National, subnational and risk attributed burden of chronic respiratory diseases in Iran from 1990 to 2019

GBD 2019 estimates the burden of 369 diseases and injuries in seven super regions, 21 regions, and 204 countries and territories from 1990 to 2019 in terms of incidence, prevalence mortality, YLL, YLD, and DALYs. It also reports the burden of disease attributed to 87 behavioral, environmental, metabolic, and occupational risk factors. Details on the data components, data gathering, resources, analytics, and population health metrics for the GBD 2019 have been discussed in detail elsewhere [10, 11]. All burden measures are presented as count and age-standardized based on the GBD reference population to remove the effect of age structure [12]. All counts and rates are reported with a 95% uncertainty interval taking into account the potential errors in measurement, modelling and possible biases. Decomposition analysis was applied between 1990 and 2019 to identify the effect of age structure, population growth, and incidence rate changes on the observed incidents cases [13]. This study is in accordance with the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER).

According to the GBD dictionary CRDs include the following five categories: asthma, COPD, interstitial lung disease and pulmonary sarcoidosis, pneumoconiosis (including silicosis, asbestosis, coal worker pneumoconiosis, and other pneumoconioses), and other chronic respiratory diseases. International Classification of Diseases (ICD)-10 was mapped to define CRDs; mortality and non-fatality (Additional file 1) [10, 14]. We obtained our data from the GBD results tool, https://​ghdx.​healthdata.​org/​gbd-results-tool.

GBD provides a comprehensive estimation of the burden attributed to 87 risk factors in 204 countries, via the comparative risk assessment (CRA); it’s a systematic and comparable approach to risk factor quantification that offers a valuable tool for synthesizing evidence on risks and risk-outcomes associations. GBD CRA updates each GBD round, integrating new data on risk-outcome pairs, risk exposure levels, and risk-outcome associations. There are critical steps to CRA: inclusion of risk-outcome causal connections in the analysis; estimation of relative risk as a function of exposure; estimation of exposure levels and distributions; determination of the counterfactual level of exposure, the level of exposure with minimum risk called the theoretical minimum risk exposure level (TMREL); decomposition of population attributable fractions and attributable burden; and estimation of mediation of different risk factors [11]. Since 2010, the risk-outcome pairs have been bound to fulfill the World Cancer Research Fund criteria in order to prove their causal relationship [11]. The identified risk factors for CRDs were ambient ozone pollution, ambient particulate matter pollution, high body mass index, high temperature, household air pollution from solid fuels, low temperature, occupational asthmagens, occupational exposure to asbestos, occupational exposure to silica, occupational particulate matter, gases, and fumes, second hand smoke and smoking. To calculate the burden of each risk factor, the level of exposure to each risk factor known as population attributable fraction, was multiplied by the whole burden of CRD by age, sex, location and year [11].

SDI is a composite indicator that is calculated from the geometric mean of three measures, including average years of educational attainment in individuals older than 15 years, income per capita, and total fertility rate in individuals younger than 25 years in a country which is expressed on the scale of zero to one with one correlating with a lower rate of fertility, higher income and educational level [10] All of the provinces were classified based on SDI into high SDI, high-middle, middle, low-middle and low SDI. The SDI strongly correlates with variables such as mortality, life expectancy, and DALY thus acts as a tool to predict health outcomes and compare regional health outcomes [15].

Counts and rates of incidence, mortality, and DALY are reported as the primary measurements of CRD's burden. Data from the death registry, disease registry, and scientific literature were the input for the original estimations of mortality by GBD [9]. Incidence was calculated by dividing deaths by death-to-incidence ratio. Prevalence estimations were provided using Dis-Mod-MR version 2.1, which is a Bayesian regression tool used by the GBD study. We calculated ten year prevalence and then multiplied the prevalence of each period by its disability weight to estimate the YLD. We used the death number by age and normative worldwide age expectancy to calculate YLL. And at last, DALYs were calculated by adding of YLDs and YLLs to each other.

We applied a universal age structure from 2019 to calculate Age-standardized rates (ASR). We reported the ASR of each burden measure per 100,000 population using a population distribution with the age composition of the GBD reference population [12]. Uncertainty intervals (UIs) were defined as the 2.5th, and 97.5th percentiles of the uncertainty distribution by randomly selecting 1000 draws from the posterior distribution Differences between point estimates were reported as significant if more than 95% of values for the difference were either positive or negative.

In order to perform decomposition analysis, two scenarios were calculated in order to determine the proportion of changes in incidence which is attributed to population growth, age structure, and age-specific incidence rate in the country; at first the number of mentioned elements in 1990 was applied to the whole population of 2019. The difference between the obtained number and the number of cases in 1990 was associated with population growth. In the following scenario, the age-specific rates in 1990, were applied to the population size, age structure and sex structure of 2019. The gap between the second scenario and the number of cases in 2019 was associated to age-specific rates, and the difference of these two scenarios was associated to population aging.

R software version 4.0.5 was used to perform Data visualization and calculation.

The number of people living with CRDs has increased 48.9% (38.2 to 61.1) and reached 4,076,810 (3,634,959 to 4,620,503) in 2019 (Table 1). 65.09% of which was due to asthma, 37.93% due to COPD, 0.70% due to Interstitial lung disease, pulmonary sarcoidosis, and due to 0.14% pneumoconiosis. It is of note that counts and age-standardized prevalence rate (ASPR) of all CRD types increased, except for asthma (− 22.6% (− 26.7 to − 18.0)) (Additional file 5: Table S1).

In 2019, there were 731,231 (633,931 to 845,833) incident cases of CRD with an age-standardized incidence rate (ASIR) of 932.1 (799.7 to 1091.5) per 100,000 people, and it was more common in males (948 (805 to 1112.1)) than females (915.4 (797.9 to 1063.5)). While the number of incident cases increased by 34.1 (22.2 to 48.7), the ASIR decreased fell slightly compared to 1990 (948.8 (820.2 to 1108.5) (Table 1, Fig. 1A).

In 2019, south Khorasan was the province with the highest ASIR (1017.8 (884.3 to 1185.5)), while Hormozgan possessed the lowest ASIR (858.7 (728.5 to 1019.2)). In almost all provinces, ASR of CRDs incidence displayed a falling pattern until 2015; however, it acquired a rising pattern after 2015 (Additional file 2: Fig. S2). Moreover, in all provinces except North Khorasan, the incidence rate was slightly higher in males compared to females (Additional file 6: Table S2).

Like prevalence, the highest incident CRD was asthma with an ASIR of 3280.1 (2717.3 to 3978.4). However, the second most common occurring CRD was interstitial lung disease and pulmonary sarcoidosis (245 (198.5 to 298.1)), which is followed by COPD (103,602 (94,771 to 113,265)) (Additional file 5: Table S1).

With regards to SDI quantiles in 2019 in high and high-middle SDI quantiles, the ASIRs were relatively similar within each quantile. However, the gap between ASIRs of provinces of all quantiles became narrower during the studies period, while the ASR of incidence in provinces of middle, low-middle, and low SDI quantiles was more scattered. Generally, the highest ASIRs belonged to low SDI provinces such as South Khorasan (1017.8 (884.3 to 1185.5)) and Sistan and Baluchistan (Sistan and Baluchistan, 1015.1 (883.1 to 1176.3)) (Fig. 2A). However, Sistan and Baluchistan had the most prominent percentage of decrease during the studies period (− 8.7 (− 13.6 to − 4) followed by South Khorasan (6.6 (− 11.4 to − 1.6)).

Concerning age groups, the highest incidence rate belonged to the + 70 age group with an ASIR of 2407.9 (2072.2 to 2784.9) per 100,000. Moreover, the incidence of CRDs had another peak in the under five age group with an ASIR of 1546.7 (929.5 to 2416.1) but decreased until the age 30, and after that, it started rising again. It is of note that in younger age groups the CRDs occur more in males, while it is more common in females in older age groups (Fig. 3A). All of provinces also displayed the national trend for incidence (Additional file 3: Fig. S3).

Decomposition analysis revealed that the incidence of CRDS has increased 34.1% during the period between 1990 to 2019; population growth was responsible for 44%, expected new cases were responsible for − 5.7%, and at last incidence rate change accounted for − 4.2% of the observed difference. Nearly all provinces displayed a similar pattern. However, in Ardebil (− 4.4%) and Hamadan (− 4.5%) the overall incidence had decreased compared to 1990, and in Gilan, the age structure had a minimal positive effect (0.2%), and in Lorestan (1.6%), Tehran (1.3%), Ilam 0.7% and the incidence rate contributed positively to the incidence change (Table 2).

The number of CRD-attributed deaths more than doubled and raised to 16,835 (14,588 to 18,193) deaths in 2019. Meanwhile, the age-standardized death rate (ASDR) declined to 26.9 (23.2 to 29.1) deaths in 100,000 people with a -36.4% (− 50.7 to − 27.3) change.

ASDR was significantly higher in males (31.4 (27.6 to 34)) compared to females (22.4 (17.5 to 26.1)) in 2019, and concordantly, the change increments were more in females (− 38.1 (− 55.1 to − 10.1)) in comparison to males (− 36.4 (− 50.7 to − 27.3)) (Table 1, Fig. 1C).

Similar to the national trend, all provinces showed a decreasing ASDR. Furthermore, in 2019, the province with the highest ASDR was Kerman, with 58.5 (29.4 to 68.7)) deaths per 100,000 people, while Tehran possessed the lowest ASDR (14.5 (11.9 to 17.6)), which decreased − 47.1% (− 62.2 to − 26.4) compared to 1990. Moreover, deaths attributed to CRDs were higher in males of all provinces in comparison to females (Additional file 6: Table S2).

The CRD causing the most deaths in 2019 and 1990 was COPD, with an ASDR of 20.3 (17.7 to 22.1) per 100,000 people. The death rates due to all CRDs remained relatively stable with negligible increases, except for asthma which its ASDR experienced a − 69.7% (− 78.9 to − 59.7) fall and decreased to 5.6 (4.8 to 6.2) (Additional file 5: Table S1).

In 2019, In all SDI quantiles, ASRs of mortality attributed to CRD lessened compared to 1990, and provinces within each quantile had relative values with some exceptions. Kerman (58.54 (29.42 to 68.73)), as a middle-SDI, showed a higher ASDR than other provinces in the same quantile; also, east Azarbayejan (49.3 (34.2 to 57.8)) had a higher ASDR than other low-middle SDI provinces (Fig. 2C, Additional file 6: Table S2). Regarding the changes in mortality rate, Sistan and Baluchistan (− 50.9% (− 62.7 to − 33.8) and Chahar mahaal and Bakhtiari (− 48.1% (− 61.5 to − 32.8) showed the highest change increments.

Regarding the age groups, unlike incidence that had a peak in younger age groups, death rates were insignificant until age 40. After that, the death rates increased and peaked in the + 70 age group (329.9 (279.8 to 359)). Moreover, the ASDRs are reduced in all age groups compared to 1990. It is noteworthy that in all age groups, deaths attributed to CRDs are more in males, while in the older age groups, CRDs were more common in females (Fig. 3). Interestingly, the same pattern was detected at the subnational level in all provinces (Additional file 1: Fig. S1).

In 2019, CRDs were responsible for 587,911 (521,418 to 661,392) DALYs, which is significantly higher than the associated DALYs in 1990 (365,429 (315,783 to 425,585)). In contrast, the ASR of DALYs was substantially lower in 2019 (794.1 (705.2 to 886.5)) compared to 1990 (1113.2 (983.7 to 1275.4)) (Fig. 1D).

Furthermore, it was significantly higher in males (334,963 (296,527 to 374,789) than females (252,949 (219,784 to 289,416)) at a national level. A similar pattern to DALYs was observed for both sexes combined and separated in YLLs and YLDs except for a subliminal increase in YLDs (both sexes: 3.4% − 1.4 to 8.7; females: 1.4 (− 3.7 to 6.7); males: 5.5 (0.3 to 12)) (Table 1).

All provinces displayed a downward trend for DALYs, with Kerman and Tehran having the highest (1371 (885.8 to 1571.6)) and lowest (553.2 (470.1 to 644.2)) DALYs, respectively. YLLs also exhibited a similar pattern to national level. On the contrary, when studying YLDs, provinces such as Alborz (8.3 (− 13.3 to − 2.8), Kermanshah (− 3.3 (− 8.6 to 2.7)), Markazi (− 2.3 (− 8.4 to 4.8)), Qazvin (− 1.5 (− 7.3 to 4.6)), Qom (− 5.2 (− 10.5 to 0.4)), Sistan and Baluchistan (− 3.2 (− 8.7 to 3.4)), Semnan (− 2.8 (− 8.7 to 3.6)) and west Azarbayejan (3.8 (− 9.7 to 2.6)) had a opposite trend to national trend, and their ASR YLDs decreased. Furthermore, males suffered from non-negligibly more DALYs than females in all provinces (Additional file 6: Table S2).

The contribution of YLLS to DALYs was slightly more in comparison to YLDs in both 2019 and 1990 (Table 1). The same trend was observed in all provinces in both years, except Tehran, with a higher YLD than YLL in 2019. (Additional file 4: Fig. S4).

The CRD that resulted in the highest DALY was COPD with an ASR of 517.2 (471 to 560.8), followed by asthma (232.3 (185 to 299.3)), other chronic respiratory diseases (26.8 (19.6 to 32.5), Interstitial lung disease and pulmonary sarcoidosis (14.7 (10.3 to 17.7)) and Pneumoconiosis 3.1 (2.6 to 3.7). YLDs also exhibited the same order, but in the context of YLLS, Interstitial lung disease and pulmonary sarcoidosis (11.2 (6.9 to 13.4)) accounted for more YLLs than other chronic respiratory diseases (9.6 (3.9 to 13.2) (Additional file 5: Table S1).

The number of DALYs increased gradually with the increasing age groups until the 50–69 age group (189,923 (168,499 to 209,733), then declined briefly to 185,770 (165,389 to 200,540) in the + 70 age group. Meanwhile, the DALYs rate was the highest in the + 70 age group with 5350.6 (4763.6 to 5776) per 100,000 people. It is of note that in almost all age groups, males suffered from more DALYs than females in the national landscape and also in most of the age groups in all provinces (Fig. 3D).

Regarding the SDI quantiles, the DALY of provinces in all SDI quantiles decreased compared to 1990. The highest DALYs belonged to middle (Kerman (1371(885.8 to 1571.6)) and low (Sistan and Baluchistan (1300.1 (923.2 to 1513.1)) and low-middle SDI (East Azarbayejan (1114.5 (894.4 to 1264.6) provinces, while the lowest DALY was observed in the high SDI province Tehran (553.2 (470.1 to 644.2)). Furthermore, the most noticeable decreases were detected in the low SDI quantile countries such as Sistan and Baluchistan (− 42.6 (− 53.9 to − 21.6)). (Fig. 2, Additional file 6: Table S2).

At the national level, the number of all DALYs attributed to the risk factors experienced a 114.5% (84.9 to 137.9) change and reached 310,187 (274,601 to 344,635) years in 2019. In contrast, the rate of DALYs declined over the period by − 21.9% (− 33.1 to − 13) and fell to 423.3 (376.3 to 468.4). The major part of DALYs attributed to risk factors is formed by males (561.5 (496.6 to 616)) rather than females (285 (240.5 to 332.7)) (Table 3).

In 2019, the risk factor with the most associated rate of DALYs was smoking (216 (189.9 to 240.8)), followed by Ambient particulate matter pollution (117.9 (88.1 to 149.4)), high body mass index (BMI) (57 (36.3 to 81.8)) and Occupational particulate matter, gases, and fumes (54 (44 to 64.7)). The rate of DALYs attributed to high temperature, ambient particulate matter pollution, and occupational exposure to silica and asbestos grew over the studied period, whereas the other eight risk factors had a declining pattern. All risk factors except high BMI, occupational asbestos exposure, and household air pollution from solid fumes imposed more DALYs on males (Additional file 7: Table S3). In a subnational landscape, smoking was similarly the leading risk factor. The percentage of DALYs attributed to household air pollution from fuels decreased over the studied period in all provinces, but for other risk factors was nearly the same (Fig. 4).

This study had its strengths and limitations. One of the strengths of GBD 2019 was the implementation of improved health information coded using the ICD system. GBD 2019 also prevents compositional bias of national estimates by adjusting variance and weighting [39].

One of the limitations of this study which was specific to CRDs, is the controversies over clear case definitions, which often happen as underdiagnoses, particularly in patients at an early stage of the disease, and even over-diagnosis in some groups [40]. Also, since our data is from the GBD study, we could not consider the coexistence of several CRDs simultaneously. Furthermore, due to the existing lag in data acquisition, the data for recent years are projected from the previous years' trends; therefore, more precise data improves the quality of the estimations.