Age-period-cohort analysis of ischemic stroke deaths attributable to physical inactivity in different income regions

This study assessed the global and regional burden of IS (ischemic stroke) deaths due to LPA (low physical activity) from 1990 to 2019, analyzed regional, sex, and age differences in ASMR (age-standardized mortality rate), and provided a comprehensive understanding of the impact of age, period, and cohort on low physical activity related ischemic stroke ASMR. We conducted an APC (age-period-cohort) analysis of the global and four World Bank income level regions’ IS mortality data attributed to LPA from 1990 to 2019, using the GBD2019 database, and the results showed that the global net drift of the Ischemic stroke age-standardized mortality attributable to low physical activity was − 1.085%[95% CI: − 1.168, − 1.003].The ASMR drop is most pronounced in the high-income zone, with a net drift of − 2.473% [95% CI: − 2.759, − 2.187] across the four income groups. The influence of age on mortality is increasing in the worldwide old population, while the period and cohort effects are decreasing. We also performed a Joinpoint regression analysis, which revealed that the specific time of considerable drop in ASMR of IS in the global LPA population was 2002–2007, with an APC of -2.628%. The specific period of considerable drop in ASMR in high-income regions with the highest variation was 1999–2007, with an APC = − 4.726%. The global burden of public health deaths caused by LPA is diminishing, with the most notable progress observed in high-income regions. However, in low and lower-middle income areas, the situation continues to deteriorate. Within the global elderly population, the effects of age on mortality is increasing, while the effects of period and cohort are diminishing. These trends vary across income levels, highlighting the necessity for enhanced international collaboration to formulate context-specific public health strategies aimed at enhancing cardiovascular health on a global, regional, and national scale.


LPA
Low physical activity IS Ischemic stroke APC Analysis Age-period-cohort analysis ASMR Age-standardized mortality rate The World Health Organization (WHO) has initiated a global Action plan on non-communicable diseases (NCDS) with the objective of diminishing premature mortality resulting from NCDS by 25% by the year 2025 1 .The third Sustainable Development Goal of the United Nations endeavors to reduce untimely deaths caused by non-communicable diseases by one-third by the year 2030 2 .Cardiovascular disease (CVD) stands as the primary cause of morbidity, disability, and mortality on a global scale, thereby leading to significant health disparities 3 .The impact of Stroke as a disease burden should not be underestimated.
Stroke is a prevalent medical condition characterized by a substantial occurrence rate, significant disability rate, elevated mortality rate, and a notable recurrence rate, commonly referred to as the "four high" characteristics.Globally, it is estimated that one in every four individuals will experience a stroke, resulting in a distressing fatality every six seconds and a disabling event every six seconds.Consequently, stroke imposes considerable economic burdens and physical and psychological suffering upon the affected individuals and their families.

Assessment methods of IS disease burden attributed to various risk factors
For each risk factor, the GBD2019 study quantifies the proportion of disease burden that could have been prevented if exposure levels were maintained at the minimum risk level, defined as the theoretical minimum risk exposure level (TMREL) 36,37 .The population attributable fraction (PAF) is obtained by comparing the theoretical minimum risk exposure level with the exposure level of specific population on the premise that the exposure level of other risk factors remains unchanged 38 .According to the framework of comparative risk assessment theory, GBD 2019 calculates the attributable disease burden caused by the main risk factors of IS with indicators such as PAF and DALYs of IS disease burden.
The number of IS deaths attributed to LPA can be calculated by multiplying the PAF of LPA by the number of IS deaths.Age-standardized mortality rates for different regions, years and genders were calculated using the world standard population structure for 2000-2025 39 .

Statistical analysis
APC An age-period-cohort (APC) model was used to analyze trends in IS deaths attributable to LPA globally and in four World Bank Income Levels regions, and the effects of age, period, and cohort on mortality.This approach has been widely used in descriptive epidemiological analysis of various chronic diseases, helping to describe and explain long-term trends in chronic disease mortality over time, as well as to identify disease trends that are related to age-related biological factors, social influences, and technological contributions 28,[40][41][42] .APC analysis can estimate the trend effect of age, period and cohort at the same time, and eliminate the mutual interference factors among them, effectively avoiding the problem of complete collinearity among the three dimensions of age, period and cohort.
In the APC model, the input data were the IS mortality rates attributed to LPA globally and in 4 World Bank Income Levels regions from 1990 to 2019.Age and period intervals in the APC model have to be equal, so we've determined six periods with 5-year intervals: 1990-1994, 1995-1999, 2000-2004, 2005-2009, 2010-2014, and  2015-2019.The 14 age groups over 25 that were chosen at 5-year intervals from the GBD 2019 database are referred to as age groups.For the birth cohort, due to different model algorithms, there are two situations.In the age, period and cohort interaction study, the birth cohort is divided into groups with 5-year intervals, consisting of 19 cohort groups from 1900-1904 to 1990-1994; in the age, period and cohort effect study, the birth cohort is divided into groups at 10-year intervals, from 1895-1904 to 1985-1994 28,40 .But this difference in algorithm does not affect the results of the study.
We used the APC model to calculate net drift, local drift, longitudinal age curves, and ratios rate (RR) of IS mortality across periods (period effects) and cohort (cohort effects).The net drift represents the annual percentage change over time in the age-standardized mortality rate of IS, indicating an overall log-linear trend by period and cohort 43 .Local drift represents the annual percentage change over time in age-standardized mortality rates of IS for each age group.The longitudinal age curve represents the expected age-specific rate in the reference cohort adjusted for period effects 43 .Period effects refer to the relative risk of IS mortality in different periods in the study area.PRR (period rate ratios, p/p 0 ) is the ratio of age-specific rates in period p relative to reference period p 0 .Cohort effects refer to the relative risk of IS mortality in different birth cohorts in the study area.CRR (cohort rate ratios, c/c 0 ) is the ratio of age-specific rates in cohort c relative to reference cohort c 0 .In addition, the reference period of this study is 2000, and the reference age is 50 years old.
We performed our analysis through the National Cancer Institute's Age Cycle Network tool, which uses weighted least squares statistical methods and assumes that counting data follow a Poisson distribution 28 .

Joinpoint
A joinpoint regression model was used to o investigate temporal trends in the burden of IS deaths attributable to LPA between 1990 and 2019 28 .Through this method, we can identify the significant change points ((join points) in the time change trend, and obtain the APC and AAPC values 28 .AAPC acts as an aggregate measure of trends in the burden of IS deaths over a fixed time interval and is calculated as a weighted average of annual percentage change (APC) 28 .AAPC values and their corresponding 95% confidence intervals (CI) reflect trends in age-standardized mortality 44 .In addition, the Wald χ 2 test was given in the joinpoint regression model to assess statistical significance.All statistical tests were double-tailed and statistical significance was defined as P < 0.05.Data analysis was performed using R software version 4.2.3, while the join point regression program version 4.9.1.0was used to perform the join point regression model and trend analysis 28,44 .

Ethical statement
The data used in this study are all publicly available data in GBD2019, and the GBD study was reviewed and approved by the University of Washington's Institutional Review Board.

Ethical committee
The study was compliant with the Guidelines for Accurate and Transparent Health Estimates Reporting, and the University of Washington Institutional Review Board reviewed and approved the waiver of informed consent for GBD 2019.

Global and regional trends in the mortality of Ischemic stroke attributable to Low Physical Activity, 1990-2019
Figure 1, Table 1, Supplementary Tables 1 and 4

Trends in IS mortality attributable to LPA by age and cohort
The pattern of the IS death rate rising with age is depicted in Fig. 2a.We may observe that while the death rate from IS has grown somewhat internationally overall for those over 80, it has fallen dramatically when compared to the 1990-1994 period.The pattern of increase is similar in high-income areas.The age group over 85 years old in the upper-middle income area showed a notable growth between 2000 and 2004.The age group over 85 saw a notable increase in the low-income area between 2000 and 2004, albeit at a slower rate than in the uppermiddle income area.The IS mortality rate for people over 80 in low-income areas started to exhibit a notable rising trend in all 6 periods, while each period's difference was not very noticeable.
With the exception of low-income regions, Fig. 2b demonstrates that the death rate from IS in the birth cohort of patients of all ages decreased globally and in the other three World Bank income level regions, suggesting a relatively low risk of IS death in these regions' recently born cohort of all ages.All of these regions' birth cohorts, both male and female, showed differing degrees of reduction.Nonetheless, the birth cohort of patients aged 75-94 years showed an increase in the death rate of IS in low-income areas, suggesting a higher death rate in the most recent birth cohort in these age groups.In low-income areas, birth cohorts for both genders, aged 75-94, indicated a little increase trend.It implies that IS patients between the ages of 75 and 94 have a higher chance of dying in low-income locations, regardless of gender.

The age, period, and cohort effects on Ischemic stroke mortality attributable to Low Physical Activity were analyzed by APC (web)
Figure 3, Supplementary Tables 4 and 5  Regarding local drift, the global local drift values for every age group of females were less than zero, suggesting a global trend of declining IS mortality among females of all ages related to LPA.Furthermore, we can observe that the drop in IS death rates for both males and females over the age of 70 has slowed globally.Furthermore, we discovered that only the upper-high income regions-of the four World Bank income levels regions-had local drift values for both male and female that were less than 0. This suggests that the IS death rate in these regions declined to varying degrees for all age groups, with the rate of decline slowing down more rapidly from 35 to 64 years old and progressively slowing down beyond 65.Local drift values for the IS death rate between 25 and 54 years old were higher than 0 in low-income areas, suggesting an increasing trend to varying degrees for this age group; local drift values for the IS death rate between 55 and 69 years old were lower than 0, suggesting a decreasing trend to varying degrees for this age group; local drift values for the IS death rate over 70 years old are higher than 0. It demonstrated that, to varying degrees, the death rate of IS over 70 exhibited a growing tendency.In the other two regions, the IS death rate of the low age group increased in different degrees, and the IS death rate of the high age group decreased in varying degrees.The difference lies in the age point where the mortality trend turns from rising to decreasing.In high-income areas, the mortality rate of IS between 25 and 44 years old showed an increasing trend, while that of IS above 45 years old showed a decreasing trend.In the lower-middle income areas, the death rate of IS between 25 and 59 years old showed an increasing trend, and the death rate of IS over 60 years old showed a decreasing trend (Fig. 3a).
Longitudinal age curves show that IS mortality attributable to LPA increases gradually with age in all age groups and all study regions.It shows similar patterns of increase for male and female.There is a remarkable similarity in the longitudinal age curves of the global and 4 different World Bank Income levels regions, namely, the IS mortality over 80 years increases rapidly in low physical activity population compared with other age groups.The difference is that the IS mortality increases peaked at different levels, with the lowest in high-income regions (28.603% [95% CI: 26.484, 30.892]), especially among male in high-income regions (20.748% [95% CI: 19.049, 22.599]), and the highest in lower-middle income regions (120.663%[95% CI: 115.894, 125.627]) (Fig. 3b, Supplementary Table 6).
As shown in Fig. 3c and Supplementary Table 7, the global period RR for both male and female showed a downward trend, from 1.043% [95%CI: 1.030, 1.057] to 0.815% [95%CI: 0.804, 0.827].Although the global period RR for male, female and the population as a whole differed slightly in 1990 (both male's and female's period RR were slightly higher than that of the whole population period RR), the situation crossed in 2000, and then the period RR for female IS declined more rapidly than that of male.The period RR in high-income regions also showed a much larger decline than the global, falling from 1.341% [95%CI: 1.293, 1.391] to 0.759% [95%CI: 0.725, 0.956].Furthermore, we observe that the period RR in high-income regions stayed relatively steady from 2010 to 2019, with only a slight increase during the period.In the upper-middle income regions, the period RR for both male and female and the overall population showed a decreasing trend, from 1.093% [95%CI: 1.067, 1.120] to 0.734% [95%CI: 0.714, 0.754].The period RR of the male, female, and overall period RR in the lower-middle income districts, however, exhibited the traits of first growing and then falling, with a small overall change amplitude.
As shown in Fig. 3d and Supplementary Table 8, the cohort RR for both male and female in the global showed a decreasing trend, from 2.298% [95%CI: 2.231, 2.367] in 1895 to 0.953% [95%CI: 0.776, 1.170] in 1994.Although the global cohort RR for female was higher than that for men in 1990, the situation crossed in 1945 and then showed a trend of higher cohort RR for male than for female.The cohort RR in high-income regions has shown a trend of first declining, then rising, and then slightly declining.In general, it declined from 6.233% [95%CI: 5.731, 6.779] in 1895 to a bottom of 0.876% [95%CI: 0.763, 1.005] in 1960, and then rose to 0.999% [95%CI: 0.427, 2.338].The change pattern of cohort RR in upper-middle income regions is similar to that of the global cohort RR, which intersects in 1945.Before 1945, the cohort RR of female was higher than that of male, while after 1945, the opposite was true, and the cohort RR of male was higher.In lower-middle income regions, the cohort RR changed relatively slightly, slowly declining from 1895 to 1944 and gradually rising after 1945.The cohort RR in low-income areas maintained a relatively stable trend from 1895 to 1944, and showed an oscillating rise after 1945, but the rise was still small.

Joinpoint
Figure 4 (Joinpoint regression model) shows ASMR trends from 1990 to 2019 for IS (Ischemic stroke) attributable to LPA by sex and region.It is evident that, with the exception of low-income regions, death rates are falling globally and in each of the other three World Bank income levels regions for both genders.
In particular, the global IS ASMR attributed to LPA during the last 30 years was divided into 5 segments, 3 of which had a considerable reduction, and the most notable change took place between 2002 and 2007 (APC = − 2.628%).Over the full 1990-2019 period, the overall AAPC was − 1.34%.
The ASMR of IS attributable to LPA in high income regions is divided into four parts across the four World Bank income levels regions, with the most notable decline occurring between 1999 and 2007 (APC = − 4.726%).The region's total AAPC is -2.957%.Similar to this, there is a notable decrease in the ASMR of IS related to LPA between 2001 and 2019 (APC = − 1.797%) in the upper middle income regions, which is divided into two categories.The region's total AAPC is − 1.065%.The ASMR of IS attributable to LPA in lower middle income areas was divided into 6 segments, among which the trend of significant increase was from 1990 to 1993 (APC = 1.436), and the trend of most significant decrease was from 2008 to 2011 (APC = − 3.418%).The overall AAPC in the region is − 0.422%.Ultimately, the ASMR of IS related to LPA in low-income areas was divided into six segments, with the largest increase occurring from 1999 to 2004 (APC = 0.812%) and a notable decline from 1995 to 1999 (APC = − 0.356%).The region's overall AAPC is 0.053% (Fig. 4a, Supplementary Tables 2, 3).
Furthermore, there exist notable disparities in ASMR patterns between genders on a global scale as well as within the four World Bank income level regions.There were four distinct segments to the global ASMR trend of female IS, with the most notable drop occurring between 2001 and 2007 (APC = − 2.785%).The region's total AAPC of ASMR for females is − 1.487%.There were five segments to the ASMR trend of male IS globally, with the most notable reduction occurring between 2003 and 2007 (APC = − 2.051%).For males in the region, the overall AAPC of ASMR is − 1.094%.In high income areas, the ASMR of female IS was divided into 6 segments, the reduction in this segment was greatest between 2003 and 2006 (APC = − 5.267%).The region's total AAPC of ASMR for females is − 2.906%.In contrast to female ASMR, male IS ASMR was divided into 4 segments, with the largest reduction occurring between 1999 and 2007 (APC = − 4.901%).Nonetheless, it displayed an increasing tendency (APC = 1.495%) from 2015 to 2019.For males in the area, the total AAPC of ASMR is − 2.960%.The ASMR of female IS in upper middle income regions was divided into three segments: 1990-2001 showed a significant increase trend (APC = 0.425%), and 2001-2016 showed a significant decrease trend (APC = − 2.355%).In this area, the total AAPC for ASMR in females was − 1.111%.Male IS's ASMR was broken down into six parts, and between 1990 and 1994, there was a noticeable rising tendency (APC = 0.81%).The biggest drop occurred between 2004 and 2007 (APC = − 2.622%).Among this area, the total AAPC for ASMR among males was − 0.695%.The ASMR of female IS in lower middle income areas was separated into four segments: the most significant declining trend occurred between 2008 and 2011 (APC = − 3.085%), while the increasing trend occurred between 1990 and 2002 (APC = 0.678%).In this area, the total AAPC for ASMR in females was − 0.462%.The male IS's ASMR was broken down into four segments, each of which displayed a noteworthy increase trend between 1990 and 1993 (APC = 1.36%) and a substantial decreasing trend between

Discussion
This study is the first to examine trends in IS-related mortality among individuals with LPA globally and in four distinct World Bank income levels regions over the previous three decades using an age-period-cohort model and Joinpoint regression analysis based on the GBD2019 database.Over the past three decades, there has been an overall annual increase in the number of IS deaths and crude mortality rates in the global LPA population; also, the IS death rate for females has consistently been higher than that of males.(Supplementary Figs. 1, 2) According to related studies, female tend to be less physically active than male, with an average inactivity rate of 31.7% for female compared to 23.4% for male 10 .The substantial sex difference between ASMR and crude mortality can be explained by this set of data.Figure 1 illustrates that the ASMR of IS caused by LPA is declining annually, and ASMR of females is larger than that of males, which is consistent with the gender disparity previously discussed (Fig. 1).However, The World Health Organization's Global Status Report on Physical Activity 2022 estimates that 1.4 billion adults globally-or roughly 27.5% of the adult population-do not engage in enough physical activity, a statistic that has essentially stayed constant over time 15,16 .In the context of LPA data remaining largely unchanged and global aging, the mortality rate of IS is still declining.This is most likely due to population expansion, the dramatic improvements in health care brought about by economic development, and the fact that the decline in mortality in high-income regions far outweighs the continued deterioration in low-and middle-income countries 45 .Additionally, four distinct World Bank income level regions have high prevalence of this trait.
The ASMR of IS in the global LPA population generally showed a positive change, and the net drift value was less than 0, according to the results of the APC analysis's net drift and local drift.In contrast, the decline in ASMR was more significantly in women, with a downward trend in women of all ages.Globally, statistics reveal that females tend to be less physically active than males, with an average inactivity rate of 31.7% in females vs. 23.4% in males 10 .The underlying genetic and molecular mechanisms responsible for the positive effects of exercise training in both sexes remain to be fully elucidated 46.Nevertheless, this suggests a potential sex disparity of IS outcomes attributable to physical activity, warranting further investigation.We discovered that ASMR declined over time in the global LPA populations, which is consistent with the preceding findings, based on the age-specific mortality rates by period.It is important to note that this tendency is especially substantial in the older age group of 80 years and older (Fig. 2a).This may be due to targeted healthcare and preventive measures implemented in some regions to improve the quality of life of older IS groups.Because to a combination of factors including rising lifespan, falling birth rates, and a huge number of individuals reaching old age, the globe is currently witnessing unprecedented "population aging, " the most pervasive and dominant global demographic trend 10 .Given that IS is a prevalent illness among the elderly, it is crucial to take the age distribution of IS-related mortality into account when assessing IS-related death 28 .Accordingly, all regions should actively respond to the global aging trend and actively carry out appropriate physical exercise programs for chronic patients such as IS, especially middle-aged and elderly patients, so as to improve the quality of life of the global middle-aged and elderly IS population and effectively reduce the global burden of IS disease 28 .
We also looked at the ASMR of IS attributed to LPA across 4 different World Bank income levels regions and explored regional differences in IS ASMR across different income levels.Because the degree of economic development and income level of a region largely determine the medical level, medical quality, perfection of the medical security system and the accessibility of medical services in the region 8 .According to our research, high-income areas have had the lowest ASMR in 2019, at 1.182%, compared to the global (2.08%) and other regions with three different income levels: upper-medium (2.805%), lower-medium (2.31%) and low (1.644%) (Table 1).The highest ASMR, however, is seen in upper-middle income areas-second only to high-income areas-rather than low-income ones.Then followed by lower middle income areas.The huge population bases in these two regions, along with the variations in their historical and cultural backgrounds and customs, may be the cause of the higher ASMR in upper-middle income and lower-middle income regions.These nations have higher rates of LPA population, which leads to a comparatively large burden and death rate from IS.The net drift results explain the difference between the upper-middle income region and the lower-middle income region, where ASMR improved over the 30-year period, showing a significant downward trend, while the lower-middle income region saw a slight increase.In low-income areas, ASMR of IS attributed to LPA has been consistently low and has only increased slightly over the 30-year period.However, according to relevant studies, ASMR of allcause IS has always shown a high trend in low-income areas 47 .This could be because the majority of low-income countries in the region are found in Africa, which makes up just 9% of the world's population 48 .Because of the limited population base and the underdeveloped economy, the majority of people are forced to work manual labor in order to make ends meet.As a result, there are very few LPA.However, the degree of population aging in low-income areas is relatively low because women's education levels are generally lower there, especially in some African countries, and because local traditional concepts have a strong influence.Women also tend to have higher birth rates after marriage.The fact that low and middle income nations have few health resources and make insufficient investments in healthcare is another crucial factor.The ASMR of IS caused by LPA is lower in low-income communities, which might be partially explained by the aforementioned factors.There is an increasing trend because the net drift outcomes for low-income areas are not encouraging.Increasing the amount of basic healthcare and prevention should be the first line of defense.The second should be the creation and long-term implementation of efficient programs that promote physical activity in low-income communities.
Longitudinal analysis using an APC model provides insight into the effects of age, period, and cohort on IS-related mortality due to LPA, helping to identify causal factors and inform targeted prevention and treatment strategies 28 .The results of the age effects revealed a noteworthy trend: the ASMR of IS rose with age.This tendency is especially notable in the LPA population after the age of 60, increasing gradually from 60 to 79 years of age, and increasing rapidly from 80 years of age.This situation is largely attributed to the influence of aging population, especially in the middle and low income areas, the increase rate of ASMR in the age group over 80 is quite large, reflecting the relatively high degree of population aging in these two areas.Therefore, it is critical to develop effective prevention strategies for elderly patients with IS.The results of the period effects reflect the significant improvement in ASMR in high-income countries over the 30-year period, and combined with the Joinpoint regression model, there is a more favorable improvement in IS-related deaths in high-income regions than in other regions including the global and the 4 regions studied (AAPC: − 2.957%) (Fig. 4a).These findings imply that high income areas have made great efforts to lower the death rate of IS in the LPA population.These efforts have included raising health spending and investment, enhancing the standard of diagnosis and treatment, developing sensible physical activity plans, promoting healthy lifestyles, and implementing other policy preventive and rehabilitation measures.Over the course of the 30-year period, ASMR in lower-middle and low-income areas did not alter considerably (AAPC: − 0.422%, − 0.053%).These two places had the lowest ASMRs in 1990, but throughout the course of the following 30 years, the persistent efforts of high-income and upper-middle income areas further decreased the ASMR of IS in the LPA population in these two areas, making it even lower than that of lower-middle and low-income areas.It does, in part, reflect the pressing global public health issue that low-income and lower-middle areas ought to be actively addressing.Cohort effects show a clear trend that, globally, those born earlier with LPA have higher ASMR.This is especially true in high-income countries, where the higher mortality rate of IS among people with LPA may be due to the limitations of early medical technology, low public awareness, and unhealthy lifestyles.In lower-middle and low-income areas, on the other hand, the opposite is true, with those who were born into LPAs more recently having greater ASMR than those who were born earlier.This could be because low and lower-middle income communities have been experiencing economic growth recently, along with a slow but steady urbanization process that inevitably results in sedentary lifestyles.Another possible reason is that the burden of stroke due air pollution and environmental risks, tobacco smoke, dietary risks, and hypertension is significantly higher in low and lower-middle income regions, where the impact of LPA is smaller, whereas in high-income regions the impact of LPA is much greater 49 .

Limitations
This study has certain shortcomings.First, mortality estimates, APC analyses, and joinpoint regression analyses are significantly unclear when limited or poor quality raw data is used for low and lower-middle income regions or small-population countries 28 .Second, it can be challenging to identify whether a single cause of death contributed to an elderly person's death because they frequently have several co-existing diseases 19 .In these situations, IS is regarded as a co-factor and the cause of death is frequently reported as a direct cause of death distinct from IS.This lowers the death rate linked to IS to some extent 50 .Third, the examination of the IS disease burden attributable to LPA was carried out globally and comprehensively across four distinct income levels areas; it did not delve deeper into the variations among individual countries within each region.

Conclusion
This study highlights the significance of low physical activity as a modifiable risk factor for IS by illustrating changes in IS-related mortality attributable to LPA over the past three decades for the globe and four World Bank income levels regions.The findings indicate that the burden of IS-related deaths due to LPA is declining globally, with high-income areas seeing the greatest reduction.In some regions, such as low and lower-middle income regions, there is still a trend toward deteriorating.This requires strengthening international cooperation to formulate effective public health interventions and policies, jointly achieve public health goals, and promote the development of public health 8,51 .However, since the population ages faster, more research is needed to comprehend the factors that affect age groups, gender differences, and regional and national variances 8 .To enhance cardiovascular health on a national, regional, and international level, we ought to encourage the creation of more focused public health initiatives.

Figure 2 .
Figure 2. Age-specific (a) and Cohort-specific (b) mortality rates of Ischemic stroke attributable to Low Physical Activity in Global and 4 World Bank Income Levels regions,1990-2019.

Figure 3 .
Figure 3.The net drifts and local drifts (a), age effects (b), period effects (c), and cohort effects (d) of Ischemic stroke attributable to Low Physical Activity in the Global and 4 World Bank Income Levels regions, 1990-2019.

Figure 4 .
Figure 4. APC and AAPC in in sex-specific mortality of Ischemic stroke attributable to Low Physical Activity across Global and 4 World Bank Income Levels regions, 1990-2019.
income areas.Similar to this, the global and the other three areas with the World Bank Income levels exhibited a declining trend in terms of ASMR, with the exception of low income regions (0.128% [95% UI: 0.071, 0.185]), and high income regions saw the biggest reduction in ASMR (− 3.470% [95% UI: − 3.646, − 3.294]).