Global, Regional, and National Burden of Chronic Obstructive Pulmonary Disease–Associated Heart Failure: A Systematic Analysis of the Global Burden of Disease Study from 1990 to 2021 with Projections up to 2050

Abstract

Background

Chronic obstructive pulmonary disease (COPD) is strongly linked to heart failure (HF), but comprehensive global evaluations of COPD-associated HF remain limited.

Methods

Using Global Burden of Disease Study 2021 data, we assessed prevalence and years lived with disability (YLDs) of COPD-associated HF. We applied decomposition analysis, Joinpoint regression, and Bayesian age-period-cohort modeling to evaluate trends by sex, age, sociodemographic index (SDI), and region, and to project burden to 2050.

Results

In 2021, global COPD-associated HF cases exceeded 3.61 million, a 1.4-fold increase since 1990. Age-standardized prevalence rates (ASPR) and YLD rates (ASYR) increased overall. Absolute cases and YLDs rose in both sexes, but age-standardized rates declined in males and rose in females. Burden peaked at ages 70–74 and grew fastest among those aged ≥95. The Middle SDI region had the highest counts; East Asia had the highest regional burden, and China the highest national absolute burden. Only High-SDI regions showed rising ASPR and ASYR; most other regions declined, notably Eastern Europe. Projections indicate global increases in ASPR and ASYR from 2022 to 2050, with slightly higher rises in females than males.

Conclusion

The global burden of COPD-associated HF continues to rise with substantial heterogeneity and is projected to increase further. Coordinated international efforts and targeted prevention and management are urgently needed.

Background

Chronic obstructive pulmonary disease (COPD) is a prevalent, treatable, and preventable respiratory condition characterized by progressive airflow limitation and persistent respiratory symptoms.¹ As one of the leading causes of global mortality, particularly in low- and middle-income countries, COPD accounts for approximately 5% of all deaths worldwide annually.² A systematic analysis estimated the global prevalence of COPD at 10.3% (95% CI: 8.2–12.8%),³ with projections suggesting nearly 600 million cases globally by 2050.⁴ This progressive chronic disease significantly impairs patients’ quality of life and imposes substantial annual costs on healthcare systems globally. The financial burden is largely driven by the management of co-existing conditions (comorbidities), which also exacerbate health-related social inequities.^5,6 Consequently, COPD represents a major and growing public health challenge due to its high morbidity, mortality, and associated healthcare costs.

Heart failure (HF) is a clinical syndrome marked by structural or functional heart abnormalities that impair the heart’s filling or pumping capacity. Key symptoms include breathlessness, fatigue, and fluid retention.⁷ Both HF and COPD are highly prevalent in older adults and significantly influence each other’s progression through interconnected pathways. COPD can accelerate HF development and worsen its prognosis. Chronic low oxygen levels, systemic inflammation, and changes in lung blood vessels associated with COPD contribute to heart damage and dysfunction. Conversely, HF negatively impacts lung function through mechanisms like pulmonary hypertension, fluid buildup in lung tissues (edema), and elevated lung venous pressure. These changes can mimic or mask COPD symptoms and increase the frequency, severity, and progression of COPD flare-ups.^8,9 The overlapping symptoms—particularly breathlessness and reduced exercise tolerance—often complicate diagnosis and delay appropriate treatment. Accurately estimating the overall public health impact of the comorbidity between COPD and HF is of paramount importance, given the intricate pathophysiological relationships and diagnostic challenges previously described. The development of optimal, individualized treatment plans—which aim to reduce hospital readmissions and mortality while enhancing long-term quality of life and survival—necessitates improved detection and management of this condition.

Epidemiological evidence indicates a significant bidirectional overlap between COPD and HF. The prevalence of HF among individuals with COPD ranges from 10% to 50%, and HF prevalence rises annually with global population aging. Concurrently, 25–50% of HF patients have coexisting COPD.¹⁰ Despite this increasing global burden, a comprehensive evaluation of COPD-associated HF, particularly across international, regional, and national levels, remains lacking. To address this gap, we conducted a systematic analysis of COPD-associated HF burden from 1990 to 2021 using GBD 2021 data. We quantified the global epidemiological burden by examining core metrics, including prevalence and years lived with disability (YLDs). These analyses were stratified by Socio-demographic Index (SDI) quintiles to assess variations across different levels of socioeconomic development. This study provides essential baseline evidence by mapping the burden of COPD-associated HF across socio-demographic strata and tracking long-term trends. Our findings aim to guide targeted public health interventions and inform evidence-based resource allocation.

Methods

Data Sources

The GBD is a systematic, scientific global health research program that uses standardized methods to provide quantitative assessments of the burden of major diseases, injuries, and health risk factors worldwide across time periods, geographic regions, and health conditions.¹¹ This study utilized data from the GBD 2021 database (https://vizhub.healthdata.org/gbd-results/). The GBD 2021 database documents disease burden metrics for 288 causes of death, 371 diseases and injuries, and 88 risk factors across 204 countries and territories—categorized into 21 regions, 7 super-regions, and 5 SDI quintiles. Core epidemiological indicators include YLDs, Years of Life Lost (YLLs), Disability-Adjusted Life Years (DALYs), and Health-Adjusted Life Expectancy (HALE).^12–14

Definitions

Within the GBD study, HF is defined using the structured Framingham or European Society of Cardiology (ESC) diagnostic criteria.¹⁵ The Framingham criteria require meeting either two major criteria or one major and two minor criteria. Conversely, the ESC guidelines emphasize the presence of typical symptoms and signs of cardiac dysfunction.¹⁵ Based on a systematic analysis of multinational data, literature, and expert opinion—and by ensuring quantifiable and mutually exclusive causal attribution—HF was etiologically linked to 27 specific causes. These causes include ischemic heart disease, hypertensive heart disease, COPD, interstitial lung disease, pulmonary hypertension, and various pneumoconioses.¹⁶ COPD-associated HF refers to clinical scenarios in which COPD is the underlying cause directly leading to the onset or progression of HF. According to the Global Initiative for Chronic Obstructive Lung Disease (GOLD), COPD is operationally defined by a post-bronchodilator forced expiratory volume in one second to forced vital capacity (FEV1/FVC) ratio of less than 0.7.¹⁷ Alternative case definitions incorporate pre- and post-bronchodilator lower limit of normal (LLN) values and European Respiratory Society (ERS) guidelines, with relevant ICD-10 codes spanning J41 to J44.9.¹⁸

Socio-Demographic Index

The SDI is a composite development metric that incorporates three standardized components and has a strong correlation with population health outcomes: (1) lagged per capita income adjusted for purchasing power parity, (2) mean educational attainment among individuals aged ≥15 years, and (3) total fertility rate among females under 25 years.¹⁹ With an SDI ranging from 0 (theoretical minimum health-related development) to 1 (theoretical maximum), nations and territories are divided into five quintiles: high, high-middle, middle, low-middle, and low.

Decomposition Analysis

The proportionate contribution of variances in particular factors to observed differences in overall results is measured using decomposition analysis.²⁰ By applying Das Gupta’s demographic decomposition method, we were able to quantify variations in the burden of COPD-associated heart failure (1990–2021) over time, separating the contributions of age-specific epidemiological rates, population size, and population age structure.

Joinpoint Regression Analysis

Joinpoint regression methodology divides time-series data into discrete linear segments with different slopes in order to find statistically significant change points in temporal trends. Model optimization used the Grid Search Method (GSM), Bonferroni correction to control the family-wise error rate (α=0.05), and Monte Carlo permutation testing (10,000 iterations) for significance evaluation.²¹ Joinpoint regression results included average annual percentage change (AAPC) throughout the whole study period, annual percentage change (APC) with 95% CI per trend segment, and the number and temporal positions of discovered joinpoints. APC measures the strength of the age-standardized rate trend within particular periods, whereas AAPC summarizes the trend over the whole analysis period. Joinpoint regression was used to determine the significance of the trend: a Significant rise in APC/AAPC if the lower 95% CI was more than 0; a significant fall in APC/AAPC < 0 with upper 95% CI less than 0; and stable if 95% CI contained 0.²²

Predictive Analytics

Future disease burden can be predicted using a multivariate statistical framework called the Bayesian Age-Period-Cohort (BAPC) model. It uses the Integrated Nested Laplace Approximation (INLA) to approximate posterior marginal distributions, assumes that disease incidence follows a log-linear Poisson distribution, and models the temporal dynamics of age, period, and cohort effects using second-order random walk (RW2) smoothing priors. By avoiding convergence and mixing problems, this method drastically lowers computing cost as compared to conventional Markov Chain Monte Carlo (MCMC) techniques. In order to predict future trends in disease burden for COPD-associated HF prevalence and YLDs, we applied the BAPC model using the R packages INLA (version 23.09.09) and BAPC (version 0.0.36).²³

In model fitting, we adopted the default RW2 before smoothing the age, period, and cohort effects, which mitigates the risk of overfitting.²⁴ Model performance was evaluated using the Watanabe-Akaike information criterion (WAIC) to confirm its superiority over simple linear models.²⁵ Although formal post-hoc predictive validation was not performed, the model’s goodness-of-fit was assessed through a qualitative visual inspection of the historical data.²⁶ This assessment revealed a strong alignment between the predicted values and the observed values from the GBD 2021 study. All estimates presented in this study, including historical fits and future projections, are reported with 95% uncertainty intervals (UI). These intervals were derived by repeated sampling from the joint posterior distribution, approximated using the INLA algorithm, thereby quantifying the predictive variability attributable to model parameter uncertainty.²⁷

Data Analysis

Descriptive statistical analyses of global COPD-associated HF burden metrics (1990–2021) were performed using Excel 2021. Distributions were stratified by period, age group, sex, and COPD-related HF subtypes, with analysis of prevalence, Age-standardized prevalence rates, YLDs, and Age-standardized YLDs rates. APC and AAPC were computed using Joinpoint Regression Software (Version 5.1.0), with segment-specific trends deemed statistically significant at α = 0.05. Statistical analyses and data visualization were performed using R software (version 4.4.3) with the following packages: dplyr, ggplot2, reshape2, and readxl.

Results

Global Disease Burden and Trends of COPD-Associated HF

As of 2021, as demonstrated in Table 1, there were 3,613,136.36 cases (95% UI: 2,666,977.72–4,801,973.44) of COPD-associated HF worldwide, representing an approximately 1.4 times increase from the 1,504,707.39 cases (95% UI: 1,141,643.67–1,942,359.75) recorded in 1990. From 1990 to 2021, the age-standardized prevalence rate (ASPR) per 100,000 population showed an upward trend, with an AAPC of 0.07 (95% CI: 0.03–0.12) (Table 1). Concurrently, the number of YLDs in 2021 was 321,752.54 (95% UI: 198,202.27–486,477.44), reflecting a 1.4 times increase from the 134,013.06 YLDs (95% UI: 83,437.22–196,502.92) reported in 1990. The age-standardized YLDs rate (ASYR) per 100,000 population also exhibited an upward trend, with an AAPC of 0.09 (95% CI: 0.05–0.13) (Table S1).

Table 1.

The Cases of Prevalence and ASPR of COPD-Associated HF in 1990and 2021, and Its Temporal Trends from 1990 to 2021

Location	Prevalent Cases		Age-Standardized Prevalent Rates		AAPC
	1990_Thousands (95% UI)	2021_Thousands (95% UI)	1990_per 100 000 (95% UI)	2021_per 100 000 (95% UI)
Global	1504707.39 (1141643.671942359.75)	3613136.36 (2666977.724801973.44)	41.93 (32.19,53.41)	43.24 (32.08,57.68)	0.07(0.03–0.12)
Sex
Male	149853.51 (101636.82207870.7)	1936132.08 (1434629.532548693.69)	53.19 (41.57,67.34)	52.66 (39.28,68.05)	−0.0(−0.08 - −0.03)
Female	130343.61 (87815.52188352.36)	1677004.28 (1246355.172247751.35)	34.13 (25.51,44.98)	35.98 (26.72,48.23)	0.12(0.03–0.21)
SDI
High-middle SDI	389597.81 (295021.57503347.12)	702721.48 (539545.31902323.65)	43.01 (32.85,54.98)	42.44 (31.57,56.64)	−0.09 (−0.16,-0.03)
High SDI	241899.16 (178011.5325813.91)	835012.66 (613179.061127713.64)	21.25 (15.89,28.39)	30.47 (23.57,39.07)	1.16 (1.11,1.21)
Low-middle SDI	245225.3 (189804.58314448.24)	224290.39 (161566.56306383.85)	53.08 (40.93,69.15)	51.86 (38.82,69.6)	−0.10 (−0.16,-0.05)
Low SDI	107003.06 (77655.08144395.77)	611352.13 (454131.32813750.94)	59.57 (43.49,79.84)	54.17 (38.28,74.95)	−0.31 (−0.36,-0.27)
Middle SDI	520265.4 (398674.57669706.86)	1238224.91 (904296.371675160.02)	63.31 (49.07,79.96)	51.95 (38.11,69.24)	−0.67 (−0.74,-0.61)
Region
Andean Latin America	3205.78 (2563.493998.89)	10244.73 (7957.7912983.22)	17.41 (13.54,22.15)	18.28 (14.1,23.32)	0.20 (0.15,0.24)
Australasia	7346.75 (5436.779718.5)	28079.85 (22518.1534838)	30.18 (22.62,39.67)	48.05 (38.76,59.46)	1.53 (1.47,1.59)
Caribbean	3444.21 (2649.184445.94)	10591.22 (8168.813627.52)	13.94 (10.64,17.95)	19.54 (15.05,25.15)	1.11 (1.07,1.15)
Central Asia	6330.79 (4692.958399.81)	6116.71 (4388.878354.64)	14.58 (10.47,19.64)	8.81 (6.1,12.08)	−1.59 (−1.66,-1.51)
Central Europe	17351.52 (12473.9923735.3)	28443.95 (20879.3337413.33)	12.22 (8.96,16.72)	12.09 (9.09,15.64)	−0.04 (−0.11,0.04)
Central Latin America	20619.92 (16153.226001.84)	70581.35 (53940.3390914.74)	28.79 (22.31,36.27)	30.11 (22.63,38.83)	0.13 (0.1,0.16)
Central Sub-Saharan Africa	8902.78 (6060.8713061.59)	17364.99 (11849.125478.55)	48.16 (31.8,72.54)	37.81 (24.3,56.65)	−0.79 (−0.85,-0.73)
East Asia	704891.3 (529474.92913805.97)	1544998.56 (1114446.592104094.12)	107.27 (83.04,135.44)	75.75 (55.2,101.64)	−1.19 (−1.26,-1.11)
Eastern Europe	34666.53 (24119.1848215.41)	26686.45 (17692.3638777.04)	13.26 (9.46,18.25)	7.43 (4.96,10.79)	−1.97 (−2.2,-1.74)
Eastern Sub-Saharan Africa	40629.6 (27806.9257066.8)	70880.06 (48416.6798838.93)	60.81 (40.74,87.16)	46.02 (30.48,67.15)	−0.89 (−0.95,-0.83)
High-income Asia Pacific	15214.21 (9713.9821713.53)	58006.4 (41356.5880674.19)	8.55 (5.53,12.17)	9.81 (7.16,13.11)	0.44 (0.38,0.49)
High-income North America	94775.07 (65837.28130098.09)	290804.85 (211305.3396862.11)	25.52 (17.86,34.77)	41.2 (30.02,55.92)	1.59 (1.51,1.66)
North Africa and the Middle East	23640.33 (19486.7128989.95)	68985.37 (55674.0484400.91)	15.75 (12.7,19.56)	17.22 (13.76,21.55)	0.30 (0.23,0.37)
Oceania	1301.3 (1006.791667.73)	3101.61 (2412.33967.9)	67.37 (52.21,86.18)	61.9 (48.21,80.15)	−0.28 (−0.3,-0.25)
South Asia	256306.78 (195446.29331249.12)	727877.91 (529292.45985792.89)	63.48 (48.74,82.49)	61 (44.96,82.77)	−0.17 (−0.22,-0.11)
Southeast Asia	70968.71 (56014.5190772.35)	164794.12 (128041.51213007.47)	35.59 (27.86,46.22)	31.36 (24.24,40.57)	−0.43 (−0.49,-0.37)
Southern Latin America	5870.02 (4362.027631.21)	15713.37 (11145.621097.8)	13.42 (10.01,17.61)	17.51 (12.52,23.4)	0.88 (0.82,0.94)
Southern Sub-Saharan Africa	19574.7 (13911.826722.09)	33484.61 (23051.2146823.89)	76.4 (52.83,107.13)	64.43 (43.76,90.62)	−0.56 (−0.69,-0.43)
Tropical Latin America	21287.02 (16089.1727589.19)	74682.62 (53714.11103293.46)	28.4 (21.75,36.68)	30.58 (22.07,42.12)	0.21 (0.15,0.28)
Western Europe	110988.76 (80677.55147527.4)	296463.15 (228807.39374549.98)	18.34 (13.47,24.05)	28.44 (22.15,35.86)	1.40 (1.32,1.48)
Western Sub-Saharan Africa	37391.32 (25516.453159.2)	65234.49 (44565.3992176.67)	46.35 (30.97,67.34)	34.94 (22.9,50.94)	−0.90 (−0.94,-0.87)

Abbreviations: COPD, Chronic obstructive pulmonary disease; HF, heart failure; UI, uncertainty interval; AAPC, average annual percentage change.

Disease Burden and Trends of COPD-Associated HF by Gender

The number of common cases and YLDs for both males and females increased gradually between 1990 and 2021, with males bearing a greater burden of COPD-associated HF than females. Male and female prevalence rates in 2021 were 1,936,132.08 (95% UI: 1,434,629.53–2,548,693.69) and 1,677,004.28 (95% UI: 1,246,355.17–2,247,751.35), respectively. The YLDs numbers for males and females were 172,523.00 (95% UI: 107,933.35–253,504.29) and 149,229.54 (95% UI: 91,977.64–231,965.96), respectively. However, the ASPR (AAPC: −0.06; 95% CI: −0.08 to −0.03) and ASYR (AAPC: −0.04; 95% CI: −0.07 to −0.02) for males showed modest decreases during the 1990–2021 period, while the ASPR (AAPC: 0.12; 95% CI: 0.03–0.21) and ASYR (AAPC: 0.13; 95% CI: 0.04–0.22) for females showed increasing trends (Figure 1, Tables 1, and S1).

Figure 1.

Trends in all-age cases and age-standardized prevalence and YLD rates of COPD-associated HF by sex from 1990 to 2021. (A) Prevalence number and rate; (B) YLD number and rate.

Disease Burden and Trends of COPD-Associated HF by Age

Analysis of COPD-associated HF prevalence and YLDs across different age groups revealed no recorded cases among individuals aged 0–15 years in the 2021 GBD database. A comparison of the prevalence, YLDs counts, and corresponding crude rates across age groups globally in 1990 and 2021 demonstrated that the crude prevalence rate (CPR) and crude YLDs rate (CYR) for COPD-associated HF initially increased and then decreased with age, peaking in the 90–94 age group (Figure 2). Similarly, the prevalence and YLDs counts for COPD-associated HF increased with age before declining, with the highest values consistently observed in the 70–74 age group (Figure 3).

Figure 2.

The prevalence rates and YLDs rates of COPD-associated HF by different age groups and genders globally in 1990 and 2021. (A) The prevalence rates in 1990; (B) The prevalence rates in 2021; (C) The YLDs rates in 1990; (D) The YLDs rates in 2021.

Figure 3.

The global prevalence counts and YLDs of COPD-associated HF by age group and sex, in 1990 and 2021. (A) The prevalence counts in 1990; (B) The prevalence counts in 2021; (C) The YLDs in 1990; (D) The YLDs in 2021.

Note: Although COPD is typically diagnosed in adults aged ≥40 years, all GBD age categories are presented for completeness.

Disease Burden and Trends of COPD-Associated HF by HF Subgroups

The GBD database classified HF into four subgroups based on disease severity and treatment status: treated heart failure, mild heart failure, moderate heart failure, and severe heart failure. According to age-specific analysis, the prevalence and YLDs counts for COPD-associated HF in all HF subtypes among people aged 15 and older rose between 1990 and 2021, with the largest increase seen in those aged 95 and above. The 2021 prevalence and YLDs numbers were more than six times higher than the 1990 levels in all subgroups. Of all HF cases, severe HF accounted for the highest percentage of prevalent cases and YLDs. Nonetheless, between 1990 and 2021, the ASPR for the 45–49, 60–64, 65–69, and 70–74 age groups fell, and the ASYR for the 60–64, 65–69, and 70–74 age groups reduced as well (Figure 4, Tables S2, and S3).

Figure 4.

Time trend of the burden of COPD-associated HF by age group. (A) COPD-associated treated HF; (B) COPD-associated mild HF; (C) COPD-associated moderate HF; (D) COPD-associated severe HF.

Global Joinpoint Regression Analysis of COPD-Associated HF

From 1990 to 2000, the global prevalence and YLDs burden of COPD-associated HF exhibited an upward trend, followed by a decline after 2000. Notably, an upward trend re-emerged in 2019. The most significant decline in ASPR occurred between 2000 and 2005 (APC = −1.02; 95% CI: −1.17 to −0.87; p < 0.0001), while the most pronounced decline in ASYR was observed between 2005 and 2012 (APC = −0.99; 95% CI: −1.14 to −0.84; p < 0.0001) (Table 2 and Figure 5).

Table 2.

Joinpoint Regression Analysis: Trends in Age-Standardized Prevalence and YLDs Rates (per 1000000 Persons) Among Both Sexes, Males, and Females in Global, 1990–2021

Gender	ASPR			ASYR
	Period	APC (95% CI)	AAPC (95% CI)	Period	APC (95% CI)	AAPC (95% CI)
Both	1990-2000	0.77 (0.73–0.81)	0.07 (0.03–0.12)	1990-2000	0.77 (0.73–0.81)	0.09 (0.05–0.13)
	2000-2005	−1.02 (−1.17 - −0.87)		2000-2005	−0.99 (−1.14 - −0.84)
	2005-2019	−0.24 (−0.27 - −0.22)		2005-2019	−0.23 (−0.26 - −0.21)
	2019-2021	1.58 (1.03–2.14)		2019-2021	1.64 (1.13–2.15)
Male	1990-2000	0.84 (0.82–0.86)	−0.06 (−0.08 - −0.03)	1990-2000	0.84 (0.82–0.85)	−0.04 (−0.07 - −0.02)
	2000-2005	−1.06 (−1.14 - −0.98)		2000-2005	−1.02 (−1.10 - −0.95)
	2005-2012	−0.28 (−0.33 - −0.24)		2005-2012	−0.26 (−0.30 - −0.22)
	2012-2019	−0.81 (−0.85 - −0.76)		2012-2019	−0.80 (−0.84 - −0.75)
	2019-2021	1.44 (1.15–1.74)		2019-2021	1.48 (1.21–1.75)
Female	1990-2000	0.60 (0.48–0.72)	0.12 (0.03–0.21)	1990-2000	0.60 (0.48–0.72)	0.13 (0.04–0.22)
	2000-2006	−0.94 (−1.25 - −0.64)		2000-2006	−0.92 (−1.25 - −0.60)
	2006-2015	−0.16 (−0.30 - −0.02)		2006-2015	−0.16 (−0.31 - −0.01)
	2015-2021	0.82 (0.57–1.06)		2015-2021	0.85 (0.60–1.10)

Abbreviations: ASPR, age-standardized prevalence rates; ASYR, age-standardized YLDs rates; APC, annual percent change; AAPC, average annual percent change presented for the full period; CI, confidence interval.

Figure 5.

Trend of age-standardized prevalence rates and YLDs rates by Joinpoint regression, 1990–2021. (A) age-standardized prevalence rates of COPD-associated HF; (B) age-standardized YLDs rates of COPD-associated HF. Indicates that the Annual Percent Change (APC) is significantly different from zero at the alpha = 005 level.

Disease Burden and Trends of COPD-Associated HF by SDI and GBD Regions

Among the five SDI regions, the Middle SDI region had the highest count of COPD-related HF prevalence (1,238,224.91; 95% UI: 904,296.37–1,675,160.02) and YLDs incidence (110,099.63; 95% UI: 67,649.95–166,471.31) in 2021, followed by the High SDI region. From 1990 to 2021, only the high SDI region exhibited an increasing trend in age-standardized prevalence rates (AAPC: 1.16; 95% CI: 1.11–1.21) and age-standardized YLDs rates (AAPC: 1.16; 95% CI: 1.10–1.22), while the other four SDI regions showed declining trends (Tables 1 and S1).

There were notable differences in the burden of COPD-associated heart failure among 21 geographical areas. In 2021, Oceania had the lowest incidence (3101.61; 95% UI: 2412.3–3967.9) and East Asia had the highest (1,544,998.56; 95% UI: 1,114,446.59–2,104,094.12). Likewise, the highest YLDs rates (137,636.88; 95% UI: 82,410.83–208,154.02) and lowest rates (276.14; 95% UI: 170.06–394) were found in East Asia and Oceania, respectively. Western Europe saw the most increases in YLDs rates and prevalence between 1990 and 2021, with AAPCs of 1.4091 and 1.408, respectively. Eastern Europe, on the other hand, displayed the biggest drops, with AAPC of −1.9694 for prevalence and −1.9893 for YLDs rates (Tables 1 and S1). The percentage of COPD-associated heart failure cases among people 80 years of age and older rose dramatically in 2021 compared to 1990 (Figure 6).

Figure 6.

The distribution of prevalent cases and YLDs as percentages across different age groups globally, in five territories, and across 21 GBD regions in 1990 and 2021. (A) The distribution of prevalent cases in 1990; (B) The distribution of prevalent cases in 2021; (C) The distribution of YLDs in 1990; (D) The distribution of YLDs in 2021.

Decomposition of Disease Burden of COPD-Associated HF by SDI and GBD Regions

The effects of aging, population increase, and epidemiological changes on the burden of heart disease related to COPD are depicted in Figure 7. Prevalence and YLDs have increased dramatically over the last 30 years, with population expansion having the biggest beneficial effects (74.54% and 66.66%, respectively). Only the Middle SDI region experienced favorable effects from aging, whereas the other four regions experienced negative effects. Positive effects of population growth were observed in every location. The prevalence and YLDs in high-middle, high, and low-middle SDI regions were positively impacted by epidemiological changes. The increase in YLDs was primarily caused by aging in Central Sub-Saharan Africa, whereas the increase in prevalence was primarily caused by aging in Tropical Latin America. East Asia and Central Latin America were most influenced by population expansion, while East Asia’s prevalence and Eastern Europe’s YLDs were most impacted by epidemiological changes.

Figure 7.

Population-level changes in ageing, population growth, and epidemiologic factors for prevalence and YLDs across various SDI and GBD regions from 1990 to 2021. Black dots represent the overall change from all three components. Positive values indicate a positive contribution, whereas negative values indicate a negative contribution. (A) Population-level changes for prevalence; (B) Population-level changes for YLDs.

National Trends of COPD-Associated HF

In 2021, China had the highest absolute number of prevalent cases of COPD-associated HF among 204 countries and territories (1,511,247.92; 95% UI: 1,088,761.5–2,063,874.16) and also the highest absolute number of YLDs (134,611.36; 95% UI: 80,350.8–204,108.42). In contrast, Tokelau reported the lowest absolute numbers of prevalent cases (0.35; 95% UI: 0.27–0.47) and YLDs (0.03; 95% UI: 0.02–0.05). Because absolute case numbers are strongly influenced by population size, these rankings should not be interpreted as reflecting the underlying risk or per-capita burden. Country-level comparisons based on absolute numbers therefore require caution. When age-standardized rates were used, the Independent State of Papua New Guinea had the highest ASPR in 2021 (83.09 per 100,000 population), whereas the Republic of Uzbekistan had the lowest ASPR (3.20 per 100,000 population). China had the highest ASYR (9.68 per 100,000 population), while the Republic of Uzbekistan had the lowest ASYR (0.29 per 100,000 population). During 1990–2021, ASPR increased in 40.7% of countries, while ASYR increased in 41.7% of countries. While the Republic of Belarus saw the biggest decreases (AAPC: −4.5373 and −4.4837, respectively), the Kingdom of Norway saw the biggest rises in ASPR (AAPC: 3.8126) and ASYR (AAPC: 3.8004) (Figure 8, Tables S4, and S5).

Figure 8.

Global trends in the burden of COPD-associated HF. (A) ASPR of COPD-associated HF across 204 countries in 2021; (B) ASYR of COPD-associated HF across 204 countries in 2021.

Projection of Disease Burden of COPD-Associated HF to 2050

According to the BAPC model, both males’ and females’ worldwide ASPR for COPD-associated HF would rise between 2022 and 2050. Males’ ASPR was predicted to expand from 53.04 per 100,000 in 2022 to 83.15 per 100,000 in 2050 (56.77% increase), while girls’ ASPR was predicted to increase from 36.15 per 100,000 in 2022 to 57.38 per 100,000 in 2050 (58.73% increase). Similarly, from 2020 to 2050, the global ASYR for HF linked to COPD was expected to increase, according to the BAPC model. It was predicted that the ASYR for males would climb from 4.69 per 100,000 in 2022 to 6.69 per 100,000 in 2050, while the ASYR for females would increase from 3.22 per 100,000 in 2022 to 4.92 per 100,000 in 2050 (Figure 9).

Figure 9.

Trends in age-standardized prevalence rates and YLDs rates globally, 1990–2050. (A) Predicted ASPR for Male; (B) Predicted ASPR for Female; (C) Predicted ASYR for Male; (D) Predicted ASYR for Female.

Discussion

COPD is a highly prevalent respiratory condition worldwide, frequently complicated by cardiovascular comorbidities such as HF. Both conditions impose a significant burden on patients, healthcare systems, and society.^28,29 Despite this, the global disease burden of COPD-associated HF—including its incidence, prevalence, YLDs, and long-term trends stratified by sex and age—remains insufficiently characterized, largely due to a lack of systematic analysis using comprehensive data. To address this gap, we systematically evaluated the disease burden and temporal trends of COPD-associated HF from 1990 to 2021 using the GBD 2021 database. Our analysis revealed a 1.4-fold increase in global prevalence and YLDs over this period. Both the ASPR and ASYR showed persistent upward trajectories, indicating a continual rise in the global burden of COPD-associated HF over the past three decades. This escalating burden likely stems from shared risk factors (eg, smoking, air pollution, population aging) and improved patient survival from medical advances, which extend exposure to comorbidity risks.^30–32

From 1990 to 2021, the number of people living with COPD-associated HF and the related YLDs increased steadily for both males and females globally. However, a notable sex-based divergence emerged: the ASPR and ASYR for males showed declining trends (AAPC: −0.06 and −0.04, respectively), while these rates continued to rise for females (AAPC: 0.12 and 0.13, respectively). This divergence may reflect a combination of biological, lifestyle, socioeconomic, and healthcare-related factors; however, our GBD-based analysis cannot directly evaluate these mechanisms. First, it has been hypothesized that physiological and hormonal characteristics may contribute to sex differences in susceptibility. Previous studies suggest that sex hormones, particularly estrogen, may exert protective effects on the cardiopulmonary system, potentially influencing pulmonary hypertension and right ventricular afterload—key pathways in COPD progression to HF—through regulation of endothelial function, inhibition of vascular smooth muscle cell proliferation, and reduction of systemic inflammatory responses.^33,34 In addition, some studies have reported that women with COPD may have higher circulating levels of pro-inflammatory cytokines, such as TNF-α and IL-6, compared with men with similar disease severity;³⁵ these cytokines have been proposed to contribute to myocardial injury, cardiac remodeling, and HF progression.³⁶ Second, although rising health awareness, declining male smoking rates, and improved occupational safety may have reduced COPD and related HF in men, women may face unique risks. Women may have inherently smaller airway dimensions, which could increase vulnerability to inhaled irritants.³⁷ Other risk factors, including indoor air pollution (eg, smoke from cooking fuels), environmental particulate matter, and secondhand smoke exposure, may also disproportionately affect women.³⁸ For example, one study on indoor air pollution suggested that prolonged use of biomass fuels for cooking contributes to severe indoor air pollution, which may be an important contributor to COPD in women.³⁹ Compared with men, women may also experience higher rates of delayed COPD diagnosis, which could hinder timely treatment and potentially accelerate progression to HF—a factor that may contribute to the faster growth of disease burden among women.⁴⁰ These observations underscore the need for future prevention, diagnosis, and management strategies to consider women as a priority population and to develop targeted approaches. Importantly, these mechanistic and health-system explanations are derived from previous literature and are biologically plausible, but they were not directly evaluated in the present GBD-based analysis. They should therefore be considered potential hypotheses rather than causal interpretations of the observed trends.

COPD and HF are highly prevalent conditions in older adults. Analysis of COPD-associated HF burden across age groups reveals a distinct pattern: prevalence, YLDs, and crude rates initially rise, peak in the 70–74 year age group, and subsequently decline with advancing age. It should be noted that COPD is typically diagnosed in adults aged 40 years and older; however, all GBD age categories are presented for completeness, and estimates for younger age groups (15–39 years) are generated by the GBD framework and should be interpreted with caution. This peak burden likely stems from the combined effects of COPD progression and age-related cardiac changes. Mechanistically, COPD induces structural and functional damage to pulmonary blood vessels, increasing pulmonary vascular resistance and leading to pulmonary hypertension.⁴¹ This places sustained pressure overload on the right ventricle, triggering remodeling, dysfunction, hypertrophy, and ultimately HF.⁴² The convergence of these COPD-driven cardiac stresses and natural age-related cardiac decline contributes to the peak burden observed between ages 70–74. While burden metrics decline in individuals aged 90 and older, this may reflect higher mortality among severe cases earlier in life, leaving a survivor population with comparatively milder disease. Notably, among HF subtypes, the prevalence and YLDs counts for individuals aged 95+ increased sixfold between 1990 and 2021 – a striking rise likely driven by population aging and the higher prevalence of complex, severe conditions in this demographic. Severe HF consistently accounted for the largest proportion of cases and YLDs. Given the substantial burden of COPD-associated HF in older populations, enhanced health management is critical. This includes regular monitoring of pulmonary and cardiac function to enable early detection and intervention. Prioritizing tailored management and rehabilitation strategies is essential to improve patients’ quality of life.

Over the past three decades, the global burden of COPD-associated HF has varied substantially across regions and countries. When grouped by SDI—a measure of development—only high-SDI regions showed significant increases in both the ASPR and ASYR. Although the ASPR of COPD-associated HF demonstrates a declining or stabilizing trend across multiple SDI regions globally, the absolute number of prevalent cases and DALYs continues to increase. This apparent paradox is largely attributable to population aging. Both COPD and HF are strongly age-related conditions.⁴³ Increasing global life expectancy has expanded the elderly population, thereby enlarging the pool of individuals susceptible to chronic diseases like COPD-HF.⁴⁴ Consequently, even if the incidence rate within specific age groups remains stable or declines slightly, the absolute number of cases will still rise due to overall population growth.^16,45 This is particularly critical in low-SDI regions, where improvements in healthcare may reduce age-standardized rates, but their large and growing youth populations will eventually age, projecting a substantial future chronic disease burden.⁴⁶ Thus, despite localized advances in prevention and treatment, demographic shifts continue to pose an enduring and escalating challenge to global public health systems.⁴⁷

Geographically, East Asia carried the highest absolute burden of COPD-associated HF in 2021, while Oceania had the lowest. Western Europe experienced the largest rises in prevalence and disability rates, contrasting sharply with Eastern Europe, where these rates declined most significantly. At the national level, China recorded the highest absolute numbers of prevalent cases and YLDs in 2021, whereas Tokelau reported the lowest absolute numbers. These absolute rankings are largely driven by differences in population size: China has a very large population, whereas Tokelau is a small island state. Therefore, absolute numbers should not be used to compare per-capita burden across countries. When age-standardized rates were used, Papua New Guinea had the highest ASPR and Uzbekistan the lowest; China had the highest ASYR and Uzbekistan the lowest. This distinction is important for interpreting national comparisons and for informing resource allocation. The burden of COPD-associated HF demonstrates significant geographic heterogeneity worldwide. While a declining trend is evident in Eastern Europe, the burden remains persistently high in East Asia. This variation stems from the complex interplay of behavioral risk factors, environmental exposures, healthcare infrastructure, and socioeconomic development levels.^48–50 In Eastern Europe, the declining burden may be linked to enhanced public health policies, including more effective tobacco control and improved standards of medical diagnosis and treatment. The region benefits from mature healthcare systems and well-established HF management programs, supported by a strong foundation of relevant research, reflecting significant investment in disease management.⁵¹ Further contributing to this positive trend are successful measures to manage cardiometabolic risk factors, such as obesity and diabetes.⁵² Conversely, the high burden in East Asia is primarily driven by persistent environmental pollution, notably fine particulate matter (PM2.5), high smoking rates, and an aging population.^53–55 Particularly fine particulate matter (PM2.5) pollution is a major driver of the COPD burden in East Asia.⁵⁶ Although many countries have made some progress in controlling PM2.5 pollution, their historical and current pollution levels continue to pose a threat to respiratory health.⁵⁷ Smoking remains the most significant behavioral risk factor, with high prevalence in certain East Asian countries directly contributing to elevated COPD incidence and mortality.⁵⁸ Additionally, indoor air pollution (such as from biomass fuel combustion) remains a significant issue in Asia, which is also a major risk factor for COPD.⁵⁹ Furthermore, population aging exacerbates the burden; increased life expectancy expands the susceptible elderly population, who face greater risk due to physiological decline and comorbidities like diabetes and hypertension.⁶⁰ Additionally, limited access to healthcare systems and inadequate diagnostic capabilities may contribute to underestimation of disease burden or delayed treatment in East Asia.⁵⁸ In summary, the geographic heterogeneity of COPD-associated HF burden reflects the diversity of public health challenges worldwide. Therefore, we require more robust and comprehensive strategies to alleviate the disease burden.

Projections from the BAPC model indicate that the global ASPR and ASYR of COPD-associated HF are both expected to rise steadily between 2022 and 2050 for both males and females. This trend signals a growing challenge for healthcare systems and societies worldwide in the coming decades, as the comorbidity of COPD and heart failure will place increasing demands on resources and negatively impact population health. Critically, the burden is anticipated to be amplified in low- and middle-income countries. Factors such as poverty and cultural beliefs in these regions often hinder consistent treatment for COPD, worsening disease outcomes, and contribute to the overall increase in associated heart failure.^61,62 Therefore, further analysis of the specific drivers of this increasing burden is warranted to develop effective public health strategies to mitigate the rising trends in COPD-associated HF. Consequently, a deeper understanding of the specific factors driving this increasing burden is needed. Identifying these drivers is essential to inform effective public health strategies aimed at addressing this growing challenge.

However, this study also has several limitations. First, the completeness and quality of GBD data represent one such limitation. Despite the GBD research team’s considerable efforts to collect and integrate global health data, the diversity of data sources and regional variations has resulted in disparities in data completeness and quality across different regions, diseases, and time periods.⁶³ Such data gaps or quality issues may compromise the accuracy and generalizability of research findings, particularly when conducting in-depth analyses of specific diseases or populations.⁶³ Secondly, the reliance on models and the complexity of attribution in the GBD database’s comorbidity data present another significant challenge. The coexistence of diseases is a common phenomenon, but GBD studies typically employ complex statistical models to estimate disease burden when handling comorbidity data, rather than directly measuring it.⁶⁴ These models may be based on certain assumptions, and if these assumptions do not align with actual conditions, they may lead to biased results. For example, HF often coexists with multiple other diseases (such as diabetes, chronic obstructive pulmonary disease, etc), and its etiology and clinical manifestations are multifactorial.¹⁶ Accurately attributing DALYs to heart failure itself rather than its underlying causes or comorbidities becomes extremely challenging. Third, biases from diagnostic overlap and misclassification also pose challenges to the accuracy of the GBD database. In epidemiological studies, diagnostic overlap and misclassification are common issues, particularly when symptoms overlap or comorbidities are present.⁶⁵ For example, heart failure shares similar symptoms with COPD, and the complex pathophysiological link between the two makes it difficult to distinguish during diagnosis which disease is responsible for the symptoms.¹⁰ This diagnostic ambiguity may lead to inaccuracies in disease burden estimates. Furthermore, the data sources relied upon by the GBD—such as medical records and death certificates—may themselves carry risks of misclassification. Fourth, although we discuss potential biological, behavioral, and health-system mechanisms that may explain the observed sex-specific, age-specific, and regional disparities, these explanations are based on prior literature and biological plausibility rather than direct evaluation in the present study. GBD data are aggregated and ecological in nature and do not contain individual-level clinical, hormonal, inflammatory, or diagnostic information. Therefore, the mechanisms discussed in this article should be interpreted as hypotheses to be tested in future studies, not as causal explanations of the observed trends.

In addition, this study is based on the GBD 2021 database, which, while currently the most comprehensive global burden of disease assessment available, has an inherent time lag. With the recent release of the GBD 2023 data, estimates of disease burden for some countries and regions may have shifted. Future studies should prioritize the use of the most up-to-date data to refine trend analyses and projections, using GBD 2023 data to validate and supplement our findings. Therefore, the findings of this study should be interpreted as the best estimates based on 2021 data, rather than a real-time reflection of the current disease burden. Finally, the BAPC predictive model provides a baseline forecast for future disease burden based on historical trends. However, future disease trajectories may be significantly influenced by numerous nonlinear factors not incorporated into the current model. For instance, breakthrough therapies in respiratory medicine, widespread implementation of more aggressive tobacco control and air pollution policies, or extreme weather events driven by climate change could fundamentally alter the incidence and prognosis of COPD and HF. Therefore, our projections should be viewed as an alert under a “business-as-usual” scenario, not a definitive future. It underscores the scale of potential public health challenges we face without strengthening current prevention and management efforts. Future research must integrate these dynamic risk factors into more sophisticated models, optimize data collection and analysis methods, and explore multidimensional influences on disease burden.

Compared to previous studies on heart failure associated with chronic obstructive pulmonary disease, first, we analyzed data from 204 countries and territories, 21 global regions, and 5 SDI. This enabled a multi-level (global-regional-national) and multidimensional (gender-age-SDI) assessment of the disease burden, overcoming the limitations of previous studies that typically focused on single regions or short time periods, and providing a truly global perspective. Second, our analysis transcends simple descriptive epidemiology by integrating multiple statistical methodologies. Decomposition analysis quantifies contributions from population growth, aging, and epidemiological shifts; nodal regression identifies critical inflection points in temporal trends; and BAPC modeling projects burden projections through 2050. This multidimensional approach ensures robust analysis of historical trajectories and future risk profiles. Finally, detailed stratification of results by gender, age, sociodemographic indices, and geographic location reveals disparity patterns, providing critical granular insights for designing targeted and effective public health strategy planning. These aspects distinguish our work from previous reports and establish a robust evidence base for global health policy and intervention strategies.

Conclusion

This study provides a systematic analysis of the global disease burden of COPD-associated HF from 1990 to 2021. We documented significant disparities in burden across global, regional, and national levels, as well as by age and sex. Our findings indicate that the burden of COPD-associated HF is shaped by multiple factors, including population growth, aging populations, and socioeconomic conditions. Projections based on established modeling techniques suggest this burden will continue to rise persistently over the coming decades. Addressing this growing challenge demands enhanced international collaboration and the implementation of integrated prevention and management strategies to improve patient outcomes. Future research should prioritize investigating the underlying mechanisms driving COPD-associated HF progression and evaluating the effectiveness of interventions. This will provide a stronger scientific foundation for optimizing clinical management and informing health policy.

Funding Statement

This research was supported by the Applied Basic Research Project of Yunnan Province (Grant No. 202201AY070001-114) and the Yunnan Revitalization Talent Support Program (grant number XDYC-YLWS-2024-0023).

Abbreviations

COPD, Chronic obstructive pulmonary disease; HF, heart failure;; YLDs, years lived with disability;; ASPR, age-standardized prevalence rate; ASYR, age-standardized YLDs rate; APC, annual percentage change; AAPC, average annual percentage change;; GOLD, Chronic Obstructive Lung Disease; BAPC, Bayesian Age-Period-Cohort; CPR, crude prevalence rate; CYR, crude YLDs rate; 95% UI, 95% Uncertainty Interval;; 95% CI, 95% Confidence Interval.

Data Sharing Statement

Data sharing does not apply to this article as no datasets were generated or analysed during the current study.

Ethics Approval and Consent to Participate

This study used publicly available, aggregated, and de-identified data from the Global Burden of Disease (GBD) 2021 study provided by the Institute for Health Metrics and Evaluation (IHME). The GBD data collection was approved by the Institutional Review Board (IRB) of the University of Washington. As this research involved secondary analysis of anonymized data without personal identifiers, additional ethical approval was not required. In accordance with Items 1 and 2 of Article 32 of the Measures for Ethical Review of Life Science and Medical Research Involving Human Subjects (February 18, 2023, China), this study was exempt from further ethical review. Informed consent was not required, as only publicly available, de-identified, aggregated data were used. This study was conducted in accordance with the principles of the Declaration of Helsinki. Consent for publication: Not applicable.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare that they have no competing interests.