Burden of tracheal, bronchus, and lung cancer in North Africa and Middle East countries, 1990 to 2019: Results from the GBD study 2019

Objective To provide estimates on the regional and national burden of tracheal, bronchus, and lung (TBL) cancer and its attributable risk factors from 1990 to 2019 in the North Africa and Middle East (NAME) region. Methods and materials The Global Burden of Disease (GBD) 2019 data were used. Disability-adjusted life years (DALYs), death, incidence, and prevalence rates were categorized by sex and age groups in the NAME region, in 21 countries, from 1990 to 2019. Decomposition analysis was performed to calculate the proportion of responsible factors in the emergence of new cases. Data are presented as point estimates with their 95% uncertainty intervals (UIs). Results In the NAME region, TBL cancer caused 15,396 and 57,114 deaths in women and men, respectively, in 2019. The age-standardized incidence rate (ASIR) increased by 0.7% (95% UI -20.6 to 24.1) and reached 16.8 per 100,000 (14.9 to 19.0) in 2019. All the age-standardized indices had a decreasing trend in men and an increasing trend in women from 1990 to 2019. Turkey (34.9 per 100,000 [27.6 to 43.5]) and Sudan (8.0 per 100,000 [5.2 to 12.5]) had the highest and lowest age-standardized prevalence rates (ASPRs) in 2019, respectively. The highest and lowest absolute slopes of change in ASPR, from 1990 to 2019, were seen in Bahrain (-50.0% (-63.6 to -31.7)) and the United Arab Emirates (-1.2% (-34.1 to 53.8)), respectively. The number of deaths attributable to risk factors was 58,816 (51,709 to 67,323) in 2019 and increased by 136.5%. Decomposition analysis showed that population growth and age structure change positively contributed to new incident cases. More than 80% of DALYs could be decreased by controlling risk factors, particularly tobacco use. Conclusion The incidence, prevalence, and DALY rates of TBL cancer increased, and the death rate remained unchanged from 1990 to 2019. All the indices and contribution of risk factors decreased in men but increased in women. Tobacco is still the leading risk factor. Early diagnosis and tobacco cessation policies should be improved.


Introduction
Tracheal, bronchus, and lung (TBL) cancer is the second leading cause of new cancer cases and accounts for most of the cancer-related deaths worldwide (25% of all cancer deaths) (1). In 2019, a Global Burden of Disease (GBD) study reported 2.26 million new cases of TBL cancer and 2.04 million deaths and 45.9 million disability-adjusted life years (DALYs) due to TBL cancer (1). The chance of TBL cancer incidence is 2.6-fold higher in men than in women (2). The incidence of TBL cancer increases with age, and older people has a higher death rate due to TBL cancer (3).
Behavioral risk factors, such as smoking, diet, physical inactivity, air pollution, occupational exposure, and genetic factors, are among TBL cancer risk factors (4). Although the incidence of TBL cancer in high-sociodemographic-index (SDI) countries is higher than in low-SDI countries, the agestandardized incidence rate (ASIR) in high-SDI countries has been decreasing due to tobacco control in the last decade (2,5). TBL cancer imposes a heavy economic burden to the countries globally (6-8). Non-specific and gradual clinical manifestations of TBL cancer have led to its late diagnosis and its poor prognosis, consequently. Although surgical techniques have improved in recent years, only a small portion of patients with TBL cancer undergo surgery with curative intent (9). Considering the aforementioned facts, appropriate policies could help to increase the survival rate of patients with TBL cancer (5).
As mentioned, the pattern of changes in incidence, deaths, and DALYs of TBL cancer is not the same worldwide and some regions, such as North Africa and Middle East (NAME), show an increasing trend, especially in women, in TBL burden (1). Thus, the NAME region calls for extra consideration. Improvements in the cancer registration system of the NAME region could be one of the reasons of this increasing trend. Poor tobacco control (the leading risk factor of TBL), increased number of women smokers, low level of awareness, and availability of data for recording the disease burden in NAME have led to poor TBL control in this region (10)(11)(12)(13)(14). Although TBL cancer is one of the preventable cancers worldwide, poor implementation of interventions for preventing TBL cancer in the NAME region has resulted in slow progress in cancer prevention since 2010. Despite improvements in the NAME region's health indicator factors, there is still a significant socioeconomic disparity across and within the region's countries, which has also affected the burden of TBL (15,16). For instance, the ASIR of lung cancer in Tunisia is 15-fold higher in men than in Sudan, which could be due to the high number of unreported cases. Therefore, as one of the most common cancers, TBL cancer is one of the main concerns of the NAME region's healthcare systems that needs to be addressed (10)(11)(12)(13).
Comprehensive and up-to-date data regarding the TBL cancer burden may contribute to a better policymaking and management of this cancer in the NAME region due to the lack of sufficient evidence in this region. The GBD project reports an estimate of incidence, prevalence, mortality, years of life lost (YLLs), years lived with disability (YLDs), and DALYs of 369 diseases and injuries and their attributable risk factors (17,18). Using findings of the GBD 2019, we report (a) estimates of the incidence, prevalence, deaths, YLLs, YLDs, and DALYs of TBL cancer and (b) estimates of attributable mortality, YLLs, YLDs, and DALYs for TBL cancer risk factors in the region, by country, sex, and age group between 1990 and 2019. Further, decomposition analysis was performed to calculate the proportion of responsible factors in the emergence of TBL new cases.

Materials and methods
Methodological details of GBD 2019 have previously been published (17,18), where the burden of 369 diseases and injuries as well as attributed burden to 87 risk factors in 204 countries and territories by sex and age groups in terms of incidence, prevalence, death, DALYs, YLLs, and YLDs between 1990 and 2019 has been reported. Therefore, we have focused on the methods and statistical analyses of estimating TBL cancer.

Attributable burden estimation framework
The comparative risk assessment (CRA) conceptual framework was previously described by Murray and Lopez. CRA is the systematic evaluation of the changes in population health which result from changing the population distribution of exposure to a risk factor or a group of risk factors (21). GBD 2019 used this framework to calculate the burden of several causes and impairments attributable to 87 environmental and occupational, metabolic, and behavioral risks. Deaths, DALYs, YLLs, and YLDs attributable to risk factors were assessed. Smoking, secondhand smoke, residential radon, particulate matter pollution, occupational carcinogens, high fasting plasma glucose, and a diet low in fruits were identified as TBL cancer risk factors.

Statistical analysis
All the rates were reported as age-standardized (22) based on the GBD reference population, and numbers were expressed as allages. Age groups for TBL cancers started from 10 years. Uncertainty intervals (UIs) of 95% were calculated with the 2.5th and 97.5th percentiles of 1,000 drawn by age, sex, location, and year.
Decomposition analyses (23,24) were conducted by calculating two scenarios to reveal the proportion of population growth, age structure changes, and incidence rate changes in the emergence of new cases between 1990 and 2019. First, we applied the age structure, sex structure, and age-specific rates from 1990 to the total population of the year 2019; then, we attributed the difference between the total number of cases in 1990 and the hypothetical scenario to population growth. In the second scenario, we applied the age-specific rates from 1990 to the age structure, sex structure, and population size of 2019. Differences between the second hypothetical scenario and the first hypothetical scenario were attributed to population aging and differences between the total number of cases in 2019 and the second hypothetical scenario were attributed to changes in the age-specific rates. Percent changes were calculated as the change between burden in 2019 and 1990 divided by the burden in 1990. Figures (Table 1). There has been a decrease in time difference of ASIR (5.9 vs. 3.4), ASPR (5.8 vs. 3.0), ASDR (5.9 vs. 3.5), DALYs (5.9 vs. 3.6), YLLs (5.9 vs. 3.6), and YLDs (5.3 vs. 3.2) rate by sex between 1990 and 2019. It is noticeable that all the indices had an increasing trend in women, but they decreased in men ( Figure 1). In 1990 and 2019, the incidence rate, rates of prevalence, DALYs, and deaths due to TBL cancer in both sexes increased with age. However, the rates of TBL cancer DALYs, prevalence, and incidence in men began to decrease from 74 and 80 and the rate of DALYs decreased in women over 80 years old ( Figure 2).
The decomposition analysis revealed that the incident cases grew by 146.8% (93.4 to 206.6) between 1990 and 2019; of these cases, 76.4%, 77.3%, and -7.0% were attributable to population growth, age structure change, and the disease-incidence rate change, respectively ( Table 2). In 1990, there were 4,120 (3,627 to 5,485) incident cases in women that increased by 284.4% (73.8%, 78.4%, and 132.2% attributed to population growth, age structure change, and incidence rate change, respectively) to 15,838 (13,642 to 18,196) in 2019. In comparison with women, incident cases in men increased by 124.0% (75.0 to 186.8) between 1990 and 2019, of which 78.9%, 76.9%, and -31.8% were attributable to population growth, age structure change, and incidence rate change, respectively (Supplementary Table 2).
At the national level, in 1990, Bahrain, Turkey, and Lebanon had the highest ASIR, ASDR, ASPR, and age-standardized DALYs rate, but in 2019, the respective measures were the highest in Turkey, Lebanon, and Palestine ( Figure 3)  In most of the countries of the region, an upward trend in the incidence, prevalence, death, DALY, YLL, and YLD rates of agestandardized TBL cancer in women between 1990 and 2019 was observed (Supplementary Table 3).
In all the countries of the region, the number of new cases was higher in 2019 compared with 1990. Our decomposition analysis showed that, except for Sudan and Afghanistan, age structure changes positively contributed to the increase of new incident cases. Bahrain had the highest negative new cases change cause, attributed to the incident rate change (-328.8%) ( Table 2).

FIGURE 2
The incidence, prevalence, deaths, and DALY rates by age groups and sex in region, 1990 compared with 2019 DALYs = disability-adjusted life years. Time trend of age-standardized rate of incidence, prevalence, deaths, and DALYs by sex in region DALYs = disability-adjusted life years.

Discussion
In this study, we analyzed the results of the GBD 2019 study to determine the regional and national burden of TBL cancer and its attributable risk factors in the North Africa and Middle East region. Our analysis showed that the regional ASIR of TBL cancer slightly increased and the ASDR remained unchanged, but the age-standardized DALY rate decreased between 1990 and 2019. All the indices had a marked increase in women but decreased in men.
Studies on the global and regional trends of TBL cancer from the GLOBOCAN database also revealed an increasing and decreasing trend in TBL cancer incidence in women and men, respectively (25, 26). Smoking as the main risk factor of TBL cancer is steadily increasing in women of the NAME region, which is contrary to men (27, 28). Previously, tobacco was considered taboo among women; however, due to social media, globalization, and marketing efforts, it is changing into a "normal" behavior among women in the NAME region, which is a public health concern (27). In addition, several studies have suggested that smoking has more harmful effects on women than men (29). New types of smoking, such as water pipes, are also increasing in the NAME region (30). Not only smoking but also opium, which is prevalently used in the NAME region, increases the risk lung cancer (31). Moreover, late diagnosis of TBL cancer leads to a poor prognosis in patients (32); the government and health system of the NAME region do not conduct appropriate screening tests and have failed to establish proper rules regarding primary prevention of TBL cancer (10), which may have resulted in an unchanged death rate and an increased incidence rate of TBL cancer between 1990 and 2019.
Bahrain, Turkey, and Lebanon had the highest ASIR in 1990, but the ASIR of TBL cancer in Bahrain dropped between 1990 and 2019. Palestine, Turkey, and Lebanon were the three top countries in ASIR in 2019. Except for Bahrain and Kuwait, all the indices in all the other countries had an increasing trend between 1990 and 2019 in women. Egypt had the lowest ASIR, ASDR, ASPR, and age-standardized DALY rate in 1990; however, despite many tobacco control attempts (10), all the indices increased between 1990 and 2019. The poor healthcare system, late diagnosis, high prevalence of tobacco smoking, more efficient cancer registration system, unhealthy nutritional pattern, and weak tobacco policies in Lebanon and Turkey are Age-standardized DALYs and death rates due to tracheal, bronchus, and lung cancer attributed to risk factors in the region and countries, 1990 and 2019 DALYs = disability-adjusted life years. among the reasons leading to a high incidence of TLB cancer in these countries (33)(34)(35). In addition, the prevalence of tobacco use among the adolescents aged 13-15 years has recently increased in the countries of the region (36). The smoking rate and air pollution in Egypt have increased in the last decades leading to an increase in TBL cancer incidence. Effective educational health programs, antismoking media messaging, worksite smoking bans, warning labels, etc., could contribute to tobacco cessation among the Egyptian population (37,38). The World Health Organization's (WHO) Framework Convention on Tobacco Control has been implemented in Bahrain and several other countries like Algeria, but its implementation and application remain insufficient. It seems that quit tobacco clinics (QTC) and other smoking cessation programs, such as sponsorship and advertisement ban, increasing cigarette taxes, media campaigns, and nicotine replacement therapy, have succeeded in decreasing tobacco smoking in this country (12,39,40). The decreased rate of ASIR in men and women could be attributed to these campaigns and interventions. Lung squamous cell carcinoma that is highly associated with smoking is decreasing in Bahrain, showing the crucial role of the decline in smoking in the occurrence of TBL cancer (41). Moreover, the rapid economic development of Kuwait and all the mentioned factors for Bahrain have contributed to diminishing tobacco smoking in these countries (42,43). Overall, ASDR, DALYs, YLLs, and YLDs attributable to all risk factors of TBL cancer decreased between 1990 and 2019. Improvement of healthcare systems in diagnosis and treatment of cancers has led to a decrease in YLLs and DALYs of TBL cancer from 1990 to 2019. Nevertheless, unlike men, attributed ASDR, DALYs, YLLs, and YLDs to risk factors rose in women. Nowadays, women have more exposure to occupational risks and smoke more tobacco, which can explain the increasing trend of attributed risk to death and DALY rate of TBL cancer (27, 29). Tobacco, air pollution, occupational risks, high fasting plasma glucose, residential radon, and a diet low in fruits were the main TBL cancer risk factors. All the risk factors, except high fasting plasma glucose, generally had a decreasing trend in the NAME region countries. Tobacco smoking, the most important risk factor of TBL cancer, is very prevalent in the NAME region and is growing rapidly in women (10). According to the WHO global report on trends in tobacco smoking prevalence from 2000 to 2025 (44), if tobacco control plans in the NAME region are not applied properly, the prevalence of smoking would increase and cause more diseases, such as TBL cancer. Air pollution is also one of the main problems in the NAME region and seems to increase non-smoker TBL cancer (45). Democratic development of the NAME countries could help to lessen the environmental problems (46). Additionally, the incidence of diabetes, another risk factor of TBL cancer, is increasing globally; however, genetic factors, sedentary lifestyle, health illiteracy, inadequate healthcare quality, smoking, demographic, and economic changes in the NAME region have led to a more rapid increase in diabetes prevalence in this region (47,48).
The incidence and prevalence of TBL cancer, a leading cause of death worldwide, are still increasing in the NAME region. Since TBL cancer is a preventable disease, management and interventions by the healthcare systems could play a crucial role in decreasing its incidence. In addition, the early diagnosis of this cancer should be prioritized to improve the prognosis of TBL cancer. Tobacco continues to be a health concern in this region due to its high prevalence and the increasing trend in women. The region countries are heterogeneous regarding the political stability, SDI, population, or conflict status. Hence, educational policies regarding prevention and risk factors of TBL cancer for people and country-specific intervention could contribute to decreasing the disease burden of TBL cancer in the NAME region (10,12,27).
GBD updates its data and allows policymakers to observe the disease burden over time to evaluate the impact of health control programs and monitor country-specific data. However, some  limitations should be noted. Combined burden of three causes of cancer as TBL is one of the main limitations of this study. Lack of pathological classification of TBL cancer, low quality of several data sources in several countries, primary data availability, inconsistency of mortality rates, cofounders, challenges with full representation of UIs, collinearity problem in models, mediation bias in risk factors, and sparse data of several risk factors are among our study's limitations. In addition, some data were not collected using the preferred measurement methods or case definition, and some important determinants, such as social determinants, are not included in the risk factors yet. Adding pathological classification of TBL cancer in future could also improve the quality of GBD data. The strength of our study is that it provides the most recent evaluation of death rates, DALYs, incidence rate, etc., of a disease and its related risk factors. To the best of our knowledge, our study is the first most comprehensive study on the risk factors and burden of TBL in the NAME region and its countries. GBD 2019 used doseresponse meta-regressions to determine more accurate results for risk exposure. Many other models were used to report a precise estimation of the disease burden and attributable risk factors. In addition, new data sources and risk factors were added compared with GBD 2016 (49).

Conclusion
The incidence, prevalence, and YLD rates of TBL cancer increased, and the death rate remained unchanged from 1990 to 2019 in the NAME region. It should be mentioned that all the indices decreased in men but increased in women. The contribution of risk factors decreased in men but increased in women. Smoking is still the main risk factor for TBL cancer. Early diagnosis and tobacco cessation programs should be improved in the NAME region. The GBD TBL cancer estimates can be used to improve the health condition in each country of the NAME region and contribute to establishing effective policies in TBL cancer control.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

GBD 2019 NAME Tracheal, Bronchus, and Lung Cancer Collaborators
Health Management and Economics Research Center, Iran Sciences, King Abdulaziz University, Jeddah, Egypt; Department of