Prosperity = death?
This week's epidemiology abstract.
While mortality has decreased over the past century, superimposed on that decline have been occasional spikes. This study endeavored to confirm this observation by correlating economic with mortality indices in a statistical way. I haven't read the article yet, but the abstract piques my interest.
Increasing mortality during the expansions of the US economy, 1900–1996
Author: Granados, José A Tapia
Source: International Journal of Epidemiology, Volume 34, Number 6, December 2005, pp. 1194-1202(9)
Background In Western countries mortality dropped throughout the 20th century, but over and above the long-term falling trend, the death rate has oscillated over time. It has been postulated that these short-term oscillations may be related to changes in the economy.
Methods To ascertain if these short-term oscillations are related to fluctuations in the economy, age-adjusted total mortality and mortality for specific population groups, ages and causes of death were transformed into rate of change or percentage deviation from trend, and were correlated and regressed on indicators of the US economy during the 20th century, transformed in the same way.
Results Statistically and demographically significant results show that the decline of total mortality and mortality for different groups, ages and causes accelerated during recessions and was reduced or even reversed during periods of economic expansion—with the exception of suicides which increase during recessions. In recent decades these effects are stronger for women and non-whites.
Conclusions Economic expansions are associated with increasing mortality. Suggested pathways to explain this deceleration or even reversal of the secular decline in mortality during economic expansions include both material and psychosocial effects of the economic upturns: expansion of traffic and industrial activity directly raising injury-related mortality, decreased immunity levels (owing to rising stress and reduction of sleep time, social interaction and social support), and increased consumption of tobacco, alcohol and saturated fats.
Keywords: Economic conditions; macroeconomic factors; economic recession; unemployment; mortality determinants; age-specific death rates; causes of death; demography
Showing posts with label causes of death. Show all posts
Showing posts with label causes of death. Show all posts
1/26/06
9/21/04
Pollution and cancer in China
The recent article in the Times is harrowing, starting from its title. But the scientific questions it raises are just as important as the sympathy and disquiet it evokes. There are two issues here. First, what is the relationship between the recent increased pollution in rural China and cancer rates (or death rates due to cancer) in those regions? Secondly, apart from the epidemiologic facts, why are the people interviewed in the article so sure that they have cancer?
A brief, incomplete survey of the epidemiologic literature (oh, let me come clean here: I just went to Medline, typed in "China+epidemiology+pollution," and read a bunch of abstracts. It's Google for epidemiologists!) indicates both rather more and less than is pointed to in the Times. The rural residents complain of gastric and other GI cancers, and the available literature (much of which is, not surprisingly, available only in translated abstracts of Chinese-language articles) points to increased rates of esophageal cancer after exposure to polluted water. (I should note that the tricky part of epidemiology is figuring out the contribution of many different factors. "Correlation does not imply causation" might be tattooed on the body of every epidemiologist.)But there are also a number of articles which point to a broader array of deleterious effects, from nasopharyngeal cancer to hepatitis B infection (which can lead to primary liver cancer). A few studies, as well as the Times article, mention what would seem to be a more acute endpoint: frank toxicity due to astronomical levels of various toxins, effluents, and general nasty chemicals in well- and riverwater.
The Times article implies that the rise in cancer rates due to pollution is a recent phenomenon, of the past couple of decades. I think this is true in a sense, but the sense was corrupted due to necessary abridgment for journalistic purposes. The course of events probably goes something like this: over the past few decades, maybe since the economic liberalization that was allowed during the 70s and 80s (my Chinese history is weak, but I think this is right), industrial activity has produced and allowed widespread pollution. This pollution has indeed led to increased cancer among rural Chinese -- but only because rural Chinese are living longer (due to improved nutrition and hygiene) and are thus able to develop cancer during their additional lifespan. In these polluted rural areas, it's only over the past decade or so that these cancer cases have reached a critical mass that can't be ignored.
Another possibility is that these diseases among rural Chinese, though probably to be laid at the feet of polluters (and the Chinese public-health establishment, necessarily weakened by totalitarianism -- cf. the recent SARS mess), are also caused by simple toxicity due to the ingestion of a high concentration of pollutants.
This is just rank speculation, of course, but it's speculation that I would bet Chinese public-health workers and epidemiologists have to engage in, to decide whether cancer is, in fact, pandemic in rural China and what to do about it (besides, obviously, cracking down on pollution).
* * *
It's not clear from the article in the Times that the people interviewed do in fact have cancer. (They are horribly ill, but it's not apparent from what.) I think that in some cases they might be extrapolating from changing causes of death among people they know. As the Chinese get richer, their causes of death may shift from infectious diseases (with which many rural Chinese may already be acquainted) to the so-called "diseases of affluence" -- heart disease, stroke, and cancer. As it turns out, the Chinese Red Cross recently released a study of widespread ill-health among urban Chinese. [Strangely enough, the study isn't mentioned on the CRC's Web site. Maybe it's on the Chinese site -- can anyone tell me?]
* * *
China's the most populous country in the world, with huge populations that are urban, rural, urbanizing, or migrating from city to country. I hope China deals with its public health problems in a way that will serve as a model for other systems.
The recent article in the Times is harrowing, starting from its title. But the scientific questions it raises are just as important as the sympathy and disquiet it evokes. There are two issues here. First, what is the relationship between the recent increased pollution in rural China and cancer rates (or death rates due to cancer) in those regions? Secondly, apart from the epidemiologic facts, why are the people interviewed in the article so sure that they have cancer?
A brief, incomplete survey of the epidemiologic literature (oh, let me come clean here: I just went to Medline, typed in "China+epidemiology+pollution," and read a bunch of abstracts. It's Google for epidemiologists!) indicates both rather more and less than is pointed to in the Times. The rural residents complain of gastric and other GI cancers, and the available literature (much of which is, not surprisingly, available only in translated abstracts of Chinese-language articles) points to increased rates of esophageal cancer after exposure to polluted water. (I should note that the tricky part of epidemiology is figuring out the contribution of many different factors. "Correlation does not imply causation" might be tattooed on the body of every epidemiologist.)But there are also a number of articles which point to a broader array of deleterious effects, from nasopharyngeal cancer to hepatitis B infection (which can lead to primary liver cancer). A few studies, as well as the Times article, mention what would seem to be a more acute endpoint: frank toxicity due to astronomical levels of various toxins, effluents, and general nasty chemicals in well- and riverwater.
The Times article implies that the rise in cancer rates due to pollution is a recent phenomenon, of the past couple of decades. I think this is true in a sense, but the sense was corrupted due to necessary abridgment for journalistic purposes. The course of events probably goes something like this: over the past few decades, maybe since the economic liberalization that was allowed during the 70s and 80s (my Chinese history is weak, but I think this is right), industrial activity has produced and allowed widespread pollution. This pollution has indeed led to increased cancer among rural Chinese -- but only because rural Chinese are living longer (due to improved nutrition and hygiene) and are thus able to develop cancer during their additional lifespan. In these polluted rural areas, it's only over the past decade or so that these cancer cases have reached a critical mass that can't be ignored.
Another possibility is that these diseases among rural Chinese, though probably to be laid at the feet of polluters (and the Chinese public-health establishment, necessarily weakened by totalitarianism -- cf. the recent SARS mess), are also caused by simple toxicity due to the ingestion of a high concentration of pollutants.
This is just rank speculation, of course, but it's speculation that I would bet Chinese public-health workers and epidemiologists have to engage in, to decide whether cancer is, in fact, pandemic in rural China and what to do about it (besides, obviously, cracking down on pollution).
* * *
It's not clear from the article in the Times that the people interviewed do in fact have cancer. (They are horribly ill, but it's not apparent from what.) I think that in some cases they might be extrapolating from changing causes of death among people they know. As the Chinese get richer, their causes of death may shift from infectious diseases (with which many rural Chinese may already be acquainted) to the so-called "diseases of affluence" -- heart disease, stroke, and cancer. As it turns out, the Chinese Red Cross recently released a study of widespread ill-health among urban Chinese. [Strangely enough, the study isn't mentioned on the CRC's Web site. Maybe it's on the Chinese site -- can anyone tell me?]
* * *
China's the most populous country in the world, with huge populations that are urban, rural, urbanizing, or migrating from city to country. I hope China deals with its public health problems in a way that will serve as a model for other systems.
7/21/04
Life and death by the numbers
If you have a basic fascination about the causes of disease and their distribution -- go get yourself a Ph.D. in epidemiology, and we can commiserate! Failing that, you can play around with the sortable statistics of the New York City Community Health Profiles.
I'm not actively involved in researching these issues (though I'd like to work for the Department of Health someday). Nevertheless, here are some observations I don't understand.
Highbridge and Morrisania (in the Bronx) has the dubious honor of the highest all-cause mortality rate of any city neighborhood. Mortality from heart disease is highest in the Rockaways, but stroke mortality is highest in East Harlem -- and each of these is found pretty far down the list for the other cause of death. Why is this?
Chelsea and Hell's Kitchen, sorry, I mean Clinton, in Manhattan, seem to have a higher infant morality rate than other neighborhoods in Manhattan (although there are no confidence intervals on the rates presented, so it's hard to tell). The infant hospitalization rate is also higher than many other boroughs. Again, why is this?
(Part of the answer is certainly the fact that Chelsea/Hell's Kitchen* is poorer than most other Manhattan neighborhoods, although how much longer this will stay the case is open to question.)
I could go on like this for hours. These general issues are more and more classified under what's called social epidemiology, an attempt to elucidate and interrelate the different strata of causal factors that affect disease in the individual and society. (The latest issue of Epidemiologic Reviews is devoted to this subfield, and the rigor of some articles made me feel somehwat less skeptical toward social epidemiology in general.)
*Yes, I could write "Chelsea/Clinton," but I won't. So there.
If you have a basic fascination about the causes of disease and their distribution -- go get yourself a Ph.D. in epidemiology, and we can commiserate! Failing that, you can play around with the sortable statistics of the New York City Community Health Profiles.
I'm not actively involved in researching these issues (though I'd like to work for the Department of Health someday). Nevertheless, here are some observations I don't understand.
Highbridge and Morrisania (in the Bronx) has the dubious honor of the highest all-cause mortality rate of any city neighborhood. Mortality from heart disease is highest in the Rockaways, but stroke mortality is highest in East Harlem -- and each of these is found pretty far down the list for the other cause of death. Why is this?
Chelsea and Hell's Kitchen, sorry, I mean Clinton, in Manhattan, seem to have a higher infant morality rate than other neighborhoods in Manhattan (although there are no confidence intervals on the rates presented, so it's hard to tell). The infant hospitalization rate is also higher than many other boroughs. Again, why is this?
(Part of the answer is certainly the fact that Chelsea/Hell's Kitchen* is poorer than most other Manhattan neighborhoods, although how much longer this will stay the case is open to question.)
I could go on like this for hours. These general issues are more and more classified under what's called social epidemiology, an attempt to elucidate and interrelate the different strata of causal factors that affect disease in the individual and society. (The latest issue of Epidemiologic Reviews is devoted to this subfield, and the rigor of some articles made me feel somehwat less skeptical toward social epidemiology in general.)
*Yes, I could write "Chelsea/Clinton," but I won't. So there.
Subscribe to:
Posts (Atom)
