# # all deaths and DALYs in all places
# https://vizhub.healthdata.org/gbd-results?params=gbd-api-2019-permalink/93529d6d103e026b28bab69a3f384007
#  
# https://vizhub.healthdata.org/gbd-results?params=gbd-api-2019-permalink/378e1f095d8971467245e3ea4a05b187
# 
# 
# # all deaths and DALYS related to main causes regardless of risks
# https://vizhub.healthdata.org/gbd-results?params=gbd-api-2019-permalink/2aa3b76b29dee030add6014b77929925
# https://vizhub.healthdata.org/gbd-results/?params=gbd-api-2019-public/0f324e915e915a99630021fc3dd06d60
# 
# # 
# files <- list.files(path="./IHME_data/", pattern = "\\.csv$")
# 
# 
# # create an empty data frame to hold the combined data
# combined_data <- data.frame()
# 
# # loop through the CSV files and read them into separate data frames
# for (i in 1:length(files)) {
#   filename <- paste0("./IHME_data/", files[i])
#   data <- read.csv(filename, header = TRUE)
#   # add the data to the combined data frame using rbind
#   combined_data <- dplyr::bind_rows (combined_data, data)
# }
# 
# # write the combined data to a new CSV file
# distinct(combined_data)->combined_data
# write.csv(combined_data, "combined_data.csv", row.names = FALSE)
# 
# rm(list=ls())

read.csv("combined_data.csv") %>% 
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) ->ds



# lapply(ds[, 1:8], unique)

# https://ourworldindata.org/smoking#daily-smokers
# read.csv("daily-smoking-prevalence-bounds.csv") %>% clean_names() %>% rename(smoking="daily_smoking_prevalence_both_ihme_gh_dx_2012") ->smokingPrev
# smokingPrev %>% filter(year=="2012") %>% mutate(entity=recode(entity, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria",  "United Arab Emirates"="UAE"))->sm

# smoking per gender
read.csv("comparing-the-share-of-men-and-women-who-are-smoking.csv") %>% clean_names()->smByGender


continents<- smByGender %>%
  filter(continent %in% c("Africa", "Antarctica", "Asia", "Europe", "North America", "Oceania" ,"South America")) %>% 
  distinct(entity, .keep_all = T) %>% 
  select(country=entity, continent)

# https://ghdx.healthdata.org/record/ihme-data/gbd-2019-socio-demographic-index-sdi-1950-2019
read.csv("IHME_GBD_2019_SDI_1990_2019_Y2020M10D15.csv")%>% clean_names() %>% select(sdi="x2019", "location") %>% 
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE"))   ->sdi

read_excel("IHME_GBD_2019_SDI_1950_2019_QUINTILES_Y2021M03D21.XLSX")->sdi_quantiles

sdi$sdiClass<- cut(sdi$sdi, breaks=c(sdi_quantiles$lower_bound, 1), labels=sdi_quantiles$sdi_quintile)

countries<- unique(ds$location)

# https://ghdx.healthdata.org/record/ihme-data/gbd-2019-smoking-tobacco-use-prevalence-1990-2019

read.csv("IHME_GBD_2019_SMOKING_TOB_1990_2019_PREV_Y2021M05D27.CSV")%>% 
  filter(year_id=="2019") %>% 
  filter(sex_name=="Both") %>% 
  mutate(entity=location_name, smoking=round(val*100,1), lower=round(lower*100,1), upper=round(upper*100,1)) %>% 
  mutate(Smoking= paste0(smoking, " \n(", lower, "-", upper, ")")) %>% 
  select(entity, smoking, Smoking ) %>% 
  mutate(entity=recode(entity, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria",  "United Arab Emirates"="UAE")) %>% 
  distinct(entity, .keep_all = T)->sm


source("/cloud/project/scripts.R")

smoking prevalence

A<- sm %>% 
  left_join(sdi, by=c("entity"="location")) %>% 
  mutate(col=ifelse(entity %in% countries, "black", "grey")) %>%  {.->>a} %>% 
  ggplot(aes(x=sdi, y=smoking, label=entity))+
  stat_smooth(method = "lm", formula='y ~ x', se = F)+
  geom_point(aes(color=col))+
  ggrepel::geom_text_repel(data=. %>% filter(entity %in% countries ))+
  scale_color_grey()+
  theme_classic()+
  labs(y="Smoking prevalence (%)", x="SDI")+
  theme(legend.position = "none")+
  scale_x_continuous(breaks=round(sdi_quantiles$lower_bound,2))

## Warning in left_join(., sdi, by = c(entity = "location")): Each row in `x` is expected to match at most 1 row in `y`.
## ℹ Row 33 of `x` matches multiple rows.
## ℹ If multiple matches are expected, set `multiple = "all"` to silence this
##   warning.

cor.test(a$smoking, a$sdi, method="spearman")->a

## Warning in cor.test.default(a$smoking, a$sdi, method = "spearman"): Cannot
## compute exact p-value with ties

report::report(a)

## Effect sizes were labelled following Funder's (2019) recommendations.
## 
## The Spearman's rank correlation rho between a$smoking and a$sdi is positive,
## statistically significant, and very large (rho = 0.45, S = 7.60e+05, p < .001)

smByGender %>%
  filter(!is.na(prevalence_of_current_tobacco_use_males_of_male_adults), !is.na(prevalence_of_current_tobacco_use_females_of_female_adults)) %>% 
  filter(year==2020) %>% 
  left_join(continents, by=c("entity"="country")) %>% 
  mutate(color=ifelse(entity %in% countries, "black", "gray")) %>% 
  mutate(Alpha=ifelse(entity %in% countries, 1, 0.9)) ->a



B<- a %>%   ggplot(aes(label=entity, color=color, x=prevalence_of_current_tobacco_use_males_of_male_adults, y=prevalence_of_current_tobacco_use_females_of_female_adults))+
  geom_point()+
  ggrepel::geom_text_repel(data=a %>% filter(color=="black"))+
  geom_abline(intercept = 0, slope = 1, size = 0.5) +
  xlim(0,70)+
  ylim(0,70)+
  theme_classic()+
  scale_color_grey()+
  theme(legend.position = "none")+
  labs(x="prevalence of smoking in males", y="prevalence of smoking in females")

## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

C<- read.csv("IHME_GBD_2019_SMOKING_TOB_1990_2019_PREV_Y2021M05D27.CSV")%>% 
  filter(sex_name=="Both") %>% 
  mutate(entity=location_name, smoking=val*100, lower=lower*100, upper=upper*100) %>% 
  select(entity, year=year_id, smoking ) %>% 
  mutate(entity=recode(entity, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria",  "United Arab Emirates"="UAE")) %>% 
  filter(entity %in% ds$location, entity %nin% c("Global", "United States of America","North Africa and Middle East")) %>% 
  group_by(year) %>% 
  ggplot(aes(x=year, y=smoking, group=entity))+
  geom_line(color="grey", na.rm=T)+
  geom_line(data=. %>% filter(entity %in% c("Jordan", "Lebanon", "Turkey", "Afghanistan",  "Syria", "UAE", "Qatar")), aes(color=entity))+
  geom_text(data = . %>% filter(entity %in% c("Jordan", "Lebanon", "Turkey", "Afghanistan",  "Syria", "UAE", "Qatar"), year=="2019"),
            aes(label = entity,
                x = 2020,
                y = smoking,
                color = entity), hjust=0, size=4) +
  guides(color = FALSE) + 
  scale_x_continuous(expand = expansion(mult = c(0.1,0.3)))+
  theme_clean()+
  theme(legend.position = "none", axis.title =  element_text(size=12),  axis.text =element_text(size=12))+
  labs(x="Year", y="smoking prevalence (%)")

## Warning: The `<scale>` argument of `guides()` cannot be `FALSE`. Use "none" instead as
## of ggplot2 3.3.4.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

Fig<- ggarrange(A,B, C, labels=c("A", "B", "C"), ncol=1)

## Warning: Removed 31 rows containing non-finite values (`stat_smooth()`).

## Warning: The following aesthetics were dropped during statistical transformation: label
## ℹ This can happen when ggplot fails to infer the correct grouping structure in
##   the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
##   variable into a factor?

## Warning: Removed 31 rows containing missing values (`geom_point()`).

## Warning: Removed 1 rows containing missing values (`geom_text_repel()`).

## Warning: Removed 1 rows containing missing values (`geom_point()`).

## Warning: ggrepel: 6 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps

## Warning: ggrepel: 1 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps

ggsave("Fig1.pdf", Fig, height=12)

## Saving 8 x 12 in image

## Warning: ggrepel: 2 unlabeled data points (too many overlaps). Consider
## increasing max.overlaps

read.csv("IHME_GBD_2019_SMOKING_TOB_1990_2019_PREV_Y2021M05D27.CSV")%>% 
  filter(year_id=="2019") %>% 
  filter(sex_name=="Male") %>% 
  filter(location_name=='North Africa and Middle East') %>% 
  mutate(entity=location_name, smoking=round(val*100,1), lower=round(lower*100,1), upper=round(upper*100,1)) %>% 
  mutate(smoking= paste0(smoking, " (", lower, "-", upper, ")")) %>% 
  select(entity, sex_name,  smoking ) %>% 
  distinct(entity, .keep_all = T) %>% print

##                         entity sex_name          smoking
## 1 North Africa and Middle East     Male 32.4 (31.9-32.9)

read.csv("IHME_GBD_2019_SMOKING_TOB_1990_2019_PREV_Y2021M05D27.CSV")%>% 
  filter(year_id=="2019") %>% 
  filter(sex_name=="Female") %>% 
  filter(location_name=='North Africa and Middle East') %>% 
  mutate(entity=location_name, smoking=round(val*100,1), lower=round(lower*100,1), upper=round(upper*100,1)) %>% 
  mutate(smoking= paste0(smoking, " (", lower, "-", upper, ")")) %>% 
  select(entity, sex_name,  smoking ) %>% 
  distinct(entity, .keep_all = T) %>% print

##                         entity sex_name     smoking
## 1 North Africa and Middle East   Female 5.6 (5.3-6)

# https://www.r-bloggers.com/colorspace-new-tools-for-colors-and-palettes/
some_col_func <- function(n) rev(colorspace::sequential_hcl(n, "Pubu"))
# Retrievethe map data
arab_maps <- map_data("world", region = countries)

full_join(arab_maps, sm  , by = c("region" = "entity")) -> arab_maps


# Compute the centroid as the mean longitude and lattitude
# Used as label coordinate for country's names
region.lab.data <- arab_maps %>%
  group_by(region) %>%
  summarise(long = mean(long), lat = mean(lat))






Theme<- theme_bw() +
  theme(
    legend.position = "right",
    axis.line = element_blank(),
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    panel.border = element_blank(),
    panel.background = element_blank(),
    axis.title = element_blank(),
    axis.ticks = element_blank(),
    axis.text.x = element_blank(),
    axis.text.y = element_blank()
  ) 






 A<-ggplot(arab_maps, aes(x = long, y = lat)) +
  geom_polygon(aes(group = group, alpha = smoking), fill="blue", color = "dark gray") + Theme+labs(alpha="Smoking\nprevalence\n(%)")







# https://www.r-bloggers.com/colorspace-new-tools-for-colors-and-palettes/
some_col_func <- function(n) rev(colorspace::sequential_hcl(n, "Pubu"))
# Retrievethe map data
arab_maps <- map_data("world", region = countries)
DS<- ds %>% filter(measure=="Deaths", cause=="All causes", sex=="Both", age=="Age-standardized", rei=="Smoking", metric=="Rate", year=="2019", location!="North Africa and Middle East")

full_join(arab_maps, DS , by = c("region" = "location")) -> arab_maps


# Compute the centroid as the mean longitude and lattitude
# Used as label coordinate for country's names
region.lab.data <- arab_maps %>%
  group_by(region) %>%
  summarise(long = mean(long), lat = mean(lat))






Theme<- theme_bw() +
  theme(
    legend.position = "right",
    axis.line = element_blank(),
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    panel.border = element_blank(),
    panel.background = element_blank(),
    axis.title = element_blank(),
    axis.ticks = element_blank(),
    axis.text.x = element_blank(),
    axis.text.y = element_blank()
  ) 





B<-ggplot(arab_maps, aes(x = long, y = lat)) +
  geom_polygon(aes(group = group, alpha = val), fill="red", color = "dark gray") + Theme+labs(alpha="Age-standarized\nDeaths\nper year")








# https://www.r-bloggers.com/colorspace-new-tools-for-colors-and-palettes/
some_col_func <- function(n) rev(colorspace::sequential_hcl(n, "Pubu"))
# Retrievethe map data
arab_maps <- map_data("world", region = countries)
DS<- ds %>% filter(measure!="Deaths", cause=="All causes", sex=="Both", age=="Age-standardized", rei=="Smoking", metric=="Rate", year=="2019", location!="North Africa and Middle East")

full_join(arab_maps, DS , by = c("region" = "location")) -> arab_maps


# Compute the centroid as the mean longitude and lattitude
# Used as label coordinate for country's names
region.lab.data <- arab_maps %>%
  group_by(region) %>%
  summarise(long = mean(long), lat = mean(lat))






Theme<- theme_bw() +
  theme(
    legend.position = "right",
    axis.line = element_blank(),
    panel.grid.major = element_blank(),
    panel.grid.minor = element_blank(),
    panel.border = element_blank(),
    panel.background = element_blank(),
    axis.title = element_blank(),
    axis.ticks = element_blank(),
    axis.text.x = element_blank(),
    axis.text.y = element_blank()
  ) 





C<- ggplot(arab_maps, aes(x = long, y = lat)) +
  geom_polygon(aes(group = group, alpha = val), fill="violet", color = "dark gray") + Theme+labs(alpha="Age-standarized\nDALYS")





Fig<-ggarrange(A,B,C, labels=c("A","B","C"))

ggsave("Fig2.pdf", Fig, height=8)

## Saving 7 x 8 in image

Smoking-related burden of diseases in the Middle East and North Africa region

Introduction

The Global Burden of Disease (GBD) study is a comprehensive assessment of the health impact of different risk factors, diseases, and injuries in populations across the world. The study provides a global and regional picture of the health burden and helps policymakers and public health experts to prioritize interventions and allocate resources.

In the Eastern Mediterranean Region (EMRO), smoking is a significant risk factor for various health outcomes, including lung cancer, chronic obstructive pulmonary disease, ischemic heart disease, stroke, and respiratory infections. The GBD study has shown that smoking is responsible for a substantial burden of disease in the EMRO region, and the associated health costs are substantial.

According to the GBD study, smoking is the second leading risk factor for disability-adjusted life years (DALYs) in the EMRO region, accounting for 10.7% of the total burden of disease. Furthermore, smoking is responsible for 21.4% of deaths and 10.1% of years of life lost (YLL) due to premature mortality. The study also highlights that the burden of smoking in the EMRO region is higher among men than women.

The GBD study has also shown that the health impact of smoking is not limited to the smoker but extends to the people around them. Secondhand smoke exposure is responsible for a significant proportion of the burden of disease due to smoking in the EMRO region, particularly in children.

In conclusion, the GBD study has provided valuable insights into the burden of smoking in the EMRO region, highlighting the need for effective tobacco control policies and interventions. Addressing smoking in the EMRO region can significantly reduce the burden of disease, improve health outcomes, and reduce healthcare costs.

Methods

The GBD study collects data from national records and surveys, with death certificates serving as a crucial source of information. When such data is unavailable, GBD resorts to modeling, utilizing regions with similar geographic and sociodemographic characteristics to approximate values. While the process is not without flaws, it represents the most effective method for studying disease burden.

The SDI is based on three indicators: income per capita, average years of education, and total fertility rate. Income is measured using the gross domestic product (GDP) per capita adjusted for purchasing power parity (PPP). Education is measured by the average number of years of schooling completed by individuals aged 15 years and older. Fertility is measured by the number of live births per woman. The SDI ranges from 0 to 1, with higher values indicating higher levels of socioeconomic development. The SDI is used by the GBD study to assess the impact of diseases, injuries, and risk factors on populations with different levels of socioeconomic development. It is also used to compare the burden of disease between different regions and countries.

In our study, we utilized the GBD results tool to investigate the risk of smoking and secondhand smoke, as well as all related causes of death and disability, in the 22 countries of North Africa and the Middle East (NAME). GBD predominantly employs the Disability-Adjusted Life Years (DALYs) to measure disease burden, a composite metric combining years of life lost (YLL) and years lost to disability (YLD).

To enable comparison between different regions, we included standardized-age, which provided us with the standardized incidence for death and DALYs, correcting for population differences relative to the global population pyramid. Descriptive statistics were utilized to represent our data, with all calculations made using R v4.2.3. Figures were produced using the ggplot package, while Spearman correlation analysis was used to assess the correlation between various variables, such as smoking prevalence and SDI, smoking prevalence and DALYs, and between different variables.

Results

the median prevalence of smoking in the Middle East and North Africa (MENA) region is 16.5 (IQR, 13.5-20.8), which is comparable to the global median (16.5; IQR, 10.5-23.1). However, five MENA countries exhibited prevalence rates above the upper quartile, namely Lebanon (27.0%), Turkey (26.5%), Jordan (25.8%), Tunisia (24.5%), and Kuwait (24.4%). Our data revealed a strong correlation between smoking and SDI (rho = 0.52, S = 4.29e+05, p < .001) (Fig 1A). Moreover, we observed higher smoking prevalence in males (median 29.1; IQR, 18.7-40.7) than females (median, 6.5; IQR, 2.6-17.3) in all MENA countries (Fig 1B).

Although the impact of smoking on health outcomes, such as DALYs and deaths, is widely recognized, our analysis did not reveal a statistically significant relationship between smoking prevalence and DALYs caused by smoking (rho = 0.41, S = 910.00, p = 0.067) or deaths (rho = 0.36, S = 984.00, p = 0.108), despite the large effect size. This finding may reflect the complex interplay of smoking with other factors, such as healthcare and income, that contribute to the outcome of smokers. Interestingly, countries with low and middle SDIs, including Yemen, Iraq, Syria, Egypt, and Lebanon, exhibited the highest relative age-standardized deaths and DALYs caused by smoking.

According to estimates, tobacco abuse was responsible for approximately 8.7 million deaths worldwide in 2019, comprising 15.4% of all deaths and 24.9% of deaths attributable to known risk factors. In North Africa and the Middle East (MENA), tobacco abuse was estimated to cause 450,358 deaths (CI, 401,337-504,061), accounting for 14.5% of all deaths and 23.2% of deaths linked to known risk factors. Globally, the disease burden measured by DALYs (disability-adjusted life years) was estimated to be 229 million, representing 9% of all DALYs and 18.8% of all DALYs with known risks. In NAME, the burden was estimated to be 14,013,882 (CI, 12,394,849-15,726,637), representing 8.5% of all DALYs and 19% of all DALYs with known risks.

The majority of deaths and DALYs in MENA were associated with smoking, with an estimated 374,199 deaths (CI, 335,684-417,432) and 11,475,764 DALYs (CI, 10,196,258-12,925,331) attributable to smoking, and 96,458 deaths (CI, 76,076-117,651) and 3,144,992 DALYs (CI, 2,422,510-3,855,567) attributable to secondhand smoking. Cardiovascular diseases were the leading cause of death and DALYs, accounting for 53.4% of all deaths and 50.3% of all DALYs, followed by neoplasms (24.6% of all deaths and 20.3% of all DALYs), chronic respiratory diseases (12.4% of all deaths and 11.9% of all DALYs), and respiratory infections and tuberculosis (4% of all deaths and 3.4% of all DALYs). Although musculoskeletal disorders (6.8% of all DALYs) and diabetes and kidney diseases (4.3%) were significant contributors to DALYs, they did not contribute to a significant proportion of deaths. All other causes, including neurological disorders, digestive diseases, sense organ diseases, transport injuries, unintentional injuries, self-harm, and interpersonal violence, accounted for 3.3% of deaths and 2.9% of DALYs (Figure 2).

The three most frequent causes of death related to smoking in the MENA region were cardiovascular diseases (53.4%), neoplasms (24.6%), and chronic obstructive pulmonary disease (12.4%). Similarly, the three most common causes associated with DALYs were CVD (50.3%), neoplasms (20.4%), and CRD (11.9%).

The distribution of these causes was found to vary across different age groups. Although CVD remained the leading cause of death and disability, musculoskeletal disorders contributed significantly to the burden of DALYs among younger age groups, while neoplasms played a more prominent role in both DALYs and deaths among older age groups. The total burden of smoking-related diseases increased with age, peaking in the 64-69 age group.

When compared to the rest of the world, the burden of smoking-related diseases in the MENA region was found to be comparable to that in middle SDI. However, there was one significant exception, namely CVD, where the age-standardized annual rate of DALYs and deaths associated with smoking were higher in the MENA region than in all other regions, measuring 1198 (CI, 1047.8-1379.1) and 46.3 (CI, 40.6-52.8) per 100,000, respectively. On the other hand, the burden of CRD in the MENA region was similar to that in high SDI, which was lower than the global average.

Globally, the annual death rate and DALYs attributable to smoking-related causes are declining. However, increasing rates were observed in middle SDI countries. Among the MENA region, increasing death rates and DALYs were noticed in Lebanon, Turkey, Tunisia, Syria, Libya, and the United Arab Emirates (UAE), while declining rates were most noticeable in Oman and Qatar.

In 2019, secondhand smoking was responsible for 96,458 (76076-117651) deaths, which accounted for 20.5% of all tobacco-related deaths. Additionally, 3,144,992 DALYs (2422510-3855567) were attributed to secondhand smoking, representing 21.5% of all DALYs related to tobacco abuse.

Young individuals (<35 years) were particularly affected by secondhand smoking, with high rates of DALYs and deaths observed in those under 20 years of age. Diabetes and kidney diseases, as well as respiratory infections, made a significant contribution to the burden of secondhand smoking. Notably, almost all DALYs and deaths below the age of 25 were attributed to respiratory infections and tuberculosis.

Discussion

The Global Burden of Disease Study 2019 analyzed the prevalence of smoking tobacco use and its attributable disease burden in 204 countries and territories from 1990 to 2019. The study estimated that globally, 1.14 billion individuals were current smokers in 2019, with smoking tobacco use accounting for 7.69 million deaths and 200 million disability-adjusted life-years. The study showed a significant reduction in smoking prevalence since 1990 but a significant increase in the total number of smokers due to population growth. [PMID: 34051883]

Smokers are 30 to 40 percent more likely to develop type 2 diabetes than nonsmokers. Smoking can also make managing the disease and regulating insulin levels more difficult because high levels of nicotine can lessen the effectiveness of insulin, causing smokers to need more insulin to regulate blood sugar levels.

Smoking has been found to have a negative impact on musculoskeletal health. Smoking can lead to decreased bone density and an increased risk of osteoporosis, as well as a higher risk of bone fractures. Smoking has also been associated with an increased risk of developing rheumatoid arthritis, a chronic autoimmune disease that affects the joints.

Furthermore, smoking has been shown to be a risk factor for low back pain and spinal disorders. Studies have found that smoking is associated with an increased risk of developing degenerative disc disease, a condition that can cause chronic low back pain.

Figure: A pie chart showing the distribution of smoking-related causes of death and DALYs in the EMRO region. The chart could show the proportion of deaths and DALYs related to neoplasms, cardiovascular disease, and other smoking-related causes.

# the percent of other causes is <10#
causes<-as.factor(c("Cardiovascular diseases", "Respiratory infections and tuberculosis", "Neoplasms" ,"Chronic respiratory diseases", "Musculoskeletal disorders", "Diabetes and kidney diseases"))
# ds %>% 
#   filter(measure=="Deaths", sex=="Both", age=="All ages", rei=="Smoking", metric=="Number", year=="2019") %>% 
#   select(-c(upper, lower)) %>% 
#   spread(cause, val) %>% 
#   mutate(others=`All causes`-`Cardiovascular diseases`- `Chronic obstructive pulmonary disease`-Neoplasms-`Respiratory infections and tuberculosis`) %>% mutate(pct=others/`All causes`)

A<- ds %>% 
  filter(measure=="Deaths", location=="North Africa and Middle East",  sex=="Both", age=="All ages", rei=="Smoking", metric=="Number", year=="2019",  cause%nin% c("All causes", "Chronic obstructive pulmonary disease")) %>% 
  mutate(cause= factor(cause, 
                       levels=c("Cardiovascular diseases", "Chronic respiratory diseases",  "Neoplasms", "Respiratory infections and tuberculosis",  "Diabetes and kidney diseases",  "Musculoskeletal disorders", "Digestive diseases", "Unintentional injuries", "Diabetes and kidney diseases", "Self-harm and interpersonal violence", "Neurological disorders", "Transport injuries", "Sense organ diseases"), 
                       labels=c("CVD", "CRD",  "Neoplasms", "Infections",  "DM",  rep("Others", 8)) )) %>% 
  mutate(cause=fct_relevel(cause,"CVD","Neoplasms", "CRD","DM","Infections","Others" )) %>% 
  group_by(cause) %>% 
  summarise(n=sum(val)) %>% 
  select(var=cause, n) %>% 
  # mutate(var=factor(var, levels=c("CVD","CRD", "Neoplasms", "Infections" ,  "DM" , "MSK", "Others"))) %>% 
  donut_chart()


B<-ds %>% filter(measure=="DALYs (Disability-Adjusted Life Years)", location=="North Africa and Middle East",  sex=="Both", age=="All ages", rei=="Smoking", metric=="Number", year=="2019",  cause%nin% c("All causes", "Chronic obstructive pulmonary disease")) %>% 
  mutate(cause= factor(cause, 
                       levels=c("Cardiovascular diseases", "Chronic respiratory diseases",  "Neoplasms", "Respiratory infections and tuberculosis",  "Diabetes and kidney diseases",  "Musculoskeletal disorders", "Digestive diseases", "Unintentional injuries", "Diabetes and kidney diseases", "Self-harm and interpersonal violence", "Neurological disorders", "Transport injuries", "Sense organ diseases"), 
                       labels=c("CVD", "CRD",  "Neoplasms", "Infections",  "DM", "MSK" , rep("Others", 7)) )) %>% 
  mutate(cause=fct_relevel(cause, "CVD","Neoplasms", "CRD","MSK","DM", "Infections","Others" )) %>% 
  group_by(cause) %>% 
  summarise(n=sum(val)) %>% 
  select(var=cause, n) %>% 
  # mutate(var=factor(var, levels=c("CVD","CRD", "Neoplasms", "Infections" ,  "DM", "MSK" , "Others"))) %>% 
  donut_chart()



Fig<-ggarrange(A,B, ncol=2, labels=c("A", "B"))

ggsave("Fig3.pdf", Fig, width=15, height=8)

Table: A comparison table showing the burden of smoking-related diseases in the EMRO region compared to other regions of the world (High SDI, middle SDI, low SDI, and global). The table could include columns for the number of deaths and DALYs related to smoking for each region.

places<- c("Low SDI", "Global", "North Africa and Middle East", "Low-middle SDI", "Middle SDI", "High SDI", "High-middle SDI")
ds %>% 
  filter(location %in% places, sex=="Both", age=="Age-standardized", rei=="Smoking", metric=="Rate", year=="2019") %>%
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(across(c(val, upper, lower), function (x) scales::label_comma(accuracy = .1)(x))) %>%
  mutate(val = paste0(val, "\n(", lower, "-", upper, ")")) %>%  
  select(-c(sex, age, rei, metric, year, upper, lower)) %>% 
  spread(location, val) ->a

a

measure	cause	Global	High SDI	High-middle SDI	Low SDI	Low-middle SDI	Middle SDI	North Africa and Middle East
DALYs (Disability-Adjusted Life Years)	All causes	2,412.7 (2,241.2-2,579.0)	2,065.4 (1,921.5-2,224.1)	2,612.6 (2,391.3-2,836.3)	1,958.0 (1,762.4-2,175.5)	2,662.2 (2,424.3-2,920.3)	2,488.1 (2,246.2-2,743.2)	2,455.0 (2,195.9-2,745.0)
DALYs (Disability-Adjusted Life Years)	Cardiovascular diseases	853.9 (788.2-918.0)	517.8 (489.5-549.6)	1,021.1 (936.2-1,114.8)	637.7 (563.7-725.4)	912.0 (821.0-1,007.8)	952.6 (857.6-1,054.8)	1,198.0 (1,047.8-1,379.1)
DALYs (Disability-Adjusted Life Years)	Chronic obstructive pulmonary disease	424.0 (380.2-465.7)	303.7 (274.1-334.3)	308.5 (276.1-350.3)	508.1 (428.4-583.8)	751.3 (625.2-853.0)	453.2 (400.4-508.2)	293.2 (254.4-331.1)
DALYs (Disability-Adjusted Life Years)	Chronic respiratory diseases	449.5 (403.9-493.6)	331.5 (297.7-369.5)	323.8 (289.1-368.7)	548.4 (467.9-623.1)	797.8 (669.6-905.5)	474.7 (419.5-531.1)	326.7 (280.3-368.5)
DALYs (Disability-Adjusted Life Years)	Diabetes and kidney diseases	78.1 (59.6-98.7)	74.3 (53.2-98.4)	72.8 (53.8-92.5)	62.1 (45.8-79.4)	84.7 (65.2-106.7)	86.1 (66.8-106.9)	103.0 (77.9-132.1)
DALYs (Disability-Adjusted Life Years)	Digestive diseases	18.0 (15.2-21.1)	12.1 (9.6-15.2)	18.5 (15.7-21.6)	17.7 (14.1-21.5)	28.3 (23.4-34.6)	16.2 (13.6-19.3)	13.0 (9.9-16.6)
DALYs (Disability-Adjusted Life Years)	Musculoskeletal disorders	123.2 (79.3-172.5)	228.3 (145.9-322.6)	148.1 (96.6-207.6)	65.7 (42.8-92.1)	76.6 (50.7-106.8)	95.1 (62.0-132.7)	141.1 (88.8-200.7)
DALYs (Disability-Adjusted Life Years)	Neoplasms	677.3 (616.4-740.3)	776.0 (731.4-817.1)	859.2 (773.1-952.3)	273.9 (235.2-311.1)	424.4 (380.9-474.7)	674.5 (583.0-775.7)	515.6 (454.9-585.4)
DALYs (Disability-Adjusted Life Years)	Neurological disorders	45.2 (14.1-110.9)	55.7 (21.0-127.1)	49.9 (16.1-120.1)	22.6 (5.6-59.8)	32.5 (7.5-87.8)	42.0 (11.1-107.7)	54.0 (17.8-133.9)
DALYs (Disability-Adjusted Life Years)	Respiratory infections and tuberculosis	144.9 (120.2-169.8)	53.8 (42.4-65.4)	97.4 (80.6-113.8)	307.1 (237.1-380.3)	274.1 (224.7-328.9)	122.7 (102.0-144.6)	82.5 (64.9-102.2)
DALYs (Disability-Adjusted Life Years)	Self-harm and interpersonal violence	0.2 (0.1-0.3)	0.1 (0.1-0.2)	0.3 (0.2-0.4)	0.2 (0.2-0.3)	0.3 (0.2-0.4)	0.2 (0.1-0.2)	0.1 (0.1-0.2)
DALYs (Disability-Adjusted Life Years)	Sense organ diseases	7.0 (4.7-10.0)	2.6 (1.7-3.8)	5.9 (4.0-8.4)	8.0 (5.3-11.6)	11.5 (7.8-16.5)	8.8 (6.0-12.6)	9.3 (6.2-13.4)
DALYs (Disability-Adjusted Life Years)	Transport injuries	5.2 (3.6-7.1)	2.6 (1.8-3.6)	5.1 (3.5-6.9)	5.1 (3.6-7.0)	7.0 (4.9-9.6)	6.1 (4.3-8.2)	6.5 (4.0-9.1)
DALYs (Disability-Adjusted Life Years)	Unintentional injuries	10.2 (7.0-14.4)	10.5 (6.9-15.4)	10.6 (7.2-15.2)	9.5 (6.6-12.9)	13.1 (9.5-17.9)	9.1 (6.3-12.8)	5.1 (3.5-7.3)
Deaths	All causes	95.6 (89.1-101.8)	73.4 (70.6-76.6)	101.4 (92.9-109.7)	79.0 (70.9-87.8)	111.4 (101.3-122.4)	104.6 (94.5-115.5)	91.6 (82.7-101.4)
Deaths	Cardiovascular diseases	33.0 (30.4-35.5)	19.3 (18.0-20.4)	40.0 (36.5-43.6)	24.0 (21.3-27.1)	34.9 (31.5-38.4)	38.1 (34.2-42.1)	46.3 (40.6-52.8)
Deaths	Chronic obstructive pulmonary disease	20.4 (18.1-22.6)	11.2 (9.8-12.6)	15.3 (13.3-17.9)	25.9 (21.1-30.2)	39.4 (32.2-45.1)	24.5 (21.3-27.8)	12.0 (10.0-13.9)
Deaths	Chronic respiratory diseases	21.1 (18.8-23.4)	11.3 (9.9-12.7)	15.5 (13.5-18.1)	27.5 (22.7-31.8)	41.3 (33.8-47.2)	25.2 (22.0-28.6)	13.1 (11.1-15.1)
Deaths	Diabetes and kidney diseases	1.5 (1.2-1.8)	0.9 (0.7-1.2)	1.1 (0.9-1.4)	1.6 (1.2-2.0)	2.1 (1.6-2.5)	1.9 (1.5-2.3)	2.0 (1.6-2.5)
Deaths	Digestive diseases	0.6 (0.5-0.7)	0.3 (0.3-0.4)	0.6 (0.5-0.7)	0.7 (0.5-0.8)	1.2 (0.9-1.4)	0.6 (0.5-0.7)	0.5 (0.4-0.7)
Deaths	Musculoskeletal disorders	0.0 (0.0-0.1)	0.0 (0.0-0.1)	0.0 (0.0-0.1)	0.0 (0.0-0.1)	0.1 (0.0-0.1)	0.0 (0.0-0.1)	0.0 (0.0-0.0)
Deaths	Neoplasms	30.6 (28.0-33.3)	35.4 (32.9-37.4)	37.5 (33.7-41.5)	11.8 (10.2-13.2)	18.4 (16.6-20.6)	30.7 (26.6-35.1)	22.4 (19.8-25.4)
Deaths	Neurological disorders	2.5 (0.3-7.7)	2.9 (0.4-8.5)	2.6 (0.2-8.0)	1.4 (0.2-4.5)	2.0 (0.1-6.5)	2.4 (0.2-7.5)	3.1 (0.4-9.4)
Deaths	Respiratory infections and tuberculosis	5.7 (4.7-6.8)	3.0 (2.3-3.7)	3.8 (3.1-4.6)	11.4 (8.7-14.2)	10.7 (8.6-13.0)	5.2 (4.2-6.3)	3.8 (2.9-4.8)
Deaths	Self-harm and interpersonal violence	0.0 (0.0-0.0)	0.0 (0.0-0.0)	0.0 (0.0-0.0)	0.0 (0.0-0.0)	0.0 (0.0-0.0)	0.0 (0.0-0.0)	0.0 (0.0-0.0)
Deaths	Transport injuries	0.1 (0.1-0.2)	0.1 (0.0-0.1)	0.1 (0.1-0.1)	0.1 (0.1-0.2)	0.1 (0.1-0.2)	0.2 (0.1-0.2)	0.2 (0.1-0.3)
Deaths	Unintentional injuries	0.3 (0.2-0.4)	0.2 (0.1-0.2)	0.2 (0.2-0.3)	0.4 (0.3-0.6)	0.6 (0.4-0.8)	0.3 (0.2-0.4)	0.1 (0.1-0.2)

write.csv(a, "table3.csv")

Figure: A bar graph showing the distribution of smoking-related diseases by age group in the EMRO region. The graph could show the number of deaths and DALYs for each age group, with separate bars for males and females.

ages<-c("30-34 years"  ,  "30-34 years"  ,  "35-39 years" ,    "40-44 years"      ,"45-49 years" ,     "50-54 years" ,     "55-59 years"  ,    "60-64 years"  ,   "65-69 years" , "70+ years"  )

excludeCauses<-c("Unintentional injuries" , "Transport injuries", "All causes","Digestive diseases","Self-harm and interpersonal violence", "Sense organ diseases"   )


A<-ds %>% 
  filter(measure=="DALYs (Disability-Adjusted Life Years)", location=="North Africa and Middle East", cause %nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Smoking", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col()+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.position="none")+
  scale_y_continuous(breaks=seq(0,1500000, 500000), labels=seq(0,1500,500))+
  labs(x="", y="DALYs in thousands per year", fill="Cause")


B<-ds %>% 
  filter(measure=="DALYs (Disability-Adjusted Life Years)", location=="North Africa and Middle East", cause%nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Smoking", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col(position="fill")+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.background = element_rect(fill = NA, color = NA))+
  scale_y_continuous()+
  labs(x="", y="Proportion", fill="Cause")


C<-ds %>% 
  filter(measure=="Deaths", location=="North Africa and Middle East", cause%nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Smoking", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col()+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.position = "none")+
  scale_y_continuous(breaks=seq(0,100000, 50000), labels=seq(0,100,50))+
  labs(x="", y="Deaths in thousands per year", fill="Cause")


D<-ds %>% 
  filter(measure=="Deaths", location=="North Africa and Middle East", cause%nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Smoking", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col(position="fill")+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.background = element_rect(fill = NA, color = NA))+
  scale_y_continuous()+
  labs(x="", y="Proportion", fill="Cause")


Fig<-ggarrange(A,B,C,D, labels=c("A", "B", "C", "D"))

ggsave("Fig4.pdf", Fig, width=12)

# combined_plot <- (A+B+C+D) +
#   plot_layout(ncol = 2, widths = c(1, 1), heights = c(2, 2))
#               
# combined_plot+
#   plot_layout(
#     heights = c(2, 2)
#   )

Table: A table showing the prevalence of smoking and the burden of smoking-related diseases in the EMRO region by country income group (low, middle, high). The table could include columns for the number of smokers, the prevalence of smoking, and the number of deaths and DALYs related to smoking for each income group.

locations<- c('Afghanistan', 'Algeria', 'Bahrain', 'Egypt',  'Iran', 'Iraq', 'Jordan', 'Kuwait', 'Lebanon', 'Libya',    'Morocco',  'Oman', 'Palestine', 'Qatar', 'Saudi Arabia', 'Sudan', 'Syria', 'Tunisia', 'Turkey', 'UAE',  'Yemen','North Africa and Middle East', 'Global', 'Low SDI','Low-middle SDI', 'Middle SDI','High-middle SDI', 'High SDI')

ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="Age-standardized", metric=="Rate", measure=="Deaths", year==2019, location!="United States of America") %>%
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(across(c(val, upper, lower), function (x) scales::label_comma(accuracy = .1)(x))) %>% 
  mutate(val = paste0 (val, "\n(", lower, "-", upper, ")")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>% 
  select(location, smoking, Smoking, `Secondhand smoke`, `All risk factors`) %>% 
  arrange(factor(location, levels = locations)) ->a



ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="All ages", metric=="Number", measure=="Deaths", year==2019, location!="United States of America") %>%
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(across(c(val, upper, lower), function (x) scales::label_comma(accuracy = .1)(x))) %>% 
  mutate(val = paste0 (val, "\n(", lower, "-", upper, ")")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>%
  arrange(factor(location, levels = locations)) %>% 
  left_join(sdi) %>% 
  distinct(location, .keep_all = T) %>% 
  select(Smoking, `Secondhand smoke`, `All risk factors`, sdi, sdiClass) ->b

## Joining with `by = join_by(location)`

## Warning in left_join(., sdi): Each row in `x` is expected to match at most 1 row in `y`.
## ℹ Row 22 of `x` matches multiple rows.
## ℹ If multiple matches are expected, set `multiple = "all"` to silence this
##   warning.

a<- dplyr::bind_cols(a,b)

## New names:
## • `Smoking` -> `Smoking...3`
## • `Secondhand smoke` -> `Secondhand smoke...4`
## • `All risk factors` -> `All risk factors...5`
## • `Smoking` -> `Smoking...6`
## • `Secondhand smoke` -> `Secondhand smoke...7`
## • `All risk factors` -> `All risk factors...8`

location	smoking	Smoking...3	Secondhand smoke...4	All risk factors...5	Smoking...6	Secondhand smoke...7	All risk factors...8	sdi	sdiClass
Afghanistan	10	81.7 (63.7-99.5)	40.1 (29.7-51.3)	825.8 (687.7-954.5)	10,305.9 (7,826.8-12,975.3)	6,061.6 (4,308.3-7,992.7)	124,347.2 (103,910.1-145,285.3)	0.343	Low SDI
Algeria	17.4	76.3 (63.1-91.5)	27.2 (21.3-34.1)	510.6 (441.4-589.1)	21,555.4 (17,664.2-26,056.6)	7,670.4 (5,902.8-9,679.3)	133,820.5 (114,270.7-156,756.9)	0.652	Middle SDI
Bahrain	16.2	82.7 (67.4-101.4)	24.4 (17.0-32.7)	492.9 (419.1-575.9)	545.5 (433.7-676.6)	148.0 (103.9-199.9)	2,756.9 (2,280.3-3,310.6)	0.751	High-middle SDI
Egypt	23.1	125.6 (98.6-160.2)	32.1 (23.3-41.8)	677.8 (546.0-821.9)	74,032.5 (57,380.1-96,218.7)	19,666.9 (14,110.3-25,878.6)	361,890.0 (284,838.2-449,609.1)	0.658	Middle SDI
Iran	14.9	57.4 (53.4-62.1)	15.8 (12.6-19.0)	378.8 (359.0-397.0)	39,926.0 (37,335.8-43,016.2)	10,606.7 (8,404.4-12,725.1)	243,052.0 (231,290.5-254,791.1)	0.67	Middle SDI
Iraq	20.9	119.0 (94.2-141.1)	31.7 (23.9-39.5)	591.2 (499.1-679.4)	25,199.2 (19,581.7-30,286.1)	6,635.0 (4,927.7-8,443.7)	117,127.8 (96,849.7-137,737.0)	0.671	Middle SDI
Jordan	34.3	82.4 (67.5-99.0)	17.5 (13.4-21.9)	397.2 (347.6-457.9)	4,741.6 (3,858.8-5,769.7)	1,014.3 (785.0-1,270.2)	20,707.6 (17,862.1-24,291.9)	0.731	High-middle SDI
Kuwait	20.3	61.7 (50.6-74.9)	15.3 (11.6-19.5)	289.5 (251.6-335.0)	1,524.2 (1,249.3-1,848.9)	357.2 (273.3-449.5)	6,380.4 (5,490.4-7,416.5)	0.851	High SDI
Lebanon	35.8	150.7 (131.3-171.7)	26.2 (20.8-31.9)	463.1 (406.0-510.9)	7,810.7 (6,797.3-8,896.7)	1,352.3 (1,076.1-1,643.1)	23,597.0 (20,683.3-26,095.9)	0.708	High-middle SDI
Libya	20.7	74.3 (60.5-91.9)	23.7 (18.1-30.5)	411.1 (341.4-495.6)	3,662.6 (2,961.3-4,606.3)	1,161.2 (885.5-1,500.5)	18,939.2 (15,632.2-23,058.7)	0.709	High-middle SDI
Morocco	11.7	71.1 (55.4-83.0)	23.2 (17.7-28.8)	594.6 (494.3-659.0)	21,382.2 (16,345.1-25,336.2)	6,615.8 (4,981.0-8,215.4)	156,478.9 (128,677.0-176,333.5)	0.548	Low-middle SDI
Oman	11.2	58.0 (50.9-66.3)	25.1 (19.7-31.1)	623.0 (574.3-676.1)	811.6 (702.2-944.7)	305.5 (239.6-376.9)	7,184.1 (6,556.1-7,900.4)	0.783	High-middle SDI
Palestine	22.5	102.5 (87.8-119.0)	29.4 (22.4-36.8)	559.0 (498.6-625.6)	2,185.4 (1,866.6-2,546.3)	615.3 (470.2-768.3)	10,841.5 (9,646.6-12,192.6)	0.588	Low-middle SDI
Qatar	17.5	75.1 (58.5-94.7)	28.3 (20.1-38.4)	603.0 (505.6-716.7)	418.1 (316.2-543.1)	117.1 (82.8-160.7)	2,339.2 (1,853.3-2,897.7)	0.83	High SDI
Saudi Arabia	14.4	64.3 (53.2-76.7)	23.5 (18.1-29.2)	473.8 (405.3-539.8)	12,270.5 (9,710.4-15,247.5)	4,037.2 (3,104.3-5,117.0)	68,910.4 (56,763.0-81,893.9)	0.805	High-middle SDI
Sudan	10.7	92.0 (73.6-115.3)	24.4 (18.5-31.2)	646.1 (567.7-753.3)	16,147.1 (12,643.7-20,547.5)	4,567.5 (3,417.4-5,914.5)	126,539.3 (110,035.8-146,365.7)	0.515	Low-middle SDI
Syria	22.8	115.5 (88.1-151.2)	27.2 (20.1-35.9)	581.3 (468.3-718.7)	13,239.9 (9,919.3-17,583.8)	3,013.4 (2,171.7-4,035.0)	55,579.3 (43,557.0-71,137.3)	0.619	Middle SDI
Tunisia	23.7	90.6 (69.1-116.9)	20.5 (14.4-27.7)	419.4 (330.5-525.4)	10,858.8 (8,220.1-14,108.6)	2,405.0 (1,678.8-3,267.4)	46,791.5 (36,512.3-59,188.6)	0.672	Middle SDI
Turkey	31	99.2 (80.2-120.8)	17.5 (12.8-22.9)	369.5 (310.5-439.0)	86,170.3 (69,436.0-105,168.1)	14,606.4 (10,701.0-19,076.0)	300,697.6 (251,239.3-358,560.8)	0.748	High-middle SDI
UAE	13.8	95.0 (75.8-116.5)	33.0 (24.8-42.9)	563.3 (480.7-663.3)	3,725.7 (2,735.9-4,913.2)	936.6 (660.9-1,275.3)	16,978.0 (13,081.0-21,728.5)	0.88	High SDI
Yemen	20.2	137.3 (112.4-174.4)	33.9 (26.1-44.0)	680.2 (588.7-823.2)	17,305.9 (13,742.3-22,470.5)	4,466.7 (3,353.7-5,877.0)	96,593.9 (81,359.9-117,091.8)	0.412	Low SDI
North Africa and Middle East	19.6	91.6 (82.7-101.4)	23.9 (19.0-29.0)	507.9 (461.8-553.4)	374,199.1 (335,684.2-417,432.9)	96,458.0 (76,076.9-117,651.5)	1,943,526.8 (1,750,242.8-2,135,282.9)	0.66	Middle SDI
Global	19.6	95.6 (89.1-101.8)	16.5 (12.8-20.4)	453.2 (426.4-478.8)	7,693,367.9 (7,158,449.6-8,200,590.6)	1,304,318.3 (1,006,960.8-1,605,391.4)	35,000,050.0 (32,937,413.3-36,938,465.5)	0.651	Middle SDI
Low SDI		79.0 (70.9-87.8)	16.7 (12.0-21.2)	684.4 (633.7-741.1)	366,990.3 (329,206.2-411,910.6)	90,096.7 (64,031.2-116,874.5)	4,399,291.7 (3,986,060.0-4,902,225.2)
Low-middle SDI		111.4 (101.3-122.4)	24.5 (18.2-31.0)	598.4 (553.5-644.3)	1,395,790.3 (1,265,134.5-1,536,588.8)	303,484.4 (227,265.5-380,913.5)	7,626,987.4 (7,057,191.5-8,266,727.8)
Middle SDI		104.6 (94.5-115.5)	22.0 (17.0-27.1)	474.0 (441.0-507.4)	2,421,045.8 (2,174,281.3-2,683,149.6)	483,105.1 (378,516.3-594,107.8)	10,280,091.3 (9,564,809.1-11,003,544.1)
High-middle SDI		101.4 (92.9-109.7)	15.9 (12.5-19.4)	391.3 (365.3-416.7)	2,067,130.6 (1,891,193.0-2,235,411.4)	315,042.0 (248,208.1-383,716.7)	7,609,594.7 (7,105,433.7-8,112,013.3)
High SDI		73.4 (70.6-76.6)	5.8 (4.5-7.2)	252.2 (239.9-264.2)	1,438,676.7 (1,379,109.5-1,507,374.5)	111,901.1 (85,856.8-139,225.9)	5,064,430.9 (4,766,547.4-5,328,350.0)

write.csv(a, "table1.csv")


ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="Age-standardized", metric=="Rate", measure=="DALYs (Disability-Adjusted Life Years)", year==2019, location!="United States of America") %>%
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(across(c(val, upper, lower), function (x) scales::label_comma(accuracy = .1)(x))) %>% 
  mutate(val = paste0 (val, "\n(", lower, "-", upper, ")")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>% 
  select(location, smoking, Smoking, `Secondhand smoke`, `All risk factors`) %>% 
  arrange(factor(location, levels = locations)) ->a



ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="All ages", metric=="Number", measure=="DALYs (Disability-Adjusted Life Years)", year==2019, location!="United States of America") %>%
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(across(c(val, upper, lower), function (x) scales::label_comma(accuracy = .1)(x))) %>% 
  mutate(val = paste0 (val, "\n(", lower, "-", upper, ")")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>%
  arrange(factor(location, levels = locations)) %>% 
  left_join(sdi) %>%
  distinct(location, .keep_all = T) %>% 
  select(Smoking, `Secondhand smoke`, `All risk factors`, sdi, sdiClass) ->b

## Joining with `by = join_by(location)`

## Warning in left_join(., sdi): Each row in `x` is expected to match at most 1 row in `y`.
## ℹ Row 22 of `x` matches multiple rows.
## ℹ If multiple matches are expected, set `multiple = "all"` to silence this
##   warning.

a<- dplyr::bind_cols(a,b)

## New names:
## • `Smoking` -> `Smoking...3`
## • `Secondhand smoke` -> `Secondhand smoke...4`
## • `All risk factors` -> `All risk factors...5`
## • `Smoking` -> `Smoking...6`
## • `Secondhand smoke` -> `Secondhand smoke...7`
## • `All risk factors` -> `All risk factors...8`

location	smoking	Smoking...3	Secondhand smoke...4	All risk factors...5	Smoking...6	Secondhand smoke...7	All risk factors...8	sdi	sdiClass
Afghanistan	10	2,257.5 (1,760.5-2,768.9)	1,192.7 (871.9-1,545.6)	24,772.7 (21,007.1-28,580.6)	356,460.6 (275,128.6-450,108.5)	288,743.1 (197,182.1-401,334.2)	6,811,261.2 (5,705,034.9-8,021,748.7)	0.343	Low SDI
Algeria	17.4	1,778.0 (1,495.1-2,111.1)	645.1 (500.9-810.1)	12,612.8 (10,883.5-14,584.3)	606,610.7 (508,010.7-726,610.9)	223,805.8 (171,125.6-282,892.6)	4,395,141.2 (3,774,454.8-5,090,417.6)	0.652	Middle SDI
Bahrain	16.2	1,869.8 (1,539.4-2,236.3)	536.9 (360.9-726.8)	11,641.8 (10,043.9-13,442.9)	19,850.4 (16,240.9-24,193.9)	5,371.3 (3,567.3-7,425.7)	117,584.5 (99,784.4-137,180.6)	0.751	High-middle SDI
Egypt	23.1	3,301.8 (2,642.0-4,180.2)	912.6 (656.7-1,197.2)	17,551.5 (14,310.2-21,262.5)	2,291,192.9 (1,823,108.3-2,925,788.8)	683,728.1 (487,624.6-903,539.2)	12,470,501.6 (10,160,111.4-15,231,673.2)	0.658	Middle SDI
Iran	14.9	1,581.1 (1,463.3-1,716.6)	408.6 (317.9-502.4)	10,286.0 (9,481.9-11,112.5)	1,229,499.9 (1,132,673.3-1,337,837.5)	310,766.1 (243,898.9-381,734.0)	7,686,073.8 (7,031,339.5-8,350,065.4)	0.67	Middle SDI
Iraq	20.9	3,048.3 (2,434.4-3,626.3)	815.2 (599.7-1,034.8)	15,408.9 (12,986.1-17,709.7)	750,256.7 (591,225.5-911,060.7)	211,033.8 (153,119.6-272,736.4)	4,219,818.8 (3,538,727.8-4,912,947.6)	0.671	Middle SDI
Jordan	34.3	2,214.1 (1,866.6-2,629.2)	482.9 (364.4-609.9)	10,846.4 (9,477.4-12,567.6)	160,889.0 (135,486.8-191,644.6)	36,427.3 (27,816.3-45,881.6)	840,888.1 (726,714.7-985,413.8)	0.731	High-middle SDI
Kuwait	20.3	1,657.7 (1,388.0-1,961.4)	418.4 (316.3-535.2)	8,742.1 (7,498.7-10,114.9)	57,575.4 (48,222.1-68,624.1)	13,471.0 (10,160.6-17,206.7)	286,227.2 (245,301.7-330,579.3)	0.851	High SDI
Lebanon	35.8	3,894.8 (3,352.4-4,446.4)	699.2 (551.7-862.5)	12,313.7 (10,848.8-13,850.6)	202,767.9 (174,285.9-231,472.3)	36,583.9 (28,932.5-45,004.9)	642,787.4 (566,600.4-723,243.1)	0.708	High-middle SDI
Libya	20.7	2,087.5 (1,713.2-2,556.5)	696.5 (530.6-884.7)	11,977.1 (10,104.6-14,120.9)	119,557.1 (97,680.6-147,757.8)	39,423.4 (29,948.8-50,554.0)	664,404.6 (558,672.9-785,485.8)	0.709	High-middle SDI
Morocco	11.7	1,896.2 (1,478.9-2,244.9)	600.6 (458.9-752.4)	15,659.7 (13,258.6-17,923.4)	637,126.6 (492,654.8-768,236.9)	195,276.6 (148,426.0-247,407.2)	4,878,164.1 (4,096,092.4-5,640,167.0)	0.548	Low-middle SDI
Oman	11.2	1,387.7 (1,218.7-1,582.1)	521.3 (404.7-648.6)	13,924.4 (12,756.6-15,225.4)	29,119.0 (25,180.2-33,819.6)	9,772.6 (7,564.2-12,217.0)	303,477.8 (272,135.2-338,344.1)	0.783	High-middle SDI
Palestine	22.5	2,598.8 (2,236.6-2,999.8)	731.9 (553.1-925.2)	13,748.6 (12,356.4-15,381.2)	66,223.6 (56,961.2-76,700.7)	19,385.3 (14,613.7-24,244.4)	392,835.9 (351,032.0-441,273.7)	0.588	Low-middle SDI
Qatar	17.5	1,699.4 (1,369.8-2,079.3)	574.1 (389.9-775.3)	12,717.5 (10,801.7-14,891.8)	20,817.0 (16,575.7-25,643.6)	5,735.4 (3,795.3-8,043.2)	151,496.0 (126,261.0-179,760.7)	0.83	High SDI
Saudi Arabia	14.4	1,914.7 (1,589.1-2,289.1)	631.3 (490.9-787.2)	12,658.1 (10,933.1-14,490.7)	491,512.2 (394,542.6-604,784.2)	154,346.5 (117,041.3-196,417.2)	2,956,001.5 (2,483,218.9-3,477,697.0)	0.805	High-middle SDI
Sudan	10.7	2,405.2 (1,890.5-3,032.6)	660.6 (493.8-845.9)	19,481.7 (17,082.6-22,407.0)	491,951.0 (379,108.6-629,753.3)	163,124.2 (118,820.0-211,827.0)	5,916,398.0 (5,091,382.9-6,954,740.0)	0.515	Low-middle SDI
Syria	22.8	2,993.1 (2,340.6-3,879.3)	725.0 (546.8-944.4)	14,145.1 (11,512.5-17,297.5)	399,326.6 (309,455.1-525,518.1)	94,255.2 (69,836.2-124,926.6)	1,720,441.5 (1,388,023.9-2,133,730.0)	0.619	Middle SDI
Tunisia	23.7	2,306.3 (1,802.7-2,918.9)	536.3 (380.6-719.2)	11,018.2 (8,956.2-13,375.8)	296,187.0 (230,766.6-375,928.9)	67,416.1 (47,675.8-90,324.7)	1,316,552.4 (1,066,863.8-1,609,673.4)	0.672	Middle SDI
Turkey	31	2,774.1 (2,309.5-3,319.2)	437.8 (324.8-562.5)	10,618.8 (9,213.4-12,238.9)	2,522,586.4 (2,098,418.1-3,021,845.7)	378,321.3 (279,814.8-489,291.1)	8,787,360.6 (7,613,201.9-10,205,031.7)	0.748	High-middle SDI
UAE	13.8	2,505.1 (2,024.6-3,085.8)	779.9 (574.2-1,013.2)	14,585.9 (12,402.0-17,222.1)	175,680.2 (135,615.8-227,842.2)	40,663.2 (29,095.2-55,461.4)	900,114.3 (734,034.0-1,106,053.4)	0.88	High SDI
Yemen	20.2	3,634.4 (2,922.8-4,663.8)	898.4 (677.0-1,176.7)	21,126.6 (18,148.6-25,245.6)	538,915.2 (425,460.0-705,877.7)	164,146.9 (117,933.9-220,820.1)	5,028,183.1 (4,287,259.5-5,976,706.4)	0.412	Low SDI
North Africa and Middle East	19.6	2,455.0 (2,195.9-2,745.0)	656.4 (510.6-805.2)	14,562.1 (13,021.8-16,071.7)	11,475,764.6 (10,196,258.3-12,925,331.5)	3,144,992.1 (2,422,510.0-3,855,567.5)	70,557,398.7 (62,742,478.6-78,420,491.2)	0.66	Middle SDI
Global	19.6	2,412.7 (2,241.2-2,579.0)	463.3 (356.4-565.4)	15,746.5 (14,498.4-17,192.4)	199,794,745.5 (185,451,306.1-213,815,712.3)	37,002,067.5 (28,634,295.1-45,125,997.8)	1,213,261,691.0 (1,118,887,000.3-1,319,484,142.5)	0.651	Middle SDI
Low SDI		1,958.0 (1,762.4-2,175.5)	482.5 (347.5-620.1)	24,702.1 (22,365.9-27,405.8)	10,692,265.1 (9,536,988.7-11,938,121.0)	3,811,790.6 (2,538,949.7-5,203,666.6)	252,094,285.0 (220,955,650.8-290,299,529.3)
Low-middle SDI		2,662.2 (2,424.3-2,920.3)	652.4 (483.3-817.3)	20,011.4 (18,316.0-21,906.3)	36,965,821.6 (33,555,308.6-40,646,664.8)	9,385,393.6 (7,010,276.4-11,746,134.9)	304,761,189.1 (277,928,142.0-334,160,922.3)
Middle SDI		2,488.1 (2,246.2-2,743.2)	545.0 (429.3-664.0)	13,758.1 (12,626.3-14,812.3)	63,098,685.3 (56,829,332.3-69,795,320.4)	13,176,786.7 (10,398,273.1-15,986,566.1)	326,325,612.7 (299,935,707.2-352,532,107.2)
High-middle SDI		2,612.6 (2,391.3-2,836.3)	399.7 (318.1-484.9)	10,854.7 (9,963.0-11,797.1)	53,206,858.4 (48,716,597.4-57,720,348.5)	7,724,039.0 (6,156,446.1-9,398,011.4)	199,985,973.9 (184,037,624.8-217,563,595.2)
High SDI		2,065.4 (1,921.5-2,224.1)	177.6 (137.0-218.8)	8,395.0 (7,606.5-9,252.6)	35,727,488.0 (33,440,895.3-38,249,427.4)	2,881,118.3 (2,223,968.8-3,566,995.4)	129,374,865.3 (118,130,723.1-141,323,206.8)

write.csv(a, "table2.csv")

correlation between smoking rate and burden in both sexes

exclude=c("Global", "United States of America","North Africa and Middle East")

ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="Age-standardized", metric=="Rate", measure=="Deaths", year==2019, location %nin% exclude) %>%
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>% 
  select(location, smoking, Smoking, `Secondhand smoke`, `All risk factors`) %>% 
  arrange(location) %>% 
  left_join(sdi)->a


summary(a$smoking)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   10.00   14.40   20.20   19.67   22.80   35.80       5

report::report(cor.test(a$smoking, a$Smoking, method="spearman"))

## Effect sizes were labelled following Funder's (2019) recommendations.
## 
## The Spearman's rank correlation rho between a$smoking and a$Smoking is
## positive, statistically significant, and very large (rho = 0.54, S = 712.00, p
## = 0.013)

A<- a %>% 
  ggplot(aes(label=location, x=smoking, y=Smoking))+
  geom_point(aes(color=sdiClass))+
  ggrepel::geom_text_repel(cex=3)+
  theme_classic()+
  scale_color_lancet()+
  stat_smooth(method = "lm", formula='y ~ x', se = T)+
  labs(x="Prevalence of smoking (%)", y="Age-standardized annual Deaths", color="SDI Class")

ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="Age-standardized", metric=="Rate", measure!="Deaths", year==2019, location %nin% exclude) %>% 
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>% 
  select(location, smoking, Smoking, `Secondhand smoke`, `All risk factors`) %>% 
  arrange(location) %>% 
  left_join(sdi)->a



report::report(cor.test(a$smoking, a$Smoking, method="spearman"))

## Effect sizes were labelled following Funder's (2019) recommendations.
## 
## The Spearman's rank correlation rho between a$smoking and a$Smoking is
## positive, statistically significant, and very large (rho = 0.52, S = 744.00, p
## = 0.018)

B<- a %>% 
  ggplot(aes(label=location, x=smoking, y=Smoking))+
  geom_point(aes(color=sdiClass))+
  ggrepel::geom_text_repel(cex=3)+
  theme_classic()+
  scale_color_lancet()+
  stat_smooth(method = "lm", formula='y ~ x', se = T)+
  labs(x="Prevalence of smoking (%)", y="Age-standardized annual DALYs", color="SDI Class")
  

Fig<- ggarrange(A,B, labels=c("A", "B"))

ggsave("Fig2B.pdf", Fig, height=8, width=8)

ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="Age-standardized", metric=="Rate", measure=="Deaths", year==2019) %>%
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(val = paste0 (val, "\n(", lower, "-", upper, ")")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>% 
  select(location, smoking, Smoking, `Secondhand smoke`, `All risk factors`) %>% 
  arrange(location) %>% 
  left_join(sdi)->a

## Joining with `by = join_by(location)`

## Warning in left_join(., sdi): Each row in `x` is expected to match at most 1 row in `y`.
## ℹ Row 18 of `x` matches multiple rows.
## ℹ If multiple matches are expected, set `multiple = "all"` to silence this
##   warning.

location	smoking	Smoking	Secondhand smoke	All risk factors	sdi	sdiClass
Afghanistan	10	81.7 (63.7-99.5)	40.1 (29.7-51.3)	825.8 (687.7-954.5)	0.343	Low SDI
Algeria	17.4	76.3 (63.1-91.5)	27.2 (21.3-34.1)	510.6 (441.4-589.1)	0.652	Middle SDI
Bahrain	16.2	82.7 (67.4-101.4)	24.4 (17-32.7)	492.9 (419.1-575.9)	0.751	High-middle SDI
Egypt	23.1	125.6 (98.6-160.2)	32.1 (23.3-41.8)	677.8 (546-821.9)	0.658	Middle SDI
Global	19.6	95.6 (89.1-101.8)	16.5 (12.8-20.4)	453.2 (426.4-478.8)	0.651	Middle SDI
High SDI		73.4 (70.6-76.6)	5.8 (4.5-7.2)	252.2 (239.9-264.2)
High-middle SDI		101.4 (92.9-109.7)	15.9 (12.5-19.4)	391.3 (365.3-416.7)
Iran	14.9	57.4 (53.4-62.1)	15.8 (12.6-19)	378.8 (359-397)	0.67	Middle SDI
Iraq	20.9	119 (94.2-141.1)	31.7 (23.9-39.5)	591.2 (499.1-679.4)	0.671	Middle SDI
Jordan	34.3	82.4 (67.5-99)	17.5 (13.4-21.9)	397.2 (347.6-457.9)	0.731	High-middle SDI
Kuwait	20.3	61.7 (50.6-74.9)	15.3 (11.6-19.5)	289.5 (251.6-335)	0.851	High SDI
Lebanon	35.8	150.7 (131.3-171.7)	26.2 (20.8-31.9)	463.1 (406-510.9)	0.708	High-middle SDI
Libya	20.7	74.3 (60.5-91.9)	23.7 (18.1-30.5)	411.1 (341.4-495.6)	0.709	High-middle SDI
Low SDI		79 (70.9-87.8)	16.7 (12-21.2)	684.4 (633.7-741.1)
Low-middle SDI		111.4 (101.3-122.4)	24.5 (18.2-31)	598.4 (553.5-644.3)
Middle SDI		104.6 (94.5-115.5)	22 (17-27.1)	474 (441-507.4)
Morocco	11.7	71.1 (55.4-83)	23.2 (17.7-28.8)	594.6 (494.3-659)	0.548	Low-middle SDI
North Africa and Middle East	19.6	91.6 (82.7-101.4)	23.9 (19-29)	507.9 (461.8-553.4)	0.66	Middle SDI
North Africa and Middle East	19.6	91.6 (82.7-101.4)	23.9 (19-29)	507.9 (461.8-553.4)	0.66	Middle SDI
Oman	11.2	58 (50.9-66.3)	25.1 (19.7-31.1)	623 (574.3-676.1)	0.783	High-middle SDI
Palestine	22.5	102.5 (87.8-119)	29.4 (22.4-36.8)	559 (498.6-625.6)	0.588	Low-middle SDI
Qatar	17.5	75.1 (58.5-94.7)	28.3 (20.1-38.4)	603 (505.6-716.7)	0.83	High SDI
Saudi Arabia	14.4	64.3 (53.2-76.7)	23.5 (18.1-29.2)	473.8 (405.3-539.8)	0.805	High-middle SDI
Sudan	10.7	92 (73.6-115.3)	24.4 (18.5-31.2)	646.1 (567.7-753.3)	0.515	Low-middle SDI
Syria	22.8	115.5 (88.1-151.2)	27.2 (20.1-35.9)	581.3 (468.3-718.7)	0.619	Middle SDI
Tunisia	23.7	90.6 (69.1-116.9)	20.5 (14.4-27.7)	419.4 (330.5-525.4)	0.672	Middle SDI
Turkey	31	99.2 (80.2-120.8)	17.5 (12.8-22.9)	369.5 (310.5-439)	0.748	High-middle SDI
UAE	13.8	95 (75.8-116.5)	33 (24.8-42.9)	563.3 (480.7-663.3)	0.88	High SDI
United States of America	17.6	92.5 (88.9-96.4)	6.4 (5-7.9)	318.1 (305.6-329.8)
Yemen	20.2	137.3 (112.4-174.4)	33.9 (26.1-44)	680.2 (588.7-823.2)	0.412	Low SDI

write.csv(a, "table4.csv")

ds %>% 
  filter(cause=="All causes", sex ==  c("Both"), age=="Age-standardized", metric=="Rate", measure!="Deaths", year==2019) %>% 
  mutate(location=recode(location, "Iran (Islamic Republic of)"="Iran", "Syrian Arab Republic"="Syria", "United Arab Emirates"="UAE")) %>% 
  mutate(across(c(val, upper, lower), function (x) round(x,1))) %>% 
  mutate(val = paste0 (val, "\n(", lower, "-", upper, ")")) %>% 
  select(-c(lower, upper)) %>% 
  left_join(sm, by=c("location"="entity")) %>% 
  filter(rei %in% c("Smoking", "All risk factors", "Secondhand smoke")) %>% 
  spread(rei, val) %>% 
  select(location, smoking, Smoking, `Secondhand smoke`, `All risk factors`) %>% 
  arrange(location) %>% 
  left_join(sdi)->a

## Joining with `by = join_by(location)`

## Warning in left_join(., sdi): Each row in `x` is expected to match at most 1 row in `y`.
## ℹ Row 18 of `x` matches multiple rows.
## ℹ If multiple matches are expected, set `multiple = "all"` to silence this
##   warning.

location	smoking	Smoking	Secondhand smoke	All risk factors	sdi	sdiClass
Afghanistan	10	2257.5 (1760.5-2768.9)	1192.7 (871.9-1545.6)	24772.7 (21007.1-28580.6)	0.343	Low SDI
Algeria	17.4	1778 (1495.1-2111.1)	645.1 (500.9-810.1)	12612.8 (10883.5-14584.3)	0.652	Middle SDI
Bahrain	16.2	1869.8 (1539.4-2236.3)	536.9 (360.9-726.8)	11641.8 (10043.9-13442.9)	0.751	High-middle SDI
Egypt	23.1	3301.8 (2642-4180.2)	912.6 (656.7-1197.2)	17551.5 (14310.2-21262.5)	0.658	Middle SDI
Global	19.6	2412.7 (2241.2-2579)	463.3 (356.4-565.4)	15746.5 (14498.4-17192.4)	0.651	Middle SDI
High SDI		2065.4 (1921.5-2224.1)	177.6 (137-218.8)	8395 (7606.5-9252.6)
High-middle SDI		2612.6 (2391.3-2836.3)	399.7 (318.1-484.9)	10854.7 (9963-11797.1)
Iran	14.9	1581.1 (1463.3-1716.6)	408.6 (317.9-502.4)	10286 (9481.9-11112.5)	0.67	Middle SDI
Iraq	20.9	3048.3 (2434.4-3626.3)	815.2 (599.7-1034.8)	15408.9 (12986.1-17709.7)	0.671	Middle SDI
Jordan	34.3	2214.1 (1866.6-2629.2)	482.9 (364.4-609.9)	10846.4 (9477.4-12567.6)	0.731	High-middle SDI
Kuwait	20.3	1657.7 (1388-1961.4)	418.4 (316.3-535.2)	8742.1 (7498.7-10114.9)	0.851	High SDI
Lebanon	35.8	3894.8 (3352.4-4446.4)	699.2 (551.7-862.5)	12313.7 (10848.8-13850.6)	0.708	High-middle SDI
Libya	20.7	2087.5 (1713.2-2556.5)	696.5 (530.6-884.7)	11977.1 (10104.6-14120.9)	0.709	High-middle SDI
Low SDI		1958 (1762.4-2175.5)	482.5 (347.5-620.1)	24702.1 (22365.9-27405.8)
Low-middle SDI		2662.2 (2424.3-2920.3)	652.4 (483.3-817.3)	20011.4 (18316-21906.3)
Middle SDI		2488.1 (2246.2-2743.2)	545 (429.3-664)	13758.1 (12626.3-14812.3)
Morocco	11.7	1896.2 (1478.9-2244.9)	600.6 (458.9-752.4)	15659.7 (13258.6-17923.4)	0.548	Low-middle SDI
North Africa and Middle East	19.6	2455 (2195.9-2745)	656.4 (510.6-805.2)	14562.1 (13021.8-16071.7)	0.66	Middle SDI
North Africa and Middle East	19.6	2455 (2195.9-2745)	656.4 (510.6-805.2)	14562.1 (13021.8-16071.7)	0.66	Middle SDI
Oman	11.2	1387.7 (1218.7-1582.1)	521.3 (404.7-648.6)	13924.4 (12756.6-15225.4)	0.783	High-middle SDI
Palestine	22.5	2598.8 (2236.6-2999.8)	731.9 (553.1-925.2)	13748.6 (12356.4-15381.2)	0.588	Low-middle SDI
Qatar	17.5	1699.4 (1369.8-2079.3)	574.1 (389.9-775.3)	12717.5 (10801.7-14891.8)	0.83	High SDI
Saudi Arabia	14.4	1914.7 (1589.1-2289.1)	631.3 (490.9-787.2)	12658.1 (10933.1-14490.7)	0.805	High-middle SDI
Sudan	10.7	2405.2 (1890.5-3032.6)	660.6 (493.8-845.9)	19481.7 (17082.6-22407)	0.515	Low-middle SDI
Syria	22.8	2993.1 (2340.6-3879.3)	725 (546.8-944.4)	14145.1 (11512.5-17297.5)	0.619	Middle SDI
Tunisia	23.7	2306.3 (1802.7-2918.9)	536.3 (380.6-719.2)	11018.2 (8956.2-13375.8)	0.672	Middle SDI
Turkey	31	2774.1 (2309.5-3319.2)	437.8 (324.8-562.5)	10618.8 (9213.4-12238.9)	0.748	High-middle SDI
UAE	13.8	2505.1 (2024.6-3085.8)	779.9 (574.2-1013.2)	14585.9 (12402-17222.1)	0.88	High SDI
United States of America	17.6	2657.6 (2468.5-2858.3)	203.6 (155.7-252.3)	11557.6 (10461.7-12721.8)
Yemen	20.2	3634.4 (2922.8-4663.8)	898.4 (677-1176.7)	21126.6 (18148.6-25245.6)	0.412	Low SDI

write.csv(a, "table5.csv")

Figure: A stacked bar graph showing the proportion of deaths and DALYs related to smoking in the EMRO region that are attributed to secondhand smoking. The graph could show the proportion of deaths and DALYs related to neoplasms, cardiovascular disease, and other smoking-related causes.

ages<-c("<20 years" , "20-24 years", "25-29 years" , "30-34 years"  ,  "35-39 years" ,    "40-44 years"      ,"45-49 years" ,     "50-54 years" ,     "55-59 years"  ,    "60-64 years"  ,   "65-69 years" , "70+ years"  )

excludeCauses<-c("Unintentional injuries" , "Transport injuries", "All causes","Digestive diseases","Self-harm and interpersonal violence", "Sense organ diseases"   )


A<-ds %>% 
  filter(measure=="DALYs (Disability-Adjusted Life Years)", location=="North Africa and Middle East", cause %nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Secondhand smoke", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col()+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.position="none")+
  scale_y_continuous(breaks=seq(0,300000, 100000), labels=seq(0,300,100))+
  labs(x="", y="Annual DALYs in thousands", fill="Cause")


B<-ds %>% 
  filter(measure=="DALYs (Disability-Adjusted Life Years)", location=="North Africa and Middle East", cause%nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Secondhand smoke", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col(position="fill")+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.background = element_rect(fill = NA, color = NA))+
  scale_y_continuous()+
  labs(x="", y="Proportion", fill="Cause")


C<-ds %>% 
  filter(measure=="Deaths", location=="North Africa and Middle East", cause%nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Secondhand smoke", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col()+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.position = "none")+
  scale_y_continuous(breaks=seq(0,20000,5000), labels=seq(0,20,5))+
  labs(x="", y="Annual deaths in thousands", fill="Cause")


D<-ds %>% 
  filter(measure=="Deaths", location=="North Africa and Middle East", cause%nin% excludeCauses, sex%in% c("Male", "Female"), age%in%ages, rei=="Secondhand smoke", metric=="Number", year=="2019") %>%
  mutate(age=str_remove(age, " years")) %>% 
  ggplot(aes(x=age, y=val, fill=cause))+
  geom_col(position="fill")+
  facet_wrap(.~sex)+
  scale_fill_lancet()+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, hjust=0.9, vjust=0.9), legend.background = element_rect(fill = NA, color = NA))+
  labs(x="", y="Proportion", fill="Cause")


Fig<-ggarrange(A,B,C,D, labels=c("A", "B", "C", "D"))

# combined_plot <- (A+B+C+D) +
#   plot_layout(ncol = 2, widths = c(1, 1), heights = c(2, 2))
#               
# combined_plot+
#   plot_layout(
#     heights = c(2, 2)
#   )

ggsave("Fig5.pdf", Fig, width=12)

2nd hand smoking in women

ds %>% 
  filter(measure=="Deaths", 
         location=="North Africa and Middle East", 
          sex%in% c("Female"), 
         age=="All ages", 
         rei %in% c("Smoking", "Secondhand smoke"), metric=="Number", year=="2019", cause %nin% excludeCauses) %>% 
  ggplot(aes(x=rei, y=val, fill=cause))+
  geom_col(position="fill")+
  theme_classic()+
  labs(x="", y="Proportion", fill="Cause")+
  scale_fill_lancet()

ds %>% 
  filter(
         location=="North Africa and Middle East", 
          sex%in% c("Both"), 
         age=="All ages", 
         rei %in% c("Smoking", "Secondhand smoke"), metric=="Number", year=="2019", cause=="All causes") %>% 
  group_by(measure) %>% 
  mutate(sum=sum(val)) %>% 
  mutate(percent=val*100/sum)

## # A tibble: 4 × 13
## # Groups:   measure [2]
##   measure      locat…¹ sex   age   cause rei   metric  year    val  upper  lower
##   <chr>        <chr>   <chr> <chr> <chr> <chr> <chr>  <int>  <dbl>  <dbl>  <dbl>
## 1 Deaths       North … Both  All … All … Smok… Number  2019 3.74e5 4.17e5 3.36e5
## 2 Deaths       North … Both  All … All … Seco… Number  2019 9.65e4 1.18e5 7.61e4
## 3 DALYs (Disa… North … Both  All … All … Smok… Number  2019 1.15e7 1.29e7 1.02e7
## 4 DALYs (Disa… North … Both  All … All … Seco… Number  2019 3.14e6 3.86e6 2.42e6
## # … with 2 more variables: sum <dbl>, percent <dbl>, and abbreviated variable
## #   name ¹location

You could also include a figure that shows the trends of smoking-related deaths and DALYs over time, using line graphs with separate lines for each year. The x-axis could show the years, and the y-axis could show the number of deaths or DALYs. You could also include different lines for different age groups, genders, or income groups to illustrate any patterns or differences over time.

sdis<-c("Low SDI","Low-middle SDI","Middle SDI" , "High SDI" , "High-middle SDI", "Lebanon", "Turkey", "Global", "Tunisia", "Libya", "Oman", "Syria", "UAE", "Qatar")
A<- ds %>% 
  filter(measure=="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both", location!="United States of America") %>%
  ggplot(aes(x=year, y=val, group=location))+
  geom_line(color="grey", na.rm=T)+
  geom_line(data=ds %>% filter(location %in% sdis,measure=="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), aes(color=location))+
  geom_text(data = ds %>% filter(year == 2019, location %in% sdis, measure=="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), 
            aes(label = location, 
                x = year + 1, 
                y = val, 
                color = location), hjust=0, size=3) + 
  geom_line(data=ds %>% filter(location %in% c("North Africa and Middle East","Jordan"),measure=="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), aes(color=location))+
  geom_text(data = ds %>% filter(year == 2019, location %in% c("North Africa and Middle East","Jordan"), measure=="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), 
            aes(label = location, 
                x = year + 1, 
                y = val+2, 
                color = location), hjust=0, size=3) +
          guides(color = FALSE) + 
  scale_x_continuous(breaks = unique(ds$year), expand = expansion(mult = c(0.1,0.3)))+
  theme_clean()+
  theme(legend.position = "none", axis.title =  element_text(size=12),  axis.text =element_text(size=12))+
  labs(x="", y="Annuarl death rate (per 100k)")


sdis<-c("Low SDI", "North Africa and Middle East","Low-middle SDI" , "High SDI" , "High-middle SDI", "Lebanon", "Turkey", "Tunisia", "Libya", "Oman",  "Jordan", "UAE", "Syria", "Qatar")

B<- ds %>% 
  filter(measure!="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both", location!="United States of America") %>%
  ggplot(aes(x=year, y=val, group=location))+
  geom_line(color="grey", na.rm=T)+
  geom_line(data=ds %>% filter(location %in% sdis,measure!="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), aes(color=location))+
  geom_text(data = ds %>% filter(year == 2019, location %in% sdis, measure!="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), 
            aes(label = location, 
                x = year + 1, 
                y = val, 
                color = location), hjust=0, size=3) +
  geom_line(data=ds %>% filter(location %in% c("Global","Middle SDI"),measure!="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), aes(color=location))+
  geom_text(data = ds %>% filter(year == 2019, location %in% c("Global","Middle SDI"), measure!="Deaths",  cause == "All causes",  age=="All ages", rei=="Smoking", metric=="Rate", sex=="Both"), 
            aes(label = location, 
                x = year + 1, 
                y = val+20, 
                color = location), hjust=0, size=3) +
  guides(color = FALSE) + 
  scale_x_continuous(breaks = unique(ds$year), expand = expansion(mult = c(0.1,0.3)))+
  theme_clean()+
  theme(legend.position = "none", axis.title =  element_text(size=12),  axis.text =element_text(size=12))+
  labs(x="Year", y="Annuarl DALYs rate (per 100k)")


# plotly::ggplotly(A)
# plotly::ggplotly(B)



Fig<-ggarrange(A,B, ncol=2, labels=c("A","B"))

ggsave("Fig6.pdf", Fig, height = 12, width =20)

cancer

read.csv("./cancer_data/IHME-GBD_2019_DATA-b9b57c6d-1.csv")->dc


A<- dc %>% filter(location=="North Africa and Middle East", cause!="Neoplasms", sex!="Both", rei!="Tobacco", metric=="Number", measure=="Deaths") %>% 
  mutate(cause=recode(cause, "Tracheal, bronchus, and lung cancer"="Lung cancer")) %>% 
  mutate(variable=paste(sex, rei)) %>%
  group_by(cause) %>% mutate(n=sum(val)) %>% 
  ggplot(aes(x=reorder(cause, -n), y=val, fill=variable))+
  geom_col()+
  geom_text(data=. %>% distinct(n, .keep_all = TRUE ), aes(label=round(n/1000,1), y=n), vjust=-0.3)+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, vjust=1, hjust=1), legend.position = "none")+
  scale_x_discrete(expand=expansion(mult = c(0.05,0.05)))+
  scale_y_continuous(breaks=seq(0,50000, 10000), labels=seq(0,50, 10), expand=expansion(mult = c(0,0.1)))+
  scale_fill_lancet()+
  labs(x="", y="Total deaths per year (in thousands)", fill="")

B<- dc %>% filter(location=="North Africa and Middle East", cause!="Neoplasms", sex!="Both", rei!="Tobacco", metric=="Number", measure!="Deaths") %>% 
  mutate(cause=recode(cause, "Tracheal, bronchus, and lung cancer"="Lung cancer")) %>% 
  mutate(variable=paste(sex, rei)) %>%
  group_by(cause) %>% mutate(n=sum(val)) %>% 
  ggplot(aes(x=reorder(cause, -n), y=val, fill=variable))+
  geom_col()+
  geom_text(data=. %>% distinct(n, .keep_all = TRUE ), aes(label=round(n/1000,0), y=n), vjust=-0.3)+
  theme_classic()+
  theme(axis.text.x = element_text(angle=45, vjust=1, hjust=1), legend.position=c(0.8,0.8))+
  scale_x_discrete(expand=expansion(mult = c(0.05,0.05)))+
  scale_y_continuous(breaks=seq(0,1.4e6, 1e5), labels=seq(0,1.4e3, 1e2), expand=expansion(mult = c(0,0.1)))+
  scale_fill_lancet()+
  labs(x="", y="Total DALYs per year (in thousands)", fill="")

ggpubr::ggarrange(A, B, labels="AUTO", ncol=1, common.legend = F)

ggsave("FigCancer.pdf", height=10, width=7)

Untitled

2023-05-27

smoking prevalence

correlation between smoking rate and burden in both sexes

cancer