Charles Rotter  crotter@cloudsystems.com

Reply from the Editor:

Charles, there is nothing wrong about being a layman and you ask an intelligent question.  I would trust your analysis of the article more than most "nonlaymen" who can at times be irrational.  In fact, I looked at the data again and I did misinterpret the WHO data and I therefore got the CDC data for the last 50 years and redid the numbers to find that the overall lifetime incidence of a smoker getting lung cancer is 8% and 1% for a nonsmoker. This would give us a relative risk of 8 (incidence of smokers developing lung cancer divided by the incidence of nonsmokers getting lung cancer). 

I took the 50 year data (see the updated Addendum A of the article) and extrapolated it for the 70 lifespan of a US male.  (To those nitpickers, yes the expected lifespan is now 73y/o but 70 is probably closer for the 50 year data and this is more than compensated by my not adding the compounding factor that lung cancer mortality is decreasing annually by at least 0.7% .)

After your email, I again looked through the literature and could not find any such analysis being done anywhere.  Nowhere was there any attempt done to determine the lifetime incidence (or "chance") of a smoker getting lung cancer. The definition of incidence or risk can be presented in several different ways.  Unfortunately the literature data is presented as a yearly incidence (as you thought) rather than a lifetime incidence so the determination of lifetime incidence becomes more difficult but I did the best possible based upon the 50 year US male CDC data as calculated in Addendum A of the article.

I know that this may be a long answer to your question, but you were right, I did make an error in the calculations but the answer is still lower than what we have been told and definitely not >50% which would justify the use of the word "cause."  I thank you for your observation and I will take 50 lashes of the proverbial wet noodle.

Respectfully,

I was accepting of your figure that the risk of lung cancer for a smoker is only 0.3%. Then I read further and found that lung cancer accounts for 3% of all deaths in the US. It sounds to me like lung cancer kills at least ten times as many people as actually get the disease!  If it killed only those who actually contracted the disease, but killed all of them, I would expect it to account for much less than 0.3% of the population, because much less than 100% of the population are smokers.  How do you account for the difference?

Curtis Cameron   curtc@airmail.net

Reply from the Editor:

Thank you for your good questions as it made me redo the statistics using CDC data and the lifetime incidence of dying from lung cancer was 8% for a US white male (USWM) smoker vs. 1% for a USWM nonsmoker.*   A more refined analysis would likely show that the lifetime incidence is even less since lung cancer survival has been on an increasing trend.  Also I have been told that the methodology of  the CDC/WHO data collection is questionable and probably biased which would make the smoking data appear worse.  Also when the literature data had a range to it, I took the data that would make smoking look worse (such as using the 50yr data and multiplying it by 7/5, rather than extrapolating it back 20 years earlier which would have made the statistics look better for the smokers).

The data when redone, answered your second question about the 3% annual death rate from lung cancer. The USWM overall had a 4% lifetime death rate (smokers + nonsmokers = 8%x40% + 1%x60% = 4% overall for USWM) and most when you add in the rest of the population, the rate will likely be even less.  But even if it did not come out this close, comparing lifetime death rates to an annual death rate (the data for only one year) is like comparing apples and oranges.  

There are many factors working here such as an increasing survival rate for lung cancer, lung cancer rates not correlating to smoking usage (both in the US and Europe the incidence of lung cancer is increasing while smoking has been decreasing), an increasing lung cancer rate due to causes other than smoking, pollution, other carcinogens, alterations in populations immune systems, and other reasons yet unknown.

In fact, lung cancer for all the WHO member states is only responsible for 2.3% of the "annual" deaths. The statistics are higher for the Americas because death due to infectious diseases is less here than in the world as a whole, making the annual percent of deaths due to lung cancer seem higher at 3%. The problem with these annual death rate statistics is that they don't tell us anything about lifetime risk, only of what people died from for a particular year. 

Because the media has ingrained in us all that cancer is this "big killer" and "tobacco is the root of all that evil," the numbers seem smaller than we had thought but they are not insignificant. Even so, I would never encourage someone to smoke cigarettes, not only because of the cancer risk but because of other health problems that smoking is probably a one of the risk factors. 

It all makes sense when explained, but researchers and the scientific literature, let alone the media is telling you things like lung cancer causes 17% of all cancer deaths, or that lung cancer is the #1 cancer killer, both which may be true but it is what they don't tell you that upsets me (like lung cancer only accounts for 2.3% and all cancers account for 13.4% of WHO annual deaths therefore lung cancer accounts for 17% of all annual cancer deaths).  Thank you for your catching the errors and letting me clarify the article further.

Respectfully,

Dr. Siepmann  mail@journaloftheoretics.com

* References for all of the above info and data can be found at Smoking Does Not Cause Cancer (CDC/WHO data) .

My friend and I recently read the Smoking Does NOT Cause Lung Cancer Paper, linked to by the Junk Science Home Page.  It was recently very hastily updated with new data, and I believe there are more than one error(s) in the appendix where the first table in the Appendix is the total number of MALES and FEMALES, not MALES as the Heading would suggest (US male data for all races) incidence of lung cancer for each age group per 100 000.   Well, is this males, females or a weighted average for each age group?

In the last table, appendix all of the numbers, every last one, are off by a factor of ten, right (too high)?? I hope that we can confirm that his conclusion in the appendix is based on correct "math" but using ... which numbers?

Also, we may be looking at (and pushing for!) much higher early detection rates of lung cancer in the future. If we do, the numbers in the appendix will presumably go way up, no?? My friend asked, what if lung cancer took one day to kill you--would we have tiny numbers in the appendix since nobody would have it for more than a day??

I liked reading the paper, but if there are serious labeling or math errors in the appendix, we should fix them as soon as possible before someone jumps all over us with sleazy ad hominem attacks.

A New Reader

Jesse V. Silverman

NYC

Reply from the Editor:

I am sorry for taking so long in getting back to you but you were very perceptive and correct about the populations being used being incorrect. The WHO data tables were poorly worded and the data were basically unworkable so I mainly used CDC 50 year data after your email and redid the numbers. The result was 0.5% for US white males. That analysis alone took so much time that to try and do any other population was not reasonable for this article though I hope someone will analyze the real lifetime incidence of lung cancer in smokers as well as the lifetime relative risks for all population groups. It would also be helpful if such an analysis could be done for the other diseases to which smoking has been associated and finding the lifetime incidence and relative risks. The problem with most of the statistics that we see in the media, is that they are based on only one years worth of data which is virtually meaningless.

In terms of the factor of 10 you referred to, that the incidences were per 100,000 and the population numbers were in millions. In terms of early detection, one would hope that it could make a difference but we lung cancer is unfortunately difficult to screen for. Even by doing annual chest X-rays on smokers has not shown much of an improvement. The chemotherapy though is somewhat better and the survival rate is improving but minimally so. The 5 year death rate is still about 90%.

Your point about taking just one section of time leading to faulty data and conclusions is a good one. Most lung cancer patients are those over 50 y/o which is where most of the baby boomers are and they had a smoking rate of over 40% which will make the annualized numbers higher than reality. Also if you were to take the average smoking rate of the population, you would get 40% over the last 50 years rather than using the current 23% as x/23% is 4.4 while x/40%=2.5 making the numbers lung cancer rates appear almost twice what they actually are.

If I wanted to present a biased viewpoint, I could have easily done so by taking the total number of lung cancer deaths for 1990 and then divided it by the number of current smokers: 828/(23%*102mil)=0.0035% incidence of lung cancer deaths in USWM smokers. Even though this would true, it would be misleading. That is the point. Even though the numbers you hear about in the literature and press may be true, they are often misleading or don't have any practical use or significance.

Another example is the risk of Down's syndrome with increasing maternal age. You will hear that your risk of having a child with Down's syndrome doubles at age 35, goes to 5x at age 40, and goes to 10x at age 45. Boy those are scary statistics. Now let's look at the whole picture. The incidence of Down's syndrome in a woman <30y/o is 1/1000, at 35 is 2/1000, at 40 is 5/1000, at 45 is 10/1000. So even though a 40y/o woman has 5x the risk of having a child with Down's syndrome, it is still only a 0.5% chance. Even at 45y/o, there is only a 1% chance of having a child with Down's syndrome.

My goal is to give the scientific profession, the media, and the public, complete and accurate information. And unlike the hate mail, I get from those who publish trash science, I don't mind constructive criticism as it will only make my work better, improve my intellect, and improve the information base of true Science. It is the latter that is the most important, because SCIENCE should be about accurate and complete information which can then be put in a way that the general public can understand.

Respectfully,

James P. Siepmann, MD  mail@journaloftheoretics.com

P.S. Unfortunately no matter how meticulous we are in obtaining the data and crunching there will be those who refuse to accept the results and will proceed to ad hominem attacks no matter what.  One of my quotes is, "Insults are the argument of those who have no argument."

The article on smoking is visionary and truth finding instead of contributing to popular opinion as science has begun to be today. Congratulations Dr. Siepmann and I say that not just because I am a slightly excessive smoker.

Yours Sincerely

Well written and researched, it was a breath of fresh air.

Sincerely,

Anthony D. Wilson

Thank you very much for such a straight forward and honest article about smoking. I wish more people would think for themselves instead of just taking in what ever is being pushed under their noses that day.

Since a non-smoker is not always an anti-smoker, maybe some of them will wake up in the midst of the great Fat Tax.  We are giving a lot of space to these articles now, we hope it is not to late for the average American to think for themselves.

Thanks again.

Yours,

Samantha Phillipe

The United Pro Choice Smokers Rights Newsletter

As the author of a little book disputing the connections between smoking and lung cancer, I enjoyed your editorial. You make a mistake, however, in relying upon any of the WHO statistics, e.g. the claim that 9 out of ten lung cancer victims are smokers. These stats, like the SAMMEC stats put out by the CDC and other U.S. government agencies, are essentially inventions, and are not based upon any epidemiological studies.

The last major epidemiological studies on smoking and lung cancer were six cancer society studies and Doll's study of British doctors, all done just prior to the publication of the 1964 Surgeon General's Report. In all of these studies, the smokers that were studied were much older than the general population (for example, some of the cancer society studies were deliberately restricted to smokers over the age of 50). Although the authors of these studies claimed to have adjusted the results for the age of the participants, they never explained how they did this, and I can find no evidence that they did any adjustment at all. To the contrary, when Prof. Burch re-analyzed Doll's results to compare the doctors to themselves (instead of to the much younger general population), he found no correlation between smoking habits and lung cancer.

Meanwhile, hundreds of experiments have been done on lab animals, ranging from monkeys to rats and mice, trying to induce lung cancers by painting the animals' lungs with tobacco tars or forcing the animals to inhale huge quantities of smoke. Indeed, the experiments continue to this day at UC (Davis) with whole generations of mice, who are raised in closed cabinets where smoke is pumped into the cabinets constantly, in huge quantities. However, nobody has ever been able to induce a single lung cancer in this manner. 

I believe that I've discussed every single study in my book, http://www.lcolby.com.   At least, I've tried to discuss every reported study on animals or humans. "Studies" such as those reported by WHO or CDC shouldn't qualify as "studies", because they don't reflect an effort to correlate people's smoking habits with lung cancer. Rather, they reflect assumptions based upon preconceptions - for example, the researchers assume that 90% of all lung cancer deaths are caused by smoking so, if there are 100,000 such deaths in a year, they simply calculate that 90,000 were caused by smoking. This is how the figures on tobacco related deaths in China, posted on the WHO page, were developed.

No human beings were actually studied or interviewed.

Lauren A. Colby   73270.3242@compuserve.com

You are quite right about the statistics, but I had to use something to prove my point (that the research and literature were using language and statistics inappropriately and therefore biased) and I wanted to use the most liberal (politically correct source so no one can say I fudged the numbers). Also I had to redo the numbers do to a mistake on my part in trying to get a lifetime incidence of smoking and used mostly CDC data for that (which you correctly noted are biased).

As you note there are many ways to pervert the data. The way that they collect the data makes it virtually impossible to determine a the lifetime incidence of a smoker getting lung cancer. For instance, by taking lung cancer data annually, rather than analyzing it over a 50 year period as I did, you will markedly distort the data because only 23% of the population smokes now, but we are seeing the lung cancers that smoking may have been a factor in from decades ago and from a larger smoking population. The 50 year average of the smoking population is more around 40% as I note in the revised article.

I agree whole heartily about the bias of Doll's and other's studies which is blatantly obvious and I referenced these as being obviously biased from their abstract on down. The bias includes the literature databases, because many of the articles that do not uphold the politically correct or status quo are often not published or never make it into the databases, let alone the media.

I reviewed your site and applaud you for trying to bring truth back into Science. I have added your site to our links section so others can see how Science has gotten corrupted. It is a tough battle though. You wouldn't believe (actually, you probably would) all of the unprintable email filled with diatribes and emotion but no facts. We all need to keep pushing forward in our attempts to bring truth in language, unbiased data, and logic back into Science.

Sincerely,

I received an unprintable comment from someone about getting the email on the new Journal articles who was not on the mailing list as he claimed.  Nothing scares me more than realizing that there are closed minded and ignorant people out there teaching our future scientists.  I have left out his name but he is a professor of astronomy (with the usual academic credentials and a list of published papers) at a well known institution which makes his ignorance and attitude all the more a threat to true science and society as a whole.  I kindly responded with this printable prose:

You are not on our mailing list, it must have been forwarded to you by someone who thought you were in need of objectivity. Our mailing list of subscribers consists of over 3000 scientists interested in intelligent articles, research, and debate.... If you wish to have an intelligent discussion, first read the article...if not, I am told that ignorance is bliss.

I then received the short response: "From what I can tell you are disreputable."

It is people like you who refuse to look at the facts in an unbiased fashion that have brought science to its sorry state (I hope that you can at least recognize that there are emotional, status quo, credential, and many more biases in science). Hiding your head in the sand to is bad enough but to perpetuate the perversion of such biases as indicated in your statement is a crime to true science.

I hope that I can influence others to approach science without such preconceived biases and to look beyond what they are taught. I can hold my head up high knowing that I am pursuing the noble goal of trying to rid science of these biases and bringing back logic and truth. I present science accurately and truthfully, and do not need to lower myself to name calling which is the last resort of those who have no facts to support their position. I will not lower myself to that level.

I can take the name calling, but I can not take the idea that you actually "teach." The [institution] deserves unbiased teachers which look at all facts, not just the one's they agree with. An [honorable man] would not retreat to security of their biases but recognizing that they can no longer view science objectively would retire in a noble parting gesture. If the day should come that I can no longer discuss an issue based upon the facts and logic, I hope that someone would point that out to me so I can do the honorable thing.

Striving for Purity in Science,