRs SEV across all regions (figure 1A). The more developed regions reported higher SEV over these years, varying from 29.916 (95% UI 26.968 to 32.237) in high SDI regions to 13.722 (95% UI 12.597 to 15.177) in low SDI regions, 2021. However, the total percentage change was highest in middle (+84.8%, 95% UI 72.6% to 96.0%) and low-middle SDI regions (+84.5%, 95% UI 70.5% to 95.5%). In comparison, it was less significant in high SDI regions, standing at +53.7% (95% UI 45.8% to 61.0%) (online supplemental table 1).

Figure 1. Age-standardised metabolic risks SEV, death rate of NCDs attributable to metabolic risks and DALYs percentage of NCDs attributable to metabolic risks across SDI regions from 1990 to 2021. Age-standardised SEV of (A) metabolic risks, (B) high body mass index and (C) high fasting plasma glucose. Age-standardised death rate of NCDs attributable to (D) metabolic risks, (E) high body mass index and (F) high fasting plasma glucose. Age-standardised DALYs percentage of NCDs attributable to (G) metabolic risks, (H) High body mass index and (I) High fasting plasma glucose. DALYs, disability-adjusted life-years; NCDs, non-communicable diseases; SDI, sociodemographic index; SEV, summary exposure value.

The SEV of HLDL, low bone mineral density and kidney dysfunction has largely stabilised across all regions in the recent 5 years (online supplemental figure S1B–D). Notably, the high SDI regions showed a marked reduction in SEV for HSBP and HLDL between 1990 and 2010, falling from 37.588 (95% UI 27.699 to 49.108) to 27.761 (95% UI 19.551 to 37.411) and from 60.196 (95% UI 42.232 to 82.108) to 54.092 (95% UI 37.249 to 74.637), respectively (online supplemental figure S1A,B). However, HSBP and HLDL remain the factors with the highest exposure across all regions, and SEV for HSBP showed an uptick since 2010 (online supplemental figure S1A) in high SDI regions, warranting continued focus on prevention.

The exposure to HFPG and HBMI has noticeably increased over the years (figure 1B,C). Actually, SEV of HBMI increased significantly across all regions with more developed regions facing higher exposure (SEV 2021: high 32.547, 95% UI 28.788 to 35.753; high-middle 25.430, 95% UI 22.566 to 28.993; middle 21.603, 95% UI 19.523 to 24.050; low-middle 16.898, 95% UI 15.476 to 18.516; low 12.457, 95% UI 11.556 to 13.660). The rise for HBMI was particularly dramatic in low-middle and middle SDI regions, with an increase of 108.1% (95% UI 89.8% to 121.8%) and 106.4% (95% UI 89.0% to 119.8%), respectively (online supplemental table 1). Furthermore, high SDI regions seemed better equipped at preventing low bone density, as opposed to regions with low SDI (online supplemental figure S1D).

Age-standardised death rate of NCDs

The age-standardised death rate of NCDs resulting from MRs generally followed a descending trend in all regions from 1990 to 2021 (figure 1D). This decrease was particularly striking in high (239.605 to 120.820 per 100000 population, −49.6%) and high-middle (344.113 to 229.889, −33.2%) SDI regions. Meanwhile, regions of middle, low-middle and low SDI changed slightly (middle: −15.2%, 95% UI−22.5% to−6.8%; low-middle: +1.5%, 95% UI-5.8% to 10.0%; low: −7.2%, 95% UI –14.3% to 1.7%). In these years, the death rate attributable to MRs was consistently lowest in areas with high SDI (figure 1D).

Mortality due to HSBP decreased in all regions, more markedly in high-middle (240.817 to 149.334 per 100 000 population, −38.0%) and high (156.309 to 60.461, −61.3%) SDI regions (online supplemental figure S1E, online supplemental table 1). In actuality, the mortality attributed to major MR factors saw considerable reductions in high-middle and high SDI regions. And, in 2021, mortality due to each MR factor was lowest in high SDI regions (figure 1D–F, online supplemental figure S1E–G). Contrastingly, mortality from HFPG and HBMI surged significantly in middle, low-middle and low SDI regions, where deaths resulting from HLDL and kidney dysfunction changed slightly (figure 1E,F, online supplemental figure S1F–G).

Despite the overall dip in mortality, HSBP-induced deaths maintained the top rank across all regions (online supplemental figure S1E). Furthermore, HBMI-induced deaths remained elevated in middle (+35.2% totally, 95% UI 11.2% to 50.0%), low-middle (+60.9%, 95% UI 37.6% to 79.5%) and low SDI (+36.4%, 95% UI 16.2% to 56.0%) regions (online supplemental table 1).

Age-standardised DALYs percentage of NCDs in multiple systems

Similar to the changes in mortality, the percentage of DALYs for NCDs caused by HSBP, HLDL and kidney dysfunction has stabilised in the past 10 years (online supplemental figure S1H–J). However, the contribution of HBMI and HFPG to NCDs DALYs is increasing rapidly, even in high SDI areas, where the DALYs due to HBMI are highest (figure 1H,I, figure 2A). Because of the respective ups and downs of each factor, the overall influence of MRs on NCDs DALYs did not change significantly (figure 2B).

Figure 2. Contribution of metabolic risks to age-standardised DALYs percentage of NCDs across SDI regions for 1990 and 2021. DALYs percentage of NCDs attributable to (A) HBMI and (B) combined metabolic risks for 1990 and 2021. DALYs percentage of diseases across various systems attributable to combined metabolic risks for (C) 1990 and (D) 2021. DALYs percentage of neoplasms attributable to metabolic risks in (E) high, (F) middle and (G) low SDI regions. DALYs, disability-adjusted life-years; HBMI, high body mass index; NAFLD, non-alcoholic fatty liver disease; NCDs, non-communicable diseases; SDI, sociodemographic index.

NCDs attributed to MRs encompass a wide range, including chronic respiratory diseases, neurological disorders, digestive diseases, musculoskeletal disorders, chronic kidney disease, CVD, diabetes mellitus (DM), neoplasms and sense organ diseases (figure 2C,D). Among these, CVD, DM and chronic kidney disease are the most significantly impacted diseases. Both in 1990 and 2021, the DALYs percentage attributed to MRs was consistently higher in more developed areas. However, this disparity has been narrowing over time, due to more remarkable increases in middle, low-middle and low SDI areas (figure 2C,D).

The DALYs percentage attributed to HFPG was the highest in both 1990 and 2021, primarily owing to its strong link with DM (online supplemental figure S2). The same is true of the high rankings for kidney dysfunction. DALYs linked to HBMI encompassed the broadest range of diseases, highlighting the extensive damage HBMI causes to multiple systems. Notably, from 1990 to 2021, there was a marked increase in the DALYs percentage due to HBMI across all regions (figure 1H). Furthermore, in both 1990 and 2021, the DALYs percentage of HBMI was directly proportional to the SDI, indicating a relatively heavier obesity burden in higher SDI areas (figure 2A).

Cardiovascular diseases

Between 1990 and 2021, globally the overall impact of MRs on CVD percentage increased slightly from 0.642 (95% UI 0.575 to 0.699) to 0.666 (95% UI 0.601 to 0.719) (figure 2C,D). HSBP remained the most influential risk factor. HLDL, HBMI and HFPG also contributed greatly (online supplemental figure S3A–C). Hypertensive heart disease, IHD and stroke are the most affected diseases across all regions. Notably, HBMI’s impact on these diseases increased over the years, especially in low and middle SDI regions. Both in 1990 and 2021, the higher the SDI in the region, the greater the disease burden of HBMI.

Neoplasms

HBMI and HFPG have been closely linked to various neoplasms (figure 2E–G). In 2021, uterine and pancreatic cancers were most affected. Other cancers of the digestive system also showed significant impacts. Over the years, the burden of these cancers due to MRs has increased across all SDI regions, with notable rises in middle and low SDI areas. HBMI is responsible for much of the growth. This trend calls for attention to liver cancer caused by HBMI (DALYs percentage in high SDI regions increased from 0.058 to 0.134, middle SDI from 0.025 to 0.077, low SDI from 0.026 to 0.054) in all regions. As of 2021, the more developed regions bear the greater burden of tumours caused by MRs.

DM and chronic kidney disease

The burden of T2DM is influenced by HBMI and HFPG, whereas T1DM is totally associated with HFPG (online supplemental figure S3D–F). From 1990 to 2021, there was an increase in T2DM burden due to HBMI. Nevertheless, high SDI regions are still the area with the heaviest burden. Chronic kidney disease, often secondary to DM and HSBP, showed similar regional trends and risk factor associations (online supplemental figures S3G–I).

Chronic respiratory diseases, digestive diseases, neurological disorders, sense organ diseases and musculoskeletal disorders

Gallbladder and biliary diseases, asthma, Alzheimer’s disease and other dementias, glaucoma, cataract, low back pain, gout and osteoarthritis are also impacted by MR factors (online supplemental figure S3J–L and S4). HBMI, HFPG and kidney dysfunction are contributors to these conditions. There has been a considerable increase in the burden of these diseases from 1990 to 2021, especially in less developed regions. The pattern of increase was consistent across different disease categories and regions. Similarly, the most developed regions have the highest DALYs of diseases now.

Social transformation and metabolic burden

According to GDP per capita data (in constant 2015 US dollars) from the World Bank for the years 1990 and 2021, this study highlighted the rapid economic ascension of countries such as China (US$905.032 to US$11 223.255, growth rate 11.401) and India (US$534.484 to US$1961.961, growth rate 2.671). In contrast, established economies like Switzerland (US$70 063.852 to US$88 520.322, growth rate 0.263) and Denmark (US$39 295.243 to US$59 205.647, growth rate 0.507) exhibited more consistent, moderate growth (see online supplemental GDP data).

Our findings indicate that in Switzerland and Denmark, the increase of exposure to MRs and HBMI was much lower than in China and India over the past 30 years (online supplemental figure S5A,B). Despite the difference in growth, the 2021 exposure levels in Denmark (MRs: 25.601, 95% UI 23.090 to 28.864; HBMI: 27.301, 95% UI 24.501 to 31.454) and Switzerland (MRs: 21.949, 95% UI 19.752 to 24.322; HBMI: 22.595, 95% UI 20.344 to 25.884) were still higher than those in China (MRs: 19.542, 95% UI 17.538 to 22.152; HBMI: 19.092, 95% UI 17.115 to 22.150) and India (MRs: 13.284, 95% UI 11.786 to 14.604; HBMI: 11.154, 95% UI 10.171 to 12.186).

Regarding mortality attributable to MRs, both Switzerland and Denmark have seen a notable decline, with a trend towards stabilisation in the past 5 years (table 1, figure 3A). The decrease in China has been more gradual (Switzerland −58.0% totally, 95% UI –62.7% to −54.2%; Denmark −63.3%, 95% UI –66.7% to −59.8%; China −20.1%, 95% UI –33.4% to −3.2%), while India shows a slightly growing trend (+7.4%, 95% UI –5.2% to 21.9%). Since 1996, mortality in Denmark and Switzerland has consistently been lower than those in China and India, showing an increasingly significant gap in recent years.

Table 1. Percentage change in age-standardised metabolic risks SEV, death rate of NCDs attributable to metabolic risks and DALYs percentage of NCDs attributable to metabolic risks in Denmark, Switzerland, China and India between 1990 and 2021.

	SEV 1990–2021	Death rate 1990–2021	DALYs percentage 1990–2021
Metabolic risks
Denmark	0.427 (0.311 to 0.552)	−0.633 (−0.667 to −0.598)	−0.389 (−0.469 to −0.320)
Switzerland	0.238 (0.142 to 0.336)	−0.580 (−0.627 to −0.542)	−0.267 (−0.348 to −0.200)
China	1.082 (0.910 to 1.240)	−0.201 (−0.334 to −0.032)	0.239 (0.122 to 0.344)
India	0.778 (0.616 to 0.952)	0.074 (−0.052 to 0.219)	0.199 (0.128 to 0.266)
High body mass index
Denmark	0.488 (0.361 to 0.613)	−0.339 (−0.421 to −0.229)	0.135 (0.033 to 0.233)
Switzerland	0.258 (0.162 to 0.358)	−0.399 (−0.486 to −0.270)	0.155 (0.066 to 0.250)
China	1.462 (1.135 to 1.702)	0.460 (−0.104 to 0.976)	1.559 (0.699 to 1.960)
India	1.242 (0.967 to 1.511)	1.180 (0.718 to 1.674)	1.504 (1.121 to 1.841)

Data in parentheses are 95% uncertainty intervals.

DALYs, disability-adjusted life-years; NCDs, non-communicable diseases; SEV, summary exposure value.

Figure 3. Comparison of age-standardised death rate of NCDs attributable to combined metabolic risks and high body mass index in Denmark, Switzerland, China and India from 1990 to 2021. (A) Age-standardised death rate of NCDs attributable to combined metabolic risks. (B) Age-standardised death rate of NCDs attributable to high body mass index. NCDs, non-communicable diseases.

The mortality related to HBMI in 1990 was initially much higher in Switzerland and Denmark compared with China and India. But this tendency changed over time, after 2014, the mortality becoming lower in Switzerland and Denmark. In contrast, for China and India, the mortality has been consistently rising in the last years, with expectations of steady increase (figure 3B). In addition, the contribution of HBMI to NCDs DALYs has raised significantly in China (+155.9%, 95% UI 69.9% to 196.0%) and India (+150.4%, 95% UI 112.1% to 184.1%) from 1990 to 2021 (table 1).

Discussion

Our study provides an in-depth analysis of the evolving impact of MR factors on NCDs across different SDI regions over the past three decades. The findings highlight significant trends and regional disparities, offering crucial insights for public health strategies and policies.

Increasing MRs

From 1990 to 2021, there has been a consistent rise in the age-standardised SEV of MRs globally, with a more pronounced increase in middle and low-middle SDI regions. HSBP and HLDL remain the primary exposures, despite recent stabilisation. However, the substantial rise in HBMI and HFPG, particularly in less developed regions, is concerning. This trend emphasises the urgent need for targeted interventions to combat the escalating obesity and diabetes epidemic in these areas.

Disparities in mortality and DALYs

Our analysis reveals a marked decline in age-standardised NCD death rates attributable to MRs in high and high-middle SDI regions, reflecting the effectiveness of public health measures and advanced healthcare systems. Conversely, low and low-middle SDI regions show minimal improvement, with some experiencing increases in mortality rates. This disparity underscores the ongoing challenges in healthcare access and policy implementation in these regions.

Similarly, the burden of NCDs DALYs due to MRs has decreased in high SDI regions but increased in lower SDI regions, driven primarily by HBMI and HFPG. The increase in DALYs is linked to a broad spectrum of diseases, particularly CVD, DM and chronic kidney disease. In particular, the cancer burden caused by MRs has surged over the past 30 years. The rising proportion of DALYs due to HBMI highlights the extensive impact of obesity on multiple organ systems, necessitating comprehensive obesity prevention and management strategies.

Socioeconomic development and metabolic burden

Our comparative analysis of countries with varying economic growth rates indicates that rapidly developing countries, such as China and India, have seen significant increases in MRs exposure, mortality and DALYs despite their economic progress. There are traces of the phenomenon. Countries transitioning from lower to higher incomes experience rapid urbanisation and shifts towards motorised transportation with consequent lower physical activity, higher prevalence of obesity and higher greenhouse gas emissions, which fuels the epidemic of metabolic burden.²³ On the other hand, developed regions show positive behavioural changes, emphasising balanced diets and active lifestyles.²⁴

In contrast, more stable developed economies like Switzerland and Denmark have successfully managed to control and reduce metabolic-related mortality and disability, showcasing the benefits of sustained public health efforts. Besides, the death rate of HSBP and HLDL declined significantly in high SDI regions, which further confirms the positive effects of high degree access to medical care and effective social health policies.²⁵ However, research shows that among low-middle and middle SDI countries, performance of effective coverage indicators for NCDs was far lower than levels reached for several communicable diseases and maternal and child health indicators—a pattern suggesting that many countries’ health systems and financing priorities are not moving as quickly as their epidemiological and demographic transitions.⁷ These highlight the need for these countries to integrate public health strategies that address the rising MRs alongside economic development.

Public health implications

The persistent and growing burden of MRs, especially HBMI and HFPG, calls for region-specific public health interventions, particularly in low and middle SDI regions. The growing epidemic of HBMI and HFPG has been attributed mainly to rapid transitions in diets, featuring a preference for highly processed foods and decreased consumption of high fibre foods including fruit and vegetables during the development of society.^{26 27} Take the USA as an example. Its grain processing industry solved the domestic grain surplus problem by mass-producing low-priced and high-calorie foods, resulting in a significant increase in the consumption of high-sugar and high-fat foods and driving up the obesity rate.²⁸ Social and cultural factors have also played a significant role. The promotion of high-calorie foods by modern communication means such as advertising and marketing has exacerbated the global obesity problem. For instance, in Latin America, the advertising volume of sugary beverages is significantly positively correlated with the increase in obesity rates.^{29 30} Low physical activity and high levels of mental stress also contribute.

Perhaps lessons can be learnt from the control of HSBP in developed countries, the mortality of which has continuously declined in the past years. A range of strategies, including primary care management and reductions of sodium intake, is known to be effective in decreasing the burden of it.^{31 32} For obesity, despite the global challenges, there are examples to draw from. For instance, Japan maintains relatively low obesity rates due to its low-fat and low-sugar dietary habits.³²,³⁶ In Latin America, particularly in Brazil, recent measures like taxing sugary beverages, promoting physical activity and providing dietary guidelines have gained international recognition in combating obesity,²⁹³⁷,³⁹ providing a model for other nations.

Based on existing evidence, measures to alleviate metabolic-related burden should focus on improving dietary habits, increasing physical activity and enhancing healthcare access. Our findings also suggest that policy efforts should not only address current burdens but also anticipate future trends. The stabilisation of certain risk factors in high SDI regions indicates that similar trends can be achieved in other regions with appropriate measures.

In detail, the first and foremost thing is to raise public awareness of MRs by various campaigns.^{40 41} Second, develop diet guidelines to promote healthy diet, emphasising raised fruit and vegetable proportion and reduced intake of unhealthy foods.⁴¹,⁴³ The effect will be better with some regulatory policies like consumption taxes, limitations on unhealthy food advertising and enhanced food tags.⁴⁴,⁴⁷ Besides, construct more public fitness facilities and provide additional sports facilities to encourage active participation in physical activities.^{40 48} In addition, formulate policies to guarantee quality healthcare facilities, strengthen the approachability and capacity of primary healthcare institutions, and foster the development of preventive healthcare,^{45 46 49} especially for less developed regions. Finally, effective monitoring and evaluation systems need to be established worldwide to promote international cooperation,^{50 51} collectively addressing the global challenges posed by MRs.

Our research focuses on the disparities of metabolic burden across regions with different levels of economic development, pinpointing specific challenges faced by region and suggested targeted solutions. This study delves into the relationship between societal transition and the burden of MRs, highlighting the inadequate attention to NCDs caused by MRs and the insufficient healthcare coverage during the rapid economic development process. This warrants attention from developing countries worldwide.

Limitations

However, our study has limitations. First, the GBD database calculates the combined effects of MRs, posing challenges regarding the coexistence of risk factors, mediating effects and calculation formulas. To address these challenges and enhance credibility, GBD incorporates the mediation factor to adjust for interactions.¹ But this adjustment method cannot fully capture all the complex interactions, especially the differences among regions and populations. Second, the impact of other modifiable risks such as diet and exercise on MRs is not taken into consideration in this analysis, which is closely related to MRs. Therefore, ignoring these factors may lead to an insufficiently comprehensive assessment of MRs and their impacts. Third, the urban–rural disparity and the limitations of SDI grouping. There are significant differences between urban and rural areas in terms of economic development, lifestyle and medical resources, especially in rapidly urbanising countries like China. Furthermore, the adoption of the SDI five-point system failed to capture the gradient differences within rapidly urbanising countries. These deficiencies may limit the interpretation and application of the research results. Finally, this study did not fully discuss the influence of factors such as region-specific culture and resources on MRs.

In order to assess MRs and related disease burden more accurately, future studies should further develop MR composite indices to replace single risk superpositions, so as to assess the combined effect of MR factors more accurately; establish the subgroup analysis framework divided by urban and rural areas and cultural circles; Integrate multidimensional data analysis of environmental factors and behavioural factors to provide more comprehensive and in-depth analysis results and suggestions for prevention and control. Furthermore, although we conducted the Joinpoint analysis, we failed to identify the key time points with statistical significance. This might be related to the heterogeneity among the multiple regions involved in the study. Therefore, we did not present this part of the results in the article.

Conclusions

In conclusion, our study offers a comprehensive analysis of the trends and impacts of MRs on NCDs globally over the past three decades. The significant disparities in metabolic burden across different SDI regions underscore the need for tailored public health strategies. Understanding these patterns will enable policy-makers and public health practitioners to develop effective interventions to reduce the economic and social burdens of NCDs, ultimately improving global health outcomes. The ongoing rise in MRs, particularly in less-developed regions, necessitates immediate and sustained action to prevent and mitigate the associated health challenges.

Data availability statement

Data are available in a public, open access repository.