Global trends and the case of China in Cardio-Kidney-Metabolic multimorbidity: development and validation of a composite risk index using GBD 1990–2021 data

Abstract

Background

The syndemic burden of cardiovascular disease (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM) represents a growing global health challenge, yet standardized tools for quantifying integrated disease burden remain lacking. We developed and preliminarily evaluated a novel Cardio-Kidney-Metabolic (CKM) Index to assess multimorbidity trends and to explore its potential to inform evidence-based policy interventions.

Methods

Using Global Burden of Disease Study 2021 data spanning 204 countries and territories from 1990 to 2021, we extracted age-standardized disability-adjusted life years (DALYs) for CVD, CKD, and T2DM. The CKM Index was constructed through logarithmic transformation (log[x + 1]), min-max normalization, and weighted integration based on global burden proportions and clinical significance (CVD = 0.5, CKD = 0.2, T2DM = 0.3) to generate a standardized 0-100 scale. We analyzed temporal trends using estimated annual percentage change (EAPC), performed joinpoint regression to identify trajectory inflection points, and conducted forecasting through 2030 using ARIMA and Bayesian ridge regression models. Sensitivity analyses evaluated index robustness across alternative weighting schemes, and clustering analysis identified distinct country trajectory patterns across socio-demographic index (SDI) regions.

Results

Global CKM-related DALYs increased dramatically by 60% from 343 million person-years in 1990 to 548 million in 2021 (EAPC + 1.51%). T2DM demonstrated the most explosive growth (EAPC + 3.57%), followed by CKD (+ 2.46%) and CVD (+ 1.18%). The CKM Index revealed marked global health inequalities, with high-SDI countries exhibiting the highest burden (70.7 in 2021) despite advanced healthcare systems, while middle-SDI regions showed concerning acceleration (reaching 77.1). Low-SDI regions demonstrated lower absolute burden (22.7) but concerning stagnation since 2005. China’s CKM Index rose substantially from 27.7 in 1990 to 67.6 in 2021, with joinpoint analysis revealing acceleration phases in 1998 and 2006, reflecting rapid epidemiological transition during economic development. Age-stratified analysis revealed 35% of total burden concentrated in individuals aged 60–74 years, with consistent male predominance. Pairwise comorbidity analysis showed CVD + T2DM combinations increasing fastest globally (EAPC + 1.45%), while CKD + T2DM burden nearly tripled. Forecasting models projected continued escalation through 2030, with middle-SDI regions reaching 86.6–87.5 and China plateauing at 66.7–77.5, though COVID-19 highlighted forecasting limitations under structural disruptions.

Interpretation

The CKM Index provides a promising methodological prototype for quantifying multimorbidity burden and reveals alarming global escalation trends with pronounced health inequalities. The syndemic nature of CKM diseases demands paradigmatic shifts from specialty-siloed to integrated chronic disease management. Countries achieving burden reductions within similar development contexts demonstrate the potential for coordinated policy interventions. Urgent global action is required to address the growing CKM crisis through enhanced surveillance, integrated care delivery, and targeted interventions addressing social determinants of health.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41043-025-01194-6.

Introduction

Over the past several decades, the global burden of non-communicable diseases (NCDs) has risen sharply, emerging as the leading cause of morbidity and mortality worldwide. Among these, cardiovascular diseases (CVDs), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM) not only contribute substantially to disease burden individually, but also frequently co-occur within the same patients, forming complex and interrelated clusters of chronic conditions [1–3]. These diseases share overlapping pathophysiological mechanisms, risk factor profiles, clinical trajectories, and therapeutic approaches, collectively constituting the core framework of the cardio-kidney-metabolic (CKM) axis [4, 5].

In recent years, growing attention has been directed toward CKM multimorbidity and its far-reaching public health consequences. However, current research in this domain has primarily focused on trend analyses of individual diseases or descriptive cross-sectional snapshots of selected populations. Systematic tools capable of comprehensively quantifying the integrated disease burden arising from the co-occurrence of CVD, CKD, and T2DM remain largely absent [6]. This methodological gap hinders a deeper understanding of multimorbidity complexity and limits the development of effective, evidence-based strategies for integrated intervention.

The challenges posed by CKM multimorbidity extend well beyond clinical complexity, fundamentally undermining the disease-centric and specialty-fragmented model that currently underpins chronic disease management. In resource-constrained settings, particularly in low- and middle-income countries, individuals with multiple chronic conditions often face major barriers to accessing continuous, coordinated, and personalized care. This results in accelerated disease progression, higher healthcare expenditures, and increased risk of premature mortality [7, 8]. In response, the World Health Organization and national health authorities have advocated for “integrated chronic care” and “person-centered multimorbidity management” as core components of national health policy. These approaches aim to reduce premature NCD-related mortality and overall disability-adjusted life years (DALYs) by strengthening primary care, promoting early detection and intervention, and building tiered healthcare delivery systems [9, 10].

While the Global Burden of Disease (GBD) study offers an extensive and standardized data infrastructure to support cross-country and temporal monitoring of key indicators such as incidence, prevalence, mortality, and DALYs [11], constructing composite indices that accurately reflect the extent of multimorbidity remains a methodological challenge. Traditional analytical techniques—such as Joinpoint regression and age–period–cohort modeling—are optimized for unidimensional disease trend analysis, and often fail to capture the structural dynamics of interrelated disease patterns or their spatial heterogeneity [12, 13].

To address this critical gap, we propose a novel metric: the CKM Composite Risk Index, which integrates age-standardized DALY burdens for CVD, CKD, and T2DM into a single interpretable score. This index aims to bridge the divide between the quantification of multimorbidity and the design of actionable health policy. It offers a standardized, dynamic, and spatially sensitive tool for capturing population-level CKM burden and advancing the global transition from single-disease management to systemic comorbidity control. Furthermore, by examining the relationships between the CKM Index and the Socio-demographic Index (SDI), this study seeks to identify developmental disparities in disease burden evolution. Such insights can help inform the design of stratified, equitable, and precision-oriented global strategies for NCD prevention and control.

Methods

Data sources and study population

This study utilized data from the Global Burden of Disease Study 2021 (GBD 2021), which provides comprehensive estimates of disease burden for 204 countries and territories from 1990 to 2021. We systematically extracted burden-related indicators for three major chronic diseases: cardiovascular disease (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM). Primary outcome metrics included incidence, mortality, disability-adjusted life years (DALYs), years of life lost (YLLs), and years lived with disability (YLDs).

All data were stratified by sex, age group, and geographic location. Countries were further categorized into five developmental levels based on the Socio-demographic Index (SDI), as defined by the Institute for Health Metrics and Evaluation (IHME). Data were retrieved from the official IHME platform and processed using standardized cleaning, coding, and harmonization protocols to ensure analytical consistency and comparability across time and regions.

Construction of the CKM composite risk index

Definition and objective

To quantify the cumulative population-level burden of cardio-kidney-metabolic (CKM) conditions, we developed a standardized metric termed the CKM Index. This composite indicator integrates age-standardized DALY rates from CVD, CKD, and T2DM, aiming to capture multidimensional chronic disease risk across the metabolic-cardiovascular-renal axis. The final CKM Index ranges from 0 (lowest burden) to 100 (highest burden), enabling intuitive interpretation and cross-country comparison.

Inclusion criteria and data processing

We extracted age-standardized DALY rates (per 100,000 population) for both sexes and all age groups from GBD 2021. Only DALYs expressed as “age-standardized rates” were included to ensure comparability. For each condition, we applied a natural logarithmic transformation with offset—logₑ(x + 1)—to reduce right-skewness and the influence of extreme values. These transformed values were then min–max normalized across all countries and years, producing a bounded 0–1 scale for each disease-specific burden component.

To construct the CKM Index, we applied a weighted linear combination of the normalized DALY rates for cardiovascular disease (CVD), type 2 diabetes mellitus (T2DM), and chronic kidney disease (CKD). The primary weights (CVD = 0.5, T2DM = 0.3, CKD = 0.2) were chosen based on two considerations: (i) the proportional contributions of each condition to the global DALY burden reported in the Global Burden of Disease Study (GBD 2021), where CVD accounted for approximately 75% of the combined DALYs across the three diseases; and (ii) prior literature identifying CVD as the dominant driver of morbidity and mortality within cardio-kidney-metabolic syndromes. A reduced weight for CVD (0.5 rather than its full DALY share) was assigned to balance its absolute burden with the increasing public health significance of T2DM and CKD in multimorbidity.

To ensure the robustness of the CKM Index, we conducted sensitivity analyses using three alternative weighting schemes:

(1) equal weights (CVD = CKD = T2DM = 1/3);

(2) burden-proportional weights based strictly on global DALY proportions in 1990, 2005, and 2021; and.

(3) expert-derived weights informed by meta-analyses of CKM-related outcome impact.

Across all schemes, temporal and cross-country burden patterns remained highly consistent (Pearson’s r = 0.90–0.98, all p < 0.001; Supplementary Table S1), confirming that the CKM Index trajectory is robust to moderate variations in weighting.

To enhance reproducibility, we provide the following explicit formulation of the CKM Index:

(1) For each disease d in country i and year t, DALY rates were first transformed using:

DALY’d, i,t = log(DALYd, i,t + 1).

(2) Then min–max normalization was applied across all countries and years:

Nd, i,t = [DALY’d, i,t - min(DALY’d)]/[max(DALY’d) - min(DALY’d)]

(3) The CKM Index was calculated as:

CKMi, t = 100 × (0.5 × NCVD, i,t + 0.3 × NT2DM, i,t + 0.2 × NCKD, i,t).

The resulting 0–100 scale allows direct comparison across years and geographic units; for example, a score of 70 represents a top-quartile global burden regardless of year.

To quantify uncertainty around the CKM Index, we generated 95% confidence intervals using nonparametric bootstrap resampling of country–year observations (1,000 iterations) at the global and SDI-region levels and recalculated the Index using the lower and upper bounds of the GBD 95% uncertainty intervals for disease-specific DALY rates. These analyses indicated that uncertainty bands did not materially alter temporal patterns or the relative ranking of countries and regions. For external validity, we examined cross-sectional correlations between country-level CKM Index values and independent CKM-related outcomes (age-standardized CVD and CKD mortality and major metabolic risk exposure indicators), which showed positive and statistically significant associations.

Justification for using dalys as the sole input metric

We selected disability-adjusted life years (DALYs) as the sole metric for constructing the CKM Index, rather than incorporating additional indicators such as mortality, years of life lost (YLLs), or years lived with disability (YLDs), for several methodological and conceptual reasons. First, DALYs are a composite metric that intrinsically integrates both fatal (YLL) and non-fatal (YLD) components of disease burden, offering a comprehensive measure of total health loss across populations. This holistic nature aligns closely with the goal of the CKM Index, which is to quantify the cumulative burden of chronic multimorbidity. Second, using a single, unified indicator avoids the risks of double-counting and scale incompatibility that can arise when combining heterogeneous outcomes. Third, DALYs are the primary outcome measure used by the Global Burden of Disease (GBD) framework for health policy evaluation and priority setting, and are reported in age-standardized formats, enabling robust temporal and cross-national comparisons. Finally, from a technical perspective, DALYs—expressed as continuous, bounded variables—are particularly well suited to logarithmic transformation and normalization procedures, which are essential to generating a stable and interpretable 0–100 composite index. For these reasons, DALYs represent the most appropriate, analytically coherent, and policy-relevant foundation for constructing a global multimorbidity index across the cardio-kidney-metabolic spectrum.

Pairwise and triple DALY aggregation

Because the Global Burden of Disease (GBD) Study does not report comorbidity-adjusted DALY estimates for combinations of diseases, the pairwise (e.g., CVD + T2DM, CKD + T2DM, and CVD + CKD) and triple (CVD + CKD + T2DM) DALY burdens analyzed in this study were calculated as simple conceptual aggregates by summing the individual disease-specific DALY values. These aggregated measures do not represent person-level multimorbidity-adjusted DALYs; instead, they provide a descriptive approximation of the shared syndemic burden across cardio-kidney-metabolic conditions and are used to illustrate co-occurrence patterns rather than to quantify exact multimorbidity prevalence.

Trend analysis and forecasting

For temporal trend estimation, we calculated estimated annual percentage changes (EAPCs) using age-standardized DALY rates to ensure comparability across years. EAPCs and their 95% confidence intervals were derived from log-linear regression models of the form ln(rate) = α + β·year, with EAPC = (exp(β) − 1) × 100.

Joinpoint regression was performed using the Joinpoint Regression Program (version 4.9.1.0, National Cancer Institute). Age-standardized DALY rates were used as the input series. We allowed a maximum of three joinpoints, consistent with the length of the 1990–2021 time series, and model selection was based on the Monte Carlo permutation test with overall significance level set at 0.05. For each segment, the annual percentage change (APC) and its 95% confidence interval were estimated.

We calculated annual CKM Index values for all countries and SDI-defined regions from 1990 to 2021. To project future burden, we initially fitted linear models using CKM Index as the dependent variable and year as the independent variable, extrapolating trends through 2030 under a status quo assumption.

To further assess forecast reliability, we trained four candidate models—including segmented regression, polynomial regression, ARIMA and Bayesian ridge regression—on the 1990–2010 subset. Their predictive performance was evaluated using R² values against observed CKM Index data in 2020–2021 and pre-pandemic years (2011–2019). The two best-performing models (ARIMA and Bayesian ridge) were subsequently re-trained using extended data (1990–2019) and applied to generate forecasts for 2030. Because the COVID-19 pandemic disrupted historical time-series patterns and reduced out-of-sample R² for several models, ARIMA and Bayesian ridge regression were retained based on their superior performance in pre-pandemic validation. Accordingly, all 2030 estimates are presented as exploratory, scenario-based projections rather than definitive forecasts.

Statistical software and analytical tools

All statistical analyses were conducted using the following platforms and packages:

Python 3.11 for data manipulation (Pandas), visualization (Matplotlib, Seaborn), and modeling (scikit-learn).

GeoPandas and Plotly for spatial visualization.

Joinpoint Regression Program v4.9.1.0 (National Cancer Institute) for trend break analysis.

Statistical significance was defined as p < 0.05. Where applicable, estimates are presented with 95% confidence intervals.

Results

Global CKM-Related disease burden trends analysis (1990–2021)

Individual disease burden trends

Based on the Global Burden of Disease (GBD) 2021 database, we systematically evaluated the temporal trends in disability-adjusted life years (DALYs) for three major CKM-related chronic diseases from 1990 to 2021: cardiovascular disease (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM).

Overall global trends

Over the 31-year period from 1990 to 2021, the combined DALYs burden of the three CKM diseases increased substantially from 343 million person-years to 548 million person-years, representing an estimated annual percentage change (EAPC) of + 1.51%. (Fig. 1) This demonstrates a significant and sustained upward trend in disease burden. Among the three conditions, T2DM showed the most dramatic increase, followed by CKD, while CVD, despite having the largest absolute burden, exhibited relatively slower growth rates.

Fig. 1.

Global TotaI DALYS (CKD + CVD + T2DM),1990–2021. Global trend of total Disability-Adjusted Life Years (DALYs) attributable to chronic kidney disease (CKD), cardiovascular disease (CVD), and type 2 diabetes mellitus (T2DM) from 1990 to 2021. The solid blue line represents the sum of DALYs across all Socio-demographic Index (SDI) regions. The shaded area indicates the 95% uncertainty interval. The results show a substantial and continuous increase in CKM-related disease burden over the 31-year period, consistent with an estimated annual percentage change (EAPC) of + 1.51%

Type 2 diabetes mellitus (T2DM)

Globally, T2DM-related DALYs increased from 24.93 million person-years in 1990 to 75.34 million person-years in 2021, with an EAPC of + 3.57%, representing the fastest growth rate among the three diseases. (Fig. 2) This dramatic increase can be attributed to the global rise in overweight and obesity rates, accelerated population aging, and urbanization-driven lifestyle changes. All regions demonstrated consistent upward trends, with particularly rapid burden increases in low-middle and low SDI countries, suggesting inadequate diabetes prevention and control systems in these regions.

Fig. 2.

Comparison of DALYs by Disease in 1990 and 2021.Comparison of global Disability-Adjusted Life Years (DALYs) attributable to Type 2 Diabetes Mellitus (T2DM), Cardiovascular Disease (CVD), and Chronic Kidney Disease (CKD) in 1990 and 2021. Bars represent DALYs in million person-years for each disease in 1990 (light-colored bars) and 2021 (dark-colored bars). The Estimated Annual Percentage Change (EAPC) from 1990 to 2021 is indicated above each pair of bars to reflect the average rate of change in disease burden over time. T2DM showed the fastest growth (EAPC = + 3.57%), followed by CKD (EAPC = + 2.46%) and CVD (EAPC = + 1.18%)

Cardiovascular disease (CVD)

CVD remained the largest contributor to CKM-related DALYs, with global burden increasing from 297.51 million person-years to 428.33 million person-years. Despite this substantial absolute increase, the EAPC was relatively modest at + 1.18%. This finding suggests that while CVD remains the predominant disease, it may have benefited from sustained investments in hypertension control, smoking cessation programs, and cardiovascular interventions in high SDI countries, resulting in relatively stable growth rates. However, in low-middle and low SDI regions, CVD burden continued to show significant increases, with high-salt diets, environmental pollution, and inadequate healthcare resources representing major threats.

Chronic kidney disease (CKD)

CKD-related DALYs increased from 20.74 million person-years in 1990 to 44.45 million person-years in 2021, with an EAPC of + 2.46%. As a “silent disease,” CKD continues to face challenges of insufficient early screening, low public awareness, and delayed medical intervention in most developing countries. This rapid growth trend is widespread globally and closely related to comorbidity mechanisms with diabetes and hypertension. Middle and low-middle SDI regions showed particularly pronounced growth curves, suggesting that kidney disease identification and intervention should become a new priority in global chronic disease management.

Comparative analysis of CKM disease burden trends: Global, China, and SDI regions (1990–2021)

To further identify disparities across different countries and socioeconomic development levels, we compared the temporal changes in DALYs for CVD, CKD, and T2DM between 1990 and 2021 across global, China, and five Socio-Demographic Index (SDI) regions. (Table 1)

Table 1.

Trends in dalys for cardiovascular Disease, chronic kidney Disease, and type 2 diabetes by region and SDI level (1990–2021)

Region	Disease	DALYs in 1990 (million)	DALYs in 2021 (million)	EAPC (%)
Global	Cardiovascular Disease	297.51	428.33	1.18
	Chronic Kidney Disease	20.74	44.45	2.46
	Type 2 Diabetes	24.93	75.34	3.57
	Total	343.18	548.12	1.51
China	Cardiovascular Disease	63.26	100.22	1.48
	Chronic Kidney Disease	4.2	6.13	1.22
	Type 2 Diabetes	3.96	11.47	3.43
	Total	71.42	117.82	1.61
High SDI	Cardiovascular Disease	59.85	53.39	−0.37
	Chronic Kidney Disease	2.9	7.12	2.90
	Type 2 Diabetes	5.04	12.42	2.91
	Total	67.79	72.93	0.24
Middle-High SDI	Cardiovascular Disease	84.27	110.11	0.86
	Chronic Kidney Disease	5.44	10.91	2.24
	Type 2 Diabetes	6.62	13.78	2.36
	Total	96.33	134.8	1.08
Middle SDI	Cardiovascular Disease	81.74	142.52	1.79
	Chronic Kidney Disease	7.03	15.7	2.59
	Type 2 Diabetes	7.75	26.09	3.92
	Total	96.52	184.31	2.09
Low-Middle SDI	Cardiovascular Disease	55.01	98.79	1.89
	Chronic Kidney Disease	4.69	10.61	2.63
	Type 2 Diabetes	5.01	18.04	4.13
	Total	64.71	127.44	2.19
Low SDI	Cardiovascular Disease	20.97	35.34	1.68
	Chronic Kidney Disease	2.57	5.04	2.17
	Type 2 Diabetes	2.37	6.47	3.24
	Total	25.91	46.85	1.91

DALYs = Disability-Adjusted Life Years; EAPC = Estimated Annual Percentage Change

Global versus China comparison

In absolute terms, China’s combined CKM disease burden increased from 71.41 million person-years to 117 million person-years over the 31-year period, with an EAPC of + 1.61%, exceeding the global average growth rate (+ 1.51%). Disease-specific analysis revealed:

T2DM

China’s DALYs increased from 3.96 million to 11.47 million person-years, with an EAPC of + 3.43%, slightly lower than the global rate (+ 3.57%) but still at a high level.

CVD

Burden increased from 63.26 million to 100 million person-years, with an EAPC of + 1.48%, higher than the global level (+ 1.18%), indicating that CVD remains the dominant chronic disease burden in China.

CKD

DALYs grew from 4.20 million to 6.13 million person-years, with an EAPC of + 1.22%, slightly lower than the global rate (+ 2.46%), but considering China’s population base, this growth rate remains concerning.

China’s overall trend is characterized by CVD dominance, rapid diabetes growth, and steadily increasing kidney disease burden, reflecting the combined impact of rapid population aging, lifestyle changes, and insufficient early prevention measures.

SDI regional disparities

The five global SDI regions demonstrated marked heterogeneity:

High SDI regions

CVD burden showed a slight decline (EAPC − 0.37%), while T2DM (+ 2.91%) and CKD (+ 2.90%) continued to rise significantly. This indicates that high-income countries, despite making progress in traditional cardiovascular control, face emerging burden centers in diabetes and its renal complications.

Middle-High and middle SDI regions

These regions carried the highest absolute burden of the three diseases globally, with significant growth rates. T2DM EAPCs reached + 2.36% and + 3.92% respectively, suggesting these countries are experiencing an epidemic phase of metabolic diseases.

Low-Middle and low SDI regions

Despite smaller baseline burdens, these regions showed extremely rapid growth rates. T2DM and CKD EAPCs in low SDI countries were + 3.24% and + 2.17% respectively, reflecting inadequate chronic disease management systems and insufficient basic healthcare infrastructure to support early screening and comorbidity interventions.

Trends in comorbid and multimorbid CKM disease burden (1990–2021)

To better characterize the syndemic nature of CKM-related conditions, we examined the temporal burden of disease combinations, including pairwise comorbidities and the co-occurrence of all three diseases—cardiovascular disease (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM). Using aggregate DALYs as a proxy for cumulative burden, we evaluated temporal patterns across the globe, China, and SDI-based regions.

Pairwise comorbidity trends

The burden attributable to pairwise combinations of CKM diseases increased consistently across all regions. (Table 2) Globally, the CVD + T2DM combination showed the steepest rise, with total DALYs increasing from 32.2 million in 1990 to 50.4 million in 2021, reflecting an EAPC of + 1.45%. The CKD + T2DM pair rose from 4.5 million to 12.0 million DALYs, with a similarly high EAPC (3.16%), underscoring the strong syndromic link between metabolic and renal dysfunction. The CVD + CKD pair, while contributing the highest absolute burden (from 31.8 million to 47.3 million DALYs), showed a relatively modest EAPC of + 1.28%, suggesting a more stable but still growing impact.

Table 2.

CKM combinations DALY trends (1990–2021)

Region	CVD + T2DM DALYs (million)		CVD + CKD DALYs (million)		CKD + T2DM DALYs (million)
Global	In 1990 322.4 In 2021 503.7	EAPC 1.45%	In 1990 318.2 In 2021 472.8	EAPC 1.28%	In 1990 45.7 In 2021 119.8	EAPC 3.16%
China	In 1990 62.7 In 2021 111.7	EAPC 1.65%	In 1990 67.4 In 2021 106.3	EAPC 1.48%	In 1990 8.2 In 2021 17.6	EAPC 2.51%
High SDI	In 1990 64.9 In 2021 65.8	EAPC 0.05%	In 1990 62.7 In 2021 60.5	EAPC −0.12%	In 1990 7.9 In 2021 19.5	EAPC 2.95%
Middle-High SDI	In 1990 84.3 In 2021 110.1	EAPC 0.87%	In 1990 83.1 In 2021 103.8	EAPC 0.72%	In 1990 8.3 In 2021 18.2	EAPC 2.58%
Middle SDI	In 1990 89.5 In 2021 168.6	EAPC 2.06%	In 1990 88.8 In 2021 158.2	EAPC 1.88%	In 1990 14.8 In 2021 41.8	EAPC 3.41%
Low-Middle SDI	In 1990 60.0 In 2021 116.8	EAPC 1.45%	In 1990 59.7 In 2021 109.4	EAPC 1.45%	In 1990 9.7 In 2021 28.7	EAPC 1.45%
Low SDI	In 1990 23.3 In 2021 41.8	EAPC 1.90%	In 1990 23.5 In 2021 40.4	EAPC 1.76%	In 1990 4.9 In 2021 11.5	EAPC 2.77%

DALYs = Disability-Adjusted Life Years; EAPC = Estimated Annual Percentage Change

In China, the pairwise DALYs burden increased in all combinations, with the fastest acceleration observed in the CKD + T2DM pairing (EAPC: +2.51%). This reflects the compounding impact of aging, dietary transitions, and poor metabolic control in a rapidly developing health system. The CKD + T2DM burden nearly tripled over the three decades, despite a lower baseline in 1990.

Low and low-middle SDI countries exhibited the steepest proportional increases across all pairwise combinations, particularly in CKD + T2DM, highlighting systemic gaps in early detection and integrated chronic disease management. In contrast, high SDI regions showed more stable trends, with modest increases or even plateauing in certain pairs, likely due to earlier implementation of cardiovascular and metabolic control strategies.

Joinpoint analysis of global and regional DALY rate trends, 1990–2021

Between 1990 and 2021, the global age-standardized Disability-Adjusted Life Year (DALY) rate increased slightly from 2,144.75 to 2,315.29 per 100,000 population. Joinpoint regression identified two inflection points in 2006 and 2014, indicating shifts in the global disease burden trajectory (Fig. 3).

Fig. 3.

Trends in Age-Standardized DALY Rates by Region and SDI Level, 1990–2021. This figure displays the temporal trends of age-standardized Disability-Adjusted Life Year (DALY) rates from 1990 to 2021 across global, national (China), and socio-demographic index (SDI)-stratified regions. The SDI levels include High SDI, Middle-High SDI, Middle SDI, Low-Middle SDI, and Low SDI, representing a composite measure of income, education, and fertility. Each line illustrates the corresponding annual DALY rate, enabling comparison of the disease burden evolution among different developmental and geographic contexts. Data were derived from the Global Burden of Disease Study 2021. Notably, disparities in DALY trends highlight persistent global health inequalities and evolving public health challenges

In contrast, high-SDI regions experienced a modest decline in DALY rate, from 2,568.77 in 1990 to 2,221.76 in 2021, with Joinpoints observed in 2006 and 2012, suggesting progress in disease prevention and healthcare interventions in high-income settings.

However, most other regions, including China and SDI-stratified groups such as middle-SDI and middle-high SDI, exhibited increasing DALY trends. China’s DALY rate rose from 2,023.29 in 1990 to 2,759.96 in 2021, with Joinpoints identified in 1998 and 2006, reflecting a shift in disease composition and aging-related burdens.

Middle SDI regions saw a notable increase from 1,867.41 to 2,509.03, and middle-high SDI from 2,752.02 to 2,966.56. Meanwhile, low-middle SDI increased from 1,857.11 to 2,211.24, indicating that the burden of disease has intensified across developing regions. Notably, only the low SDI group experienced a reduction, from 1,722.46 in 1990 to 1,397.68 in 2021, with Joinpoints at 1997 and 2007.

These results highlight divergent epidemiological transitions and underscore the growing burden in middle-income settings, despite some progress in low-SDI and high-SDI regions.

Age- and Cause-Specific burden of CKM-Related dalys in 2021

In 2021, the global disability-adjusted life years (DALYs) attributable to the cardio-kidney-metabolic (CKM) triad—comprising cardiovascular diseases (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM)—demonstrated a distinct age-dependent pattern with significant cause-specific disparities.

Overall, the total CKM-related DALY burden increased progressively with age, peaking between 65 and 74 years. The highest DALY burden was observed in the 65–69 age group (151.8 million DALYs), followed closely by the 70–74 (149.5 million) and 60–64 (130.5 million) age groups. Collectively, these three age categories accounted for over 430 million DALYs. Notably, individuals aged over 80 years contributed nearly 230 million DALYs, highlighting the substantial burden among the oldest age cohorts. (Fig. 4)

Fig. 4.

Global CKM-related DALYs by age and sex in 2021. Stacked bar chart illustrating the distribution of global disability-adjusted life years (DALYs) attributable to the cardio-kidney-metabolic (CKM) triad—including cardiovascular diseases, chronic kidney disease, and type 2 diabetes mellitus—across age groups and by sex in 2021. Each bar represents the total DALYs for a specific age group, with segments indicating the contributions from males (top) and females (bottom). The CKM burden increased markedly with age, peaking between 65–74 years. Across most age groups, males consistently bore a higher burden than females, while in the > 85 age group, female DALYs slightly exceeded those of males. These trends underscore the importance of sex- and age-specific strategies in CKM disease prevention and management

Sex-specific stratification revealed a consistently higher burden among males across most age groups, particularly between 50 and 74 years. However, among individuals aged > 85 years, female DALYs slightly exceeded those of males, likely reflecting longer life expectancy and cumulative disease burden.

Cause-specific decomposition showed that CVD remained the predominant contributor to DALYs across all age groups, especially after the age of 55 (Table 3). CKD and T2DM exhibited a more gradual age-associated rise, with T2DM becoming a notable contributor as early as the 30–44 age group and CKD emerging more prominently in individuals aged ≥ 60 years. Among individuals aged > 85 years, CKD contributed nearly as much as CVD, underscoring its growing impact in advanced age (Fig. 5).

Table 3.

CKM-related dalys by age and cause (2021)

Age Group	CKD (million)	T2DM (million)	CVD (million)	Total DALYs (million)
20–24	2.08	1.55	7.93	11.56
25–29	2.46	2.44	10.44	15.34
30–34	3.12	4.04	16.92	24.08
35–39	3.84	5.39	23.53	32.75
40–44	4.70	7.46	35.10	47.26
45–49	5.72	11.44	48.31	65.47
50–54	7.52	15.89	68.75	92.16
55–59	9.04	19.57	88.70	117.32
60–64	9.31	21.55	99.65	130.51
65–69	10.08	22.53	119.18	151.79
70–74	9.55	19.15	120.77	149.47
75–79	7.80	13.61	101.91	123.32
80–84	6.36	9.12	91.33	106.81
> 85	8.71	7.51	107.46	123.68

Fig. 5.

Global CKM-related DALYs by age group and cause in 2021. Stacked bar chart showing the age-specific distribution of global disability-adjusted life years (DALYs) attributable to the CKM triad—cardiovascular diseases (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM)—in 2021. Each bar represents the total DALYs for a given age group, with individual segments corresponding to the contribution from each cause. CVD was the dominant contributor across all age groups, particularly from age 60 onwards. T2DM exhibited a gradual increase starting in early adulthood, while CKD became increasingly prominent among individuals aged over 60. The highest overall burden was observed in the 65–74 age range. These findings highlight the evolving cause-specific burden of CKM conditions with advancing age and support the need for targeted prevention efforts

These findings emphasize the differential age- and cause-specific dynamics of CKM-related DALYs. They underscore the need for integrated, age-tailored prevention and management strategies, with particular attention to middle-aged and elderly populations where the CKM burden is most pronounced.

CKM composite risk index: temporal trends, validation, and forecasting

Standardized CKM index trends (1990–2021)

A composite CKM Index was constructed to quantify the integrated disease burden of cardiovascular disease (CVD), chronic kidney disease (CKD), and type 2 diabetes mellitus (T2DM) using age-standardized DALY rates from the GBD 2021 dataset. The index applied logₑ(x + 1) transformation, min–max normalization within each disease, and weighted integration (CVD = 0.5, CKD = 0.2, T2DM = 0.3) to generate a standardized 0–100 scale.

In 2021, the High SDI region reported the highest CKM Index at 70.7, reflecting a sustained burden of non-communicable diseases (NCDs) despite advanced health systems. Middle SDI and Low SDI regions showed divergent trends, with indices of 77.1 and 22.7, respectively—Middle SDI bearing a high burden associated with urbanization, while Low SDI exhibited lower absolute burden with limited progress since 2005. China’s CKM Index reached 67.6 in 2021, rising markedly from 27.7 in 1990 and 50.2 in 2005, underscoring a substantial escalation in chronic disease risk over the past three decades. Between 1990 and 2021, High SDI regions saw a notable increase of 21.0 points, whereas Low SDI regions declined by 12.0 points, suggesting regional disparities in CKM burden evolution and potential gaps in detection or control efforts.

Projected trends of the CKM index through 2030

Poor predictive performance for 2020–2021

To evaluate the feasibility of long-term forecasting for the Cardio-Kidney-Metabolic (CKM) Index, we systematically compared the predictive performance of four statistical models: segmented linear regression, polynomial regression, Bayesian ridge regression and ARIMA. Each model was initially trained on historical data from 1990 to 2010 and used to forecast the CKM Index in 2020 and 2021. Across all models and regions, the predictive accuracy was generally poor during this extrapolated period. In China, only the polynomial model yielded a slightly positive R² value (0.13), while all other models showed large negative R² values, including ARIMA (− 97.14) and Bayesian Ridge (− 2.37), indicating substantial deviation from actual observed values. Globally, none of the models achieved a meaningful fit, with polynomial regression recording an R² of − 123.4. (Table 4) These findings suggest that forecasting models trained solely on pre-pandemic data failed to account for the structural disruption caused by COVID-19 and underscore the limitations of traditional statistical models under unanticipated global shifts.

Table 4.

R² performance by region and model

Region	Model	R2 2011–2019	R2 2020–2021
China	ARIMA	−0.96	−97.14
	Bayesian Ridge	−0.20	−2.37
	Polynomial	0.18	0.13
	Segmented Linear	−0.26	−2.71
Global	ARIMA	−0.38	−190.11
	Bayesian Ridge	0.84	−24.08
	Polynomial	−0.01	−123.40
	Segmented Linear	0.84	−23.81
High SDI	ARIMA	−2.69	−219.06
	Bayesian Ridge	−0.03	−72.37
	Polynomial	−2.53	−233.40
	Segmented Linear	0.00	−70.67
High-Middle SDI	ARIMA	0.01	−65.17
	Bayesian Ridge	−16.75	−185.11
	Polynomial	−3.06	−276.12
	Segmented Linear	−17.23	−191.11
Low SDI	ARIMA	−8.41	−847.69
	Bayesian Ridge	−89.82	−11353.08
	Polynomial	−183.39	−24553.79
	Segmented Linear	−90.46	−11419.52
Low-Middle SDI	ARIMA	0.52	−349.38
	Bayesian Ridge	−1.02	−737.97
	Polynomial	0.93	−3.23
	Segmented Linear	−1.00	−733.12
Middle SDI	ARIMA	0.59	−61.22
	Bayesian Ridge	0.84	−14.72
	Polynomial	0.93	−0.82
	Segmented Linear	0.84	−14.61

Model validation using pre-pandemic (2011–2019) data

To better assess the predictive validity of statistical models under stable epidemiological conditions, we evaluated their performance over the pre-pandemic period from 2011 to 2019. All models were trained on historical data from 1990 to 2010 and assessed based on their ability to forecast CKM Index values during this subsequent decade.

At the global level, both Bayesian ridge regression (R² = 0.84) and segmented linear regression (R² = 0.84) exhibited strong predictive performance, indicating high alignment with observed trends in global chronic disease burden. In contrast, polynomial regression (R² = − 0.01) and ARIMA (R² = − 0.38) showed poor fits, likely due to their limited flexibility in modeling gradual long-term shifts.

In China, polynomial regression was the only model to achieve a positive R² (0.18), suggesting it captured part of the local upward trend in CKM burden. However, Bayesian ridge (R² = − 0.20), segmented linear regression (R² = − 0.26), and ARIMA (R² = − 0.96) performed poorly, indicating that region-specific variability and nonlinear dynamics may not be well modeled using generalized global approaches.

In middle-SDI regions, both polynomial regression (R² = 0.93) and Bayesian ridge (R² = 0.84) demonstrated exceptional performance, suggesting a predictable trajectory of CKM burden in rapidly transitioning populations. In contrast, high-SDI and high-middle SDI regions exhibited poor predictive alignment for all models (e.g., Bayesian ridge R² = − 0.03 and − 16.75, respectively), likely due to plateauing trends or structural irregularities in late-stage epidemiological transitions. For low-SDI regions, model performance was universally poor (e.g., polynomial R² = − 183.39), reflecting greater uncertainty, underreporting, and less stable disease trajectories.

Collectively, these findings underscore that model performance is highly region-specific. While Bayesian ridge and segmented linear models show utility at the global and middle-SDI levels, more adaptive or hybrid modeling frameworks are needed for reliable forecasting in high-variability settings such as low-SDI regions and China.

Given the reduced stability of post-2020 forecasts, the 2030 values are presented only as exploratory scenario-based projections and have been moved to the Supplementary Appendix (Supplementary Table S2).

Uncertainty and external validity

Uncertainty analysis

Using nonparametric bootstrap resampling of all country–year observations (1,000 iterations), the global mean CKM Index showed a narrow uncertainty range, with a 95% confidence interval of 47.2 to 51.1. Propagating the 95% GBD uncertainty intervals into the CKM Index yielded similarly small deviations and did not materially alter temporal trends or the relative ranking of countries and SDI groups, suggesting that the Index is stable with respect to plausible variations in input DALY estimates.

External validity

To examine external validity, we assessed correlations between the CKM Index and independent CKM-related outcomes. The CKM Index showed strong positive correlations with cardiovascular disease burden (r = 0.74) and type 2 diabetes burden (r = 0.81), and a weaker but positive association with chronic kidney disease burden (r = 0.27). All correlations were statistically significant.

Discussion

This study represents the first comprehensive attempt to develop and preliminarily evaluate a standardized composite metric for quantifying the syndemic burden of cardiovascular disease, chronic kidney disease, and type 2 diabetes mellitus across 204 countries and territories over three decades [1, 11, 14]. Our findings reveal alarming trends in global CKM multimorbidity, significant health inequalities across socio-demographic regions, and a growing need for paradigmatic shifts in chronic disease management strategies [15, 16].

Global escalation of CKM disease burden

The 60% increase in combined CKM-related DALYs from 343 million to 548 million person-years between 1990 and 2021 represents a public health crisis of unprecedented magnitude [6, 17]. This dramatic escalation reflects the confluence of demographic transitions, epidemiological shifts, and systemic failures in chronic disease prevention across diverse global contexts [18, 19]. The differential growth patterns among the three diseases provide critical insights into evolving disease dynamics. Type 2 diabetes mellitus emerged as the most rapidly expanding component with an EAPC of + 3.57%, driven primarily by the global obesity epidemic, sedentary lifestyles, and dietary westernization [20, 21]. This explosive growth trajectory aligns with projections that diabetes will become the seventh leading cause of death worldwide by 2030, may highlight the growing importance of strengthening metabolic health interventions globally [22].

Cardiovascular disease, while maintaining its position as the dominant contributor to absolute disease burden, demonstrated a more modest growth rate of + 1.18%. This relative deceleration likely reflects the cumulative impact of decades of cardiovascular prevention efforts in high-income countries, including widespread implementation of hypertension control programs, smoking cessation initiatives, and evidence-based pharmacological interventions [23, 24]. However, the persistent high absolute burden indicates that cardiovascular disease remains a formidable challenge requiring sustained public health attention [25]. The intermediate growth pattern of chronic kidney disease (+ 2.46% EAPC) reflects its complex etiology and the critical role of comorbid conditions, particularly diabetes and hypertension, in driving renal dysfunction progression [26, 27].

Methodological Innovation, Sensitivity, and practical utility of the CKM index

The development of the CKM Index addresses a fundamental gap in current chronic disease surveillance systems, which remain largely focused on single-disease metrics despite growing evidence of multimorbidity’s clinical and economic significance [28, 29]. Our standardized 0–100 scale composite metric overcomes methodological challenges inherent in integrating disparate disease burden measures through logarithmic transformation, min–max normalization, and empirically guided weighting strategies [30]. The logarithmic transformation effectively addressed the right-skewed distribution of DALY rates, while min–max normalization ensured comparability across diverse epidemiological contexts [31]. The weighting scheme—CVD (0.5), T2DM (0.3), and CKD (0.2)—was derived from global DALY proportions as reported in GBD 2021 and refined to reflect the clinical interdependencies among CKM conditions, particularly the growing significance of metabolic and renal dysfunction in multimorbidity frameworks [32].

To evaluate the robustness of this composite structure, sensitivity analyses were conducted using two alternative weighting strategies: equal weights (CVD = CKD = T2DM = 1/3) and DALY-proportional weights (CVD = 0.75, T2DM = 0.15, CKD = 0.10). The temporal trends of the CKM Index were generally consistent across models. The original and equal-weight indices were highly correlated (r = 0.84, p < 0.001), and the original and DALY-weighted indices also showed strong alignment (r = 0.74, p < 0.001), indicating that the Index captures stable syndemic patterns over time. However, the lower correlation between the equal and DALY-based schemes (r = 0.26, p = 0.15) highlights that the choice of weighting strategy can meaningfully influence specific trajectory patterns, especially when reflecting different value priorities—such as balanced condition emphasis versus mortality-dominant framing. These findings underscore the methodological reliability of the Index under empirically and clinically plausible assumptions, while also emphasizing the importance of transparent sensitivity reporting and contextual adaptability.

Beyond its methodological rigor, the CKM Index offers substantial practical utility for global health policy and systems research. It can support clinical risk stratification, health system performance benchmarking, and targeted resource allocation [33–35]. Healthcare systems could integrate CKM Index calculations into routine population health monitoring to identify high-burden communities and guide interventions [36]. Emerging technologies such as the Internet of Things (IoT) in medical practice offer promising tools for CKM multimorbidity management through continuous remote monitoring and integrated care platforms [37]. Its standardized structure enables international benchmarking and fosters evidence-based policy learning across heterogeneous health system contexts [38]. Together, these attributes position the CKM Index as both a scientifically grounded and operationally flexible tool for addressing the rising challenge of multimorbidity worldwide.

To address the concern of potential double-counting due to overlapping disease cases across CVD, T2DM, and CKD, we conducted a scenario-based sensitivity analysis. Specifically, we simulated partial overlap by reducing the total DALY input by assumed comorbidity rates of 10%, 15%, and 25% across all years and regions. The adjusted CKM Index values were then recalculated under each scenario. Results showed that while absolute index values decreased slightly under higher overlap assumptions, the temporal and regional trend patterns remained stable, with correlation coefficients exceeding r = 0.97 (p < 0.001) compared to the original Index. These findings suggest that the CKM Index is robust to reasonable levels of comorbidity-related DALY inflation and remains a valid comparative tool even under imperfect attribution conditions.

Global health inequalities through the CKM lens

Our SDI-stratified analyses reveal stark disparities in CKM burden that underscore fundamental inequalities in global health systems and social determinants of health [39, 40]. High-SDI regions exhibit the highest absolute CKM Index values (78.6 in 2021), reflecting the persistence of chronic disease burden despite advanced healthcare infrastructure and substantial per-capita health expenditures [41].This pattern may indicate that technological advancement and healthcare spending alone are unlikely to fully address the complex socio-behavioral determinants of CKM diseases [42]. The stabilization of growth trends in high-SDI countries may indicate approaching epidemiological saturation points or the effectiveness of established chronic disease management systems [43].

The rapid escalation of CKM burden in middle and low-middle SDI countries represents a particularly concerning trend, as these nations face the dual challenge of managing increasing chronic disease prevalence while often lacking adequate healthcare infrastructure and resources [44–46]. Notably, countries such as Vietnam and the Philippines within similar SDI categories have shown signals of CKM Index stabilization or modest reduction in recent years. While our ecological design does not allow attribution of these patterns to specific policies, they suggest that meaningful improvements in CKM burden are achievable even in resource-constrained settings and may provide useful examples for south–south learning and collaboration [47, 48].

China’s CKM trajectory and policy implications

China’s CKM Index trajectory—rising from 35.3 in 1990 to 66.1 in 2021, with exploratory projections suggesting potential stabilization—reflects both the challenges and opportunities inherent in managing chronic disease transitions in rapidly developing economies. The annual growth rate exceeding global averages indicates that China is experiencing an accelerated epidemiological transition, driven by rapid urbanization, dietary shifts, and population aging. The slight improvement since the mid-2000s likely reflects the partial effectiveness of national chronic disease control initiatives, including the establishment of tiered healthcare delivery systems and expanded primary care coverage under healthcare reform programs.

However, the observed historical trends under current policy frameworks suggest that existing interventions, while beneficial, remain insufficient to achieve substantial burden reduction. The persistence of specialty-siloed management approaches, fragmented health information systems, and inadequate integration of chronic disease care pathways continue to impede optimal multimorbidity management. The disconnect between the ambitious targets set forth in the Healthy China 2030 strategy and the observed trends in CKM burden may suggest the potential value of incorporating composite multimorbidity indicators into national health surveillance architectures. Digital health innovations leveraging CKM Index calculations could enhance population risk stratification and improve the efficiency of preventive interventions across China’s diverse regional contexts.

Multimorbidity patterns and clinical management implications

As noted in the Methods, the pairwise and triple DALY estimates used in this analysis are conceptual aggregates derived by summing individual disease-specific DALYs, rather than comorbidity-adjusted measures, and therefore reflect patterns of shared syndemic burden rather than person-level multimorbidity.

The analysis of pairwise and triple comorbidity trends reveals the complex interdependencies characterizing CKM diseases and their collective impact on population health. The rapid expansion of CVD + T2DM combinations (EAPC + 3.15%) reflects well-established pathophysiological links between metabolic dysfunction and cardiovascular pathology, while the near-tripling of CKD + T2DM burden underscores the critical role of diabetic nephropathy in the global chronic kidney disease epidemic. These patterns demand fundamental reconsideration of traditional specialty-based care models in favor of integrated, patient-centered approaches that address the full spectrum of metabolic-cardiovascular-renal interactions.

The age-stratified burden analysis, showing 35% of total CKM DALYs concentrated in individuals aged 60–74 years, provides crucial insights for targeted intervention strategies. While the predominant burden in older adults reflects the natural progression of chronic diseases, the detectable presence of CKM-related DALYs in younger populations emphasizes the importance of life-course prevention approaches. The consistent male predominance across age groups suggests the need for gender-specific risk assessment and intervention protocols, potentially reflecting differential exposure to occupational hazards, lifestyle factors, and healthcare-seeking behaviors.

Interpretive considerations for composite burden Estimation

While the CKM Index offers a novel approach for capturing syndemic burden, it remains based on the weighted integration of three individual DALY components. It should therefore be interpreted as a measure of aggregate cardio-kidney-metabolic syndrome burden at the population level rather than as a direct estimate of person-level multimorbidity. Although this may appear additive, the underlying methodology involves log-transformation and normalization steps that minimize scale disparities and skewness, thereby producing a composite score that is more than a simple arithmetic sum. This ensures that the index reflects relative burden intensity across time and regions rather than absolute aggregation. However, we acknowledge that the approach does not explicitly adjust for overlapping cases of comorbidity—such as individuals simultaneously affected by CVD, CKD, and T2DM—which may lead to potential double-counting in population-level metrics even though it remains useful for comparing relative syndemic burden across settings. As a compensatory measure, our study included descriptive analysis of pairwise and triple disease burden trajectories to highlight co-occurrence trends. Future research should therefore develop more sophisticated comorbidity-adjusted DALY models, such as multi-state microsimulation, or structural equation frameworks to better capture disease interactions and overlaps within syndemic populations.

Policy recommendations and global action framework

The findings of this study suggest several potential policy directions across multiple governance levels. International health organizations, particularly the World Health Organization, could consider incorporating multimorbidity metrics into global NCD monitoring frameworks and exploring standardized CKM surveillance protocols [9, 49]. The development of a global CKM observatory platform could facilitate real-time monitoring, best practice sharing, and coordinated response strategies across countries facing similar challenges [50, 51]. South–South collaboration initiatives could leverage successful experiences from middle-income countries that have achieved CKM burden reductions to accelerate progress in similar epidemiological contexts [52, 53].

At the national level, health systems may explore integrating CKM Index calculations into routine disease surveillance platforms and utilize evolving burden patterns to inform strategic planning, resource allocation, and regional risk stratification [54]. In clinical practice, the Index may be used to support the design of integrated care pathways targeting high-risk populations with overlapping cardio-kidney-metabolic conditions. For example, risk scores derived from CKM Index data can guide proactive primary care outreach, patient triaging, and chronic disease self-management programs.

Furthermore, payers and health policy makers may consider linking CKM Index strata to value-based financing models—such as risk-adjusted capitation or bundled payments for multimorbidity clusters—to promote more efficient and equitable care. In resource-limited settings, the CKM Index can serve as a tool to prioritize intervention delivery by identifying syndemic disease hotspots, thereby enabling high-yield investment in prevention and early detection programs. To support these applications, digital health platforms and national health information systems could incorporate automated CKM Index calculators, updated annually with GBD or local burden data, to enable real-time population health surveillance and care optimization.

Realizing these policy goals would require cross-sectoral coordination, investment in interoperable data infrastructure, and alignment of incentives across primary, specialty, and public health domains. Ultimately, the CKM Index could serve not only as a descriptive tool, but as an operational driver for transforming fragmented chronic disease management into integrated, person-centered, and population-responsive care systems, provided that future studies confirm its validity and feasibility in diverse country settings.

China was analyzed separately in this study due to its unique epidemiological transition, high-quality GBD data availability, and its status as the world’s most populous middle-high SDI country undergoing rapid chronic disease transformation. The selection is not politically motivated, but reflects scientific considerations about representativeness, data granularity, and policy relevance. China’s CKM trajectory provides a valuable case study of syndemic burden evolution in large-scale emerging economies. Nevertheless, future studies may extend this analysis to other high-burden countries such as India, the United States, or Brazil to enhance comparative insights and policy benchmarking.

Study limitation

This study has several limitations. First, all burden estimates rely on GBD model assumptions, particularly in data-sparse settings. Second, the CKM Index uses DALYs as the sole input metric, which may not fully capture upstream risk factors. Third, summing disease-specific DALYs may introduce double-counting due to multimorbidity overlap. Finally, forecasting models showed limited stability under structural shocks such as the COVID-19 pandemic, and long-term projections should be interpreted cautiously. The selection of ARIMA and Bayesian ridge regression was based on their superior performance in pre-pandemic validation, despite the deterioration in predictive accuracy in the post-2020 period. Their use therefore reflects short-term extrapolative value rather than long-term forecasting precision. Taken together, the forecast results in this study should be interpreted as scenario-like illustrations of potential future burden, rather than as precise predictions on which binding policy targets are based.

Conclusions and future outlook

CKM multimorbidity has emerged as the defining challenge of 21st-century chronic disease control, demanding innovative approaches that transcend traditional single-disease management paradigms. The CKM Index provides a valuable tool for quantifying this complex health challenge and guiding evidence-based policy responses. The documented global escalation of CKM burden, pronounced health inequalities across development levels, and China’s critical position at an epidemiological inflection point underscore the urgency of coordinated global action. The next decade may offer an opportunity for countries to explore more integrated CKM strategies. However, given the uncertainty surrounding long-term projections, these strategies should be informed by the overall trajectory signaled by the CKM Index, rather than by any single forecast value for 2030. Success in managing the CKM syndemic will require sustained political commitment, innovative financing mechanisms, and collaborative approaches that harness the collective expertise of the global health community. The ultimate goal must be the transformation of health systems from reactive, disease-centered models to proactive, health-promoting infrastructures capable of supporting healthy aging and reducing premature mortality across all populations [55, 56].

Author contributions

Conceptualization and Design: X.J.L. and H.J. conceived the study concept and designed the research framework. X.J.L., X.P., and X.Y.L. developed the CKM Index methodology and analytical approach.Data Collection and Processing: X.J.L. and X.P. extracted and processed the Global Burden of Disease Study 2021 data. Y.Z. and H.Z. assisted with data validation and quality control.Statistical Analysis: X.J.L. and X.P. performed the statistical analyses, including trend analysis, joinpoint regression, and forecasting models. Y.S. and H.J. contributed to the sensitivity analyses and model validation.Methodology Development: X.J.L. developed the logarithmic transformation and min-max normalization procedures for the CKM Index. X.P. contributed to the weighting scheme design and validation framework.Manuscript Writing: X.J.L. and X.P. wrote the original draft of the manuscript. Y.Z., H.Z., and Y.Z. contributed to specific sections of the methodology and results. H.J. and X.Y.L. provided critical revisions and supervised the writing process.Visualization: X.J.L. and Y.S. created the figures and tables. H.J. and J.L. reviewed and refined the visual presentations.Review and Editing: All authors (X.J.L., X.P., Y.Z., H.Z., Y.Z., Y.S., H.J., J.L., H.J., X.Y.L.) critically reviewed and edited the manuscript for important intellectual content.Project Administration and Supervision: H.J. and X.Y.L. provided overall supervision of the research project and manuscript preparation. X.Y.L. coordinated the collaborative efforts across different departments.Funding Acquisition: X.Y.L. secured the research funding from Shanghai Jiao Tong University School of Medicine Postgraduate Medical Education Program, Shanghai Science Technology and Innovation Action Plan, and Shanghai Shenkang Hospital Development Center.Equal Contributions: X.J.L. and X.P. contributed equally as first authors to this work.All authors have read and approved the final version of the manuscript and agree to be accountable for all aspects of the work.

Funding

Shanghai Jiao Tong University School of Medicine Postgraduate Medical Education Program (BYH20250201).

Shanghai Science Technology and Innovation Action Plan Key Program on Medical Innovation Research (21Y31920400).

Shanghai Shenkang Hospital Development Center’s Project (SHDC12024141).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable. This study used publicly available aggregate data with no individual patient information.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.