Research Article
Print
Research Article
Years of Life Lost due to Premature Mortality in Russia, 1990-2021
expand article infoAnastasiya I. Pyankova, Timur A. Fattakhov§, Mikhail B. Denisenko
‡ HSE University, Moscow, Russia
§ Lomonosov Moscow State University, Moscow, Russia
Open Access

Abstract

According to the Global Burden of Disease, in Russia in 2019, the standardised rate of years of life lost from premature mortality reached its lowest value since the early 1990s. Still, it was 1.5 and 1.3 times higher than the similar rates for men and women in the WHO European Region. The authors sought to trace the evolution of the structural characteristics of years of life lost in Russia from 1990 to 2021 and identify the factors that led to such a significant gap in the level of losses from premature mortality.

Estimates of the absolute number of years of life lost (YLL), age-specific (AYLL) and age-standardised rates (ES1976) of years of life lost (SYLL) for each sex were made based on Rosstat data for 1990-2021 on the distribution of deaths by sex, by five-year age groups (0, 1-4, 5-9...85+), and causes of death (statistical form C-51). A table for life expectancy at birth at 92.6 years was used as a standard life table. Redistribution of garbage codes of causes of death and correction for polymorbidity were not performed.

Estimates of years of life lost are comparable to WHO estimates for Russia in absolute values by sex and age, while only partially so by causes of death. From 1990-2019, SYLL declined in both sexes, by a quarter. In 2019, SYLL for men was 374 per 1,000, 2.3 higher than that for women. Increased losses during the COVID-19 pandemic levelled up these gains. The maximum inequality in years of life lost for both sexes was characteristic of external causes of death (ECD) and respiratory diseases (RD), while the minimum, of neoplasms (NP). From 1990 to 2021, SYLL declined in both sexes from CD, NP, ECD, and RD. In the pre-pandemic period, there was an increase in losses from digestive diseases (DD), infectious diseases (ID) and a group of all other classes of causes of death.

The approach we used enabled us to focus more on causes of death with a low standardised death rate (SDR), such as HIV, liver disease, and pancreatic conditions. While these causes contribute less to the SDR, deaths from them typically occur at a younger age, thus raising the total number of years of life lost. The analysis allowed us to reevaluate the impact of COVID-19, accountable for 1/7 and 1/5 of all years of life lost for men and women in 2021, respectively. Therefore, if women’s life expectancy decline was more significant than men’s, the SYLL for men during both years of the pandemic was higher than that for women.

Keywords

premature mortality, years of life lost, COVID-19, causes of death, WHO Global Health Estimates, burden of diseases

JEL codes: I14, I18, J10, J17, J18

Introduction

Origin and evolution of the measure

In 1947, American epidemiologist Mary Dempsey introduced Potential Years of Life Lost (PYLL) as a measure of premature mortality from tuberculosis [Dempsey, 1947]. Potential years of life lost or, as it is sometimes referred to in the Russian-language literature, ‘premature years of life lost’, is calculated as the sum of deaths in each age group multiplied by the number of years that members of the group did not live to reach a specific cut-off age, after which death is no longer considered premature. In general terms, the formula for calculating PYLL is as follows:

P Y L L = x = k L [ ( L - x ) × d x ] (Formula 1)

where: x – age of death or midpoint of the age group; k – lower cut-off age of ‘premature deaths’; L – upper cut-off age of ‘premature deaths’; dx – the number of deaths at age x.

To estimate the years of potential life lost, it is first necessary to address the question: What age is the cut-off, after which all deaths are considered premature? Researchers determine the cut-off age of premature death based on current and projected mortality dynamics and research objectives. There is no consensus in practice regarding the cut-off age for premature death: Eurostat sets it at 70, and the OECD at 75 (as of 2019). Depending on practical objectives, premature deaths can be calculated starting from birth, or over age 15, or for persons of working or reproductive age. In 1948, Thomas Greville, in his review of M. Dempsey’s paper, proposed an alternative way of calculating ‘potential years of life lost’. As weights for those who died at a certain age, the life expectancy corresponding to this age will be used from the standard life table [Greville, 1948]. In this case, Formula 1 can be re-written as follows:

P Y L L = x = k L [ E x × d x ] (Formula 2)

where: Ex – life expectancy of those surviving to age x, k – lower cut-off age of ‘premature deaths’; L – upper cut-off age of ‘premature deaths’, dx – the number of deaths at age x.

Then W. Haenszel proposed to calculate standardised values to allow a correct comparison of populations with different age-sex structures [Haenszel, 1950].

However, the selected cut-off age is arbitrary, making it difficult to compare results across studies. Meanwhile, the ‘potential years of life lost’ (PYLL) has been widely used and implemented in public health statistics in many countries, including the USA [National Center for Health Statistics, 2017], Canada [Statistics Canada, 2022], European Union member states, and the Organisation for Economic Co-operation and Development [OECD, 2022].

In the early 1990s, a response to this situation was provided by the Global Burden of Disease (hereafter GBD) project, which estimated a modified indicator, Years of Life Lost (YLL), that does not use the upper and lower cut-off age for ‘premature mortality’. The lower cut-off age equals 0, and the upper one equals the top age in the standard life table. In general, the formula for calculating YLL can be written as follows:

Y L L = 0 ω [ E x × d x ] (Formula 3)

where: Ex – life expectancy of those surviving to age x, ω – top age in the standard life table, dx – the number of deaths at age x.

In the first 1990 GBD estimates, the standard life table for men and women differed. For women, a table was adopted with the highest life expectancy at birth, which at that time was 82.5 years for Japanese women, and for men, with a life expectancy at birth of 80 years. The 2010 revision used a unified life table with life expectancy at birth equal to 86 years for both sexes. The GBD project is currently being implemented by the Institute for Health Metrics and Evaluation (hereafter GBD IHME). Its guidelines only describe the principle of generating a standardlife table – using the lowest observed age-sex-specific death rates for territories with a population of more than five million inhabitants in 2016 – but the resulting table is not provided. It is also not given in the next edition of 2024, only mentioning that the principle of its generation has been retained.

Since the early 2010s, the World Health Organisation (hereafter WHO) has sought to collaborate on population health estimates with the GBD IHME. However, estimates of years of life lost due to premature mortality published by the two organisations differ, sometimes significantly. Their complex relationship has been described in detail in [Mathers, 2020]. Since 2017, IHME has used its own population estimates and age-specific fertility rates [Schumacher et al., 2024], which differ from those of the UN Population Division and, by extension, from the WHO Global Health Estimates (hereafter GHE WHO). Within its framework, WHO uses a different standard life table, also subject to periodic revision in response to changes in projected life expectancy in the UN population projections. In the 2013 revision, WHO used a table with a maximum projected life expectancy of 91.9 years by 2050 [World Health Organisation, 2013]. The subsequent revisions specified that the standard life table is based on the lowest age-specific death rates projected by 2045-2050 in a country and proposed a life table with life expectancy of 89.9 years [World Health Organisation, 2020], while the most recent revision used a life table with life expectancy at birth of 92.6 years [World Health Organisation, 2024]. Whenever the UN population projections are updated, so will the standard life table used in the Global Health Estimates and published as part of the revised guidelines, which will require a recalculation of past estimates.

The main difference between WHO and IHME in their approaches to the standard life table is that WHO uses the lowest projected age-specific death rates, whereas IHME uses observed age-specific death rates. Another difference is that, unlike the IHME, the WHO publishes an abridged standard life table, which allows the results to be reproduced1. It is important to stress that, despite the differences in the standard life tables used for calculation, ‘years of life lost’ is the measure employed within both estimates.

As a result, WHO and IHME estimate the burden of premature mortality for all countries based on ‘years of life lost’ (YLL), whereas national estimates could be based on different methodological approaches (parametric or non-parametric) and indicators (YLL or PYLL). An important purpose of its use is to fully assess losses from various causes of death. ‘Years of life lost’, being one possible measure of the ‘burden’ [Rubo, Czuppon, 2023], that society bears because of premature deaths, is easily integrated into health planning systems, and helps translate mortality losses into economic losses [Gökler, Metintaş, 2022; Chiabai et al., 2018; Kolesnikova et al., 2016]. In Russia, the assessment of the burden of years of life lost due to premature mortality has drawn the attention of researchers, but none of the indicators has been included in the system of state health statistics.

In this paper, the authors aimed to trace the evolution of the structural characteristics of years of life lost in Russia from 1990 to 2021. They sought to identify those contributing to the significant gap in premature mortality losses between Russia and the WHO European Region. Additionally, the authors aimed to highlight the importance of national estimates by comparing their findings with estimates produced by international organizations, such as the Global Burden of Diseases (GBD) by the Institute for Health Metrics and Evaluation (IHME) and the Global Health Estimates (GHE) by the World Health Organization (WHO).

Years of life lost and classic mortality measures

In the context of ‘years of life lost’ indicators, the significance of deaths at younger ages is greater than that of deaths at older ages, when compared to the classical mortality measures. Therefore, the contribution of younger ages to the resulting value of this indicator is greater than, for example, to life expectancy or standardised death rates (SDR), where no weights per se are assigned to ages. The earlier the death, the greater the number of years of life lost prematurely.

The high value of the rate(s) of years of life lost for the infant age group compared to the age-specific death rates is a notable observation (Fig. 1). The maximum convergence of the rates is observed in childhood ages 1-9 years, which indicates minimal losses from premature mortality. However, after 10 years of age, the age-specific rates of the years of life lost begin to increase much earlier than the age-specific death rates. The divergence of rates for men starts as early as age 10-14, for women from 15-19, and increases sharply after age 25. In older ages, the rates of years of life lost reach a peak at 80–85 for men), following which they begin to decline, in women, there is no decrease in this age grouping, while the age-specific death rates continue to increase following the mortality laws for older ages.

Figure 1.

Age-specific years of life lost and death rates per 1,000 men and women, Russia, 2019. Source: authors’ calculations based on the data described in the “Methods and data” section.

Since deaths at younger ages are considered more significant in estimating years of life lost than those at older ages, the proportion of causes of death with a lower mean age of death (such as external causes, digestive diseases, and infectious diseases) is higher in SYLL compared to the SDR (Table 1).

Table 1.

Structure of age-standardised years of life lost (SYLL) and death rates (SDR) by major causes of death classes , Russia, 2019, %

Men Women
SYLL SDR SYLL SDR
All causes 100 100 100 100
Circulatory diseases 38.1 44.2 36.6 45.8
External causes of death 17.6 11.3 9.3 5.1
Neoplasms 16.2 17.5 20.9 17.6
Other classes of causes of death 13.4 14.6 18.9 21.5
Digestive diseases 6.7 5.6 7.7 5.9
Respiratory diseases 4.3 4.5 2.6 2.2
Infectious diseases 3.8 2.3 3.9 1.9

Notably, in women, the proportion of neoplasms in the SYLL structure is higher than the share of this cause in the SDR, indicating greater premature loss. ‘Years of life lost’ allows us to address and adapt the health system’s priorities in a slightly different way.

Two approaches to estimation – two sets of estimates

There are two measures for estimating the burden of disease due to premature mortality: years of potential life lost (represented by Formula 1 or Formula 2) and years of life lost (represented by Formula 3). This has led to some confusion in terminology within the Russian research community. Presumably several studies have taken the revised 2010 GBD methodology, which uses a standard life table with life expectancy at birth of 86 years for Japanese women, in terms of ‘years of potential life lost’ [Vatolina et al., 2014; Samorodskaya et al., 2015]. There are some incorrect statements that ‘the global burden of disease indicator (DALY) consists of Potential Years of Life Lost (PYLL) and Years Lived with Disability, YLD)’ [Drapkina et al., 2019: 23]. Whereas DALY consists of Years of Life Lost (YLL) and Years Lived with Disability (YLD) [Denisenko, 2011; World Health Organization, 2020]. The guidelines for using ‘years of potential life lost’ were intended to even out the discrepancies [Krasilnikov et al., 2014]. As a result, most Russian studies follow these guidelines to estimate ‘years of potential life lost’ (PYLL) even when referring to the GBD methodology [Samorodskaya, Semenov, 2021] or refer to the indicator to be estimated as ‘years of life lost’ using its corresponding English abbreviation YLL [Savvina et al., 2019]. The use of the cut-off age, after which death is no longer considered premature, indicates the estimation of ‘years of potential life lost’ (PYLL) rather than ‘years of life lost’ (YLL).

In those Russian studies where PYLL was estimated, the cut-off age for premature mortality varied. In some of them, the retirement age was chosen as the cut-off age [Savvina et al., 2019], in others, the age of 70 years, like in the above-mentioned guidelines [Shchepin, Shishkin, 2019; Yumaguzin, Vinnik, 2015]. There are studies where estimation was done for the population in economically active age, 15–72 years2 [Bolotova et al., 2021; Drapkina et al., 2019; Kontsevaya et al., 2018, 2019].

Only a few studies have been identified using ‘years of life lost’ (YLL) and standard life table with a life expectancy at birth of 86 years [Tulenkov, 2015] or 91.9 years [Pyankova, Fattakhov, 2017; Fattakhov, Mironova, 2021]. However, these studies focus either on a specific cause of death (road traffic accidents), a particular region (Arkhangelsk Oblast), or specific population group (estimation of years of life lost of women in custody). Their results are difficult to compare in the Russian context, much less internationally.

The cut-off age for determining premature mortality may vary using the “years of potential life lost” measure based on Formula 1. In contrast, Formula 2 considers the cut-off age of premature mortality alongside the standard life table. When estimating “years of life lost,” only the standard life table may change. To show the possible range of estimates both according to the approach chosen and within one of those, namely, the standard life table, we estimated YLL due to premature mortality based on years of potential life lost (Formula 1) and years of life lost (Formula 3) and using two standard life tables for Russia for 2021 (Table 2).

Table 2.

The number of years of life lost (YLL) and years of potential life lost (PYLL) across all causes of death and from ischaemic heart disease (IHD) using different standard life tables, Russia, 2021, million person-years

Both sexes Men Women
Total incl. IHD Total incl. IHD Total incl. IHD
2021 (Life table with life expectancy and сut-off age for premature mortality = 89.99)
YLL 53.2 9.2 30.3 5.3 22.9 3.87
PYLL 47.0 7.75 28.0 4.8 19.0 2.95
YLL/PYLL 1.13 1.19 1.08 1.11 1.20 1.31
2021 (Life table with life expectancy and cut-off age for premature mortality = 92.65)
YLL 57.1 9.9 32.5 5.7 24.6 4.2
PYLL 53.4 9.1 31.0 5.4 22.4 3.6
YLL/PYLL 1.07 1.09 1.05 1.06 1.10 1.14

First, the number of years of life lost (YLL) are higher than the years of potential life lost, as illustrated by the YLL/PYLL ratio, which is everywhere greater than 1 in Table 2, and is also been corroborated by other studies [Egunsola et al., 2019]. Second, the difference between the sexes in the YLL/PYLL ratio is clear, including when using different standard life tables. If a life table with life expectancy at birth of 92.65 years and the corresponding cut-off age is selected, then in 2021, the YLL are 10% higher than the corresponding PYLL for women, whereas for men, the same are 5% higher. When differentiated by causes of death most prevalent in older ages, such as Ischaemic Heart Disease (IHD), the ratio increases for both sexes, but they do so more intensely for women, by 4 percentage points (p.p.), compared to 2 p.p. for men. If a life table is selected with life expectancy at birth of 89.9 years, the difference between the sexes in the YLL/PYLL ratio is more pronounced. Third, years of life lost is higher if a standard life table is selected with a higher life expectancy at birth and corresponding cut-off age of premature mortality. Fourth, the gap in estimates between YLL and PYLL also depends on the standard life table and cut-off age selected – the higher the life expectancy at birth in the standard life table and the corresponding cut-off age, the smaller the gap between YLL and PYLL. The gap in the estimates is reduced due to a greater increase in the years of potential life lost (PYLL) as the cut-off age for premature mortality increases.

The greater the differentiation of the subject of study, the greater the variation in estimates depending on the selected measure and its varying parameters. A minimum variation in years of life lost estimates is among men from all causes of death, while the maximum variation is among women from IHD .

In summary, the presented range of estimates indicates that the results are sensitive to the selection of the measure and its varying parameters and, consequently, the importance of their explicit description.

Methods and data

Annual absolute number (YLL), age-specific (AYLL) and standardised rates of years of life lost (SYLL) were calculated by sex and cause of death for Russia from 1990 to 2021 using the following formulas:

Absolute number of years of life lost,

Y L L ( c , s , x , t ) = E ( x ) × D ( c , s , x , t ) (Formula 4)

Age − specific years of life lost rate,

A Y L L ( c , s , x , t ) = Y L L ( c , s , x , t ) N ( s , x , t ) × 1000 . (Formula 5)

Age standardised years of life lost rate,

S Y L L ( c , s , t ) = x = 0 x = 85 + A U Y L L ( c , s , x , t ) × S ( x ) (Formula 6)

where D (c,s,x,t) – number of deaths from a specific cause c at age x of the sex s in year t; E (x) – life expectancy at age x from the standard life table; N (s,x,t) – population size of sex s, at age x in year t; S (x) – proportions of relevant age groups in the total population taken as the standard (1976 European standard for the age-sex structure of the population).

The standard life table used was from WHO’s Global Health Estimates project, with a life expectancy at birth of 92.6 years [World Health Organisation, 2024] (Table A1).

Calculations are based on Rosstat data for 1990-2021 on the age-sex distribution of deaths by causes of death (statistical form C-51) separately for men and women, by five-year age groups (0, 1-4, 5-9, ..., 85+) for the main causes-of-death classes (circulatory diseases (I00-I99); neoplasms (C00-D48); external causes of morbidity and mortality (V01-Y98); diseases of the respiratory system (J00-J99); diseases of the digestive system (K00-K93); certain infectious and parasitic diseases (A00-B99); COVID-19 (U07,1); other classes of causes of death3 in total; as well as for narrower groups of causes of death (e.g. HIV, tuberculosis, some oncological nosologies; please refer to Table A4 for their full list). This paper often uses larger age groups, sometimes combining them into 7, 6, or 3, based on the study’s objectives or comparison sources. Results in this paper are given for the following reference years: 1990, 2000, 2010, 2019, 2021.

Results

In Russia, the total number of years of life lost has been steadily declining for both men and women since 2005, following fluctuations in the 1990s and early 2000s. The 1990 figures were overcome in 2014 for women and in 2016 for men. The decline continued until 2019, when the lowest values were recorded for both sexes, at 43.5 million person-years.

Age groups contributed differently to the changes in the absolute number of years of life lost. From 1990 to 2021, young men experienced a decrease in losses, while those aged 35 and older experienced an increase in the absolute loss (Figure 2). The decomposition by age groups of the changes over different periods in YLL allows us to see the improvements achieved during 2005-2019 (bottom right panel), when only some older age groups experienced an increase in losses from premature mortality. Setbacks occurred during the pandemic. Although some younger groups experienced improvements, older groups experienced significant losses (bottom left panel).

Figure 2.

Age contribution to changes in the years of life lost between different periods for Russian men. Source: authors’ calculations based on the data described in the “Methods and data” section.

Absolute values by age do not indicate the process’s intensity, as the number of deaths in any age group depends on the size of that group.

Absolute number of years of life lost: comparison of estimates

The estimates of years of life lost in absolute terms are closest to the WHO Global Health Estimates, which exceed the results of this study by 0.5% to 1.6%, depending on sex and year. The proximity of the estimates is probably due to the use of a unified standard life table. This contrasts the estimates by the Institute for Health Metrics and Evaluation (hereinafter IHME4), which are lower than ours by 2.2% to 5.5%, depending on sex and year (Table 3).

Table 3.

Absolute number of years of life lost due to premature mortality in Russia as estimated by certain organisations, million person-years

1990 2000 2010 2019 2021
WHO Global Health Estimates (2024)
Total 65.93 55.83 44.00 57.80
Men 42.13 34.64 26.54 32.79
Women 23.80 21.19 17.47 25.01
Institute for Health Metrics and Evaluation (2024)
Total 45.31 62.34 54.18 41.71 55.11
Men 26.76 39.46 33.55 24.83 31.03
Women 18.55 22.87 20.63 16.88 24.08
This study
Total 47.80 65.17 55.47 43.47 57.08
Men 28.33 41.47 34.32 26.18 32.47
Women 19.47 23.70 21.14 17.28 24.61

When comparing the WHO estimates and the results of this study by age and sex, the following can be noted. In 2000, WHO estimates were higher than ours in most age groups for both men and women, except for the oldest group (70+). By 2021, they were lower for those aged 0 to 30-49 and higher for those aged 50-70+. However, the differences do not exceed 15%, although they increase over time. The IHME estimates for all age groups are lower than our study’s, but not by more than 15%, and only occasionally slightly higher in the 70+ age group (Table A2).

The situation is quite different if we compare the WHO estimates by sex and cause of death with the findings of this study. For some causes of death, such as neoplasms or cerebrovascular diseases, the differences are not large, while for others, the estimates are dramatically different, such as for HIV-related deaths, intentional self-harm, including suicide, and certain other causes among men (Table A3). This may be due to several factors. First, a different grouping of causes of death. WHO offers estimates of years of life lost due to premature mortality, divided into three major blocks: 1. Infectious, maternal, perinatal and nutritional conditions; 2. Non-communicable diseases; 3. Accidents. Thus, the causes of death making up the customary class ‘respiratory diseases’ are split between the 1st and 2nd major blocks; certain poisonings from the class ‘external causes of death’ are moved to the 2nd major block under ‘mental disorders and substance abuse’, etc. Table 3 in the Annex gives the WHO estimates, converting them to the usual classification of causes of death to allow comparison with our results. However, the differences remain due to another factor which is the redistribution of deaths carried out by the WHO, to which codes are assigned that can be attributed to poorly defined and imprecisely designated causes of death5.

As a result, the findings of this study are comparable with the WHO estimates for Russia in terms of absolute values by sex and age, while only partially so by cause of death.

Dynamics of standardised rate of years of life lost

In 2019, the standardised rate of years of life lost for men was 374 per 1,000, which was 2.3 times higher than for women. During 1990 and 2021, the period including the two years of the COVID-19 pandemic, SYLL for men decreased only marginally, by 8%, while for women no positive variation was seen at all (Fig. 3, Table 4). During the period of the COVID-19 pandemic, there was a significant increase in SYLL, resulting in the 2021 rates returning to the 2013 levels for men and the early 2010s or 1990s levels for women. The setback to the early 1990s level for women was due to a more intense increase in the SYLL during the COVID-19 pandemic, especially in its second year (Figure 3).

Table 4.

The ratio of age-specific rates of years of life lost by periods

0–14 15–29 30–44 45–59 60–75 75+
Men
2021/1990 0.3 0.6 1.0 0.9 1.0 0.9
2019/1990 0.3 0.6 1.0 0.8 0.8 0.7
2021/2000 0.3 0.3 0.6 0.6 0.8 0.8
2019/2000 0.3 0.3 0.6 0.5 0.6 0.7
Women
2021/1990 0.3 0.8 1.3 1.0 0.8 0.9
2019/1990 0.3 0.7 1.2 0.8 0.7 0.8
2021/2000 0.3 0.5 0.9 0.7 0.7 0.8
2019/2000 0.3 0.5 0.8 0.6 0.6 0.7
Figure 3.

Dynamics of the age-standardised rate of years of life lost from premature mortality by sex per 1,000 population and their ratio in Russia. Source: authors’ calculations based on the data described in the “Methods and data” section.

Excluding the pandemic period, some improvements can be observed: between 1990 and 2019, there was a 25% and 28% decrease in the SYLL for men and women, respectively (Figure 3, Table 4).

The level of losses due to premature mortality in Russia varies significantly by sex; in some years (2005-2006), the SYLL for men was 2.6 times higher than for women. However, after the maximum gap in the mid-2000s, there was a convergence of rates between the sexes due to a faster decline in losses among men. A more intense increase in SYLL among women compared to men during the COVID-19 pandemic led to an even greater reduction of the gap. As a result, in 2021, women’s SYLL was 2.1 times lower than men’s.

Variation in age-sex rates of years of life lost

Before 2019, a decline in age-specific rates of years of life lost was observed in almost all age groups among both men and women; however the intensity of the decline was uneven across ages (Table 4, Figure 4). The variations associated with the COVID-19 pandemic deserve special mention, as all previous long-term trends were disrupted.

Figure 4.

Changes of the age profile of years of life lost, per 1,000 population in each age group, Russia. Source: authors’ calculations based on the data described in the “Methods and data” section.

Age-specific rates of years of life lost decreased most intensely in infancy and childhood (0-14 years) among both men and women. The variations in this age group were the most stable. In the group of young adults aged 15-29, the decline was also intense. In 30-44-year-olds, the progress in reducing losses from premature mortality was least significant. Thus, over 1990-2019, AYLL in this age group did not change for men, while for women these even increased relative to 1990. The dynamics of AYLL at ages 45-59 and 60-75 for men and women are similar: the rates for men have decreased by 20% by 2019 in both groups relative to 1990 and by 50% and 40% relative to 2000, respectively; for women, by 30% and 40% relative to 1990, and by 40% in both groups relative to 2000. By 2019, in the oldest age group of men 75+, the age-specific rate(s) declined by 30% relative to both 1990 and 2000; and a comparable decline was observed in women.

However, comparing the variation in age-specific rates of years of life lost, including the years of the COVID-19 pandemic, with the 1990 baseline, we find a backslide in the reduction of losses in the group of young adult men and women (30-44 years). It is most pronounced for the youngest women, as well as for women aged 45-59 years. Comparing with 2000, while excluding the 1990s, there was some progress for men, especially in the 30-44 and 45-59 age group characterised by their riskiest behaviours. For young women aged 30-44, the reduction was small, with rates declining by 10% between 2000 and 2021.

Years of life lost, by sex and cause of death

The causes-of-death structure of the SYLL is sex-specific. In 2019, circulatory diseases (hereinafter referred to as CD), external causes of death (hereinafter referred to as ECD) and neoplasms (hereinafter referred to as NP) collectively accounted for 72% of the SYLL in men. For women, the top three causes of death, accounting for 76% of the SYLL, are slightly different: CD, NP, and group of other classes of causes of death. Respiratory (hereafter RD) and digestive (hereafter DD) diseases as well as infectious diseases (hereafter ID) together accounted for 15% of the SYLL for men and 14% for women in 2019 (Table 5).

Table 5.

Age-standardised years of life lost rate by cause of death per 1,000 population, and its structure, 2019

men % women % men/women
All causes 374 100 161 100 2.3
Circulatory diseases 142 38 59 37 2.4
External causes of death 66 18 15 9 4.4
Neoplasms 61 16 34 21 1.8
Other classes of causes of death 50 13 30 19 1.6
Digestive diseases 25 7 12 8 2.0
Respiratory diseases 16 4 4 3 3.7
Infectious diseases 14 4 6 4 2.3

In 2019, the highest gender inequality in SYLL was observed in external causes of death and respiratory diseases, with male rates exceeding female rates by a factor of 4.4 and 3.7 , respectively. In contrast, the least inequality was found in neoplasms and the ‘all other causes’ group, where male rates were 1.8 and 1.6 times higher than female rates.

Over the entire period, taking into account the years of the COVID-19 pandemic, both sexes saw a reduction in losses from such ‘large’ (in terms of level and contribution to the standardised rates) classes of causes of death as CD, neoplasms, external causes of death, and respiratory diseases. Before the pandemic, the greatest reduction in losses among women was characteristic (in descending order) of RD, external causes of death, CD, only then followed by neoplasms, while among men the least progress in the reduction of losses was achieved from CD.

From 1990-2019, the SYLL declined for both sexes from such ‘large’ (in terms of level and share in the SYLL) classes of causes of death as CD, neoplasms, external causes of death, and respiratory diseases. Before the COVID-19 pandemic, the most significant reduction in SYLL among women occurred (in descending order) from RD, external causes of death, CD, and among men – from RD, external causes of death, neoplasms. Losses among women from neoplasms decreased less intensively, and among men – from CD.

Table 6.

Classification of the major causes of death by the evolution pattern and contribution to the age-standardised years of life lost rate (SYLL) before the pandemic

1990–2019 Evolution
Men Women
Reduction Growth or slight reduction Reduction Growth or slight reduction
Proportion of SYLL, 2019 Three major causes (>70% of SYLL) CD, ECD, NP - CD, NP All other classes
Other causes RD All other classes, DD, Infectious ECD, RD DD, Infectious

The pandemic years had a different impact on the long-term trends of SYLL by these causes of death. For example, the pandemic did not interrupt the systematic and steady decline in SYLL from neoplasms; SYLL from external causes of death, which had been declining since the early 2000s, stagnated during the pandemic; SYLL from circulatory diseases, which had also been declining gradually since the early 2000s, went up; and SYLL from respiratory diseases raised sharply.

The greatest years of life lost from premature mortality are caused by circulatory diseases (Fig. 5). Age-specific years of life lost rates from CD decrease from high of 0 years, with rates of 6.7 and 5.2 per 1,000 men and women, respectively. These rates reach a minimum at 5-9 years, after which they increase steadily. Losses similar to those at 0 years in men are reached at 20-24 years, and in women, at the next age group, after which they increase steadily. Excessive losses due to CD in men start to increase from an earlier age, they are two-fold higher than women’s losses starting from ages 10-14, and three or more times higher from ages 30-34 up to 60-64, after which the gap narrows due to an intensive growth at older ages in women.

Figure 5.

Age-standardised years of life lost rate for men and women, Russia, per 1,000 population. Source: authors’ calculations based on the data described in the “Methods and data” section.

The reduction in neoplasm losses has been persistent (Figure 5) and is characteristic of most locations for which losses were calculated (22 locations, Table A4). The most impressive reduction in losses occurred from a location with initially very high loss rates in both sexes, gastric malignant neoplasms (MNP). Over 1990-2021, the loss rate from this location decreased 3-fold in men and 3.6-fold in women. Another equally important achievement is the reduction in losses among men from the location with the highest loss rates: tracheal, bronchial, and lung MNP (2-fold, down to 12.4 person-years per 1,000 men in 2021). Women have high rates of loss from breast cancer (5 person-years per 1,000 women in 2021) and other female genital organs (3.7 person-years per 1,000 women in 2021); there was also a moderate reduction in these locations, by 19% and 24%, respectively. Losses from skin malignant neoplasm increased in both sexes; in women – from lip, oral cavity, and cervix MNP; and in men, from colon MNP. However, the rates of losses from these locations were initially very low or relatively low, so their negative dynamics did not disrupt the overall reduction of losses from neoplasms.

External causes of death are the second-largest class of causes of death for men and the fourth for women (Figure 5). The reduction in losses from this class of causes is one of the most impressive for both men and women. However, it stopped in 2019, although the potential for further reduction is still high. The different positions of this cause of death in the standardised rates of years of life lost for men and women are due to a huge gap in their levels. The maximum gap occurs at the most productive ages, reaching five-fold by the age of 25-29 and staying at this level until the age of 65-69, then starting to rapidly decline.

Most of the progress in the reduction in premature mortality from respiratory disease occurred between 2003 and 2019, but the COVID-19 pandemic set back mortality rates among women to the level of the early 2000s. The setback among men because of the pandemic did not appear as impressive as among women due to the initially significantly higher loss rates among men. Over 1990-2019, the gender gap in loss rates increased smoothly, peaking in 2005 at 4.1 times, but the apparent reduction was not sustainable. Thus, in 2019, as in the previous 10 years, it remained one of the highest (3.7 times) compared to other major classes of causes of death, excluding external causes.

Over 1990-2019, an increase in losses occurred from such classes of causes of death as digestive diseases, infectious diseases, and other causes-of-death classes (Figure 5 and Table A4).

The only causes-of-death class that showed an increase, albeit wave-like, of losses over the 1990-2021 is that of digestive diseases. The SYLL for this cause for both sexes more than doubled over the period, including the COVID-19 pandemic, when the increase in losses accelerated dramatically. The increase in losses from this class begins at age 30 for both men and women and is characteristic of each subsequent age group. There was a reduction before age 30, but the level of losses before that age was low, which did not affect the SYLL trend. Liver diseases, pancreatic diseases, cholelithiasis and cholecystitis, and peptic ulcer account for 80% of all DD losses. However, they demonstrate both different levels of losses and a variety of trends. Thus, cholelithiasis and peptic ulcer disease have both a low standardised coefficient of years of life lost compared to the other two DD causes, and its steady reduction until 2013, after which there was an upward or stagnant trend. Liver diseases, including alcoholic liver diseases, and pancreatic diseases were characterised by a wave-like increase in losses over 30 years: the SYLL for men increased from 3 to 3.5 times, for women, from 2.5 to 4 times.

Infectious diseases are a noteworthy cause-of-death class, as they show different trends for men and women. Men, for example, experienced a sharp increase in the SYLL until 1999, after which there was first an unstable and then an increasingly pronounced decline, including during the pandemic. Whereas for women, a steady SYLL increase began after 1999 and continued until 2017, after which its reduction began, extending, as in men, through the pandemic period. However, underlying these dynamics are asynchronous variations in the two major causes of death in this class, HIV, and tuberculosis. From 1990 to 2005, the SYLL from tuberculosis increased for men and especially for women, with the increasing by 2.6 and 4.5, respectively. After 2005, there was a steady reduction in losses from this cause of death. However, from 2000, the standardised rates of years of life lost to HIV for both sexes began to grow exponentially, peaking in 2018. Thus, the driver of the increase in losses among women was the growing HIV epidemic, with a low level of losses from tuberculosis. In contrast, the driver of the decrease in losses for men was an intense reduction in losses from tuberculosis starting from a very high level (Table A4).

The group comprised of other causes of death is a heterogeneous one (see Footnote 3 for its composition), which includes all other 13 causes of death apart from the six major ones and COVID-19. In 2019, this group of causes of death ranked third in terms of proportion of the standardised rates of years of life lost for women, ahead of external causes of death, and fourth for men. For women, there was a gradual reduction in the rate of loss from this group from 1994; for men, it started later, from 2003. The reduction continued until 2011, when the trend reversed, accelerating after 2014. Concerns about the origins of the increase in mortality from causes within this group were raised by researchers earlier [Andreev, 2016; Vasin, 2015]. However, the COVID-19 pandemic introduced a dramatic change in the dynamics of losses from this group of causes, when the rates of years of life lost increased dramatically among both men and women (Fig. 5), which raises even more significant concern. The heterogeneity of the causes of death that make up this group lies in the mortality level and the direction and intensity of its variation6. Another aspect of the differences between the classes of causes of death that make up the group under consideration is the explanatory value (definiteness and/or specificity) of the class. On the one hand, the group includes a class that entirely belongs to the so-called ‘garbage’ codes of causes of death7 [Naghavi et al., 2010] – symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified. On the other hand, the group includes quite definite causes of death, such as endocrine, nutritional and metabolic diseases, and other classes of causes of death.

Losses from premature mortality because of COVID-19 in the first year of the pandemic in men were 25.3 person-years per 1,000 men and in women, 13.5 person-years per 1,000 women; in the second year it increased by a factor of 2.6 and 3.6, respectively. In 2021, losses from COVID-19 in men were high (66 person-years per 1,000) and comparable to those from the entire class of external causes of death. However, in terms of the structure, they accounted for 1/7 of all years of life lost among men. Among women, COVID-19 losses were lower (49 person-years per 1,000) than among men, but they accounted for 1/5 of all years of life lost by women in 2021, which was comparable to the sum of losses from the two classes of causes of death, neoplasms and external causes of death.

Limitations of the study

One limitation of this study that affects its methodological comparability with the WHO and IHME estimates is the different approach to the redistribution of deaths coded with ‘garbage’ codes for cause of death. The WHO and IHME estimates involve the redistribution of deaths from this group of causes by meaningful causes of death. The list of ‘garbage’ codes for causes of death used by these organisations and the approach to allocating their numbers to other headings differ. In the context of GBD estimates, the list of ‘garbage’ codes of causes of death is subdivided into four levels according to the degree of their impact on the quality of mortality statistics by cause and the possibility of their more accurate attribution to the cause of death of a higher taxonomic level. In Russia, it is estimated that the proportion of deaths from major ‘garbage’ codes of causes of death (levels 1 and 2) increased from 8% in 1990 to 15% in 2017 [Abbafati et al., 2020]. In our study, there was no redistribution of deaths that were assigned a cause-of-death code that could be considered poorly defined or ‘garbage,’ which may affect the estimates of losses by cause of death. Given the negative upward trend in the proportion of these causes of death, their role is expected to increase.

The poor quality of estimates for ages over 85 for both sexes should be acknowledged as a data limitation, which is reflected in the reduced years of life lost in this age group and probably leads to some underestimation of losses at older ages. This is apparently due to the overestimation of the denominator, the population size, in this age group and, consequently, underestimated coefficients for these ages. This issue with the quality of Russian statistics was highlighted on numerous occasions [Papanova et al., 2018; Andreev, 2012].

Conclusion

Notwithstanding the above-mentioned limitations, the study demonstrated that its findings are comparable with the WHO estimates for Russia in terms of absolute values, sex, and age. However, this is only partially the case in terms of causes of death, due to the non-distribution of ‘garbage’ causes of death and a slightly different grouping of causes of death in the WHO Global Health Estimates. It is therefore recommended that further steps be taken to improve the accuracy of estimates of years of life lost due to premature mortality and to enhance the international comparability of estimates. These should include a study of the role of ‘garbage’ codes of causes of death and an assessment of their impact on the quality of mortality statistics in Russia.

Furthermore, it has been demonstrated that a reduction in the standardised rates of years of life lost for both men and women did occur during the pre-pandemic period, while maintaining a two-fold gender gap in favour of women. However, the losses incurred during the pandemic negated these gains. It has been demonstrated that the greatest gender inequality in the extent of losses is observed in external causes of death and respiratory diseases, indicating that men still exhibit a relatively low level of self-preservation behaviour. Furthermore, it is notable that throughout the period, there was a wave-like increase in losses among both sexes from digestive diseases.

The study has shown that the heterogeneous group ‘other causes of death’ has become increasingly important in recent years, accounting for more than 1/3 of all years of life lost by women in 2021. The growth of losses from this group, which began in 2011, is also due to causes that can be classified as ‘garbage’ or poorly defined causes of death, which reflects a growing problem of quality of Russian mortality statistics by cause of death. However, the increase in losses within this group from quite definite causes of death (diseases of the nervous system, mental disorders and behavioural disorders, endocrine system, nutritional and metabolic disorders) may be due to quite different reasons:

  1. Updating of the rules for coding mental and behavioural disorders under ICD-10, which should have been implemented as of 2005, but in Russia it apparently only took effect in 2014 [Danilova, 2021].
  2. An increased prevalence of diseases, in particular diabetes [Dedov et al., 2018], as well as its better detectability, which, in turn, may be associated with the establishment of the federal register of diabetes mellitus.
  3. Increased diversity of regional practices of coding causes of death, which may be responsible for higher losses from diseases of the nervous system, when the variation of SDRs from diseases of the nervous system by region increased in 2017 compared to 2007, when the level was regionally comparable [Danilova, 2021].
  4. A potential impact of the COVID-19 pandemic, for example, on increased losses from premature deaths resulting from pregnancy, childbirth and the puerperium, as well as maternal mortality, which increased from 9 women in 2019 to 34.5 women per 100,000 live births in 2021. An increase in maternal mortality during the COVID-19 pandemic, especially in its second year, was also recorded in several other countries (USA, Brazil) [Hoyert, 2023; Guimarães, 2023], but the Russian specificity lies in the scale of the increase, as the initial levels were lower than in the above examples.

The ‘years of life lost’ enabled a more detailed examination of causes of death with a low SDR relative to the ‘large’ classes of causes of death, where deaths occur at an earlier age, thus increasing the contribution of these causes to the total years of life lost. To illustrate, the pronounced surge in losses from infectious diseases, notably HIV, commences at ages 25-29 to peak at ages 35-39. This pattern is similar, except for a five-year shift, to that observed in losses from digestive diseases, particularly those of liver and pancreas.

The ‘years of life lost’ also provided new insights into losses from ‘major’ causes such as COVID-19. During the pandemic, women’s life expectancy declined more than men’s, dropping by 3.66 years compared to a 2.73-year decrease for men. However, throughout both years of the COVID-19 pandemic, SYLL was lower for women than for men.

Besides, the study examines the differences between two nonparametric methods for assessing years of life lost due to premature mortality: years of potential life lost (YPLL) and years of life lost (YLL). While these methods have been well developed and are widely used internationally, their distinction has not yet been clearly established in domestic research.

Moreover, even within a single approach (such as years of life lost), the results can vary significantly based on the standard life table used, the type of death rates included (projected or observed), the criteria for determining “garbage” codes for causes of death, and how the deaths coded as such are re-classified. These factors contribute to variability in estimates of losses from premature mortality and can lead to the identification of different priority risk groups, both in terms of age and causes of death.

References

  • Andreev EM (2012) O tochnosti rezul’tatov rossiiskikh perepisei naseleniya i stepeni doveriya k raznym istochnikam informatsii [On the accuracy of the results of Russian population censuses and the degree of trust in various sources of information]. Voprosy statistiki (11): 21–35.
  • Bolotova E, Samorodskaya I, Dudnikova A (2021) The structure of mortality and potential years life lost from respiratory diseases of the population of economically active age (15–72 years) of the Russian Federation in 2019. Vrach (The Doctor) (11): 5–10. https://doi.org/10.29296/25877305-2021-11-01
  • Chiabai A, Spadaro J, Neumann M (2018) Valuing deaths or years of life lost? Economic benefits of avoided mortality from early heat warning systems. Mitigation and adaptation strategies for global change 23(7): 1159–1176. https://doi.org/10.1007/s11027-017-9778-4
  • Dedov II, Shestakova MV, Vikulova OK, Zheleznyakova AV, Isakov MА (2018) Diabetes mellitus in Russian Federation: prevalence, morbidity, mortality, parameters of glycaemic control and structure of glucose lowering therapy according to the Federal Diabetes Register, status 2017. Diabetes mellitus 21(3): 144–159. https://doi.org/10.14341/DM9686
  • Denisenko MB (2011) Integral’nye pokazateli zdorov’ya [Integral health indicators]. In: Gridasov GN, Denisenko MB et al. (Eds) Mediko-sotsial’nye posledstviya demograficheskogo stareniya (na primere Samarskoi oblasti) [Medical and social consequences of demographic aging (on the example of the Samara region)]. Samara: Volga-Business, 168-200.
  • Drapkina OМ, Samorodskaya IV, Bolotova EV, Starinskaya MA (2019) The nosological structure of years of potential life lost in the economically active age in the Russian Federation in 2016. The Russian Journal of Preventive Medicine 22(1): 22–28. https://doi.org/10.17116/PROFMED20192201122
  • Egunsola O, Raubenheimer J, Buckley N (2019) Variability in the burden of disease estimates with or without age weighting and discounting: a methodological study. BMJ Open 9 (8): e027825. http://dx.doi.org/10.1136/bmjopen-2018-027825
  • Fattakhov TA, Mironova AA (2021) Lost years of life as a result of premature mortality of the population of municipal districts of the Arkhangelsk region in 2010-2019. Regional Research 1(71): 96–106. https://doi.org/10.5922/1994-5280-2021-1-8
  • Greville TNE (1948) Comments on Mary Dempsey’s Article on “Decline in Tuberculosis: The Death Rate Fails to Tell the Entire Story. ” American Review of Tuberculosis 57(4): 417–419. https://doi.org/10.1164/art.1948.57.4.417
  • Haenszel W (1950) A standardized rate for mortality defined in units of lost years of life. American Journal of Public Health and the Nations Health 40: 17–26. https://doi.org/10.2105/AJPH.40.1.17
  • Kontsevaya AV, Drapkina OM, Balanova YuA, Imaeva AE, Suvorova EI, Khudyakov MB (2018) Economic burden of cardiovascular diseases in the Russian Federation in 2016. Rational Pharmacotherapy in Cardiology 14(2): 156–166. https://doi.org/10.20996/1819-6446-2018-14-2-156-166
  • Kontsevaya AV, Mukaneyeva DK, Balanova YuA, Khudyakov MB, Drapkina OM (2019) Economic burden of respiratory diseases and chronic obstructive pulmonary disease in Russian Federation, 2016. Pulmonologiya 29(2): 159–166. https://doi.org/10.18093/0869-0189-2019-29-2-159-166
  • Naghavi M, Makela S, Foreman K et al. (2010) Algorithms for enhancing public health utility of national causes-of-death data. Population Health Metrics (8): 9. https://doi.org/10.1186/1478-7954-8-9
  • Papanova EK, Shkolnikov VM, Andreev EM, Timonin SA (2018) High Life Expectancy of Muscovites Over Age 80: Reality or a Statistical Artifact? Advances in Gerontology 8: 86–95. https://doi.org/10.1134/S2079057018020145
  • Samorodskaya IV, Vatolina MA, Boytsov SA (2015) Methodological issues and results of assessment of the global burden of disease (literature review). Russian Journal of Preventive Medicine 18(1): 4045. https://doi.org/10.17116/PROFMED201518140-45
  • Samorodskaya EV, Semenov VYu (2021) Years of potential life lost from cardiovascular diseases of the economically active Russian population in 2013-2019. Russian Journal of Cardiology 26(5): 4161. https://doi.org/10.15829/1560-4071-2021-4161
  • Savvina NV, Bessonova OG, Vinokurova II, Grjibovski AM (2019) Analysis of working years life lost due to alcohol in the republic of Sakha (Yakutia). Health care of the Russian Federation 63(4): 180-185. URL: https://doi.org/10.18821/0044-197X-2019-63-4-180-185
  • Schumacher AE (2024) Global age-sex-specific mortality, life expectancy, and population estimates in 204 countries and territories and 811 subnational locations, 1950–2021, and the impact of the COVID-19 pandemic: a comprehensive demographic analysis for the Global Burden of Disease Study 2021. The Lancet 403: 10440, https://doi.org/10.1016/S0140-6736(24)00476-8
  • Tulenkov AM (2015) The lost years of life (YLL) as a result of premature mortality of prisoners of the Volga federal district for 2014. Problems of women health 10 (4): 44–47. URL: https://www.elibrary.ru/item.asp?id=25482486
  • Vatolina MA, Samorodskaya IV, Boitsov SA (2014) Poteryannye gody zhizni i smertnost’ v rezul’tate bolezni Al’tsgeimera v Rossii [Years of life lost and mortality due to Alzheimer’s disease in Russia]. Psikhiatriya (1): 49–53. URL: https://www.psychiatry.ru/stat/296
  • Vos T, Lim SS. et al. (2020) Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet 396(10258): 1204–1222. https://doi.org/10.1016/S0140-6736(20)30925-9
  • Yumaguzin VV, Vinnik MV (2015) Mortality from external causes in Russia and in the OECD countries: assessment of premature loss and conditions of their reduction. Vestnik Bashkirskogo universiteta 20(3): 896–902. https://www.elibrary.ru/item.asp?id=24999540

Other sources of information

Danilova I (2021) Problems of comparability of data on causes of death over time within the framework of ICD-10. URL: https://demogr.hse.ru/data/2021/05/16/1382799417/Danilova_ICD_130521.pdf [Accessed on 29.10.2024]

Hoyert DL (2023) Maternal mortality rates in the United States, 2021. NCHS Health E-Stats. https://doi.org/10.15620/CDC:124678

Krasilnikov IA, Ivanova AE, Semenova VG, Sabgaida TP, Evdokushkina GN (2014) Methodological recommendations for the use of the indicator “Years of potential life lost” (YPLL) to substantiate priority health problems of the Russian population at the federal, regional and municipal levels. URL: https://mednet.ru/images/stories/files/materialy_konferencii_i_seminarov/2010/kadry2014/sessiya/metod.pdf [Accessed on 29.10.2024]

National Center for Health Statistics (2017) Years of potential life lost before age 75 for selected causes of death. by sex. race. and Hispanic origin: United States. Selected years 1980–2016. National Center for Health Statistics. https://www.cdc.gov/nchs/hus/contents2017.htm#018

OECD (2022) Potential years of life lost (indicator). OECD. https://doi.org/10.1787/193a2829-en

Statistics Canada (2022) Statistics Canada. Table 13–10–0157–01 Mortality and potential years of life lost. by selected causes of death and sex. five-year period. Canada and Inuit regions. Statistics Canada. https://doi.org/10.25318/1310015701-eng

Vikulova OK (2019) Diabetes mellitus registry: results of work 2018. requirements for filling. critical errors. URL: https://www.endocrincentr.ru/sites/default/files/all/EVENTS2019/NEWSSUM/13.03.19 Prof.komissia/VIK_Exp_13.03.19.pdf [Accessed on 29.10.2024]

WHO (2014) WHO methods and data sources for global burden of disease estimates 2000-2011. Geneva. World Health Organization

WHO (2020a) Global Health Estimates 2020: Disease burden by Cause. Age. Sex. by Country and by Region. 2000-2019. Geneva. World Health Organization; https://www.who.int/data/gho/data/themes/mortality-and-global-health-estimates/global-health-estimates-leading-causes-of-dalys

WHO (2020b) WHO methods and data sources for global burden of disease estimates 2000-2019. Geneva. World Health Organization/ http://www.who.int/gho/mortality_burden_disease/en/index.html (дата обращения 29.10.2024)

https://cdn.who.int/media/docs/default-source/gho-documents/global-health-estimates/ghe2021_daly_methods.pdf?sfvrsn=690b16c3_1WHO (2024) WHO methods and data sources for global burden of disease estimates 2000-2021. Geneva. World Health Organization. URL: https://cdn.who.int/media/docs/default-source/gho-documents/global-health-estimates/ghe2021_daly_methods.pdf?sfvrsn=690b16c3_1ghe2021_daly_methods.pdf (who.int) (дата обращения 29.10.2024)

Acknowledgements

A.I. Pyankova, M.B. Denisenko – This work is an output of a research project implemented as part of the Basic Research Program at the National Research University Higher School of Economics (HSE University).

T.A. Fattakhov – The study was carried out within the framework of research work «The reproduction of the population in socio-economic development».

Information about the authors

Anastasiya Ivanovna Pyankova – Candidate of Sociology, Research fellow, Associate Professor Vishnevsky Institute of Demography, HSE University, Moscow, 109028, Russia. Email: apyankova@hse.ru

Timur Asfanovich Fattakhov – Senior Research Fellow, Population Department Faculty of Economics of Lomonosov Moscow State University, Moscow, 119991, Russia. Email: timur300385@mail.ru

Mikhail Borisovich Denisenko – Candidate of Economics, Department Head, Associate Professor Vishnevsky Institute of Demography, HSE University,109028, Moscow, Russia. Email: mdenissenko@hse.ru

Annex

Table A1.

Abridged life table ever used in estimating years of life lost from premature mortality by different organisations in different revisions

Organisation GHE WHO GBD IHME
Year of revision 2024 2020 2013 2010 1990 men 1990 women
Neonatal 92.65 89.99 91.93 86.01 79.94 82.43
Post-neonatal 92.21 89.55 91.55 85.68 78.85 81.36
1–4 89.74 87.07 89.41 83.63 77.77 80.28
5–9 85.25 82.58 84.52 78.76 72.89 75.47
10–14 80.26 77.58 79.53 73.79 67.91 70.51
15–19 75.27 72.60 74.54 68.83 62.93 65.55
20–24 70.28 67.62 69.57 63.88 57.95 60.63
25–29 65.31 62.66 64.60 58.94 52.99 55.72
30–34 60.34 57.71 59.63 54.00 48.04 50.83
35–39 55.38 52.76 54.67 49.09 43.10 45.96
40–44 50.43 47.83 49.73 44.23 38.20 41.13
45–49 45.51 42.94 44.81 39.43 33.38 36.36
50–54 40.61 38.10 39.92 34.72 28.66 31.68
55–59 35.74 33.33 35.07 30.10 24.07 27.10
60–64 30.92 28.66 30.25 25.55 19.65 22.64
65–69 26.21 24.12 25.49 21.12 15.54 18.32
70–74 21.62 19.76 20.77 16.78 11.87 14.24
75–79 17.19 15.65 16.43 12.85 8.81 10.59
80–84 13.08 11.96 12.51 9.34 6.34 7.56
85+ 7.28 7.05 7.60 5.05 3.82 3.59
Table A2.

Comparison of absolute number of years of life lost by sex and age group from the findings of this study with WHO, IHME estimates, million person-years

Men Women
WHO IHME This Study WHO IHME This Study
2000
All 42.1 39.5 41.8 23.8 22.9 23.8
0–4 1.3 1.3 1.3 1.0 0.9 0.9
5–14 0.5 0.5 0.5 0.3 0.3 0.3
15–29 5.3 4.7 5.2 1.3 1.2 1.4
30–49 13.3 12.5 13.2 3.8 3.5 3.7
50–59 6.9 6.5 6.9 2.7 2.5 2.7
60–69 8.8 8.2 8.6 5.0 4.7 5.0
70+ 5.9 5.8 6.0 9.7 9.7 9.8
2010
All 34.6 33.5 34.4 21.2 20.6 21.2
0–4 0.9 0.9 0.9 0.7 0.7 0.7
5–14 0.2 0.2 0.2 0.1 0.1 0.1
15–29 3.3 3.3 3.4 1.0 1.0 1.1
30–49 9.9 9.7 9.7 3.2 3.2 3.2
50–59 8.0 7.8 8.0 3.4 3.2 3.3
60–69 5.6 5.3 5.6 3.2 3.0 3.2
70+ 6.6 6.3 6.6 9.6 9.4 9.6
2019
All 26.5 24.8 26.3 17.5 16.9 17.3
0–4 0.5 0.5 0.5 0.4 0.4 0.4
5–14 0.2 0.2 0.2 0.1 0.1 0.1
15–29 1.2 1.2 1.3 0.4 0.4 0.5
30–49 7.2 6.8 7.1 2.7 2.5 2.6
50–59 5.3 5.0 5.3 2.3 2.1 2.3
60–69 6.9 6.3 6.8 3.8 3.5 3.7
70+ 5.2 4.9 5.1 7.8 7.9 7.8
2021
All 32.8 31.0 32.6 25.0 24.1 24.7
0–4 0.4 0.4 0.4 0.3 0.3 0.3
5–14 0.2 0.2 0.2 0.1 0.1 0.1
15–29 1.1 1.1 1.3 0.4 0.4 0.5
30–49 8.2 8.0 8.3 3.3 3.2 3.3
50–59 6.2 5.7 6.0 3.1 2.9 3.0
60–69 9.1 8.4 8.9 6.0 5.7 6.0
70+ 7.7 7.3 7.4 11.7 11.5 11.4
Table A3.

Comparison of absolute number of years of life lost by cause from the findings of this study with WHO estimates, 2021, thousand person-years

Men Women
WHO This Study WHO This Study
All causes 32788 32466 25009 24613
Infectious and parasitic diseases 1723 862 487 423
Tuberculosis 321 197 86 54
HIV 1324 552 333 293
Hepatitis 1 54 0 28
All other infectious diseases 77 58 67 47
All respiratory diseases 7914 6312 6616 6641
COVID-19 6955 4646 6232 5581
Neoplasms 4535 4061 3711 3270
Malignant neoplasms 4526 4010 3702 3212
Other neoplasms 9 50 10 58
Cardiovascular diseases 9582 10943 8122 8169
Hypertension 127 155 145 167
Ischaemic heart disease 5153 5749 4183 4172
Cerebrovascular diseases 2536 2576 2822 2536
Digestive diseases 1695 2009 1204 1357
Peptic ulcer disease 164 193 100 107
Liver diseases 980 1163 681 773
including alcoholic liver disease 469 257 226 152
Cholelithiasis and cholecystitis 28 24 41 34
Pancreatic diseases 282 330 130 145
External causes of death 3343 4701 945 1194
Road injury 552 597 184 190
Accidental poisonings (x40, x43, x46-49) 122 - 39 -
All accidental poisonings (x40-44, x45-49) - 757 - 152
Accidental falls 262 188 96 73
Accidents caused by fire 103 92 43 38
Accidental drownings 174 186 31 34
Injuries with undetermined intent - 1530 - 376
Intentional self-harm. including suicide 1108 568 237 104
Interpersonal violence 362 210 112 62
Table A4.

Dynamics of standardised rates of years of life lost by cause of death, person-years per 1,000 persons

1990 2000 2010 2019 2021 1990 2000 2010 2019 2021
Men Women
Total 496 668 526 374 459 223 268 216 161 223
Infectious and parasitic diseases (ID) 10 21 17 14 12 4 5 5 6 5
Tuberculosis 7 17 11 3 3 1 2 3 1 1
Viral hepatitis 0 0 1 1 1 0.2 0.2 0.2 0.4 0.3
HIV 0 0 4 9 7 0 0 1 4 4
Neoplasms (NP) 93 83 71 61 57 44 42 39 34 32
malignant neoplasms (MNP) of lip, oral cavity, pharynx 4 4 4 3 3 0 0 1 1 1
esophageal MNP 4 3 2 2 2 1 0 0 0 0
gastric MNP 17 12 9 6 5 7 5 4 2 2
MNP of small bowel, including duodenum 0 0 0 0 0 0 0 0 0 0
Colon MNP 3 4 4 4 3 3 3 3 2 2
Rectum, rectosigmoid, anus MNP 3 4 3 3 3 2 2 2 2 2
MNP of other digestive organs 8 7 6 7 7 4 3 3 3 3
larynx MNP 3 3 2 1 1 0.1 0.1 0.1 0.1 0.1
trachea, bronchi, and lungs MNP 30 25 20 15 14 3 2 2 2 2
MNP of other respiratory organs and intrathoracic organs 1 1 1 0 0 0 0 0 0 0
MNP of bone and articular cartilage 1 1 1 0 0 1 0 0 0 0
Skin MNP 1 2 2 2 2 1 1 1 1 1
mammary MNP - - - - - 7 8 7 6 5
cervix MNP - - - - - 2 2 3 3 2
MNP of other female genitalia, unspecified - - - - - 5 5 5 4 4
prostate MNP 2 3 4 4 4 - - - - -
MNP of other male genitalia 0.4 0.5 0.4 0.3 0.3 - - - - -
urinary organ MNP 5 5 5 4 3 1 1 1 1 1
MNP of other or unspecified locations 4 6 7 4 4 3 4 4 3 2
Leukaemia 3 4 4 3 2 2 2 1 1 1
Other MNP of lymphatic and hematopoietic tissues 2 0 0 2 2 1 0 0 1 1
Benign neoplasms и misclassified neoplasms 1 1 1 1 1 1 1 1 1 1
Circulatory diseases (CD) 191 262 221 142 154 102 124 98 59 66
Hypertensive disease (cardiac and/or renal involvement) 1 4 4 2 2 1 3 3 1 1
Ischaemic heart disease 107 138 119 74 81 44 52 46 28 32
including myocardial infarction 19.9 18.1 16.3 11.2 11.3 6.0 5.9 5.6 3.6 3.7
Cerebrovascular diseases 58 77 56 35 37 41 50 33 18 20
Other heart diseases 10 24 32 12 14 4 8 11 4 5
Respiratory diseases (RD) 32 40 26 16 24 11 11 7 4 10
Pneumonia 7 18 16 7 15 4 5 4 2 7
Digestive diseases (DD) 13 22 29 25 28 6 9 14 12 14
Peptic ulcer disease 3 3 3 2 3 1 1 1 1 1
Liver diseases 5 10 18 14 16 2 5 9 8 9
including alcoholic liver disease 0 2 5 3 4 0 1 3 2 2
Cholelithiasis and cholecystitis 1 0 0 0 0 1 1 0 0 0
Pancreatic diseases 1 3 4 4 5 1 1 1 1 2
External causes (ECD) 112 176 108 66 67 26 40 26 15 15
Traffic accidents 25 21 15 9 9 6 6 5 3 3
Suicide 21 33 19 9 8 4 5 3 1 1
Murder 12 22 10 4 3 3 6 3 1 1
Accidental poisonings 15 32 17 9 10 4 8 4 2 2
including alcohol poisoning 9 19 9 4 4 2 5 2 1 1
Injuries with undetermined intentions 9 23 20 20 22 2 5 4 4 5
Accidental drownings 8 10 8 2 3 1 2 1 0 0
Accidental falls 3 6 4 3 3 1 1 1 1 1
Accidents caused by fire 2 5 4 1 1 1 1 1 0 0
Other classes of causes of death (OCCD) 44 64 53 50 119 31 37 27 30 80
Old age 2 4 3 3 3 2 4 3 3 3
Symptoms and misclassified conditions 6 21 20 14 16 2 5 5 4 4
Alcoholic psychosis 0.1 0.4 0.3 0.1 0.1 0.0 0.1 0.0 0.0 0.0
Chronic alcoholism / addiction syndrome 1 2 2 2 1 0 1 1 0 0
COVID-19 - - - - 66 - - - - 48
Table A5.

Dynamics of standardised rates of mortality by sex and causes of death included in the group ‘all other causes of death’, 2011-2021

2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism, (D50-D89) M 29.5 28.7 29.4 33.9 35.3 35.9 34.3 39.1 36.5 37.3 33.1
W 24.1 24.3 24.6 27.3 28.8 28.5 28.2 28.6 30.5 28.8 26.5
Endocrine, nutritional and metabolic diseases, (E00-E90) M 50.5 55.7 67.3 101.8 131.4 154.3 179.9 196.1 194.8 241.9 202.0
W 63.2 74.0 84.9 124.3 161.8 185.3 209.9 224.2 223.8 275.2 234.5
Mental and behavioural disorders, (F00-F99) M 50.4 48.7 50.5 106.9 99.0 108.8 123.4 133.3 122.0 153.4 128.6
W 16.7 17.2 20.2 49.7 58.6 62.1 73.0 80.0 76.2 92.6 76.2
Diseases of the nervous system, (G00-G99) M 164.4 156.4 179.0 270.5 396.4 517.9 600.3 629.1 585.1 685.0 655.3
W 85.0 83.6 102.1 173.5 262.6 361.6 435.9 466.5 424.8 508.9 518.0
Diseases of the eye and adnexa, (H00-H59) M 0.1 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
W 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Diseases of the ear and mastoid process, (H60-H95) M 1.2 1.0 1.1 1.1 1.2 1.4 1.5 1.4 1.3 1.0 1.2
W 0.8 0.8 0.7 0.8 0.7 0.6 0.9 1.0 0.7 0.6 1.1
diseases of the skin and subcutaneous tissue, (L00-L99) M 14.7 14.2 13.7 14.9 16.1 19.9 21.5 21.3 21.9 19.7 21.0
W 9.5 9.9 10.8 11.4 12.5 15.9 17.2 17.0 16.9 15.9 16.1
diseases of the musculoskeletal system and connective tissue, (M00-M99) M 9.0 9.1 9.0 11.6 17.2 20.4 22.5 25.3 27.3 27.8 25.7
W 12.5 12.2 11.9 15.5 21.3 25.3 26.0 29.2 30.8 34.0 32.0
Diseases of the genitourinary system, (N00-N99) M 89.7 92.2 96.2 106.9 128.4 132.4 139.7 145.5 154.4 161.6 150.0
W 55.5 55.9 59.1 63.5 75.2 77.1 77.8 85.0 87.6 93.5 87.7
Pregnancy, childbirth and the puerperium, (O00-O99) M
W 3.5 2.5 2.5 2.4 1.9 2.4 1.9 1.7 1.2 1.9 6.2
Certain conditions originating in the perinatal period, (P00-P96) M 67.1 87.6 80.6 71.1 63.6 55.6 48.9 45.5 43.3 42.0 43.1
W 49.8 70.9 63.0 56.7 47.2 43.9 39.7 34.5 34.9 32.5 30.8
Congenital malformations [abnormalities], deformations and chromosomal abnormalities, (Q00-Q99) M 49.5 47.0 44.5 42.5 38.4 39.6 36.4 35.1 31.5 34.1 32.9
W 39.9 40.5 36.7 34.9 33.1 32.0 28.1 28.4 27.6 25.7 25.7
Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified, (R00-R99) M 738.9 765.0 836.7 971.1 923.9 827.5 700.3 697.4 701.2 782.8 773.4
W 434.1 465.2 536.2 647.4 635.4 611.0 530.3 505.3 497.8 556.7 526.9

1 Summary standard life tables, published as part of the estimation of years of life lost due to premature mortality, are presented in Table 1, Annex.
2 In the studies mentioned above, it is unclear whether the estimated losses pertain to the economically active population – defined as individuals aged 15 to 72 who are considered employed or unemployed during the survey week, according to Rosstat surveys – or if they refer to the entire population within that age span. The phrase “population of economically active age” is used in all cases, making it ambiguous.
3 Other classes of causes of death are a composite class of causes of death including: diseases of the blood, hematopoietic organs and selected disorders involving the immune mechanism (D50-D89), diseases of the endocrine system, nutritional and metabolic disorders (E00-E90), mental and behavioural disorders (F00-F99), diseases of the nervous system (G00-G99), diseases of the eye and its apparatus (H00-H59), diseases of the ear and mastoid, diseases of the skin and subcutaneous tissue (L00-L99), diseases of the musculoskeletal system and connective tissue (M00-M99), diseases of the genitourinary system (N00-N99), pregnancy, childbirth and the puerperium (O00-O99), certain conditions originating in the perinatal period (P00-P96), congenital anomalies [malformations], deformities and chromosomal disorders (Q00-Q99), symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified (R00-R99).
4 IHME – Institute for Health Metrics and Evaluation (https://www.healthdata.org/)
5 WHO considers the following classes of causes of death as ‘garbage’ codes: symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified (R00-R94, R96-R99); neoplasms (C76, C80, C97, C55); circulatory diseases (I10, I46, I47. 2, I49.0, I50, I51.4, I51.5, I51.6, I51.9, I70.9); mental and behavioural disorders (F19) and, within external causes of death, accidental poisoning by and exposure to noxious substances (X44), and events of undetermined intent (Y10-Y34, Y87.2) [World Health Organization, 2024].
6 See Table 5, Annex, for a more detailed overview of this group of causes of death, which presents the SDR dynamics separately for each class of causes of death for 2011-2021.
7 The concept of ‘garbage’ codes for causes of death was introduced by Naghavi and Lopez in 1990 in the Global Burden of Disease project. A ‘garbage’ cause-of-death code is defined as one that cannot or should not be used as the original cause of death [Murray, Lopez 1996]. The authors then expanded the concept of ‘garbage’ codes and produced their classification [Naghavi, Makela et al. 2010].
login to comment