Chapter 3: The World Scene

As I indicated in the last chapter, neither I nor anybody else can prove a negative. Therefore, I'm not going to try to prove that smoking does not cause hangnails, or heart disease or anything else. The burden of proof rests on those who assert that there is, in fact, a smoking/disease connection. The connection most often alleged is the connection to lung cancer. I will concentrate on that connection, in the following pages.

The Oxford Atlas of the World, ISBN 0-19-520955-9, published in 1992, gives figures for cigarette consumption in different countries during the time period 1986-1988. The figures are in annual consumption of cigarettes per capita. I have taken them from a graph and have attempted to interpolate between dividers; however, the interpolation errors should be negligible. Here are the figures:

To draw any conclusions concerning the influence of smoking upon lung cancer in these countries, we need figures on lung cancer death rates (LCDR's). Fortunately, the World Bank puts out a book which gives statistics for a number of countries which give disease statistics in a form known as "45Q15". The "45Q15" number represents the percentage risk of someone who is 15 years old dying from a particular disease by the time he or she is 60. Figures are not available for all countries; such important ones as the former USSR and India either don't report at all or don't break down deaths from cancer into different types of cancer. Never-the-less, we do have LCDR's for some of the countries for which we have smoking consumption figures. All of the following statistics are in 45Q15 format, which means they are risk figures in percentages.

In the United States, the male LCDR is 1.4%, the female risk is 0.7%. Hungary, with the highest rate of cigarette consumption of any country, has a male LCDR of 2.4; female 0.5%. Hungary shares the highest rates with its neighbor, Czechoslovakia, where the male rate is 2.4% and the female rate is 0.3%. Prima facie, these figures indicate that a high smoking rate is associated with a high LCDR. Or do they?

Let's look at Japan. As we have seen, Japan is practically tied with Hungary for the highest rate of cigarette consumption in the world. It turns out, however, that the male LCDR in Japan is 0.5%. That's approximately one-fifth the rate in Hungary; approximately one-third the U.S. rate. The LCDR for females in Japan is also astonishingly low, 0.2%.

Furthermore, although they have the highest smoking rate of any major nation, the Japanese are remarkably healthy! At birth, a Japanese male has a whopping life expectancy of 75 years (as opposed to 72 in the U.S.A.). Japanese girls, at birth, have a life expectancy of 80 years. Those are the highest life expectancies in the entire world.

Another heavy smoking nation is China. The authors of the World Bank book tell us so, and a recent PBS special concentrated on the "alarming" rate of smoking in China. In fact, in China, the government grows tobacco and receives much of its revenue from cigarette sales. In China, however, the LCDR is about the same as in Japan: 0.56% for men; 0.39% for women, in 1988, the last year for which we have World Bank information.

Interestingly, some nations in the tropical and sub-tropical belts have very low LCDR's, notwithstanding evidence suggesting that smoking is widespread in these countries. In Mauritius, an island in the Indian Ocean where tobacco is an important crop, the LCDR for males is only 0.4; for females it is 0.1. In Barbados, the male LCDR is 0.5; the female rate is zero. In the Seychelles, an island paradise in the Indian Ocean, the male LCDR is 0.4; the female LCDR is 1.0, making that nation the only one in the entire world, where the female rate exceeds the male rate.

At least one researcher has suggested that the low LCDR's in the tropical and sub-tropical countries are attributable to the exposure of the residents to sunshine, which raises vitamin D levels. That theory, however, fails to explain the very low LCDR's in China and Japan which are not tropical or sub-tropical countries.

One possible explanation may relate to the diagnosis of lung cancer. Sri Lanka (formerly) Ceylon) has the lowest male LCDR of any country in the world (0.1%), and a female rate of zero. So, if you're worried about lung cancer, you should catch the next plane to Sri Lanka. Before you do, however, you should be aware there is a disease category called "Senile and ill defined". The male death rate from "ill defined" illness in Sri Lanka is 3.4%; the female rate is 2.2%. These figures are many times greater than those for another country (for example, the male rate in the U.S. is 0.3%; in Hungary, it is zero). Clearly, the doctors in Sri Lanka are not doing a very good job of diagnosing causes of death. By comparison, in Hungary (which has the largest number of doctors per capita of any country in the world), every death is accounted for, positively. There are no deaths attributed to "ill defined" causes.

Diagnosis, alone, however, cannot be the whole answer. Japan has an excellent medical system, and cases of lung cancer are surely and accurately diagnosed. The death rate from "ill defined" illnesses in Japan is only 0.1% for males; zero per cent for females. In China, also, there is a rigorous effort to pin-point causes of death; the rates of death for males and females from "ill defined" causes are less than 0.1%. Yet, as we have seen, the LCDR's in China and Japan are very low, despite very high rates of smoking. Moreover, the LCDR figures cannot be dismissed as resulting from poor diagnosis, since the low rate of "ill defined" illness in each country proves that a vigorous effort is being made to accurately pin point exact causes of death.

Possibly, genetic factors are at work. Hungary and Czechoslovakia, each with high LCDR's, are contiguous countries, inhabited largely by fair skinned, blue eyed people. Japan and China, which have very low LCDR's, are separated only by the narrow Sea of Japan, and populated by people with relatively similar racial characteristics. Few figures are available on LCDR's in the developing nations in the tropical and sub-tropical zones, but the available figures suggest that lung cancer rates are small in these countries, which are largely inhabited by Blacks ² . Can it be that certain races of the world are genetically more susceptible to lung cancer than others?

I don't know. I can, however, say with certainty that smoking doesn't cause lung cancer in Japan and China. If it did, the LCDR's in these countries, which are populated by heavy smokers, could not possibly be so low!

Addendum (2002):

When I wrote this Chapter in 1996, the only information I was able to get on smoking and disease came from a book published by the World Bank. The World Bank used a very peculiar method of computing lung cancer death rates (LCDRs), based on a percentage likelihood that somebody would die by a certain age, from the disease. Also, the Bank didn�t have figures for many countries.

With the continued development of the Internet, figures have become available which allow for a comparison of LCDRs, smoking rates, and life expectancies in may countries. A Dutchman, Kees van der Griendt, has compiled date for 87 countries, using data from the World Health Organization and the CIA Fact Book. The complete study is at his web site:

http://www.kidon.com/smoke/index.html

It turns out that a high rate of smokers prevalence translates, in many cases, to long life expectancy and low rates of lung cancer. For males, in 1994, the country with the highest life expectancy (76.6 years) was Iceland, where 31% of the men smoked. The next runner-up was Japan, where 59% of the men smoked, and life expectancy was 76.5 years. Other countries with high rates of male smoking and long life expectancies included Israel (45%, 75.9 years); Greece (46%, 75.2 years); Cuba (49.3%, 74.7 years) and Spain (48%, 74.5 years).

Clearly, these figures rebut the hysterical claims of anti-smoking organizations. Figures bandied about in this country, and never challenged, estimate that smoking costs the smoker at least seven years of life expectancy. Figures circulated in Europe and cited on Mr. Van der Griendt�s web page, claim as much as 20 to 25 years of loss of life expectancy. But the official vital statistics from countries with high rates of smoking fail to validate these claims. To the contrary, it turns out that some of the countries with the highest rates of smoking have the longest life expectancies. This is important, not only from the standpoint of lung cancer, but also from the standpoint of heart disease. If, as is frequently claimed, smoking leads to heart attacks, the effects should be clearly show up in the form of greatly reduced life spans in countries where a lot of people smoke. They don�t.