Temporal and spatial trends of ischemic heart disease burden in Chinese and subgroup populations from 1990 to 2016 : data from the 2016 global burden of disease study

Background: Ischemic heart disease (IHD) is the leading cause of premature death which poses public health challenges worldwide. This study was to analyze how the IHD burden among Chinese and subgroup populations changes in response to temporal and spatial trends from 1990–2016. Methods : Based on data from the updated estimates in the 2016 Global Burden of Disease (GBD) study, we used years lived with disability (YLDs), years of life lost (YLLs), and disability-adjusted life years (DALYs) to describe the IHD burden. The percentage changes were applied to illustrate temporal and spatial variations of the IHD burden stratified by age, sex, and province, over the periods 1990–2016, 1990–2005, and 2005–2016. We estimate population-attributable fraction (PAF) for 24 modifiable risk factors at the provincial level. Results: YLD rates, YLL rates, and DALY rates for IHD underwent a notable increase among all age groups and increased by 119.4%, 83.3%, and 84.5% nationally from 1990–2016. In YLD rates, an evident rise was seen in females compared to males, while males experienced a more substantial increase than that in females in YLL rates and DALY rates from 1990–2016. Compared with 1990– 2005, annual average changes in overall population in YLL rates and DALY rates showed a tardier increase whereas an opposite increasing trend of YLD rates was observed between 2005 and 2016. Geographically, all provinces saw declines in the YLLs/YLDs ratio from 2005–2016, with seventeen of thirty-three provinces showing an upward trend between 1990 and 2005. Most provinces witnessed a remarkable upsurge in the age-standardised DALY rate from 1990–2016 whereas the economically advantaged region Macao saw the most marked reduction. High systolic blood pressure and high LDL cholesterol remained the two leading risk factors of IHD in all provinces in 2016. Diet high in sodium was the leading behavioral risks in twenty-eight provinces with smoking heading the list in five provinces. Conclusions: Substantial disparities in temporal and spatial trends of the IHD burden emphasize concerns for elderly men and those in economically disadvantaged regions with resource constraints. Regional differences in the IHD burden can be partly explained by modifiable risk factors.

Ischemic heart disease (IHD), the main subcategory of cardiovascular diseases (CVDs), has been identified as the main chronic noncommunicable disease (NCDs). The World Health Organization (WHO) projected that by 2020, IHD will become the primary cause of global death and disability [1].
As the chief contributor to the increase of global CVD burden, IHD caused approximately 26.2 million morbidity and 9.5 million mortality worldwide in 2016 [2,3]. In China, the world's largest developing country with >1.4 billion population, the improved medical care and living standards have raised healthy life expectancy (HALE) from 59.8 years in 1990 to 69.9 years in 2016 [4]. However, the disease burden from CVDs has become a serious public health problem with IHD being the underlying cause of these growing health concerns [5,6]. The increased IHD mortality rate was responsible for the rise in the overall CVDs deaths and IHD is projected to be the leading cause of death in China [7].
Over the past two decades, demographic shifts, transitions in life styles, and enhanced health care have had a widespread and far-reaching influence on epidemiology of IHD [8]. The temporal and spatial trends of IHD burden should therefore be major concerns; however, previous studies paid disproportionately much attention to the overall population, with variations across subgroup populations in China remaining unknown. More importantly, these studies were of inadequate perspectives with limited time frames, lacking specific and detailed trend analyses of annual average transition, and failing to provide comprehensive temporal and spatial assessments which considered both demographic characteristics and provincial disparities [9][10][11]. To bridge these persistent gaps significant to informing public health, this study focuses on the systematic evaluation of variations in age-, sex-, and province-specific IHD burdens during1990-2016, based on data from the 2016 Global Burden of Disease (GBD) study. Targeted preventive strategies and initiatives are then proposed to mitigate IHD burden and allocate health resources effectively [12].

GBD 2016 synopsis
The GBD study was a statistical report examining the health status of the global population. It provided comprehensive, dynamic, and accurate epidemiological models to estimate the burden of disease, injuries, and risk factors, which would help to inform governments of health policies to prioritize resource allocation and meet public health needs at global, regional, national, and subnational levels.
The 2016 GBD study provided years lived with disability (YLDs) of 328 diseases or injuries, disabilityadjusted life years (DALYs) of 333 diseases or injuries, 84 risk factors, and the disease burden of 2892 sequelae in 195 countries and regions around the world from 1990 to 2016 [2][3][4][5]. Chinese subnational assessments adopting locational specific methodology had been put in place since the 2013 GBD study, based on cooperation between the Chinese Center for Disease Control and Prevention (CDC) and the Institute for Health Metrics and Evaluation (IHME) at the University of Washington in the US [13]. The detailed of the study have been fully described elsewhere [2][3][4][5].

Data sources
Updated data on the causes of death were mainly obtained from the Disease Surveillance Points System, the Maternal and Child Surveillance System, and the Death Cause Reporting System of the China CDC. The incidence and prevalence data were collected from disease surveillance, relevant literature researches and national investigations. New systematic reviews and high quality metaanalyses associated with nationwide scientific researches, including the Fifth National Health Service Survey and the Chronic Disease and Risk Factor Surveillance System, were also introduced into the GBD 2016 study [14][15].
The GBD study collected globally accessible data and used unified methods to ensure the results were comparable, and nationally and regionally representative. We also conformed to the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER) to ensure the transparency and reproducibility of results. IHD was defined as per the International Statistical Classification of Diseases (10th Revision) (ICD-10) with diagnosis codes I20-I25 [3].

Estimation of disease burden
The Bayesian meta-regression model DisMod-MR 2.1 was used as the main model to estimate the prevalence of non-fatal diseases. The Cause of Death Ensemble Model (CoDEM) was adopted to appraise cause-specific mortality [4]. The indicators describing the IHD burden in this study included YLDs, years of life lost (YLLs), and DALYs. YLDs were calculated based on disability weights from previous population-surveys in the GBD 2013 and GBD 2015 studies [2,4], and YLLs were calculated by multiplying life expectancy by cause-specific deaths. DALYs were the sum of YLD and YLL. The methodology in detail for the above metrics has been published previously [2][3][4]. The age-

Risk factors
Comparative risk assessment (CRA) framework was used to estimate the excess IHD DALYs attributable to behavioral, metabolic, and environmental risk factors. We used population-attributable fraction (PAF) to estimate the avoidable proportion if the exposure to a specific risk could be reduced to theoretical minimum risk exposure level (TMREL). The methodology in detail for risk assessments have been published elsewhere [5]. 24 risk-IHD pairs in causal relationship at the provincial level in 2016 were established in this study. Table 1 shows the percentage change of YLD, YLL, and DALY rates for IHD in age-sex Chinese subgroups. From 1990 to 2016, YLD, YLL, and DALY rates in both sexes denoted an increase of 119.4%, 83.3%, and 84.5%, respectively. A marked discrepancy in the variation of the IHD burden between sexes was observed: a greater increase in the YLD rate was seen in females (124.4%) subgroup populations, there was an evident reduction in the annual average percentage change in YLL and DALY rates among population aged <70 years during 2005-2016, but an increase was observed in the YLD rate among all age groups, especially those aged 50~69 and above 70 years. No difference in YLL rates in the ≧70-year-old age group was observed over the same period.

IHD burden trends by age and sex
The longitudinal age curves of the YLD, YLL, and DALY rates by sex and year are shown in Figure 1 (A-F). For males, females, and both sexes, YLD, YLL, and DALY rates increased from ages <5 and reached at peak at 70+ years old in 1990, 2005, and 2016. YLL and DALY rates for IHD in males were higher than those in females who (except for those aged 5~14 years) had higher YLD rates.

Spatial variation in YLD and YLL
Despite higher YLLs than YLDs being consistently observed in all regions of China, a substantial decline (16.5%) in the YLLs/YLDs ratio for both sexes was observed from 28.9 in 1990 to 24.1 in 2016.
The YLLs/YLDs ratio varied across provinces. While Guizhou (6.4%), Yunnan (6.3%), and Hunan (4.9%) saw an increase in the YLLs/YLDs ratio, the remaining thirty provinces showed a decrease from 1990 to 2016. While sixteen provinces saw an increase and seventeen a decrease in the YLLs/YLDs ratio from 1990-2005, all provinces in China saw a reduction afterwards (Table 2).
Comparison maps of age-standardised YLD and YLL rates across provinces were employed to visualize the variation in IHD burden before and after 2005 (

Province-level modifiable risk factors
The top two leading risk factors at the provincial level in 2016 showed homogeneity. High systolic blood pressure and high LDL cholesterol, which are metabolic risk factors, remained the two leading risk factors in all provinces, with the former being the leading risk factor in twenty-eight provinces and the latter in the remaining five provinces (Guangxi, Hainan, Hong Kong, Xinjiang, and Yunnan).
The leading behavioral risk factor varied across provinces, with die high in sodium being the leading behavioral risks in twenty-eight provinces and smoking in Gansu, Guangxi, Hainan, Tianjin, and Heilongjiang. For environmental risk factors, IHD DALYs attributable to ambient particulate matter pollution were highest in Tianjin and Beijing, and lowest in Tibet. Conversely, household air pollution from solid fuels accounted for the highest proportion of DALYs in Tibet (15.79%), and the lowest PAF was observed in Shanghai (0.44%), followed by Beijing (0.45%) and Tianjin (0.87%) (Figure 3 and

Age and sex disparities
The burden of IHD varied between ages, sexes, and provinces among the Chinese population. Our age-sex-specific analysis demonstrated that the IHD burden among seniors, particularly those aged 70+ years, was markedly higher than those under 49 years old, indicating that age was positively associated with increased risks of IHD. Since this is related to a growing population and rapid aging, it is expected that the middle-aged and elderly individuals should be the foremost concern when managing the IHD burden in China [16,17]. Marked discrepancies in the burden of IHD between sexes were found. YLDs in females were consistently higher than those in males whereas females experienced lower YLLs and DALYs. Behavioral patterns might partly explain the different trends seen in the IHD burden between sexes. In addition to males being more likely to have dreadful habits such as smoking, alcohol consumption, and poor diet which increases their baseline mortality rate of CVD [3], women tended to be active in seeking health care because they were more concerned in general about physical and mental health [10,18]. These findings show that the IHD burden among middleaged and elderly males should be more widely publicized to manage the overall IHD burden.

Regional trends and disparities
Differences in the IHD burden existed between provinces of China. Most provinces saw a remarkable increase in IHD burden between 1990 and 2016. However, the age-standardised DALY rate experienced a sharp decline in several economically developed provinces, particularly in southeastern coastal areas such as Zhejiang, Hong Kong, and Macao. Socioeconomic status is one possible explanation for the provincial discrepancies, with poorer socioeconomic development being correlated with higher prevalence and death rates of diseases [11,12,19]. Equitable access to essential health services within provinces is a concern in China. The growing IHD burden over the past two decades is particularly obvious in certain remote provinces of less economically developed regions, such as Guizhou and Yunnan, where there is generally lower public awareness of preventive self-care, limited education, weaker health services, and inability to pay for treatment. Furthermore, north-south and west-east disparities in medical resource allocation, environmental conditions, individual metabolism, distribution of risk factors, and lifestyles and behavioral habits also exacerbated the provincial discrepancies [8,11,13,20,21]. The regional disparities were observed across countries worldwide.
High-income countries such as South Korea, Denmark, and Israel experienced a considerate reduction in age-standardised DALY rates during 1990-2016, whereas Bangladesh and Uzbekistan in lowermiddle income regions had an opposite increasing trend [4]. Tailored strategies of mitigating the IHD burden should be priorities for these countries with limited economic development.  15.4 in 2011. This rapidly rising incidence increased in its contribution to IHD prevalence [22]. Over the past two decades, improvements in health care services and medical treatments prolonged HALE of the Chinese population [13]. Unfortunately, the improvements contributed to the population growing and ageing, which increased the number of IHD events, especially among the elderly aged >65 years [4,8,20]. In addition, most old patients with IHD have comorbid peripheral atherosclerotic disease, but guideline-recommend therapies are unavailable to them [23,24]. These present considerable challenges for disability control of IHD. While the IHD

Potential risk factors of IHD
Over the past two decades, demographic shifts, with transitions in economic systems, social structures, environmental factors, life styles, and medical treatments and health-care services have had a widespread and far-reaching impact on potential IHD risk factors [8,27]. We found wide geographic variation in epidemiology of risk factors among provinces in 2016. Metabolic risks factors including high blood pressure and high LDL cholesterol were the leading contributors of IHD burdens, with diet high in sodium and smoking remaining the main sources of the IHD burden attributable to behavioral risks in China. Exposure to environmental risk factors was associated with economic development, some provinces with weak economies, such as Tibet, had the highest IHD burden due to air pollution from solid fuels, with Beijing and Tianjin, rich in resources, experiencing the highest IHD burdens attributable to ambient particulate matter. Consistent with previous studies, our findings suggest the need to take tailored steps to mitigate exposure to risk factors such as physical inactivity, hypertension, high LDL-cholesterol levels, diabetes, and the low rates of education regarding the tertiary prevention of IHD [7,[28][29][30][31][32][33].

Study strengths and implications for policies
Several studies paying attention to the increased burden of IHD in China have been carried out. Most previous studies mainly conducted at the local level with a limited time frame and failed to provide an overall temporal and geographical trend analysis considering both demographic characteristics and provincial disparities [9][10][11]16,17,34]. Our study provided comprehensive estimates of the IHD burden among Chinese and subgroup populations according to age, sex, and regions, with in-depth analyse of regional differences in risk factors at the provincial level. Through this we have uncovered some potentially valuable insights which can influence policy decisions. Firstly, we call for priorities aimed at improving the health level of general populations, particularly among middle-aged and elderly males engaging in high-risk behaviors. Better guidance in implementing targeted health policies at the provincial level needs to be emphasized, while also allowing for more accessible approaches based on different economic levels across provinces of China. Health authorities should pay more attention to the issue of "health inequality" and ensure that everyone is available to psychological well-being, regardless of their socio-economic status, gender, ethnicity, and residency.
Our findings also uncover valuable insights for informing local governments of tailored strategies and programmes to reduce the prevalence of IHD risk factors.
While it improves on previous studies, this study has some limitations. Although our analysis of IHD burden was conducted at national and provincial levels, the disease burden at the county level and urban-rural discrepancies were not examined. The gaps between assessments at national/provincial levels and district levels were somewhat limited to measure a shift in the local disease burden and effectively and specifically influence public policies, which underscores the need for future district level studies [35]. Although the GBD 2016 study had updated the epidemiological data and analytic methodology, changes in diagnostic technology over time might cause inevitable measurement errors in the acquisition of data [2][3][4][5]. With regard to the estimation of YLDs, the calculation of disability weighting is based on multinational results, and precise data on non-fatal outcomes of IHD are insufficient. These may cause uncertainty for China [2].