3
Spatial, Age, and Cause-of-Death Patterns of Mortality in Russia, 1988-1989

Sergei A. Vassin and Christine A. Costello

Introduction

Although the overall mortality level in the former Soviet Union and its republics has been well studied by both Western and Soviet demographers, much less is known about spatial differentials of mortality in these countries. In the European part of Russia (which is relatively homogeneous compared with Russia as a whole), differences in level of life expectancy at birth are still substantial. Even after considerable decline in the range of life expectancy at birth during the 1980s, in 1988 the range was 8.5 years for the rural male population and 4.4 years for the urban male population (Shkolnikov and Vassin, 1994:400-401). If both the European and Asian parts of Russia are considered, the range widens.

The health situation in Russia has been characterized by an unusually long, continuing crisis in adult mortality. Very high death rates from injuries and early cardiovascular disease have had a significant impact on the level of Russian mortality, but should also have a pronounced effect on its age pattern. As shown by Anderson and Silver (in this volume), the age patterns of mortality in Russia and the Baltic states are unusual and different from the widely used neutral West model life table pattern (Coale and Demeny, 1966; Coale et al., 1983). Given the size and diversity of Russia, and in particular the spatial mortality differentials noted above, the questions arise of how many different age patterns of mortality exist in Russia, and how they compare with mortality patterns found elsewhere around the world.

The question of the Russian age pattern of mortality, along with its underlying cause-of-death profile, is relevant in two contexts in this volume: in the use



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 66
--> 3 Spatial, Age, and Cause-of-Death Patterns of Mortality in Russia, 1988-1989 Sergei A. Vassin and Christine A. Costello Introduction Although the overall mortality level in the former Soviet Union and its republics has been well studied by both Western and Soviet demographers, much less is known about spatial differentials of mortality in these countries. In the European part of Russia (which is relatively homogeneous compared with Russia as a whole), differences in level of life expectancy at birth are still substantial. Even after considerable decline in the range of life expectancy at birth during the 1980s, in 1988 the range was 8.5 years for the rural male population and 4.4 years for the urban male population (Shkolnikov and Vassin, 1994:400-401). If both the European and Asian parts of Russia are considered, the range widens. The health situation in Russia has been characterized by an unusually long, continuing crisis in adult mortality. Very high death rates from injuries and early cardiovascular disease have had a significant impact on the level of Russian mortality, but should also have a pronounced effect on its age pattern. As shown by Anderson and Silver (in this volume), the age patterns of mortality in Russia and the Baltic states are unusual and different from the widely used neutral West model life table pattern (Coale and Demeny, 1966; Coale et al., 1983). Given the size and diversity of Russia, and in particular the spatial mortality differentials noted above, the questions arise of how many different age patterns of mortality exist in Russia, and how they compare with mortality patterns found elsewhere around the world. The question of the Russian age pattern of mortality, along with its underlying cause-of-death profile, is relevant in two contexts in this volume: in the use

OCR for page 66
--> of an appropriate standard pattern of mortality to assess data quality and in the use of such a pattern to establish deaths for the purpose of assessing potential years of life lost to premature mortality. Several of the chapters in the first part of this volume address quality-of-data issues. Two of the chapters—those by Kingkade and Arriaga and by Murray and Bobadilla—examine the potential years of life lost to premature mortality. Generally, potential years of life lost is calculated by comparing actual with expected deaths, where expected deaths are based on a standard age pattern of mortality appropriate to a region. To determine health priorities within the different regions of Russia, appropriate standards on which to base expected deaths must be selected, unless an arbitrary age at which all life ceases is chosen. Previous work has investigated levels of life expectancy, spatial differentials, and age-specific components of life expectancy within European Russia (Shkolnikov and Vassin, 1994). This paper expands that spatial analysis to include both the European and Asian parts of Russia, introduces data on cause of death, and examines differentials not only in the level but also the shape (age pattern) of mortality profiles for the years 1988-1989. Establishing the underlying age patterns of Russian mortality at the end of the 1980s is particularly pertinent in the context of long-term change within Russia. The election of Gorbachev to the position of General Secretary occurred in March 1985. Less than two months later (May 7, 1985), a resolution for ''actions against drunkenness and alcoholism" was issued. Three weeks later, the anti-alcohol campaign had begun. This campaign, which lasted to the end of 1987 (see Nemcov, 1995), had the effect of reducing regional differences in mortality as well as the characteristically high excess mortality among middle-aged Russian males. The collapse of the political and economic structure of the Soviet Union occurred over the period 1989-1991. Thus the years 1988-1989 serve as a benchmark for mortality patterns in Russia. Moreover, since the subsequent social upheaval has been suspected of disrupting the state statistical system, it is appropriate to use the years 1988-1989 as baseline data against which to measure future change. In addition, the census of 1989 provides the most reliable age structure by province, and the use of two years of death data improves the reliability of the estimates. Finally, this period preceded the accelerated mortality observed during 1992-1993. The next section of this chapter describes the data sources and methods used for the analysis. This is followed by discussion of the variation in mortality levels in Russia in 1988-1989 by selected causes of death: injuries, cardiovascular disease, and neoplasm. The discussion addresses basic differentials in male-female and rural-urban mortality, regional and provincial variation in mortality levels, and regional variation in cause-specific mortality. Differences in underlying age patterns of mortality are then investigated through the use of cluster analysis for the mathematical grouping of provincial life tables, which results in a set of "typical" age patterns of mortality for males and females for different

OCR for page 66
--> regions of the country. Next, the chapter compares the Russian mortality patterns with Coale-Demeny model life tables, other standard mortality patterns, and other European and U.S. patterns. The chapter ends with a summary and conclusions. Data Sources and Methods The basic data for administrative units of Russia used for this analysis are numbers of deaths in 1988 and 1989 in Russia by age, sex, cause of death, urban or rural location, and administrative unit. The data are from statistical reports of the State Committee of Statistics of the former Soviet Union for provinces, special districts, and autonomous republics (respectively, oblast, krai, and autonomous republics, hereafter referred to as provinces). Population data are from the 1989 census. For the analysis, the entire data set consists of 292 observations: for each sex, 146 observations cover the urban population of 73 provinces, including Moscow and Leningrad cities; the rural population of 71 provinces; and the total urban and rural population of all of Russia. For the analysis of spatial variation in mortality levels, we use life expectancies at birth and cause-specific death rates for three causes of death: injuries, cardiovascular disease, and neoplasm. Death rates are standardized by age to the European standard (Waterhouse et al., 1976) for each of four subpopulations: male urban and rural, and female urban and rural (Table 3-1). The underlying provincial life expectancies and age-standardized cause-specific death rates for three causes of death are included in Annex 3- 1. Percentiles for the life expectancies and cause-specific death rates were calculated and quintiles assigned. Quintiles were used to allow comparisons within a province of the rankings in different causes of death. The lowest quintile, 1, represents a situation of low mortality, while the highest, 5, represents high mortality. Quintiles of life expectancy and cause-specific death rates are also given in the annex. For the analysis of age patterns of mortality, 2-year multiple decrement life tables for 1988-1989 were calculated. These life tables are based on the above data and were constructed by Chiang's (1978) method. 1Thus, the data set consists of 292 life tables, one for each sex and administrative unit. We examine variation in the age-sex profiles by constructing typical profiles through clustering of the provincial mortality profiles. We use a formal approach based on a generalized concept of profile structure developed a number of years ago (Cronbach and Gleser, 1953), which allows the use of cluster analysis to find mortality curves with identical shape (Wunsch, 1984). This approach is discussed more thoroughly in a later section of the chapter. Given the data quality issues addressed by many of the authors in this volume, a word about the reliability of the data used for this analysis is in order. Certainly, provincial-level data are subject to greater error than national estimates. We note some of these problems with certain provinces in the course of

OCR for page 66
--> TABLE 3-1 Rural-Urban Variation in Life Expectancy and Mortality Rates from Selected Causes of Death in Russia, 1988-1989   Life Expectancy at Age Standardized Death Rate per 100,000 Population 0 15 35 60 Neoplasm Cardiovascular Disease Injuries Average of Provincial Levels Male, rural 61.8 49.1 32.3 14.6 289.4 927.3 266.4 Male, urban 64.4 51.3 33.5 14.7 327.8 882.3 191.8 Female, rural 73.4 60.3 41.4 19.7 114.6 604.3 64.4 Female, urban 74.3 60.8 41.6 19.3 149.2 589.9 50.4 Standard Deviation Male, rural 1.8 1.7 1.3 1.0 57.8 122.4 55.6 Male, urban 1.2 1.2 0.9 0.8 44.7 100.3 34.7 Female, rural 1.8 1.7 1.6 1.3 27.9 78.7 22.1 Female, urban 1.2 1.1 1.0 0.9 19.0 64.4 11.7 Coefficient of Variation (%) Male. rural 2.8 3.4 4.2 7.0 20.0 13.2 20.9 Male, urban 1.9 2.3 2.8 5.7 13.6 11.4 18.1 Female, rural 2.5 2.8 3.8 6.4 24.4 13.0 34.3 Female, urban 1.6 1.8 2.5 4.7 12.7 10.9 23.2 NOTE: Standardized death rate determined by direct method of standardization using European standard.

OCR for page 66
--> the discussion. We find extremes in life expectancy levels to be represented by the Northern Caucasus autonomous republics, with high life expectancies, and the more isolated provinces of the Northern, Far Eastern, and Eastern Siberian autonomous republics, with low life expectancies. The age pattern of mortality in these provinces is unusual as compared with other provinces, thus raising the question of whether these unusual patterns are a consequence of unusual life conditions or of doubtful statistics. In general, however, we believe the data for a sufficient number of provinces to be of high enough quality to support the analysis undertaken herein. Variation in Mortality Levels, 1988-1989 This section examines variation in mortality levels in Russia during 19881989, focusing first on female-male and rural-urban differentials, then on regional and provincial variation, and finally on regional variation in cause-specific mortality. Female-Male and Rural-Urban Differentials in Mortality One well-known feature of Russian mortality is the significant difference in life expectancy at birth [e(0)] between males and females. Among rural populations, male life expectancy is 11 years less than female e(0), and among urban populations it is 10 years less, based on the average of provincial life tables (Table 3-1). Much higher death rates among males due to neoplasm, cardiovascular disease, and injuries contribute significantly to these differentials, which are similar across all provinces and administrative units of Russia (see Annex 3-1). Differentials in rural and urban mortality are also marked in Russia, although to a much lesser extent than those between the sexes. Rural males have a life expectancy at birth 2.6 years less than urban males. The difference in remaining life expectancy at higher ages diminishes with increasing age, but does not disappear until age 60. Rural females also have a lower life expectancy than urban females, although on average the difference is less than 1 year. The difference among females diminishes at younger ages, nearly disappearing by about age 35. At younger adult ages, rural mortality is higher, but crossover effects are somewhat evident in higher urban mortality at older adult ages. (Crossover effects are discussed by Anderson and Silver, in this volume.) Overall, higher mortality in rural areas is found consistently in nearly all provinces, particularly for males, for whom only one province shows a reversal of the differential. For females, only nine provinces show higher urban than rural mortality. Among rural males, the absolute range in life expectancy at birth across European and Asian provinces is over 11 years; among urban males, the range is 8 years (Annex 3-1). However, the absolute range misrepresents to a certain extent the spatial differentiation in mortality in Russia. Relative measures of

OCR for page 66
--> variation, such as the standard deviation, demonstrate that differences in level of life expectancy are rather moderate (Table 3-1). The standard deviation in provincial levels of life expectancy at birth for both sexes is 1.2 years in urban areas and 1.8 years in rural areas. Thus, the majority of the population live in areas of fairly similar mortality levels, but variation in life expectancy is greater among the rural than the urban population. Rural areas overwhelmingly and consistently have higher rates of death due to injury than urban areas, by a substantial margin. For males, rural morality rates due to injury are on average 38 percent higher than in urban areas. This differential is particularly marked in several European regions of Russia: the Northern, Northwestern, Central, Volga-Vyatka, Central Blackearth, and Baltic regions. For females, differentials between rural and urban morality rates due to injury are on average smaller—28 percent higher in rural than in urban areas. The differential for females is most marked in the Northwestern, Central, and Volga-Vyatka regions and in part of the Ural region. The differential between rural and urban areas is much less marked for cardiovascular disease than for injury. Generally, mortality rates due to cardiovascular disease are on average 5 and 3 percent higher in rural than in urban areas for males and females, respectively. For males, the differentials are the greatest in the Northern and Northwestern regions and in parts of the Central Region, where rural mortality from cardiovascular disease is 10 to 20 percent higher than urban. In selected provinces of the Volga, North Caucasus, and Ural regions, the differential reverses, with higher urban than rural rates. For females, only a few provinces in high-mortality regions show high differentials. In the lower-mortality regions (Central Blackearth, Volga, and North Caucasus), urban mortality from cardiovascular disease is higher than rural in many provinces. Neoplasm demonstrates the opposite pattern from injuries and cardiovascular disease, showing a larger rural-urban differential among females than among males. In general, urban rates of mortality from neoplasm are nearly always greater than rural rates—on average 12 percent higher for males and 23 percent higher for females. Within each region, there are single provinces where rural mortality from neoplasm is 30 to 40 percent lower than urban. In the Volga-Vatkya region, the differential among males is fairly large and consistent across provinces, with 20 to 40 percent lower mortality due to neoplasm in rural areas. For females, a larger differential is most frequent in provinces of the low-mortality regions, especially the Central Blackearth and Volga regions. In the high-mortality regions of Siberia and the Far East, the differential is small or reversed. Regional and Provincial Variation in Mortality Life expectancies at birth and the relevant quintiles relating to mortality levels of the 1988-1989 provincial life tables are shown for males and females in Annex 3-1 for urban and rural populations. Annex 3-1 also presents age-stan-

OCR for page 66
--> dardized cause-of-death rates and quintiles for injury, cardiovascular disease, and neoplasm within each of the four subpopulations. In general, the Northern and Northwestern regions in the north of European Russia, the northern part of the Ural region, a large part of Siberia, and the Far East include the territories with the lowest life expectancies. The unusually high variation in life expectancy in an industrialized country such as Russia is due to high mortality in these more remote regions of the country. High mortality is particularly evident in the less-populated regions of the Eastern Siberian and Far Eastern regions, in both rural and urban areas, with life expectancies between 55.7 and 63.7 years for males and 65.1 and 73.5 years for females. Certain provinces in Western Siberia and in the northern part of the Ural region show moderately high mortality at quintile 4. In most areas of the Northwestern region, mortality is high for males and moderately high for females. The notable exception is exceedingly low mortality for males, but not females, in the city of St. Petersburg. Also in this general geographic zone is the Northern region, which shows a more moderate level of overall mortality for both sexes. A northeastern to southwestern gradient, moving from higher to lower mortality levels across regions, is evident in the 1988-1989 levels of life expectancy. This gradient is most evident for urban areas and has been described in the literature (Andreev, 1979; Shkolnikov and Vassin, 1994). In general, the high-mortality areas are sparsely populated, but because they are rich in minerals, they are territories of intensive industrial development. (These development areas are classified as urban in the present analysis.) Migrants and prisoners are an important part of the population of these regions. Certainly, the extremely severe climatic conditions, absence of an advanced social infrastructure, and housing shortages that characterize these regions are not attractive to prospective inmigrants. However, the Soviet government stimulated the flow of labor into these areas through a special system of privileges. Although absence of freedom of movement characterized the Soviet Union, migrants who worked for extremely long periods of time in difficult conditions in these areas were granted the privilege to settle in any part of the country. As a rule, those who went to work in the Northern and Asian parts of Russia had particularly good health. Migrants had to be certified by a special state medical commission as being fit to move. Thus, the migration stream into the Northern and Far Eastern regions of the country consisted of younger, healthy in-migrants, while the return stream consisted of older, less healthy, but wealthier migrants moving to the more prosperous regions of the country. Through this exchange, the poor health of the north was spread over the more favored regions of the south. Of course, there was a prevalence of men among the migrants, and thus the composition of the Russian Northern and Far Eastern populations demonstrates significant sex disproportions. The most favorable levels of mortality are found in the North Caucasus

OCR for page 66
--> region, where there are four autonomous republics. These favorable levels are questioned by some, who suggest that they are a product of underregistration of deaths rather than good health conditions, but conclusive work on this topic is not yet in the literature. High levels of life expectancy are also found in the southern areas of the country. The Central Blackearth (Chernozem) region and much of the Volga region have life expectancies between 60.7 and 66.4 years for males and 71.3 and 75.7 years for females. Few areas in these regions have mortality levels at or above the median (with one exception). The Central Blackearth and Volga regions are the main grain regions of Russia, with the best natural and climatic conditions for agriculture (New Russia, 1994). Parts of these regions are frequently called the "granary of Russia," and "Chernozem" soils are considered the embodiment of fruitfulness. Thus, many of the provinces of these rural areas are relatively wealthy and productive. In the remaining regions, the provinces generally exhibit intermediate levels of mortality, but within each region there are pockets of high mortality. The Central region, spread over a large area, shows numerous pockets of moderately high mortality, particularly in rural areas. The rural population in the northern and central areas of European Russia suffers particularly high mortality, largely because of the poor living conditions of the Nechernozem zone. The Nechernozem (meaning poor agricultural conditions) zone includes 23 areas and 6 autonomous republics in the Northern, Northwestern, Central, and Volga-Vyatka economic regions; the Baltic region (Kaliningrad); and the Sverdlovsk, Perm, and Udmurt autonomous republics in the Ural region. Agriculture and living conditions in these territories deteriorated greatly during implementation of the program "Prospective Villages," begun in the 1970s at the initiative of the Central Committee of the Communist Party. Under this program, villages were divided into prospective and nonprospective groups. Prospective villages were favored for infrastructure development and were targeted to attract migrants from the nonprospective villages. However, the majority of the prospective villages were never fully developed. Rural migrants moved from both types of villages to urban areas, rather than to the prospective villages. The population left behind tended to be older and less well off. Thus the nonprospective villages were actually doomed to extinction, and the prospective villages failed to thrive. The Nechernozem zone became synonymous with a "dying" countryside and very poor living conditions of the rural population. The last days of one of these "condemned" villages are chronicled by a contemporary author in the novel Farewell to Matyora (Rasputin, 1991). Regional Variation in Cause-Specific Mortality In the Far Eastern region, cause-specific mortality rates are generally high from injury, cardiovascular disease, and neoplasm. However, rural males do not

OCR for page 66
--> exhibit as extreme rates as the other three subpopulations (urban males and rural and urban females). In Eastern Siberia, one sees fairly high rates2 of injury at quintiles 4 or 5 among males and females in most rural and urban areas. Neoplasm rates are generally high for females and more moderate for males. Low rates of mortality from cardiovascular disease are evident among males in particular, with rates at quintiles 1 or 2 for the male-urban and male-rural populations. Rates are more moderate for the female-urban population; rural females in Eastern Siberia, however, have moderately high levels of cardiovascular disease. Western Siberia has high rates of injury for urban females, and moderate to high neoplasm rates in more than half of the areas for all four subpopulations. In contrast, in three provinces of the Ural region, there are high levels of injury among all four subpopulations, but generally lower rates of cardiovascular disease and neoplasm. In the Northern region, high rates of cardiovascular disease are found for all subpopulations, but injury is not predominant. In the Northwestern area, there are generally moderately high mortality rates from all three causes, with the noted exception among males in Leningrad. In the Central Blackearth and Volga regions, there are generally low rates from injury, cardiovascular disease, and neoplasm, with selected provinces as exceptions. In general, cause-specific mortality rates from injury, cardiovascular disease, and neoplasm vary consistently with the overall level of mortality in four regions, but the relative importance of these causes varies in the remaining regions. Correlations between life expectancy and injury levels are -.63 (urban male) and -.88 (rural male), with rural and urban females at -.74. For cardiovascular disease, the correlations with e(0) are -.45, -.61, -.70, and -.72 for urban and rural males and urban and rural females, respectively. For neoplasm, the equivalent correlations are -.57,-.45, -.62, and -.72, respectively. Correlations between causes of death are also fairly high. The relation between mortality rates from injury and cardiovascular disease is on the order of 0.45 for three of the four subpopulations studied, although for urban males, the correlation between those rates is only 0.16. Yet given the regional variations in mortality patterns documented in this chapter, it is not sufficient to pay particular attention to areas of high mortality and assume a similar underlying cause-of-death structure. Rather, particular causes of death contributing to high mortality in a region need to be identified before regional health issues can be characterized.

OCR for page 66
--> Russian Age Patterns of Mortality General Approach to Classifying of the Shape of the Mortality Curve Many mortality studies have demonstrated that the level is the major component of "explained" variation in mortality curves (Ledermann and Breas, 1959; Bourgeois-Pichat, 1962; Messinger, 1980). Its impact is so considerable that it prevents closer inspection of weaker, but not less informative and important, differences in the shape of mortality curves. In particular, the shape of the mortality curve, or the age pattern of mortality, may reflect specific conditions of life among a population better than does the level. For example, elevated child mortality relative to other ages is a feature of the Coale-Demeny South model life table that reflects increased risk of intestinal infections produced by climatic, sanitary, dietetic, cultural, and behavioral features of the lifestyle of Southern Europeans at the end of the nineteenth and first half of the twentieth centuries (Coale and Demeny, 1966; Coale et al., 1983). Social class groupings are distinguished not only by level, but also by the shape of the mortality curve, which is thus useful for the study of social inequality in mortality (Anson, 1994). The shape of the mortality curve can also reflect peculiarities of the process of the epidemiological transition (Vassin, 1994). The shape of the mortality curve is supposed to be more stable than the level, for even when the level varies considerably, a relative shape is maintained (Valaouras, 1974). This adherence to an underlying shape enables the construction of regional model life tables, and also emphasizes that the shape is more strongly connected to the specific character of a social situation than the level. To analyze mortality profiles by the shape of the curve, it is necessary to eliminate differences in level and to identify the underlying typical patterns. There are different approaches to classifying life tables according to mortality profiles. A major classification effort was carried out by Coale and Demeny (1966; Coale et al., 1983), resulting in the widely used four regional families of model life tables. To find typical mortality patterns, Coale and Demeny visually analyzed several hundred mortality patterns. In this chapter, we employ a more formal approach based on a generalized concept of profile structure developed a number of years ago by Cronbach and Gleser (1953).3 This concept allows the use of cluster analysis to find mortality curves with identical shapes. According to this concept, any profile consists of three components: elevation, scatter, and shape. The level is equivalent to an average of the profile (expressed, e.g., as a simple or geometric average). Scatter is a measure of variation (like variance), whereas shape is something that remains in the profile after the first two components have been removed, similar to the product-moment of correlation between profiles. In demographic practice, the shape of a profile is understood to be all that remains after elimination of differences in level only. We have not departed

OCR for page 66
--> from this usual practice and have accepted the concept of shape as that which incorporates two components—scatter and shape. To be certain that cluster analysis would be reliable for grouping of life tables, we tested the approach on the entire range of Coale-Demeny and U.N. model life tables.4 Variation in the level of mortality in such a sample of model life tables is enormously high (life expectancy varies from 20 to 80 years, with a standard deviation of 18 years and a coefficient of variation of 36 percent). If, in spite of this variation, the method manages to classify all model life tables appropriately into their own families, this means the method is able to ignore differences in the level and to classify mortality curves effectively and properly according to their shape. Results of this test were the correct classification of the entire set of male life tables and the misclassification of only 2 among 279 female life tables, proving that this method is suitable for the classification of life tables by the shape of the mortality curve. Application of Cluster Analysis to Russian Provincial Life Tables To determine whether there were natural clusters of age patterns of mortality in Russia, and the number of such clusters, we first used the same method of classification as that used in the test on model life tables. This searching for natural clusters showed that for males and females, there exist only two natural clusters: one urban and one rural. This means that despite its vast territory, Russia is relatively homogeneous in the shape of its mortality curves and that there are two salient patterns of mortality: rural and urban. However, moderate differences in the shape of Russian regional curves can still be reasonably important within the country. To investigate within-country differences in mortality, we took further steps to break the urban and rural patterns down into more detail by using the Ward method of cluster analysis (Wunsch, 1984). This method has two features that should be noted. First, since the method is vulnerable to outliers, we excluded 15 of the most unusual mortality curves from the analysis to get more reliable results. Second, with this method, the researcher defines an arbitrary final number of clusters. We set the number of clusters equal to six both for males and females. However, as will be shown below, the proper number of clusters is less than six. The typical age patterns of mortality resulting from the cluster analysis are shown graphically in Figures 3-1a for males and 3-1b for females. The cluster profiles shown are the average of all members of each cluster.5 Scores of the double standardized logits of the probability of death are shown as dashed lines on the right y-axis. 6 The deviations of scores specify whether mortality is higher or lower for each cluster relative to the average profile of the whole set of logit scores of q(x). Negative deviations indicate that mortality is below average, and positive, that it is higher. The sum of deviations from the average is equal to zero.7

OCR for page 66
-->   Males - Rural Areas Province e(0) Q Injury Q CVD Q Neopl. Q Northern Region Arhangelsk 60.9 4 273.3 3 1090.6 5 314.6 4 Karelia 61.8 3 251.2 3 1115.4 5 353.4 5 KOMI 60.8 4 294.9 4 1079.4 5 277.5 2 Murmansk Vologda 62.1 3 227.3 1 1057.7 5 290.2 3 Northwestern Region Lenin. Obl. 62.0 3 265.4 3 1031.6 5 351.8 5 Leningrad Novgorod 57.8 5 361.3 5 1116.6 5 345.4 5 Pskov 60.2 5 297.8 4 1058.8 5 292.3 3 Central Region Bryansk 62.1 3 244.3 2 938.8 3 261.7 2 Ivanovo 60.5 5 265.5 3 1126.2 5 320.4 4 Jaroslav 60.1 5 298.8 4 1033.2 5 297.0 3 Kalinin 58.6 5 361.1 5 1144.9 5 299.2 4 Kaluga 59.8 5 289.5 4 1004.7 4 316.7 4 Kostroma 61.2 4 274.2 3 1088.5 5 298.3 4 Moscow Obl. 62.8 2 267.3 3 955.7 4 361.2 5 Moscow Orlovskay 61.5 4 298.8 4 892.4 3 269.1 2 Ryazan 61.1 4 309.9 5 938.7 3 290.3 3 Smolensk 60.6 4 289.6 4 988.0 4 299.2 4 Vladimir 61.6 3 245.0 2 1027.7 4 337.9 5 Volga-Vyatka Region Chuvashia 62.4 2 320.6 5 773.5 1 172.7 1 Gorkovskaya 61.7 3 258.2 3 920.1 3 273.4 2 Kirovskay 61.3 4 304.4 5 893.7 3 253.4 2 Maryiskay 60.7 4 345.5 5 870.0 2 185.1 1 Mordva 63.7 1 204.8 1 902.6 3 234.3 1 Central Blackearth Region Belgorod 63.8 1 228.1 1 841.0 2 219.6 1 Kurskay 61.7 3 265.3 3 946.8 4 254.4 2 Lipezk 62.2 2 270.4 3 938.7 3 286.2 3 Tambov 61.1 4 274.8 4 911.3 3 296.0 3 Voronej 63.3 1 226.1 1 806.5 1 239.9 1

OCR for page 66
-->   Males - Urban Areas Province e(0) Q Injury Q CVD Q Neopl. Q Volga Region Astrahan 63.8 4 210.8 4 924.0 4 351.7 4 Kalmyikia 61.2 5 233.1 5 873.3 3 314.9 2 Kuibyishevsk 64.9 2 169.2 2 855.5 2 347.7 4 Penza 65.2 2 184.1 3 861.8 3 315.4 2 Saratov 64.5 3 175.5 2 917.7 4 333.0 3 Tataria 65.6 1 183.1 3 832.4 2 288.1 1 Ulyanovsk 65.2 2 175.1 2 936.8 4 326.5 3 Volgograd 65.4 1 171.6 2 800.1 1 338.5 3 North Caucasus Region Chechny 64.9 2 125.9 1 820.3 1 249.3 1 Dagestan 68.0 1 110.6 1 583.5 1 212.9 1 Kabarda 65.8 1 136.5 1 766.8 1 242.8 1 Krasnodar 64.9 2 185.6 3 851.5 2 283.7 1 Osetia 66.6 1 144.7 1 821.2 1 228.6 1 Rostov 65.0 2 156.0 1 855.9 2 281.4 1 Stavropol 65.9 1 162.3 2 813.9 1 287.2 1 Ural Region Bashkiria 65.1 2 178.2 2 839.8   280.2 1 Chelyabinsk 64.9 2 183.0 3 781.7 1 345.5 4 Kurganskay 64.4 3 200.7 4 812.0 1 362.8 5 Orenburg 64.9 2 190.7 3 833.6 2 330.1 3 Perm 64.2 4 212.8 5 912.4 4 309.5 2 Sverdlovsk 64.2 4 200.9 4 879.4 3 327.0 3 Udmurtia 64.0 4 223.9 5 885.8 4 269.3 1 Western Siberia Region Altai 63.8 4 221.5 5 792.9 1 361.6 5 Kemerovo 63.1 5 256.4 5 893.5 4 310.2 2 Novosibirsk 64.0 4 182.9 3 873.2 3 342.5 4 Omsk 64.7 3 195.7 3 782.7 1 374.6 5 Tomsk 64.4 3 173.9 2 863.8 3 349.1 4 Tumen 64.9 2 195.9 3 825.7 2 292.6 2 Eastern Siberia Region Buryatia 63.0 5 226.5 5 795.8 1 347.5 4 Chita 63.4 5 229.4 5 788.4 1 296.4 2 Irkutskay 62.8 5 244.6 5 850.2 2 326.8 3 Jakutia 63.0 5 63.5 1 904.5 4 374.5 5 Krasnoyarsk 63.2 5 204.1 4 830.3 2 337.5 3 Tuva 59.8 5 344.9 5 756.5 1 381.9 5

OCR for page 66
-->   Males - Rural Areas Province e(0) Q Injury Q CVD Q Neopl. Q Volga Region Astrahan 63.8 1 209.4 1 940.7 3 346.2 5 Kalmyikia 61.8 3 234.4 2 841.2 2 268.5 2 Kuibyishevsk 62.8 2 241.1 2 903.8 3 282.8 3 Penza 63.0 2 250.6 2 876.9 2 274.2 2 Saratov 62.1 3 239.0 2 907.1 3 322.4 4 Tataria 63.8 1 233.3 2 844.1 2 222.5 1 Ulyanovsk 63.0 2 228.9 1 1008.6 4 275.4 2 Volgograd 63.3 1 232.8 2 813.9 1 327.1 5 North Caucasus Region Chechny 64.6 1 128.9 1 666.1 1 230.8 1 Dagestan 67.1 1 120.0 1 561.5 1 147.4 1 Kabarda 65.5 1 152.0 1 711.6 1 216.8 1 Krasnodar 62.9 2 253.3 3 862.6 2 266.7 2 Osetia 64.1 1 193.1 1 860.0 2 224.0 1 Rostov 63.6 1 216.0 1 844.0 2 256.8 2 Stavropol 63.8 1 207.3 1 882.4 2 260.8 2 Ural Region Bashkiria 63.3 1 249.5 2 816.9 1 218.6 1 Chelyabinsk 62.9 2 217.8 1 832.8 1 333.1 5 Kurganskay 62.6 2 250.3 2 813.0 1 304.8 4 Orenburg 64.5 1 195.3 1 853.1 2 254.9 2 Perm 60.1 5 322.7 5 975.5 4 251.4 1 Sverdlovsk 60.7 4 288.4 4 904.1 3 297.0 3 Udmurtia 61.3 4 325.2 5 809.1 1 221.3 1 Western Siberia Region Altai 62.2 2 253.3 3 825.3 1 297.6 4 Kemerovo 60.4 5 324.0 5 946.4 4 281.7 3 Novosibirsk 62.3 2 235.1 2 885.0 3 298.4 4 Omsk 62.8 2 235.0 2 869.0 2 318.2 4 Tomsk 61.6 3 252.7 3 942.4 4 326.3 4 Tumen 61.9 3 277.0 4 860.0 2 232.2 1 Eastern Siberia region Buryatia 61.5 4 277.1 4 827.3 1 291.4 3 Chita 62.0 3 283.3 4 830.3 1 286.1 3 Irkutskay 59.1 5 332.3 5 880.7 2 295.7 3 Jakutia 61.1 4 263.7 3 799.1 1 379.8 5 Krasnoyarsk 60.5 5 287.2 4 853.3 2 281.7 3 Tuva 55.7 5 455.8 5 951.7 4 357.7 5

OCR for page 66
-->   Males - Urban Areas Province e(0) Q Injury Q CVD Q Neopl. Q Far Eastern Region Amurskay 63.7 4 216.6 5 889.3 4 311.6 2 Habarovsk 62.2 5 233.2 5 1010.3 5 370.7 5 Kamchatka 62.7 5 218.6 5 1376.2 5 433.8 5 Magadan 63.1 5 157.6 1 1119.1 5 496.0 5 Primorski 63.4 5 227.5 5 941.6 5 332.5 3 Sahalin 62.5 5 233.9 5 1072.2 5 377.5 5 Baltic Region Kaliningrad 65.3 1 192.9 3 876.5 3 346.8 4 Summary Statistics Mean 62.7   183.1   870.2   323.3   Std Deviation 10.5   48.8   142.5   58.2  

OCR for page 66
-->   Males - Rural Areas Province e(0) Q Injury Q CVD Q Neopl. Q Far Eastern Region Amurskay 62.1 3 239.3 2 954.8 4 271.4 2 Habarovsk 60.1 5 293.6 4 1009.8 4 349.4 5 Kamchatka 60.2 5 249.8 2 1304.9 5 323.6 4 Magadan 60.9 4 403.1 5 1105.2 5 552.0 5 Primorski 60.9 4 287.7 4 1001.7 4 293.1 3 Sahalin 61.9 3 276.8 4 1124.8 5 369.0 5 Baltic Region Kaliningrad 59.4 5 352.3 5 990.0 4 369.2 5 Summary Statistics Mean 61.8   266.4   927.3   289.4   Std Deviation 1.8   55.2   121.5   57.3  

OCR for page 66
-->   Females - Urban Areas Province e(0) Q Injury Q CVD Q Neopl. Q Northern Region Arhangelsk 74.7 3 45.9 3 619.3 4 145.5 3 Karelia 74.0 4 52.6 3 667.6 5 151.6 4 KOMI 73.1 5 62.5 5 667.5 5 151.9 4 Murmansk 74.6 3 41.6 2 656.6 5 140.2 2 Vologda 74.6 3 40.4 2 625.9 5 139.9 2 Northwestern Region Lenin. Obl. 73 .8 4 61.0 5 636.3 5 173 .4 5 Leningrad 74.1 4 53.2 3 577.7 3 197.2 5 Novgorod 74.1 4 53.9 4 613.5 4 146.2 3 Pskov 74.2 4 47.2 3 632.5 5 154.4 4 Central Region Bryansk 75.3 1 35.8 1 586.1 3 141.6 2 Ivanovo 74.1 4 41.6 2 656.6 5 139.2 2 Jaroslav 75.0 2 20.3 1 579.3 3 147.6 3 Kalinin 74.7 3 48.8 3 598.8 4 138.6 2 Kaluga 74.8 2 38.0 1 577.5 3 150.7 4 Kostroma 74.4 3 47.4 3 637.1 5 147.8 3 Moscow Obl. 74.6 3 43.2 2 594.7 4 165.4 5 Moscow 74.2 4 43.7 2 556.1 2 188.6 5 Orlovskay 75.4 1 45.3 3 533.1 1 140.4 2 Ryazan 75.5 1 39.1 1 529.1 1 145.4 3 Smolensk 75.0 2 39.4 2 565.9 2 156.7 4 Vladimir 75.0 2 37.2 1 597.4 4 142.3 2 Volga-Vyatka Region Chuvashia 75.6 1 60.6 5 514.2 1 121.2 1 Gorkovskaya 74.8 2 39.7 2 587.6 3 149.3 3 Kirovskay 74.7 3 57.5 4 598.5 4 115.0 1 Maryiskay 75.1 2 56.7 4 529.6 1 124.4 1 Mordva 75.7 1 39.2 1 541.7 1 130.5 1 Central Blackearth Region Belgorod 74.9 2 36.5 1 557.9 2 143.3 2 Kurskay 74.9 2 41.4 2 569.9 2 135.1 1 Lipezk 75.4 1 37.8 1 564.6 2 143.7 2 Tambov 74.9 2 38.6 1 556.9 2 147.9 3 Voronej 75.7 1 20.4 1 529.0 1 128.4 1

OCR for page 66
-->   Females - Rural Areas Province e(0) Q Injury Q CVD Q Neopl. Q Northern Region Arhangelsk 73.5 3 59.1 3 661.7 5 106.0 3 Karelia 73.2 4 54.5 2 763.9 5 115.7 3 KOMI 72.1 4 75.9 4 738.0 5 94.2 1 Murmansk 73.7 3 51.2 2 762.6 5 115.7 3 Vologda 74.5 2 48.7 1 627.6 4 100.4 2 Northwestern Region Lenin. Obl. 74.0 3 69.7 4 595.1 3 136.1 5 Leningrad Novgorod 72.9 4 79.1 4 637.2 4 114.0 3 Pskov 72.5 4 81.0 5 642.3 4 116.6 4 Central Region Bryansk 74.7 1 48.3 1 569.3 2 91.5 1 Ivanovo 73.5 3 53.9 2 670.9 5 106.0 3 Jaroslav 74.3 2 61.8 3 581.4 3 112.7 3 Kalinin 72.4 4 82.5 5 665.2 5 104.5 2 Kaluga 72.7 4 62.0 4 632.9 4 118.9 4 Kostroma 73.4 4 54.9 2 686.8 5 111.2 3 Moscow Obl. 74.4 2 54.9 2 612.1 3 144.3 5 Moscow Orlovskay 74.1 3 60.7 3 572.8 2 99.0 2 Ryazan 74.3 2 61.1 3 563.2 2 105.8 2 Smolensk 73.4 4 64.5 4 616.6 4 115.9 4 Vladimir 73.6 3 56.6 3 639.9 4 119.4 4 Volga-Vyatka Region Chuvashia 73.1 4 115.7 5 549.6 1 76.2 1 Gorkovskaya 74.6 2 50.0 2 570.2 2 102.5 2 Kirovskay 73.3 4 90.8 5 562.4 2 92.9 1 Maryiskay 71.3 5 133.3 5 612.8 4 80.8 1 Mordva 75.5 1 49.6 2 538.5 1 82.91   Central Blackearth Region Belgorod 75.7 1 42.7 1 537.1 1 84.2 1 Kurskay 74.3 2 54.2 2 587.7 3 95.4 2 Lipezk 75.1 1 60.9 3 550.8 1 92.4 1 Tambov 74.6 2 51.3 2 556.2 2 104.8 2 Voronej 75.0 1 42.4 1 527.5 1 96.4 2

OCR for page 66
-->   Females - Urban Areas Province e(0) Q Injury Q CVD Q Neopl. Q Volga Region Astrahan 74.4 3 48.2 3 593.0 3 158.8 4 Kalmyikia 71.3 5 61.1 5 597.4 4 137.4 2 Kuibyishevsk 74.6 3 44.2 2 569.8 2 158.4 4 Penza 75.5 1 45.3 3 561.2 2 134.6 1 Saratov 74.6 3 43.6 2 602.8 4 149.0 3 Tataria 75.6 1 45.4 3 534.3 1 128.8 1 Ulyanovsk 75.1 2 44.2 2 572.3 3 138.3 2 Volgograd 75.1 2 43.0 2 544.6 1 162.4 5 North Caucasus Region Chechny 74.1 4 36.8 1 551.6 2 135.5 1 Dagestan 77.5 1 31.0 1 383.6 1 104.9 1 Kabarda 76.1 1 35.7 1 494.3 1 125.2 1 Krasnodar 74.4 3 47.7 3 595.6 4 147.9 3 Osetia 75.9 1 36.3 1 531.4 1 133.0 1 Rostov 74.2 4 41.2 2 612.3 4 145.5 3 Stavropol 75.3 1 37.5 1 557.8 2 147.0 3 Ural Region Bashkiria 74.8 2 51.8 3 546.9 1 130.0 1 Chelyabinsk 74.6 3 51.3 3 544.0 1 152.4 4 Kurganskay 74.9 2 54.1 4 533.8 1 152.9 4 Orenburg 74.8 2 42.3 2 575.8 3 146.1 3 Perm 73.9 4 62.1 5 611.0 4 136.5 2 Sverdlovsk 74.1 4 58.1 4 604.1 4 143.8 3 Udmurtia 74.4 3 63.3 5 587.5 3 119.1 1 Western Siberia Region Altai 74.2 4 62.8 5 559.8 2 162.3 5 Kemerovo 73.2 5 78.1 5 620.7 5 142.1 2 Novosibirsk 74.0 4 53.9 4 577.0 3 154.9 4 Omsk 74.8 2 56.4 4 518.9 1 177.1 5 Tomsk 73.7 4 57.9 4 572.9 3 168.5 5 Tumen 74.6 3 57.9 4 582.4 3 130.1 1 Eastern Siberia Region Buryatia 73.5 5 54.6 4 561.5 2 166.9 5 Chita 73.4 5 54.8 4 580.1 3 149.9 3 Irkutskay 73.2 5 63.6 5 584.4 3 164.4 5 Jakutia 72.2 5 45.3 3 619.7 4 183.9 5 Krasnoyarsk 73.5 5 55.9 4 570.8 2 158.2 4 Tuva 70.0 5 96.8 5 603.6 4 193.7 5

OCR for page 66
-->   Females - Rural Areas Province e(0) Q Injury Q CVD Q Neopl. Q Volga Region Astrahan 73.9 3 56.0 3 602.7 3 135.2 5 Kalmyikia 72.6 4 49.2 1 581.2 2 103.7 2 Kuibyishevsk 74.3 2 55.9 2 589.3 3 108.5 3 Penza 75.2 1 50.8 2 536.5 1 93.0 1 Saratov 74.5 2 51.6 2 559.9 2 111.4 3 Tataria 75.8 1 46.7 1 504.0 1 85.0 1 Ulyanovsk 74.3 2 53.1 2 616.7 4 102.1 2 Volgograd 74.3 2 52.8 2 562.5 2 127.4 5 North Caucasus Region Chechny 75.5 1 27.8 1 426.5 1 94.8 2 Dagestan 76.3 1 26.8 1 355.8 1 58.5 1 Kabarda 76.1 1 29.1 1 469.4 1 93.5 1 Krasnodar 74.2 2 56.1 3 590.5 3 122.0 4 Osetia 76.2 1 30.9 1 502.7 1 113.4 3 Rostov 74.8 1 45.4 1 580.7 2 107.2 3 Stavropol 74.1 3 46.1 1 603.3 3 122.4 4 Ural Region Bashkiria 75.0 1 56.4 3 522.9 1 83.5 1 Chelyabinsk 73.7 3 57.6 3 543.0 1 125.0 4 Kurganskay 74.5 2 59.8 3 540.8 1 122.0 4 Orenburg 74.9 1 49.0 1 554.8 2 101.1 2 Perm 71.7 5 88.7 5 679.1 5 103.1 2 Sverdlovsk 72.6 4 84.1 5 600.1 3 116.1 4 Udmurtia 73.4 4 87.4 5 594.9 3 85.0 1 Western Siberia Region Altai 73.4 4 63.9 4 575.2 2 124.4 4 Kemerovo 71.8 5 99.4 5 644.6 4 109.4 3 Novosibirsk 73.8 3 58.8 3 580.0 2 113.0 3 Omsk 73.3 4 56.1 3 604.8 3 125.2 4 Tomsk 71.9 5 67.2 4 659.4 5 142.2 5 Tumen 73.7 3 75.3 4 569.7 2 98.7 2 Eastern Siberia Region Buryatia 71.4 5 69.0 4 614.3 4 151.1 5 Chita 71.3 5 63.3 4 642.6 4 159.7 5 Irkutskay 71.2 5 79.7 4 623.6 4 127.1 5 Jakutia 69.6 5 47.4 1 627.8 4 229.3 5 Krasnoyarsk 72.3 4 76.2 4 584.6 3 115.6 3 Tuva 65.1 5 138.4 5 764.5 5 212.2 5

OCR for page 66
-->   Females - Urban Areas Province e(0) Q Injury Q CVD Q Neopl. Q Far Eastern Region Amurskay 73.5 5 54.8 4 657.6 5 139.8 2 Habarovsk 72.8 5 57.5 4 681.6 5 161.1 5 Kamchatka 71.9 5 66.1 5 850.3 5 166.5 5 Magadan 71.6 5 70.5 5 784.6 5 216.0 5 Primorski 73.2 5 64.1 5 671.6 5 157.2 4 Sahalin 72.4 5 64.2 5 755.5 5 156.5 4 Baltic Region Kaliningrad 74. 3 3 62.4 5 557.9 2 160.8 4 Summary Statistics Mean 74.3   49.6   589.9   149.2   Std Deviation 1.2   12.4   63.9   18.9  

OCR for page 66
-->   Females - Rural Areas Province e(0) Q Injury Q CVD Q Neopl. Q Far Eastern Region Amurskay 71.3 5 70.5 4 718.9 5 116.8 4 Habarovsk 70.8 5 73.8 4 763.5 5 151.1 5 Kamchatka 69.8 5 83.7 5 818.0 5 186.3 5 Magadan Primorski 71.7 5 83.7 5 719.5 5 125.6 4 Sahalin 72.2 4 80.6 5 638.4 4 135.5 5 Baltic Region Kaliningrad 71.3 5 122.2 5 596.7 3 131.7 5 Summary Statistics Mean 73.4   64.4   604.3   114.6   Std Deviation 1.8   21.9   78.2   27.7