"Now to tip my hat to myself, ..."
Brilliant reasoning, beats Banzhaf. If I was ASH, I would hire you on the spot to make you their chief strategic advocate 
.
"My policy doesn't make smoking a condition of employment."
I'm sure ASH doesn't like this idea at all, it would put a serious stop at smoker-bashing.
While exsmokers face elevated (but gradually reduced) risks of lung cancer for 25 years after quitting smoking, excess risks of heart disease decline rapidly upon quitting smoking and approach that of nonsmokers within one year.
The relative risk numbers don't add up.
Here's what I get from the lung cancer alliance site;
Total diagnosed cases per year: ~160,000
Of those 10-15% are never smokers,
50% are former smokers, and 35-40% are current smokers.
These are all approximate numbers, but they are enough to get a rough estimate of relative risk for smokers, non-smokers, and ex-smokers:
There are approximately 300,000,000 people in America, and the average life expectancy is roughly 80. That means I should expect on the order of 4,000,000 deaths per year from all causes.
Now lets break those deaths down into smokers, ex-smokers, and current smokers. Smokers constitute 20% of the adult population, but ex-smokers who are dying today (mostly people in their 70's and 80's) probably account for 30% (since smoking rates were significantly higher in the 1950's). Never smokers probably constitute ~50% of the population in their 70's and 80's. This group (i.e. never smokers) should then be dying at a rate of roughly 2,000,000 per year. Match that up with the lung cancer alliance's 160,000 deaths per year and 10-15% of those being never-smokers. That gives 16,000 - 24,000 never-smoker lung cancer deaths per yer. Or roughly 1% of never-smoker deaths.
Now take the ex-smokers. If they are 30% of the elderly population, they should be dying at a rate of 1,200,000 per year. But 50% of 160,000 is 80,000, or 7.5%. That's a relative risk of 7.5 compared to never smokers.
As for the smokers, if they constitute 20% of the adult population, they should be dying at a rate of 800,000 per year. But 35-40% of 160,000 gives ~60,000 deaths per year, which is also ~7.5%. That's also a relative risk of ~7.5 compared to never smokers.
I can believe relative risk numbers for ex-smokers and current smokers in this ball-park, i.e. 5 to 10. I can believe that I somewhat underestimated the amount of never smokers and that the relative risk numbers should be 8-9 instead of 7.5.
But I simply don't believe relative risk numbers of 26.9 and 50.7. To get those relative risks you would need never-smokers to constitute over 100% of the population. The numbers just don't add up.
The other thing that makes me dubious of the CPS-II numbers rather than the Lung Cancer Alliance numbers is the (in my opinion) false precision implied in their relative risk rates.
The Lung Cancer Alliance numbers are quoted as "10-15% never smokers" and "30-35% current smokers". I think these figures show a reasonable understanding of the uncertainties inherit in measuring such things. If you extrapolate these uncertainties to the relative risks, you get numbers on the order of:
Risk of lung cancer death for never smokers: 1.0% +/- 0.3%
Risk of lung cancer death for current smokers: 8% +/- 2%
etc.
You don't get numbers as accurate as "Relative risk = 50.7" or "Relative risk = 26.9", which imply being able to estimate such things to better than 1 part in 100.
Dave,
It depends on the way these Odd Ratios or Relative Risks are calculated. What one would expect is that the RR is calculated as:
% of lung cancer in smokers
------------------------------------- *100%
% of lung cancer in non-smokers
but that is NOT the way these risks are calculated. I described the actual way these numbers are calculated briefly in a previous post. The essential difference between the two methods is that in the second case there's an implicit assumption entered into the equation that the cancer cases MUST have it's source in the behavior (smoking) or exposure (ETS).
Statistics programs do the rest and use the scientist's belief that the cause must be somewhere in the 'abnormal behavior'.
O.K. Thanks. I don't think I totally understood the post you refered to, other than that 'relative risk' is a complicated statistical construct which doesn't mean what I THINK it should mean. (or what MOST people would think it should mean.)
But what I still don't understand is even if there is some bizarre and misleading definition of relative risk which gives these ridiculously high numbers, how can they possibly claim such precision? Do those numbers represent an upper limit to a fit with a 95% confidence level or something like that? Surely the statistics programs that produce these numbers give some kind of estimate of the goodness of the fit, or the error in the final number or SOMETHING. Where on earth does that implied level of precision come from? Or am I completely misunderstanding what's going on?
Aw never mind the quibbling over stats. Siegel's idea is brilliant. It means that companies like Weyco who demand their employees quit as a condition of employment, should immediately fire them the moment they DO.
According to ACS Facts and Figures, there were an estimated 172,570 new cases of lung and bronchus cancer in 2005, and 163,510 deaths of lung and bronchus cancer.
But there's a huge difference between the total number of lung cancer death, and age adjusted lung cancer death rates, the latter of which is a more accurate, reliable and appropriate indicator of lung cancer mortality trends.
Bill, I bet if you took the age adjusted lung cancer death rates and the age adjusted rates of what percent of the population are smokers, ex-smokers and never smokers, you'd see a similar thing. Ex-smokers and smokers would have a risk of lung cancer 6-8 times higher than non-smokers, not 30-50 times higher.
Discussing lung cancer stats is really playing on Bill's field.
For centuries smoking has been branded as harmful. Since 1964 the world has been warned, advised, educated, and beaten over the head in school and at home (through our TVs) about the health risks of smoking. Quit smoking programs are offered -- for free!
No one in this world doesn't know that primary smoking might cause lung cancer (this is a finding I accept) or whatever else anti-tobacco wants to pin primary smoking on the smoker on.
Here's the rub... I DON'T CARE.
I heard you. I CHOOSE to smoke anyway. And if the Bill's of the world think it's not a choice then acts forcing me to "behave" is wrongheaded. Cut it out and advocate making tobacco illegal. If you won't do that then leave me alone to my CHOICE.
How I choose to live my QUALITY of life or how I risk dying is none of anyone's business to stop other than to warn me.
And if I'm going to be another lung cancer stat one day.... so what?!? It was borne out of how I chose to live my life, not by someone else's demands.
I drink lots of coffee (17 carcinogens). Cancers have been blamed on coffee. Caffeine is "addictive." I like to burn the skin on my chicken when I bbq it. Carcinogens (PAHs) galore. And if I want to eat nothing but french fries for breakfast, lunch and dinner, then that's my choice too.
If all of this puts me at risk for my death being from cancer then that's MY problem, not the "public health" tyrants who have decided that my PRIVATE life -- and death -- is their business.
If nothing Bill does makes me stop will Bill advocate locking me up "for my own good?"
Cutting smoking-related lung cancer deaths is a matter for advice, warnings and services for those who seek it -- not forcing people to stop living the way they want so that the number of cases can be cut.
The only reason to count cases is to know where warnings might be called for. Nothing else.
My body is not property of the state or property of Bill. Who made you G*d, Bill?
"Thus the cardinal moral truths are these: 'that all Men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty, and the Pursuit of Happiness...' We are all created equal, as defined by our natural rights; thus, no one has rights superior to those of anyone else. Moreover, we are born with those rights, we do not get them from government[or Bill]... And our rights to life, liberty, and the pursuit of happiness imply the right to live our lives as we wish [not as Bill says we should] -- to pursue happiness as we think best [not how Bill defines it], by our own lights -- provided only that we respect the equal rights of others to do the same. [something Bill will have no part of]"
As for the socialist argument of "cost to society" Bill will throw up as a shield... we'll all pay our own costs or die without getting care, okay? But everyone gets sick and dies. No one gets off. Might as well say we all pay for each other.
dr. mike,
fyi, i think there is a typo in your article. second paragraph down, last sentence, after 'the rest of the story'. your piece is very provocative.
for the record, i totally agree with JustTheFacts. nobody's business but my own.
JustTheFacts: Thanks for spelling it out so clearly.
From the Universal Declaration of Human Rights:
"No one shall be subjected to arbitrary interference with his privacy, family, home or correspondence, nor to attacks upon his honour and reputation."
"Everyone has the right to work, to free choice of employment, to just and favourable conditions of work and to protection against unemployment.
Everyone, without any discrimination, has the right to equal pay for equal work."
Brandz-
Thanks for catching the typo, which I've fixed. It's nice to have someone reading my whole post instead of just seeing that I am disagreeing with the opinion of an anti-smoking group and therefore accusing me of being funded by Big Tobacco.
Mm those statistics are really out of place. If you do some research you'll find if a person quits at a younger age theirrisk of lung cancer falls short to someone who has never smoked. You can't compare someone who has smoked for 2-5 years then quits to someone who has smoked 15+ who quits.
2 Visitors Online
Commenting by HaloScan
