Economic change, crime, and mortality crisis in Russia: regional analysis
BMJ 1998; 317 doi: https://doi.org/10.1136/bmj.317.7154.312 (Published 01 August 1998) Cite this as: BMJ 1998;317:312
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
All these articles are based on the official USSR/Russia mortality statistics. Authors of the JAMA article 4 re-coded Russian codes of death causes to ICD-9 codes, and used as 'other alcohol-related causes' codes 303 (alcohol dependence), 305 (non-dependent use of alcohol, it is important if the idea about 'devil drinking' is correct), and E860 (accidental alcohol poisoning). Authors of recent BMJ article 1 included in this group also code 291 (alcohol psychoses) and combine this group with all cirrhoses (alcohol related and not alcohol related, as well as all diseases of the pancreas, codes 571.0-571.3, 571.5-571.6, 577). In the same time, the code 305 was connected to group of 'other' death causes 1, which may lead to serious underestimation of alcohol related death.
Analysis by cause of death in 1 may be misleading because it was done only for selected causes of death: 'Other' do not include all diseases not mentioned in specific groups and 'Total' includes only sum of selected groups of diseases, not total mortality. In addition to selective non-inclusion of non-dependent alcohol consumption some causes of death are not included (e.g. heart diseases codes 390-398, hypertension, 401-409). This omission of other death causes may lead to overestimation of alcohol related death causes.
allberg et al. are correct stating that data they use may be biased, but it is wrong to explain distortions by 'culture of distortion of certain statistics at local level during the Soviet period'. It was not culture, but policy and not a local one, but directed by State. It is well known for many aspects of Soviet times' life including health statistics 6. Having in mind possible 'correction' of health statistics during Soviet times it is possible to think that at least part of the increase in the frequency of alcohol related death may be explained by more liberal use of these death codes during 1990-1994 period. Experience with regional statistics in Russia show that it is impossible to use official death causes statistics for any serious analysis today as well as from Soviet times. Omissions and biases of health statistics were well described in relation to cancer statistics 6.
The reason to state that alcohol related diseases are most important cause of increase of mortality, is the largest percent increase of 'other alcohol related causes' between 1990 and 1994 - 266.4% (both sexes), while all causes increase is only 32.6% 4. It is wrong conclusion, because this largest percent increase is not responsible for the size of increase in mortality. For men general mortality increase is 602.2 cases per 100 000 and alcohol related causes input is only 53.4, while only diseases of the heart (codes 390-398, 401-429) give 146.4.
Another argument used by authors of all these articles is some parallelism of changes of mortality and alcohol consumption. Authors of the last article 1 reasonably refuse to use this argument because of weakness of the alcohol consumption statistics. One more argument is based on the time trends - it is the coincidence of the mortality decline in USSR with the Gorbachev's times' anti-alcohol campaign. It is completely wrong argument, because anti-alcohol campaign had place only in the USSR, while decrease of mortality had place in all former Socialist countries during yearly years of reforms, as well as subsequent increase.
Last thing worth to comment is that 'ecological' analysis of USSR' data by Wallberg et al. 1 is not first one. Parallels of changes of mortality and transition impact in former republics of the USSR were described In book published in 1997 7. Author of this book explained the increase of mortality by stress of disappointment with reforms. He did not find the direct mechanism of the realization of the stress, but demonstrated that changes in income, alcohol consumption, nutrition, smoking, or health care are not responsible for large part of mortality changes. Data by 1 support the idea that some 'stress of reforms' is responsible for increase in mortality, but it is precocious to conclude that alcohol drinking is leading 'cause'.
While we can only use inconclusive correlation analysis (including time trend and ecological comparisons), it is important to abstain from simple and early explanations. The explanation of increase of mortality in former Soviet countries by devil drinking in Russia is not supported by data. After 1994 death rate is decreasing. This decrease in mortality is developing without any successes in reforms (not in democratization, nor in economic development) and no data says that alcohol consumption decreased. It is another argument against connection of mortality increase to alcohol drinking.
V. Vlassov, Dr. Med. Sci.
Saratov, Russia, vvvla@sgu.ssu.runnet.ru
Reference List
1. Walberg P, McKee M, Shkolnikov VM, Chenet L, Leon DA. Economic change, crime, and mortality crisis in Russia: regional analysis. Br.Med.J. 1998;317:312-318.
2. Leon DA, Shkolnikov VM, Chenet L. Huge variation in Russian mortality rates 1984-1994: artifact or alcohol or what? Lancet 1997;350:383-388.
3. Lincoln T. Death and the demon drinking in Russia. Nature 1997;388:723-723.
4. Notzon FC, Komarov YM, Ermakov SP, Sempos CT, Marks JS, Sempos EV. Causes of declining life expectancy in Russia. J.A.M.A. 1998;279,#10:793-800.
5. Leon DA, Shkolnikov VM. Social stress and the Russian mortality crisis. J.A.M.A. 1998;279,#10:790-791.
6. Rahu M. Cancer epidemiology in the former Soviet Union. Epidemiology 1992;3.- # 5:464-470.
7. Gundarov IA. Paradoxes of Russian Reforms [Russian]. Moscow: URSS, 1997;158 p.
Competing interests: No competing interests
Critical comments on Walberg et al's study
Editor - Walberg et al's study of the regional variations in the
health indicators during social and economical transition in Russia (1990
-1994) [ 1 ] merit some critical comments. Analysis of the regional
diversity during transitional period can not be full
without considerations of the basic regional variations existed
before. It must be stress that some Russian's regions have had a
distinctive starting position and conditions during transitional pacing,
what requires more closely social homogenity. Thus athors have included
in analysis two largest Russian cities Moscow and St.Petersburg,
which have administrative status of the region ( all other regions have
both urban and rural population ). Standard of living in these cities
is significantly higher than middle for the contry, what connected with
centralisation of political power and
economical resourcies. Thus in 1994 per capita income and doctors rate
(per 10.000 of population) in Moscow was 691.1 and 75.5,
respectively, versus 206.3 and 45.1 for all Federation. Amongs other
distinctive ("deviant") regions may be considered Moscow region, which
composed the metropolitan area with Moscow city, and Murmansk
region, moustly advanced to the north, with the
higherst rate of urban population in the contry ( 92.4 % versus 73.0
%). Analysis of regional statistic published by the Russian state
statistical committee (Goskomstat) have demonstated that regions with
the higherst mortality rate from the violent causes and crimes before
1990 have a lagest incriase of both indices
during transition ( Table 1 ). Walberg et al's regional gradation by
change in male life expectancy [ 1 ] was used for comparasion of two polar
group of regions: First, with the smallest decrease in life expectancy
( 13 regions ), and, Second, with the largest decrease ( 9 regions,
except Moscow and St.Petersburg cities and Moscow and Murmansk
regions). Regional differencies in mortality
and crime registered for extended period suggests existence of some
background, closely connected with behavioral component, which,
possibly, have influence on the pace of changes in the health status
and reflects the specificity of the regional "reactivity."
There are noticeble south to north gradient in the mortality
variations in Russia. Thus the 8 from the 9 regions with the largest
decrease in life expectancy were located to the north of latitude 56
degrees, and all 13 regions with the smallest decrease were situated
to the south of latitude 56 degrees. South-north mortality
variations are typical for all European continent, what allows to
consider enviromental [ 2 ], or genetical [ 3 ] factors mostly
responseable of a such variations. The latest studies of the East
European gene pool have demonstrated a certain association of
geographic variation in the gene pool with the latitudinal zonality
in disease incidence [ 4 ]. It is possible, that genetical ( or
genetical-enviromental ) gradient on the European mortality map may have
direct and indirect manifestation ( e.g. through behavioral component
).
Table 1. Comparison of the 13 Russian regions with smallest decrease
in male life expectancy and 9 regions with largest decrease
regions regions largest
with the region with the region p mean /
smallest SD largest SD smallest
decrease, decrease, mean
mean mean (ratio)
---------------------------------------------------------------------
1980
accidents* 144.7 31.4 213.0 15.5 .0001 1.47
1985
accidents 126.5 19.1 186.2 27.9 <_.0001 _1.47="_1.47" p="p"/> crimes** 860.7 176.3 1140 211.8 .004 1.32
1990
accidents 130.1 14.5 150.1 13.8 .004 1.15
crimes 1042 197.7 1358 209.3 .002 1.3
1994
accidents 193.0 27.0 334.5 30.5 <_.0001 _1.73="_1.73" p="p"/> crimes 1423 243.9 2090 316.8 .0001 1.47
---------------------------------------------------------------------
--
accidents* - death rate from accidents, traumas and poisoning
(per 100.000 of population),
crimes** - rate of registered crimes per 10.000 of population
References
1. Walberg P, McKee M, Shkolnikov V, Chenet L, Leon D. Economic
change, crime, and mortality crisis in Russia: regional analysis.
BMJ 1998; 317: 312-8.
2. Grimes DS, Hindle E, Dyer T. Albanian paradox, another example of
protective effect of Mediterranean lifestyle? (letter) Lancet 1998; 351:
835-6.
3. Kondrichin S. Albanian paradox, another example of protective
effect of Mediterranean lifestyle? (letter) Lancet 1998;351:836.
4. Rychkov Yu G, Zhukova OV, Ogryzko EV, Shneider Yu V. East
European gene pool and diseases in rural population of European Russia.(in
Russian) Genetika 1988;34:1138-1150.
Competing interests: ---------------------------------------------------------------------