Life Expectancy at Birth in Iran and Asia: A Six-Decade Comparative Trend Analysis

Background: Life expectancy is one the key indicators for investigating the overall health status of a population. Thus, analyzing the trend of this demographic summary measure is of great importance for planning health and social services in different societies. In this study, we aimed to model the long-term trend of life expectancy at birth in Iran and also compare the pattern of life expectancy in Iran with the whole population across Asia over the past six decades. Methods: The annual life expectancy at birth data sets were extracted for Iran and total Asia between 1960 and 2020 from the database provided by the Our World in Data website. The trend analysis was performed using the joinpoint regression model. Results: During the study period, Iranians and Asians have, respectively, experienced about 32 and 28.6 year increase in life expectancy. In addition, the obtained results from the joinpoint analysis showed that, despite the noticeable decline in life expectancy during the Iran-Iraq war, the average annual percent change of life expectancy in Iranian people was about 0.1 higher than total Asian population (0.9% vs. 0.8%). Conclusions: Despite protracted wars, poverty and social inequality in some parts of Asia, life expectancy has drastically increased in this continent during the last decades. However, life expectancy in Iranians and Asians is still much lower than those who live in wealthier parts of the world. To elevate life expectancy to a higher level, the policy makers in Asian countries should put more efforts into improving the standards of living and access to health facilities in their societies.


Background
Life expectancy (LE) is one of the most important measures for assessing health status in different populations. It estimates the average number of additional years that an individual of a given age is expected to live. Life expectancy at birth (LEB) is the most commonly used metric which is de ned as the average number of years a newborn is expected to live if the current age-speci c mortality rates at the time of its birth do not change in the upcoming years (1). As another indicator, some researchers use the healthy life expectancy (HALE) at birth (years) to measure the average number of years that a newborn could expect to live in "full health" by taking into account mortality and disability (2,3). Recent reports show that the average LEB has doubled in the entire world from the beginning of the twentieth century (4). In addition, the World Health Organization (WHO) reports show about a 10% increase in the average LEB in the last two decades (from 66.8 years in 2000 to 73.4 years in 2019) (5).
Life expectancy in a population is related to its socio-economic condition, level of public health and health care resources (6). Regarding these in uential factors, there are still remarkable disparities across and even within the world countries. According to the estimates in 2020, the average LEB was 79 years for males and 82 years for females in developed countries, while males and females had an average LEB of 69 and 73 in less developed territories, respectively (7). The reported estimates by the World Bank reveal a signi cant gap between different world regions based on the average LEB. For instance, countries in Sub-Saharan Africa had an average LEB of 62 years in 2019, while an average LEB of 81 years was reported for countries in the European Union region in the same year. Among the world countries, the lowest LEB estimates were reported for some African nations such as the Central African Republic, Chad, Lesotho, Nigeria, and Sierra Leone (all with an estimated LEB of less than 55 years) in 2019. In contrast, the highest LEB values were related to some high-income countries such as Hong Kong and Japan (about 85 years) in this year (8). As can be observed, there was an estimated gap of about 30 years in LEB between more developed countries and those in poorer areas of the world. As mentioned, these disparities can also be observed in different sub-populations within each world country. For instance, a study by Singh and Lee in the USA indicated that people with higher levels of education, lower poverty levels, professional or managerial occupations and those who were house owners had considerably higher LE than the others (9).
Previously, many researchers investigated the trend of LE and its determinants over the past few decades in our country, Iran. These studies have generally focused on different time intervals after Islamic Revolution in 1979 and all of them indicated an upward trend of life expectancy due to improvements in healthcare access and quality, literacy rate, and socioeconomic status (10)(11)(12). Regarding this, the novelty of the present research lies in achieving two main goals. First, to model the long-term trend of LEB in the Iranian population from two decades before the Islamic revolution (1960) to four decades after this historical event (2020). Second, to compare the trend of LEB in Iran and the total Asian population in the described period (1960-2020).

Data sources
In practice, there are two common methods for calculating life expectancy for different populations which lead to cohort and period life expectancy. The cohort life expectancy is commonly de ned as the average life length of a particular cohort (for instance a group of individuals who were born in a particular year). To calculate the cohort life expectancy at birth, we should follow a cohort of newborns in a speci c year and record the exact date of death for each one of them. Then the cohort's life expectancy at birth can be easily calculated as the average ages of all participants. Alternatively, some researchers use a hypothetical cohort to estimate life expectancy. In this approach, we postulate that this hypothetical cohort is exposed to the observed death rates at a particular period of time (for example a year). The estimated life expectancy using this methodology is generally known as period life expectancy. Using the period approach is a more common methodology in estimating life expectancy because the cohort data may be inaccessible or incomplete. Most of the available data sources, such as those provided by the United Nations, World Bank, and World Health Organization, are based on the period life expectancy method.
Here, it should be noted that in some circumstances, for instance when we wish to estimate life expectancy for world regions, computing the life expectancies based on either cohort or period data is almost impossible. In these cases, a combination of period and cohort data sets might be applied to construct the relevant life tables and estimate the related life expectancies (13,14,1).
The life expectancy data sets for all world countries and regions in different periods of time are available in some data sources provided by Our World in Data, World Bank, and WHO. Comparing the reported life expectancies in these data sets shows that the differences between these estimates are quite negligible. In the present study, we extracted the life expectancy data for Iranian and total Asian population in the period 1960-2020 from the data source provided by the Our World in Data (4).

Statistical analysis
In this study, we used the joinpoint regression analysis to model the pattern of changes in life expectancies at birth in Iran and total Asia by gender over the study period. To brie y describe this regression modeling approach, suppose a set of n observations (t 1 , y 1 ), (t 2 , y 2 ), ... , (t n , y n ). The joinpoint regression model is a statistical tool for relating the study time points (t i ) with the response data (y i ) so that: In addition, the weighted mean of the estimated APCs can be reported as the Average Annual Percent Change (AAPC) (15). The joinpoint software version 4.8.0.1 was used to t the joinpoint regression models and estimate the parameters (16).

Results
In this research, the LE data for Iranian and total Asian populations in 61 years (from 1960 to 2020) was analyzed. Table 1 shows the LE values for the Iranian and Asian populations by sex in the study period.
For Iranian male, female and total population, respectively 65.8%, 76.9%, and 71.2% rise in LE can be observed in this 60-year time interval. In the same time period, the Asian male, female and total population had experienced a 61.6%, 64.6%, and 63.2% increase in LE, respectively. Iranian people have experienced higher growth in LE than Asians in the last six decades (about 4.2%, 12.3%, and 8%, respectively for male, female and total population). Figure 1 displays the trend of LE in Iranian (Figure 1-a) and Asian (Figure 1-b) populations by sex over the period 1960-2020.
As a notable point in Figure 1, it can be seen that while all Asian populations (male, female and total) have experienced a continuous rise in LE, Iranian males (and subsequently total Iranian people) had a considerable fall in LE during the late seventies and early eighties followed by a remarkable increase until early nineties.
In the next step, we utilized the joinpoint regression analysis to model the trend of LE in Iran and the Asia continent in the study period. Table  2 and Figure 2 show the obtained results from tting joinpoint regression models to the LE data by gender in Iranian and Asian populations.  1983-1992 (1983-1991). The average annual percent change of life expectancy for both male and total populations was 0.9, which means that Iranian people have annually experienced about a 1% rise in LE during the study period.
Unlike the estimated trends for Iranians, the Asian men, women, and total populations have experienced quite similar trends in the last six decades. For total Asian people, the joinpoint regression model showed ve different joinpoints in 1962,1969,1975,1985, and 2014. The estimated APCs for the corresponding segments were 1.2, 2.3, 1.2, 0.8, 0.5 and 0.3, respectively. The estimated AAPC of 0.8 for the total Asian population tells us that these people have annually experienced less than 1 percent increase in LE during the study period.

Discussion
In this study, we examined the six-decade (1960-2020) trend of LEB in Iran and compared this trend with the total Asian population. Overall, when we summarize the obtained ndings via the estimated summary measure of AAPC, it can be concluded that both the Iranian and Asian populations have experienced a rather similar trend of LEB over the study period. On the other hand, a more precise comparison between these two trends indicates that Iranian males (and subsequently total Iranian population) have drastically experienced a downward LEB trend in the period of 1976-1983, while the Asian population had a rising trend of LEB with an annual percent change of 0.8 over the same time interval. This remarkable decrease in LEB of the Iranian population is directly the consequence of the revolutionary struggles in Iran and the beginning of the Iran-Iraq war in 1980.
A quick look at the reported life expectancies in the results section reveals that Iranian people have experienced about 32 years increase in LEB during the study period. Now, to describe our ndings from another point of view, we can divide the study period into two different time intervals: before and after the victory of the Islamic revolution and the beginning of the Iran-Iraq war (i.e. 1960-1980 and 1980-2020). When we compare the LEB estimates between 1980 and 1960, 9.8%, 34.4%, and 20.4% increase can be observed, respectively in males, females, and the total Iranian population. Moreover, comparing the LEB estimates between 1980 and 2020 tells us that Iranian male, female and total populations have experienced about 51.0%, 31.6%, and 42.2% rise in LEB, respectively. It seems the sharper increasing slope of LEB in Iranian men compared to Iranian women from 1980 to 2020 has compensated for the considerable gap between males and females in 1980. In other words, an estimated difference of more than 9 years in LEB in 1980 has changed to about 2 years difference in 2020. These ndings are completely in line with a research by Sepanlou et al. based on the reported data from the Global Burden of Disease (GBD) 2015 study (11). In addition, our reported LEB values are quite similar to those presented by other studies which used different data sets (for example national and subnational data sources) to estimate or predict life expectancy in the Iranian population (10,12). Regardless of the destructive effect of the Iran-Iraq war on life expectancy in the early 80, the rising trend of life expectancy in Iran could be attributed to several factors such as launching a reformed primary health care system (PHC) after the Islamic revolution which decreased health inequality between urban and rural populations, increased socio-economic growth and literacy, improvement in access to health insurance, safe water, and sanitation as well as a decreased rate of malnutrition (10,12,(17)(18)(19).
According to the presented LEBs, we can observe that the total Asian population has experienced about 28.6 years rise in life expectancy from 1960 to 2020. In addition to the identical average annual percent changes for both Asian genders (AAPC = 0.8), comparing the estimated APCs in different time intervals shows rather similar values for males and females. However, a closer look at the LEBs shows that Asian women had about 1.8 years higher life expectancy than Asian men in 1960, while this difference has increased to 4.3 years in 2020. As another noticeable result, although the Asian population has continuously experienced an upward trend of LEB over the 60-years study period, comparing the estimated APCs in different time intervals demonstrates that the steepest slope of trend has occurred in the 60s.
Afterward, a sharp decrease could be seen in slopes of the trend until the ending point of the study (from APC of 2.3 in the 60s to APC of 0.3 in the last decade of the study). This upward trend of life expectancy at birth in Asian territories re ects remarkable declines in death rates, especially among mothers, newborns, and infants during the previous decades. Along with wider access to health services, improved lifestyles and living standards, better access to healthy food, drinking water, and sanitation facilities, increased rate of educated people are some of the most important indicators of this gain in longevity in the whole population across Asia (11,(20)(21)(22). Now, it is time to compare the trend of LEB in Iran and Asia using the obtained estimates from the joinpoint regression analysis. At rst glance, the estimated AAPCs look quite similar in both Iranian and Asian populations (respectively, 0.9 and 0.8). It means that Iranian and total Asian people have annually experienced less than one percent increase in LEB over the past six decades. However, some apparent discrepancies between the observed LEBs and estimated APCs for Iran and Asia could be discovered with a more precise comparison. First, in 1960 Asian people had about 0.3 year higher LEB than Iranians and this difference reached more than 6 years in 1980 and then decreased to about 0.4 year in 1990. From the early 90s, the situation was reversed in favor of Iranians so that Iranian people had a LEB of more than 3 years higher than Asians at the ending point of the study. Second, while Asians had continuously positive slopes of the trend over the study period, Iranians have experienced a negative slope during 1976-1983 as a direct consequence of the Iran-Iraq war. Third, the highest slope of LEB growth in Asian populations has occurred in the sixties (APC of 2.3), while Iranians have witnessed the highest increasing slope in the period 1983-1991 (APC of 2.7). Although, this remarkable rise in LEB during the 60s could be a direct result of the above-mentioned factors in Asia, the extraordinary upward slope of trend during the 80s in Iran might be related to some other causes. Following the Islamic revolution in 1979 and the start of the Iran-Iraq war, the revolutionary government decided to launch new policies which encouraged marriage at younger ages and child-bearing. These policies along with an in ux of Afghan refugees raised the level of fertility signi cantly and led to the highest population growth rate of more than 4 percent from 1980 to 1986 in Iran (23)(24)(25)(26).
In this study, we analyzed the LEB data in the period 1960-2020 to investigate the long-term trend of this vital health metric in the Iranian population. To do this, we also used similar data from the whole population as the comparison group. To our knowledge, this is the rst work in this eld that applies powerful inferential statistical methods for discovering the pattern of changes in LEB and identifying the trend parameters in different time intervals. However, our statistical analysis (joinpoint regression model) does not allow us to assess the effect of different related factors (such as socio-economic variables and health indices) on the trend of LEB. This can be considered as one of the most important limitations of the present study. As another limitation of the present work, it should be noted that the published similar works in this eld have generally applied descriptive statistical methods to represent the pattern of LEB in different parts of the world. Since we interpreted our ndings according to inferential indices such as APC and AAPC estimates, comparing these analytic measures with descriptive statistics was rather impossible.

Conclusions
In general, our ndings revealed that life expectancy at birth in Iran has increased about 32 years between 1960 and 2020, while total Asian population have experienced less than 29 years rise in LEB during the same period of time. Comparing the pattern of change in LEB in Iran and Asia shows that Iranians have witnessed a sharper increasing trend than total Asian population over the past six decades, except during a seven-year period in late 70s and early 80s. However, the LEB in Iran and Asia is still lower than some other parts of the world such as northern America, Oceania and Europe. In order to elevate life expectancy to a higher level, the policy makers in Iran and Asian countries should put more efforts on promoting the quality of health systems, fair distribution of wealth, improvement in access to health facilities and providing proper nutrition in the upcoming decades.

Availability of data and materials
We con rm that all methods were performed in accordance with the relevant guidelines and regulations. The datasets analysed in the current study are available at: https://ourworldindata.org/life-expectancy.

Ethics approval and consent to participate
In this study, we used a freely available data from the Our World in Data website, thus the requirement for informed consent was waived. The ethical aspects of the present study were approved by the Ethics Committee of Shahid Beheshti University of Medical Sciences (Ethics code: IR.SBMU.RETECH. REC. 1399.1139).

Consent for publication
In this study, we extracted the life the expectancy estimates from the Our World in Data website and cited according to the proposed format of this database.