Laplace Approximation Research Articles

Spatiotemporal models with confounding effects: application on under-five mortality across four sub-Saharan African countries

BackgroundDifferent strategies have been developed to minimize under-five mortality (U5M) in sub-Saharan African (sSA) countries; however, it is still a major health concern for children in the region. Spatiotemporal modeling is important for areal data collected over time. However, when the number of time points and spatial areas is large and the areas are disconnected, fitting the model becomes computationally complex because of the high number of required parameters to be estimated. Therefore, the main aim of this study is to adopt a spatiotemporal dynamic model that includes the confounding effects between time, space, and their interactions with fixed covariates, with a special emphasis on U5M across disconnected sSA countries.MethodWe used nationally publicly representative Demographic and Health Survey (DHS) data for the period from 2000 to 2020. Bayesian spatiotemporal hierarchical modeling with an integrated nested Laplace approximation (INLA) program was used to model the spatiotemporal distribution of U5M among children across 37 districts located in four disconnected sSA regions: Ethiopia, Nigeria, Zimbabwe, and Ghana.ResultsA total of 170,356 under-five children from 37 districts were considered, and 15,467 died before the age of five. The relative risk of U5M in the first DHS was 2.02, which sharply decreased to 0.5 in the recent phase. The proportion of improved access to water, sanitation, clean fuel use, urbanization, and access to health facilities in the district had a significant negative association with U5M. The higher the proportion of these covariates, the lower is the prevalence of childhood mortality.ConclusionThis study revealed evidence of strong spatial, temporal, and interaction effects that influence under-five mortality risk across districts. Improving the women’s literacy index, access to improved water, the use of clean fuel, and the wealth index are associated with an improvement in the risk of mortality among under-five children across the districts. Districts in Nigeria and Ethiopia have the highest risk of U5M; hence, districts in these countries require special attention.

Pemodelan Tingkat Kemiskinan di Provinsi Aceh Menggunakan Distribusi Poisson dan Binomial Negatif

This study aims to model the poverty rate in Aceh Province using the Generalized Linear Models (GLMs) approach with the Integrated Nested Laplace Approximation (INLA) method. The models used in this study include Poisson and Negative Binomial regression. The data used includes variables of per capita expenditure, population density, average years of schooling, open unemployment rate, and labor force participation. Multicollinearity tests were conducted using Tolerance and Variance Inflation Factor (VIF) values to ensure the absence of multicollinearity problems between predictor variables. The results show that the Negative Binomial model is better than the Poisson model, as indicated by the smaller WAIC value. The average years of schooling variable has a significant negative effect, while the open unemployment rate has a significant positive effect on the poverty rate. Meanwhile, the variables of expenditure per capita, population density, and labor force participation make smaller and partially insignificant contributions in the model. Thus, increasing access to education and reducing unemployment are key factors in poverty alleviation efforts in Aceh Province. The results of this study are expected to serve as a reference for policy makers in formulating effective strategies to reduce poverty.

Joint Spatial Analysis of Low Birth Weight and Stunting in West African Countries

Introduction: The burden of childhood morbidity and mortality are still huge in most sub-Saharan African countries with West African sub-region contributing largely to the burden. Previous findings have demonstrated strong link between early life events such as low birth weight (LBW) with later events particularly malnutrition. We aim at estimating the specific and shared spatial patterns of LBW and stunting among under-five children in multiple West African countries. Methods: Data set for the study was sourced from the Demographic and Health Surveys conducted in fourteen West African countries. We used a Bayesian shared component model allows us to split the spatial surface into those specific to each of the outcomes and one shared by the two, with inference based on a Bayesian approximation procedure through the integrated nested Laplace approximation. Results: The findings show spatial clustering in the shared and specific effects of the health outcomes, demonstrating high likelihood in northern Nigeria spanning through Niger and that the spatial pattern for the shared effects are similar to those of the specific effects of stunting. Furthermore, mother’s level of education, attendance in antenatal care and household wealth index are strongly associated with the shared health outcomes. Conclusion: The study provides insight into the spatial pattern of LBW and stunting among West African children and can be useful in targeted interventions in regions with high burden of LBW and malnutrition which may include more advocacy that promote the use of antenatal care services during pregnancy.

Spatial-Structural Mechanisms of Racialized Disparities in Overdose Mortality: A Spatiotemporal Analysis.

Efforts to understand and respond to the opioid crisis have focused on overdose fatalities. Overdose mortality rates (ratios of overdoses resulting in death) are rarely examined though they are important indicators of harm reduction effectiveness. Factors that vary across urban communities likely determine which community members are receiving the resources needed to reduce fatal overdose risk. Identifying communities with higher risk for overdose mortality and understanding influential factors is critical for guiding responses and saving lives. Using incident reports and mortality data from 2018 to 2021, we defined overdose mortality ratios across Milwaukee at the census tract level. To identify neighborhoods displaying higher mortality than predicted, we used Integrated nested Laplace approximation to define standardized mortality ratios (SMRs) for each tract. Geospatial and spatiotemporal analyses were used to identify emerging hotspots for high mortality risk. Overall, mortality was highest in Hispanic and lowest in White communities. Communities with unfavorable SMRs were predominantly Black or Hispanic, younger, less employed, poorer, less educated, and had higher incarceration rates and worse mental and physical health. Communities identified as hotspots for overdoses were predominantly non-White, poorer, and less employed and educated with worse mental and physical health, higher incarceration rates, and less housing stability. The findings demonstrate that overdose mortality rates vary across urban communities and are influenced by racial disparities. A framework that enables identification of challenged communities and guides community responses is needed.

Penentuan Model Terbaik dalam Pengaruh Lapangan Usaha dan Nilai Ekspor Terhadap Pertumbuhan Ekonomi di Indonesia

This study aims to evaluate the effect of business sector and export value on economic growth in Indonesia using Generalised Linear Models (GLM) and Integrated Nested Laplace Approximation (INLA) approaches with Gamma and Gaussian distributions. The results showed that agriculture, forestry and fishing (PKP), mining and quarrying (PP) and fiscal (PK) sectors have a significant impact on economic growth, while real estate (RE) and export value (NE) are not significant. Based on the Akaike Information Criterion (AIC), the best model is the GLM with Gaussian distribution, which provides a balance between model complexity and the ability to explain the data. The INLA method provided consistent results with informative confidence intervals, but did not outperform the GLM in model evaluation. This study provides insights for policy makers to prioritise the PKP, PP and PK sectors in promoting economic growth and recommends further research with additional variables and more complex analytical methods for more in-depth results.

Mapping Drug-Resistant Tuberculosis Treatment Outcomes in Hunan Province, China.

Drug-resistant tuberculosis (DR-TB) remains a major public health challenge in China, with varying treatment outcomes across different regions. Understanding the spatial distribution of DR-TB treatment outcomes is crucial for targeted interventions to improve treatment success in high-burden areas such as Hunan Province. This study aimed to map the spatial distribution of DR-TB treatment outcomes at a local level and identify sociodemographic and environmental factors associated with poor treatment outcomes in Hunan Province, China. A spatial analysis was conducted using DR-TB data from the Tuberculosis Control Institute of Hunan Province, covering the years 2013 to 2018. The outcome variable, the proportion of poor treatment outcomes, was defined as a composite measure of treatment failure, death, and loss to follow-up. Sociodemographic, economic, healthcare, and environmental variables were obtained from various sources, including the WorldClim database, the Malaria Atlas Project, and the Hunan Bureau of Statistics. These covariates were linked to a map of Hunan Province and DR-TB notification data using R software version 4.4.0. The spatial clustering of poor treatment outcomes was analyzed using the local Moran's I and Getis-Ord statistics. A Bayesian logistic regression model was fitted, with the posterior parameters estimated using integrated nested Laplace approximation (INLA). In total, 1381 DR-TB patients were included in the analysis. An overall upward trend in poor DR-TB treatment outcomes was observed, peaking at 14.75% in 2018. Deaths and treatment failures fluctuated over the years, with a notable increase in deaths from 2016 to 2018, while the proportion of patients lost to follow-up significantly declined from 2014 to 2018. The overall proportion of poor treatment outcomes was 9.99% (95% credible interval (CI): 8.46% to 11.70%), with substantial spatial clustering, particularly in Anxiang (50%), Anren (50%), and Chaling (42.86%) counties. The proportion of city-level indicators was significantly associated with higher proportions of poor treatment outcomes (odds ratio (OR): 1.011; 95% CRI: 1.20 December 2024 001-1.035). This study found a concerning increase in poor DR-TB treatment outcomes in Hunan Province, particularly in certain high-risk areas. Targeted public health interventions, including enhanced surveillance, focused healthcare initiatives, and treatment programs, are essential to improve treatment success.

Spatial variation in housing construction material in low- and middle-income countries: A Bayesian spatial prediction model of a key infectious diseases risk factor and social determinant of health.

Housing infrastructure and quality is a major determinant of infectious disease risk and other health outcomes in regions where vector borne, waterborne and neglected tropical diseases are endemic. It is important to quantify the geographical distribution of improvements to dwelling components to identify and target resources towards populations at risk. This study aimed to model the sub-national spatial variation in housing materials using covariates with quasi-global coverage and use the resulting estimates to map predicted coverage across the world's low- and middle-income countries. Data on materials used in dwelling construction were sourced from nationally representative household surveys conducted since 2005. Materials used for construction of flooring, walls, and roofs were reclassified as improved or unimproved. Households lacking location information were georeferenced using a novel methodology. Environmental and demographic spatial covariates were extracted at those locations for use as model predictors. Integrated nested Laplace approximation models were fitted to obtain, and map predicted probabilities for each dwelling component. The dataset compiled included information from households in 283,000 clusters from 350 surveys. Low coverage of improved housing was predicted across the Sahel and southern Sahara regions of Africa, much of inland Amazonia, and areas of the Tibetan plateau. Coverage of improved roofs and walls was high in the Central Asia, East Asia and Pacific and Latin America and the Caribbean regions. Improvements in all three components, but most notably floors, was low in Sub-Saharan Africa. The strongest determinants of dwelling component quality related to urbanization and economic development, suggesting that programs should focus on supply-side interventions, providing resources for housing improvements directly to the populations that need them. These findings are made available to researchers as files that can be imported into a GIS for integration into relevant analyses to derive improved estimates of preventable health burdens attributed to housing.

Joint modelling of dyadic and monadic count outcomes: an application to modelling forced migration flows

The aim of this study is to explore the adoption of a joint modelling framework for dealing with dyadic and monadic count outcomes with excess zeros simultaneously via a common latent structure. As a case study, we consider the problem of identifying the different push and pull factors of cross-border forced migration and internal displacement. We consider a full panel data analysis and estimate a random effects joint hurdle model following the Bayesian paradigm; the resultant posterior is approximated through the integrated nested Laplace approximation.

Fast spatial simulation of extreme high-resolution radar precipitation data using integrated nested Laplace approximations

Abstract Aiming to deliver improved precipitation simulations for hydrological impact assessment studies, we develop a methodology for modelling and simulating high-dimensional spatial precipitation extremes, focusing on both their marginal distributions and tail dependence structures. Tail dependence is crucial for assessing the consequences of extreme precipitation events, yet most stochastic weather generators do not attempt to capture this property. The spatial distribution of precipitation occurrences is modelled with four competing models, while the spatial distribution of nonzero extreme precipitation intensities are modelled with a latent Gaussian version of the spatial conditional extremes model. Nonzero precipitation marginal distributions are modelled using latent Gaussian models with gamma and generalized Pareto likelihoods. Fast inference is achieved using integrated nested Laplace approximations. We model and simulate spatial precipitation extremes in Central Norway, using 13 years of hourly radar data with a spatial resolution of 1×1km2, over an area of size 6,461km2, to describe the behaviour of extreme precipitation over a small drainage area. Inference on this high-dimensional data set is achieved within hours, and the simulations capture the main trends of the observed precipitation well.

Laplace-based strategies for Bayesian optimal experimental design with nuisance uncertainty

Finding the optimal design of experiments in the Bayesian setting typically requires estimation and optimization of the expected information gain functional. This functional consists of one outer and one inner integral, separated by the logarithm function applied to the inner integral. When the mathematical model of the experiment contains uncertainty about the parameters of interest and nuisance uncertainty, (i.e., uncertainty about parameters that affect the model but are not themselves of interest to the experimenter), two inner integrals must be estimated. Thus, the already considerable computational effort required to determine good approximations of the expected information gain is increased further. The Laplace approximation has been applied successfully in the context of experimental design in various ways, and we propose two novel estimators featuring the Laplace approximation to alleviate the computational burden of both inner integrals considerably. The first estimator applies Laplace’s method followed by a Laplace approximation, introducing a bias. The second estimator uses two Laplace approximations as importance sampling measures for Monte Carlo approximations of the inner integrals. Both estimators use Monte Carlo approximation for the remaining outer integral estimation. We provide four numerical examples demonstrating the applicability and effectiveness of our proposed estimators.

Joint quantile regression of longitudinal continuous proportions and time-to-event data: Application in medication adherence and persistence.

This study introduces a novel joint modeling framework integrating quantile regression for longitudinal continuous proportions data with Cox regression for time-to-event analysis, employing integrated nested Laplace approximation for Bayesian inference. Our approach facilitates an examination across the entire distribution of patient health metrics over time, including the occurrence of key health events and their impact on patient outcomes, particularly in the context of medication adherence and persistence. Integrated nested Laplace approximation's fast computational speed significantly enhances the efficiency of this process, making the model particularly suitable for applications requiring rapid data analysis and updates. Applying this model to a dataset of patients who underwent treatment with atorvastatin, we demonstrate the significant impact of targeted interventions on improving medication adherence and persistence across various patient subgroups. Furthermore, we have developed a dynamic prediction method within this framework that rapidly estimates persistence probabilities based on the latest medication adherence data, demonstrating integrated nested Laplace approximation's quick updates and prediction capability. The simulation study validates the reliability of our modeling approach, evidenced by minimal bias and appropriate credible interval coverage probabilities across different quantile levels.

Non-Gaussian Stochastic Volatility models

Stochastic volatility models are fundamental tools in finance for accurately estimating and managing risks due to their ability to accommodate a time-varying volatility structure. However, a relevant constraint within these models is the reliance on Gaussian processes to model the latent (log-)variance, which can limit their ability to effectively capture events such as sudden jumps or spikes in the latent volatility. To address this limitation, we propose in this short paper a non-Gaussian SV model utilizing an inference procedure that combines Laplace and Variational Bayes approximations. Our study showcases the advantages of this correction in modeling the conditional variance of Bitcoin's return series.

Assessing the Impact of Geopolitical Risk on Longevity Bond Pricing: Insights from Bayesian Multivariate Regression

AbstractThis paper investigates the multivariate pricing of coupon longevity bonds (CLBs) using the Fama–French–Lee–Carter (FF–LC) five-vector model in the framework of Bayesian integrated nested Laplace approximation (INLA) in the presence of geopolitical risk (GPR). The variance-covariance and correlation matrices are utilized to capture the interdependence between factors. We prove the generalization of multivariate Bayesian INLA with the basic probability assignment which is utilized as a posterior uncertainty belief associated with the GPR uncertainty category (a rich representation of GPR uncertainty) that is an element of the frame of discernment in the CLB posterior estimation. INLA Bayesian principal component analysis (INLA-BPCA) is applied to the model prediction parameters generating a multivariate normally distributed posterior. The deviance information criterion (DIC) assesses optimal factor selection. The results show that the BPCA posterior gains a feature that allows for a balance between the goodness-of-fit and complexity in hierarchical model selection by incorporating the retained principal components (or the effective number of parameters) from the DIC formula. Furthermore, it is also evident in our results, that the DIC outperforms the Bayesian information criterion (BIC), and Watanabe–Akaike information criterion (WAIC). The DIC is more suitable for Bayesian-based parametric models with high complexity. Lastly, the INLA-BPCA-DIC is applied to select the best longevity factors that yield a low longevity price of risk for insurers and practitioners, to attenuate the risks associated with investing in CLBs in the presence of geopolitical uncertainty shocks.

Invaders break assembly rules to beat the natives: how cheatgrass cheats

Understanding how some introduced plants achieve invasive status while most simply become naturalized is a fundamental question in invasion ecology. Traditional approaches comparing native and introduced plants have linked ruderal traits such as annual life history, high fecundity, and rapid growth rates to invasiveness. However, they do not explain why other introduced species bearing similar traits fail to become invasive, possibly because generic comparisons ignore local processes that drive community assembly. Herein, we contrasted native and introduced annuals in the context of local successional processes to elucidate how introduced annual bromes like cheatgrass (Bromus tectorum) overtake perennial grasslands in the intermountain western United States. We created disturbed plots and seeded them first with annuals representing natives, naturalized species, or invasive bromes. We then seeded plots with native perennial community dominants to examine how the different annuals influenced succession. Native annuals established transient populations that facilitated perennial establishment compared to unseeded controls, enabling the shift to perennial dominance. Naturalized annuals mirrored the natives, but invasive annuals maintained robust populations at high biomass that inhibited perennial establishment and impeded succession. Mechanistically, invasive annuals reduced soil moisture and elevated plant biomass, litter, and soil N. However, only litter abundance correlated with perennial seedling recruitment across treatments. Overall, litter showed a unimodal relationship wherein lower litter abundance associated with native and naturalized annuals appeared to facilitate perennial seedling establishment while higher litter levels generated by invasives appeared to suppress perennial establishment and inhibit succession. Additional experiments provided little support for the roles of pathogen spill-over or plant-soil feedbacks favoring the introduced bromes. The domination of perennial grasslands by annual bromes may be driven by litter buildup that allows these introduced plants to break local succession rules by acting as both early seral and climax species. Traits like litter accumulation may strongly influence invasion outcomes but are indistinguishable using trait comparisons lacking community context.

Analysis of volatile compounds in Xiangjiao baijiu from different storage containers and years based on HS-GC-IMS and DI-GC–MS

The volatile compounds in 16 different storage containers and years of Xiangjiao Baijiu (XJBJ) were compared and analyzed via direct injection (DI) combined with gas chromatography–mass spectrometry (GC–MS) and headspace extraction (HS) coupled with gas chromatography-ion mobility spectrometry (HS-GC-IMS) for the first time. Through HS-GC-IMS analysis, it was found that the succession rules of 14 compounds such as furfural during aging process. A total of 60 compounds were identified using DI-GC–MS. Twenty-five of these compounds were further quantified, and 18 compounds had odor activity values (OAVs) > 1, which were important contributor to aroma of XJBJ. Among them, those with OAVs >1000 included ethyl hexanoate, ethyl octanoate, ethyl butanoate, and ethyl pentanoate. Combining the results of quantitative, OAVs and partial least squares-discriminant analysis (PLS-DA) revealed that 10 compounds such as ethyl octanoate were the important compounds that lead to the differences between different storage types of XJBJ.

Bayesian Inference for Long Memory Stochastic Volatility Models

We explore the application of integrated nested Laplace approximations for the Bayesian estimation of stochastic volatility models characterized by long memory. The logarithmic variance persistence in these models is represented by a Fractional Gaussian Noise process, which we approximate as a linear combination of independent first-order autoregressive processes, lending itself to a Gaussian Markov Random Field representation. Our results from Monte Carlo experiments indicate that this approach exhibits small sample properties akin to those of Markov Chain Monte Carlo estimators. Additionally, it offers the advantages of reduced computational complexity and the mitigation of posterior convergence issues. We employ this methodology to estimate volatility dependency patterns for both the SP&500 index and major cryptocurrencies. We thoroughly assess the in-sample fit and extend our analysis to the construction of out-of-sample forecasts. Furthermore, we propose multi-factor extensions and apply this method to estimate volatility measurements from high-frequency data, underscoring its exceptional computational efficiency. Our simulation results demonstrate that the INLA methodology achieves comparable accuracy to traditional MCMC methods for estimating latent parameters and volatilities in LMSV models. The proposed model extensions show strong in-sample fit and out-of-sample forecast performance, highlighting the versatility of the INLA approach. This method is particularly advantageous in high-frequency contexts, where the computational demands of traditional posterior simulations are often prohibitive.

Increasing Access, Equitability, and Rigor in the Assessment of Neighborhood Mortgage Discrimination.

Mortgage discrimination alters the distribution of investment, opportunity, and economic advantage-key contributors of health disparities. Leveraging Home Mortgage Disclosure Act data, we assessed mortgage denial risk in 380 U.S. urban areas. We estimated the risks by census tract-relative to the urban-specific average-using a Bayesian spatial model with conditionally autoregressive distributions fitted with integrated nested Laplace approximation. This approach borrows information through spatial and non-spatial smoothing, resulting in stable estimates in the presence of sparse data. The method, publicly accessible, allows researchers to apply our approach, fostering deeper insights into mortgage lending discrimination and systematic neighborhood disinvestment.

Iterative Methods for Vecchia-Laplace Approximations for Latent Gaussian Process Models

Latent Gaussian process (GP) models are flexible probabilistic nonparametric function models. Vecchia approximations are accurate approximations for GPs to overcome computational bottlenecks for large data, and the Laplace approximation is a fast method with asymptotic convergence guarantees to approximate marginal likelihoods and posterior predictive distributions for non-Gaussian likelihoods. Unfortunately, the computational complexity of combined Vecchia-Laplace approximations grows faster than linearly in the sample size when used in combination with direct solver methods such as the Cholesky decomposition. Computations with Vecchia-Laplace approximations can thus become prohibitively slow precisely when the approximations are usually the most accurate, that is, on large datasets. In this article, we present iterative methods to overcome this drawback. Among other things, we introduce and analyze several preconditioners, derive new convergence results, and propose novel methods for accurately approximating predictive variances. We analyze our proposed methods theoretically and in experiments with simulated and real-world data. In particular, we obtain a speed-up of an order of magnitude compared to Cholesky-based calculations and a 3-fold increase in prediction accuracy in terms of the continuous ranked probability score compared to a state-of-the-art method on a large satellite dataset. All methods are implemented in a free C++ software library with high-level Python and R packages. Supplementary materials for this article are available online, including a standardized description of the materials available for reproducing the work.

Efficient Tuning of an Isotope Separation Online System Through Safe Bayesian Optimization with Simulation-Informed Gaussian Process for the Constraints

Optimizing process outcomes by tuning parameters through an automated system is common in industry. Ideally, this optimization is performed as efficiently as possible, using the minimum number of steps to achieve an optimal configuration. However, care must often be taken to ensure that, in pursuing the optimal solution, the process does not enter an “unsafe” state (for the process itself or its surroundings). Safe Bayesian optimization is a viable method in such contexts, as it guarantees constraint fulfillment during the optimization process, ensuring the system remains safe. This method assumes the constraints are real-valued and continuous functions. However, in some cases, the constraints are binary (true/false) or classification-based (safe/unsafe), limiting the direct application of safe Bayesian optimization. Therefore, a slight modification of safe Bayesian optimization allows for applying the method using a probabilistic classifier for learning classification constraints. However, violation of constraints may occur during the optimization process, as the theoretical guarantees of safe Bayesian optimization do not apply to discontinuous functions. This paper addresses this limitation by introducing an enhanced version of safe Bayesian optimization incorporating a simulation-informed Gaussian process (GP) for handling classification constraints. The simulation-informed GP transforms the classification constraint into a piece-wise function, enabling the application of safe Bayesian optimization. We applied this approach to optimize the parameters of a computational model for the isotope separator online (ISOL) at the MYRRHA facility (Multipurpose Hybrid Research Reactor for High-Tech Applications). The results revealed a significant reduction in constraint violations—approximately 80%—compared to safe Bayesian optimization methods that directly learn the classification constraints using Laplace approximation and expectation propagation. The sensitivity to the accuracy of the simulation model was analyzed to determine the extent to which it is advantageous to use the proposed method. These findings suggest that incorporating available information into the optimization process is valuable for reducing the number of unsafe outcomes in constrained optimization scenarios.

The Trends in Prevalence of Blindness Caused by Refraction Disorders in China from 1990 to 2019 and Its Predictions: Findings from the Global Burden of Disease Study 2019

ABSTRACT Purpose To examine the burden of blindness caused by refraction disorders (BCRD) in China over the past 30 years by year, age, and sex, and to estimate future projections. Methods Data from the Global Burden of Diseases (GBD) 2019 database were used to analyze the number of cases and age-standardized prevalence rates (ASPRs) of BCRD in China from 1990 to 2019. We focused on changes over time using estimated annual percentage changes (EAPCs). In addition, we performed the Nordpred analysis and the Bayesian Age-Period-Cohort (BAPC) model with integrated nested Laplace approximations to predict the BCRD burden from 2020 to 2034. Results The number of prevalent cases due to BCRD increased from 750,956.7 (95% uncertainty interval [UI] 636,381.24–872,040.62) in 1990 to 1,145,881.76 (95% UI 931,966.43–1,342,338.18) in 2019. The ASPRs of BCRD showed a decreasing trend, with EAPCs of −0.58 (95% confidence interval [CI] −0.81–0.36). The older and female populations had a higher BCRD burden. The number of prevalent cases due to BCRD is projected to continue to increase from 1.33 million in 2020 to 1.86 million in 2034. The ASPR also showed an increasing trend over the next 15 years. Conclusion Over the past three decades, the prevalence of BCRD in China has improved in both sexes and will continue to increase in the next 15 years. This study highlights the importance of prevention of BCRD, especially for women and the elderly.